def get_FPS(self, image, test_interval):
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
#---------------------------------------------------------#
crop_img = letterbox_image(image,
[self.input_shape[0], self.input_shape[1]])
#----------------------------------------------------------------------------------#
# 将RGB转化成BGR，这是因为原始的centernet_hourglass权值是使用BGR通道的图片训练的
#----------------------------------------------------------------------------------#
photo = np.array(crop_img, dtype=np.float32)[:, :, ::-1]
#-----------------------------------------------------------#
# 图片预处理，归一化。获得的photo的shape为[1, 512, 512, 3]
#-----------------------------------------------------------#
photo = np.reshape(
preprocess_image(photo),
[1, self.input_shape[0], self.input_shape[1], self.input_shape[2]])
preds = self.get_pred(photo).numpy()
if self.nms:
preds = np.array(nms(preds, self.nms_threhold))
if len(preds[0]) > 0:
preds[0][:, 0:4] = preds[0][:, 0:4] / (self.input_shape[0] / 4)
det_label = preds[0][:, -1]
det_conf = preds[0][:, -2]
det_xmin, det_ymin, det_xmax, det_ymax = preds[0][:, 0], preds[
0][:, 1], preds[0][:, 2], preds[0][:, 3]
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
boxes = centernet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.input_shape[0], self.input_shape[1]]),
image_shape)
t1 = time.time()
for _ in range(test_interval):
preds = self.get_pred(photo).numpy()
if self.nms:
preds = np.array(nms(preds, self.nms_threhold))
if len(preds[0]) > 0:
preds[0][:, 0:4] = preds[0][:, 0:4] / (self.input_shape[0] / 4)
det_label = preds[0][:, -1]
det_conf = preds[0][:, -2]
det_xmin, det_ymin, det_xmax, det_ymax = preds[0][:, 0], preds[
0][:, 1], preds[0][:, 2], preds[0][:, 3]
top_indices = [
i for i, conf in enumerate(det_conf)
if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices], -1), np.expand_dims(
det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
boxes = centernet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.input_shape[0], self.input_shape[1]]),
image_shape)
t2 = time.time()
tact_time = (t2 - t1) / test_interval
return tact_time
def detect_image(self, image):
#---------------------------------------------------#
# 对输入图像进行一个备份，后面用于绘图
#---------------------------------------------------#
old_image = image.copy()
image = np.array(image, np.float32)
im_height, im_width, _ = np.shape(image)
#---------------------------------------------------#
# 计算scale，用于将获得的预测框转换成原图的高宽
#---------------------------------------------------#
scale = [
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0]
]
scale_for_landmarks = [
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0]
]
#---------------------------------------------------------#
# letterbox_image可以给图像增加灰条，实现不失真的resize
#---------------------------------------------------------#
if self.letterbox_image:
image = letterbox_image(image,
[self.input_shape[1], self.input_shape[0]])
else:
self.anchors = Anchors(self.cfg,
image_size=(im_height,
im_width)).get_anchors()
#-----------------------------------------------------------#
# 图片预处理，归一化。
#-----------------------------------------------------------#
photo = np.expand_dims(preprocess_input(image), 0)
preds = self.retinaface.predict(photo)
#-----------------------------------------------------------#
# 将预测结果进行解码
#-----------------------------------------------------------#
results = self.bbox_util.detection_out(
preds, self.anchors, confidence_threshold=self.confidence)
#--------------------------------------#
# 如果没有检测到物体，则返回原图
#--------------------------------------#
if len(results) <= 0:
return old_image
results = np.array(results)
#---------------------------------------------------------#
# 如果使用了letterbox_image的话，要把灰条的部分去除掉。
#---------------------------------------------------------#
if self.letterbox_image:
results = retinaface_correct_boxes(
results, np.array([self.input_shape[0], self.input_shape[1]]),
np.array([im_height, im_width]))
results[:, :4] = results[:, :4] * scale
results[:, 5:] = results[:, 5:] * scale_for_landmarks
for b in results:
text = "{:.4f}".format(b[4])
b = list(map(int, b))
# b[0]-b[3]为人脸框的坐标，b[4]为得分
cv2.rectangle(old_image, (b[0], b[1]), (b[2], b[3]), (0, 0, 255),
2)
cx = b[0]
cy = b[1] + 12
cv2.putText(old_image, text, (cx, cy), cv2.FONT_HERSHEY_DUPLEX,
0.5, (255, 255, 255))
print(b[0], b[1], b[2], b[3], b[4])
# b[5]-b[14]为人脸关键点的坐标
cv2.circle(old_image, (b[5], b[6]), 1, (0, 0, 255), 4)
cv2.circle(old_image, (b[7], b[8]), 1, (0, 255, 255), 4)
cv2.circle(old_image, (b[9], b[10]), 1, (255, 0, 255), 4)
cv2.circle(old_image, (b[11], b[12]), 1, (0, 255, 0), 4)
cv2.circle(old_image, (b[13], b[14]), 1, (255, 0, 0), 4)
return old_image
def detect_image(self, image):
image_shape = np.array(np.shape(image)[0:2])
crop_img = np.array(
letterbox_image(
image, (self.model_image_size[0], self.model_image_size[1])))
photo = np.array(crop_img, dtype=np.float32)
photo /= 255.0
photo = np.transpose(photo, (2, 0, 1))
photo = photo.astype(np.float32)
images = []
images.append(photo)
images = np.asarray(images)
with torch.no_grad():
images = torch.from_numpy(images)
if self.cuda:
images = images.cuda()
outputs = self.net(images)
output_list = []
for i in range(3):
output_list.append(self.yolo_decodes[i](outputs[i]))
output = torch.cat(output_list, 1)
batch_detections = non_max_suppression(output,
len(self.class_names),
conf_thres=self.confidence,
nms_thres=0.3)
try:
batch_detections = batch_detections[0].cpu().numpy()
except:
return image
top_index = batch_detections[:,
4] * batch_detections[:,
5] > self.confidence
top_conf = batch_detections[top_index, 4] * batch_detections[top_index,
5]
top_label = np.array(batch_detections[top_index, -1], np.int32)
top_bboxes = np.array(batch_detections[top_index, :4])
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
top_bboxes[:, 0],
-1), np.expand_dims(top_bboxes[:, 1], -1), np.expand_dims(
top_bboxes[:, 2], -1), np.expand_dims(top_bboxes[:, 3], -1)
# 去掉灰条
boxes = yolo_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.model_image_size[0], self.model_image_size[1]]),
image_shape)
font = ImageFont.truetype(font='model_data/simhei.ttf',
size=np.floor(3e-2 * np.shape(image)[1] +
0.5).astype('int32'))
thickness = (np.shape(image)[0] +
np.shape(image)[1]) // self.model_image_size[0]
for i, c in enumerate(top_label):
predicted_class = self.class_names[c]
score = top_conf[i]
top, left, bottom, right = boxes[i]
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(
np.shape(image)[0],
np.floor(bottom + 0.5).astype('int32'))
right = min(
np.shape(image)[1],
np.floor(right + 0.5).astype('int32'))
oo = [bottom - top, right - left] # 所画出框框的长和宽
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
class_score = {predicted_class: '%.2f' % score}
# print(label)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[self.class_names.index(
predicted_class)])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[self.class_names.index(predicted_class)])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
del draw
try:
return image, oo, class_score
except:
return image
def get_FPS(self, image, test_interval):
image_shape = np.array(np.shape(image)[0:2])
# ---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
# ---------------------------------------------------------#
if self.letterbox_image:
crop_img = np.array(letterbox_image(image, (self.model_image_size[1], self.model_image_size[0])))
else:
crop_img = image.convert('RGB')
crop_img = crop_img.resize((self.model_image_size[1], self.model_image_size[0]), Image.BICUBIC)
photo = np.array(crop_img, dtype=np.float32) / 255.0
photo = np.transpose(photo, (2, 0, 1))
# ---------------------------------------------------------#
# 添加上batch_size维度
# ---------------------------------------------------------#
images = [photo]
with torch.no_grad():
images = torch.from_numpy(np.asarray(images))
if self.cuda:
images = images.cuda()
outputs = self.net(images)
output_list = []
for i in range(3):
output_list.append(self.yolo_decodes[i](outputs[i]))
output = torch.cat(output_list, 1)
batch_detections = non_max_suppression(output, len(self.class_names),
conf_thres=self.confidence,
nms_thres=self.iou)
try:
batch_detections = batch_detections[0].cpu().numpy()
top_index = batch_detections[:, 4] * batch_detections[:, 5] > self.confidence
top_conf = batch_detections[top_index, 4] * batch_detections[top_index, 5]
top_label = np.array(batch_detections[top_index, -1], np.int32)
top_bboxes = np.array(batch_detections[top_index, :4])
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(top_bboxes[:, 0], -1), np.expand_dims(
top_bboxes[:, 1], -1), np.expand_dims(top_bboxes[:, 2], -1), np.expand_dims(top_bboxes[:, 3], -1)
if self.letterbox_image:
boxes = yolo_correct_boxes(top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.model_image_size[0], self.model_image_size[1]]),
image_shape)
else:
top_xmin = top_xmin / self.model_image_size[1] * image_shape[1]
top_ymin = top_ymin / self.model_image_size[0] * image_shape[0]
top_xmax = top_xmax / self.model_image_size[1] * image_shape[1]
top_ymax = top_ymax / self.model_image_size[0] * image_shape[0]
boxes = np.concatenate([top_ymin, top_xmin, top_ymax, top_xmax], axis=-1)
except:
pass
t1 = time.time()
for _ in range(test_interval):
with torch.no_grad():
outputs = self.net(images)
output_list = []
for i in range(3):
output_list.append(self.yolo_decodes[i](outputs[i]))
output = torch.cat(output_list, 1)
batch_detections = non_max_suppression(output, len(self.class_names),
conf_thres=self.confidence,
nms_thres=self.iou)
try:
batch_detections = batch_detections[0].cpu().numpy()
top_index = batch_detections[:, 4] * batch_detections[:, 5] > self.confidence
top_conf = batch_detections[top_index, 4] * batch_detections[top_index, 5]
top_label = np.array(batch_detections[top_index, -1], np.int32)
top_bboxes = np.array(batch_detections[top_index, :4])
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(top_bboxes[:, 0], -1), np.expand_dims(
top_bboxes[:, 1], -1), np.expand_dims(top_bboxes[:, 2], -1), np.expand_dims(top_bboxes[:, 3],
-1)
if self.letterbox_image:
boxes = yolo_correct_boxes(top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.model_image_size[0], self.model_image_size[1]]),
image_shape)
else:
top_xmin = top_xmin / self.model_image_size[1] * image_shape[1]
top_ymin = top_ymin / self.model_image_size[0] * image_shape[0]
top_xmax = top_xmax / self.model_image_size[1] * image_shape[1]
top_ymax = top_ymax / self.model_image_size[0] * image_shape[0]
boxes = np.concatenate([top_ymin, top_xmin, top_ymax, top_xmax], axis=-1)
except:
pass
t2 = time.time()
tact_time = (t2 - t1) / test_interval
return tact_time
def detect_image(self, image):
start = timer()
# convert img_size to input_size
new_image_size = (self.model_image_size[0], self.model_image_size[1])
boxed_image = letterbox_image(image, new_image_size)
image_data = np.array(boxed_image, dtype='float32')
image_data /= 255.
image_data = np.expand_dims(image_data, 0)
# sess.run
out_boxes, out_scores, out_classes = self.sess.run(
[self.boxes, self.scores, self.classes],
feed_dict={
self.yolo_model.input: image_data,
self.input_image_shape: [image.size[1], image.size[0]],
K.learning_phase(): 0
})
# print(out_scores)
# print(out_boxes)
print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
# starting draw bounding boxes
font = ImageFont.truetype(font='font/simhei.ttf',
size=np.floor(2e-2 * image.size[1] + 0.5).astype('int32'))
# thickness of bounding box and this thickness is changing according to img_size
thickness = (image.size[0] + image.size[1]) // 500
for i, c in list(enumerate(out_classes)):
predicted_class = self.class_names[c]
box = out_boxes[i]
score = out_scores[i]
top, left, bottom, right = box
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(image.size[1], np.floor(bottom + 0.5).astype('int32'))
right = min(image.size[0], np.floor(right + 0.5).astype('int32'))
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[c])
draw.rectangle([tuple(text_origin), tuple(text_origin + label_size)],
fill=self.colors[c])
draw.text(text_origin, str(label, 'UTF-8'), fill=(0, 0, 0), font=font)
del draw
end = timer()
print('detect time:', end - start)
return image
def detect_image(self, image_id, image):
self.confidence = 0.01
self.iou = 0.5
f = open("./input/detection-results/" + image_id + ".txt", "w")
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
#---------------------------------------------------------#
crop_img = np.array(
letterbox_image(
image, (self.model_image_size[1], self.model_image_size[0])))
photo = np.array(crop_img, dtype=np.float32) / 255.0
photo = np.transpose(photo, (2, 0, 1))
#---------------------------------------------------------#
# 添加上batch_size维度
#---------------------------------------------------------#
images = [photo]
with torch.no_grad():
images = torch.from_numpy(np.asarray(images))
if self.cuda:
images = images.cuda()
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
outputs = self.net(images)
output_list = []
for i in range(3):
output_list.append(self.yolo_decodes[i](outputs[i]))
#---------------------------------------------------------#
# 将预测框进行堆叠，然后进行非极大抑制
#---------------------------------------------------------#
output = torch.cat(output_list, 1)
batch_detections = non_max_suppression(output,
len(self.class_names),
conf_thres=self.confidence,
nms_thres=self.iou)
#---------------------------------------------------------#
# 如果没有检测出物体，返回原图
#---------------------------------------------------------#
try:
batch_detections = batch_detections[0].cpu().numpy()
except:
return image
#---------------------------------------------------------#
# 对预测框进行得分筛选
#---------------------------------------------------------#
top_index = batch_detections[:,
4] * batch_detections[:,
5] > self.confidence
top_conf = batch_detections[top_index,
4] * batch_detections[top_index, 5]
top_label = np.array(batch_detections[top_index, -1], np.int32)
top_bboxes = np.array(batch_detections[top_index, :4])
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
top_bboxes[:, 0],
-1), np.expand_dims(top_bboxes[:, 1], -1), np.expand_dims(
top_bboxes[:, 2],
-1), np.expand_dims(top_bboxes[:, 3], -1)
#-----------------------------------------------------------------#
# 在图像传入网络预测前会进行letterbox_image给图像周围添加灰条
# 因此生成的top_bboxes是相对于有灰条的图像的
# 我们需要对其进行修改，去除灰条的部分。
#-----------------------------------------------------------------#
boxes = yolo_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.model_image_size[0], self.model_image_size[1]]),
image_shape)
for i, c in enumerate(top_label):
predicted_class = self.class_names[c]
score = str(top_conf[i])
top, left, bottom, right = boxes[i]
f.write("%s %s %s %s %s %s\n" %
(predicted_class, score[:6], str(int(left)), str(
int(top)), str(int(right)), str(int(bottom))))
f.close()
return
def detect_image(self, image):
image_shape = np.array(np.shape(image)[0:2])
crop_img = np.array(letterbox_image(image, self.image_size))
photo = np.array(crop_img, dtype=np.float32)
photo = np.transpose(preprocess_input(photo), (2, 0, 1))
images = []
images.append(photo)
images = np.asarray(images)
with torch.no_grad():
images = torch.from_numpy(images)
if self.cuda:
images = images.cuda()
_, regression, classification, anchors = self.net(images)
regression = decodebox(regression, anchors, images)
detection = torch.cat([regression, classification], axis=-1)
batch_detections = non_max_suppression(detection,
len(self.class_names),
conf_thres=self.confidence,
nms_thres=self.iou)
try:
batch_detections = batch_detections[0].cpu().numpy()
except:
return image
top_index = batch_detections[:, 4] > self.confidence
top_conf = batch_detections[top_index, 4]
top_label = np.array(batch_detections[top_index, -1], np.int32)
top_bboxes = np.array(batch_detections[top_index, :4])
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
top_bboxes[:, 0],
-1), np.expand_dims(top_bboxes[:, 1], -1), np.expand_dims(
top_bboxes[:, 2], -1), np.expand_dims(top_bboxes[:, 3], -1)
# 去掉灰条
boxes = retinanet_correct_boxes(top_ymin, top_xmin, top_ymax, top_xmax,
np.array(self.image_size), image_shape)
font = ImageFont.truetype(font='model_data/simhei.ttf',
size=np.floor(3e-2 * np.shape(image)[1] +
0.5).astype('int32'))
thickness = (np.shape(image)[0] +
np.shape(image)[1]) // self.image_size[0]
for i, c in enumerate(top_label):
predicted_class = self.class_names[c]
score = top_conf[i]
top, left, bottom, right = boxes[i]
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(
np.shape(image)[0],
np.floor(bottom + 0.5).astype('int32'))
right = min(
np.shape(image)[1],
np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[self.class_names.index(
predicted_class)])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[self.class_names.index(predicted_class)])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
del draw
return image
def detect_image(self, image_id, image):
f = open("./input/detection-results/" + image_id + ".txt", "w")
self.confidence = 0.01
self.nms_threhold = 0.5
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
#---------------------------------------------------------#
crop_img = letterbox_image(image,
[self.input_shape[0], self.input_shape[1]])
#----------------------------------------------------------------------------------#
# 将RGB转化成BGR，这是因为原始的centernet_hourglass权值是使用BGR通道的图片训练的
#----------------------------------------------------------------------------------#
photo = np.array(crop_img, dtype=np.float32)[:, :, ::-1]
#-----------------------------------------------------------#
# 图片预处理，归一化。获得的photo的shape为[1, 512, 512, 3]
#-----------------------------------------------------------#
photo = np.reshape(
preprocess_image(photo),
[1, self.input_shape[0], self.input_shape[1], self.input_shape[2]])
preds = self.get_pred(photo).numpy()
#-------------------------------------------------------#
# 对于centernet网络来讲，确立中心非常重要。
# 对于大目标而言，会存在许多的局部信息。
# 此时对于同一个大目标，中心点比较难以确定。
# 使用最大池化的非极大抑制方法无法去除局部框
# 所以我还是写了另外一段对框进行非极大抑制的代码
# 实际测试中，hourglass为主干网络时有无额外的nms相差不大，resnet相差较大。
#-------------------------------------------------------#
if self.nms:
preds = np.array(nms(preds, self.nms_threhold))
if len(preds[0]) <= 0:
return
#-----------------------------------------------------------#
# 将预测结果转换成小数的形式
#-----------------------------------------------------------#
preds[0][:, 0:4] = preds[0][:, 0:4] / (self.input_shape[0] / 4)
det_label = preds[0][:, -1]
det_conf = preds[0][:, -2]
det_xmin, det_ymin, det_xmax, det_ymax = preds[0][:, 0], preds[
0][:, 1], preds[0][:, 2], preds[0][:, 3]
#-----------------------------------------------------------#
# 筛选出其中得分高于confidence的框
#-----------------------------------------------------------#
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
#-----------------------------------------------------------#
# 去掉灰条部分
#-----------------------------------------------------------#
boxes = centernet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.input_shape[0], self.input_shape[1]]), image_shape)
for i, c in enumerate(top_label_indices):
predicted_class = self.class_names[int(c)]
score = str(top_conf[i])
top, left, bottom, right = boxes[i]
f.write("%s %s %s %s %s %s\n" %
(predicted_class, score[:6], str(int(left)), str(
int(top)), str(int(right)), str(int(bottom))))
f.close()
return
def detect_image(self, image):
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
#---------------------------------------------------------#
crop_img = letterbox_image(image,
[self.image_size[0], self.image_size[1]])
#----------------------------------------------------------------------------------#
# 将RGB转化成BGR，这是因为原始的centernet_hourglass权值是使用BGR通道的图片训练的
#----------------------------------------------------------------------------------#
photo = np.array(crop_img, dtype=np.float32)[:, :, ::-1]
#-----------------------------------------------------------#
# 图片预处理，归一化。获得的photo的shape为[1, 512, 512, 3]
#-----------------------------------------------------------#
photo = np.reshape(
np.transpose(preprocess_image(photo), (2, 0, 1)),
[1, self.image_size[2], self.image_size[0], self.image_size[1]])
with torch.no_grad():
images = Variable(
torch.from_numpy(np.asarray(photo)).type(torch.FloatTensor))
if self.cuda:
images = images.cuda()
outputs = self.centernet(images)
if self.backbone == 'hourglass':
outputs = [
outputs[-1]["hm"].sigmoid(), outputs[-1]["wh"],
outputs[-1]["reg"]
]
#-----------------------------------------------------------#
# 利用预测结果进行解码
#-----------------------------------------------------------#
outputs = decode_bbox(outputs[0], outputs[1], outputs[2],
self.image_size, self.confidence, self.cuda)
#-------------------------------------------------------#
# 对于centernet网络来讲，确立中心非常重要。
# 对于大目标而言，会存在许多的局部信息。
# 此时对于同一个大目标，中心点比较难以确定。
# 使用最大池化的非极大抑制方法无法去除局部框
# 所以我还是写了另外一段对框进行非极大抑制的代码
# 实际测试中，hourglass为主干网络时有无额外的nms相差不大，resnet相差较大。
#-------------------------------------------------------#
try:
if self.nms:
outputs = np.array(nms(outputs, self.nms_threhold))
except:
pass
output = outputs[0]
if len(output) <= 0:
return image
batch_boxes, det_conf, det_label = output[:, :
4], output[:,
4], output[:,
5]
det_xmin, det_ymin, det_xmax, det_ymax = batch_boxes[:,
0], batch_boxes[:,
1], batch_boxes[:,
2], batch_boxes[:,
3]
#-----------------------------------------------------------#
# 筛选出其中得分高于confidence的框
#-----------------------------------------------------------#
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
#-----------------------------------------------------------#
# 去掉灰条部分
#-----------------------------------------------------------#
boxes = centernet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.image_size[0], self.image_size[1]]),
image_shape)
font = ImageFont.truetype(font='model_data/simhei.ttf',
size=np.floor(3e-2 * np.shape(image)[1] +
0.5).astype('int32'))
thickness = max(
(np.shape(image)[0] + np.shape(image)[1]) // self.image_size[0], 1)
for i, c in enumerate(top_label_indices):
predicted_class = self.class_names[int(c)]
score = top_conf[i]
top, left, bottom, right = boxes[i]
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(
np.shape(image)[0],
np.floor(bottom + 0.5).astype('int32'))
right = min(
np.shape(image)[1],
np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label, top, left, bottom, right)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[int(c)])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[int(c)])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
del draw
return image
def detect_image(self, image):
start = timer()
# 调整图片使其符合输入要求
new_image_size = (self.model_image_size[1], self.model_image_size[0])
boxed_image = letterbox_image(image, new_image_size)
image_data = np.array(boxed_image, dtype='float32')
image_data /= 255.
image_data = np.expand_dims(image_data, 0) # Add batch dimension.
if self.eager:
# 预测结果
input_image_shape = np.expand_dims(
np.array([image.size[1], image.size[0]], dtype='float32'), 0)
out_boxes, out_scores, out_classes = self.yolo_model.predict(
[image_data, input_image_shape])
else:
# 预测结果
out_boxes, out_scores, out_classes = self.sess.run(
[self.boxes, self.scores, self.classes],
feed_dict={
self.yolo_model.input: image_data,
self.input_image_shape: [image.size[1], image.size[0]],
K.learning_phase(): 0
})
print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
# 设置字体
font = ImageFont.truetype(font='font/simhei.ttf',
size=np.floor(3e-2 * image.size[1] +
0.5).astype('int32'))
thickness = (image.size[0] + image.size[1]) // 300
small_pic = []
for i, c in list(enumerate(out_classes)):
predicted_class = self.class_names[c]
box = out_boxes[i]
score = out_scores[i]
top, left, bottom, right = box
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(image.size[1], np.floor(bottom + 0.5).astype('int32'))
right = min(image.size[0], np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[c])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[c])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
del draw
end = timer()
print(end - start)
return image
def detect_image(self, image_id, image):
self.confidence = 0.01
f = open("./input/detection-results/" + image_id + ".txt", "w")
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
if self.letterbox_image:
crop_img = np.array(
letterbox_image(
image,
(self.model_image_size[1], self.model_image_size[0])))
else:
crop_img = image.convert('RGB')
crop_img = crop_img.resize(
(self.model_image_size[1], self.model_image_size[0]),
Image.BICUBIC)
photo = np.array(crop_img, dtype=np.float64)
#-----------------------------------------------------------#
# 图片预处理，归一化。
#-----------------------------------------------------------#
photo = preprocess_input(
np.reshape(photo, [
1, self.model_image_size[0], self.model_image_size[1],
self.model_image_size[2]
]))
preds = self.m2det.predict(photo)
#-----------------------------------------------------------#
# 将预测结果进行解码
#-----------------------------------------------------------#
results = self.bbox_util.detection_out(
preds, self.prior, confidence_threshold=self.confidence)
#--------------------------------------#
# 如果没有检测到物体，则返回原图
#--------------------------------------#
if len(results[0]) <= 0:
return image
#-----------------------------------------------------------#
# 筛选出其中得分高于confidence的框
#-----------------------------------------------------------#
det_label = results[0][:, 0]
det_conf = results[0][:, 1]
det_xmin, det_ymin, det_xmax, det_ymax = results[0][:, 2], results[
0][:, 3], results[0][:, 4], results[0][:, 5]
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
#-----------------------------------------------------------#
# 去掉灰条部分
#-----------------------------------------------------------#
if self.letterbox_image:
boxes = m2det_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.model_image_size[0], self.model_image_size[1]]),
image_shape)
else:
top_xmin = top_xmin * image_shape[1]
top_ymin = top_ymin * image_shape[0]
top_xmax = top_xmax * image_shape[1]
top_ymax = top_ymax * image_shape[0]
boxes = np.concatenate([top_ymin, top_xmin, top_ymax, top_xmax],
axis=-1)
for i, c in enumerate(top_label_indices):
predicted_class = self.class_names[int(c) - 1]
score = str(top_conf[i])
top, left, bottom, right = boxes[i]
f.write("%s %s %s %s %s %s\n" %
(predicted_class, score[:6], str(int(left)), str(
int(top)), str(int(right)), str(int(bottom))))
f.close()
return
def detect_image(self, image, trfn, carn, lrx, rrx, carnums, carlist, sy,
slx, srx, lx, ly):
run_a_red_light = 0
totalcarn = 0
#list_num=len(carlist)
#start = timer()
# 调整图片使其符合输入要求
new_image_size = (self._defaults["model_image_size"][0],
self._defaults["model_image_size"][1])
boxed_image = letterbox_image(image, new_image_size)
image_data = np.array(boxed_image, dtype='float32')
image_data /= 255.
image_data = np.expand_dims(image_data, 0) # Add batch dimension.
# 预测结果
out_boxes, out_scores, out_classes = self.sess.run(
[self.boxes, self.scores, self.classes],
feed_dict={
self.yolo_model.input: image_data,
self.input_image_shape: [image.size[1], image.size[0]],
# K.learning_phase(): 0
})
print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
font = ImageFont.truetype('font/simhei.ttf', 28)
for i, c in list(enumerate(out_classes)):
if i == 1:
trfn = trfn + 1
outputpath = trafficoutputpath
img = image.crop((lx - 15, ly - 30, lx + 15, ly + 30))
img.save(outputpath + str(trfn) + ".jpg")
temp = check1(outputpath + str(trfn) + ".jpg")
draw = ImageDraw.Draw(image)
if temp == 'red':
draw.rectangle((lx - 15, ly - 30, lx + 15, ly + 30),
outline="red",
width=2)
draw.rectangle((lx - 15, ly - 55, lx + 70, ly - 30),
fill="white")
draw.text((lx - 15, ly - 55),
"red",
fill=(0, 0, 0),
font=font)
f1 = open(resultpath, "a", encoding='utf-8')
f2 = open(run_a_red_lightpath(), "a", encoding='utf-8')
f3 = open(Road_ROOTpath() + 'all_illegal_car_info.txt',
"a+",
encoding='utf-8')
for cn, cf in list(enumerate(out_classes)):
if cf != 2:
continue
predicted_class = self.class_names[cf]
box = out_boxes[cn]
score = out_scores[cn]
top, left, bottom, right = box
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(image.size[1],
np.floor(bottom + 0.5).astype('int32'))
right = min(image.size[0],
np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
x = left
y1 = top
y2 = bottom
if cf == 2 and x < srx and (x > slx and
y2 < 1080) and y1 > 500:
carn = carn + 1
car_outputpath = caroutputpath
img = image.crop((left, top, right, bottom))
img.save(car_outputpath + str(carn) + ".jpg")
# if y>500 :
ch = cartag(car_outputpath + str(carn) + ".jpg")
f1.write(ch + "\n")
draw.rectangle((left, top, right, bottom),
outline=self.colors[cf],
width=2)
draw.rectangle([
tuple(text_origin),
tuple(text_origin + label_size)
],
fill=self.colors[cf])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
draw.rectangle((right, top, right + 150, top + 40),
fill=self.colors[cf])
draw.text((right, top, right + 30, top + 60),
ch,
fill=(0, 0, 0),
font=font)
if ((top + bottom) / 2 - 50) < sy:
if ch is not '00000' and ch[1] is not '1':
f3.write(ch + " 闯红灯" + "\n")
f2.write(ch + " 闯红灯" + "\n")
run_a_red_light += 1
draw.rectangle(
(right, top + 40, right + 100, top + 80),
fill='red')
draw.text(
(right, top + 40, right + 60, top + 140),
"闯红灯",
fill=(0, 0, 0),
font=font)
if ch is not '00000' and ch[1] is not '1':
image.save(run_a_red_light_img_path() +
ch + '_闯红灯' + ".jpg")
# f3.close()
# f2.close()
# f1.close()
else:
draw.rectangle((lx - 15, ly - 30, lx + 15, ly + 30),
outline="green",
width=2)
draw.rectangle((lx - 15, ly - 55, lx + 70, ly - 30),
fill="white")
draw.text((lx - 15, ly - 55),
"green",
fill=(0, 0, 0),
font=font)
del draw
if c == 9:
continue
predicted_class = self.class_names[c]
box = out_boxes[i]
score = out_scores[i]
top, left, bottom, right = box
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(image.size[1], np.floor(bottom + 0.5).astype('int32'))
right = min(image.size[0], np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
draw.rectangle((left, top, right, bottom),
outline=self.colors[c],
width=2)
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[c])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
draw.rectangle((left, top, right, bottom),
outline=self.colors[c],
width=2)
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[c])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
if c == 2 or c == 3 or c == 5 or c == 7:
totalcarn += 1
del draw
if c == 2 and bottom > 950 and left > slx:
carnums += 1
img = image.crop((left, top, right, bottom))
img.save(caridpath + str(carnums) + '.jpg')
carid = cartag(caridpath + str(carnums) + '.jpg')
draw = ImageDraw.Draw(image)
draw.rectangle((right, top, right + 150, top + 40),
fill="green")
draw.text((right, top, right + 30, top + 60),
carid,
fill=(0, 0, 0),
font=font)
if carid in carlist:
continue
else:
if carid != '00000' and len(carid) == 7:
carlist.append(carid)
draw = ImageDraw.Draw(image)
draw.rectangle((1300, 150, 1900, 200), fill="white")
draw.text((1300, 160, 1900, 180),
"路口通过car的数量为:" + str(len(carlist)),
fill="black",
font=font)
del draw
draw = ImageDraw.Draw(image)
draw.rectangle((1300, 50, 1900, 140), fill="white")
if totalcarn >= 10:
draw.text((1300, 60, 1900, 100),
"当前路口机动车数量为" + str(totalcarn) + "大于9" + " 拥堵",
fill="black",
font=font)
else:
draw.text((1300, 60, 1800, 100),
"当前路口机动车数量：" + str(totalcarn),
fill="black",
font=font)
draw.text((1300, 100, 1800, 140),
"当前路口闯红灯数量：" + str(run_a_red_light),
fill="black",
font=font)
del draw
return image, trfn, carn, carnums, len(
carlist), run_a_red_light, totalcarn
def detect_image(self, image):
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
if self.letterbox_image:
crop_img = np.array(
letterbox_image(image,
(self.input_shape[1], self.input_shape[0])))
else:
crop_img = image.convert('RGB')
crop_img = crop_img.resize(
(self.input_shape[1], self.input_shape[0]), Image.BICUBIC)
photo = np.array(crop_img, dtype=np.float64)
#-----------------------------------------------------------#
# 图片预处理，归一化。
#-----------------------------------------------------------#
photo = preprocess_input(
np.reshape(photo,
[1, self.input_shape[0], self.input_shape[1], 3]))
preds = self.get_pred(photo).numpy()
#-----------------------------------------------------------#
# 将预测结果进行解码
#-----------------------------------------------------------#
results = self.bbox_util.detection_out(
preds, confidence_threshold=self.confidence)
#--------------------------------------#
# 如果没有检测到物体，则返回原图
#--------------------------------------#
if len(results[0]) <= 0:
return image
#-----------------------------------------------------------#
# 筛选出其中得分高于confidence的框
#-----------------------------------------------------------#
det_label = results[0][:, 0]
det_conf = results[0][:, 1]
det_xmin, det_ymin, det_xmax, det_ymax = results[0][:, 2], results[
0][:, 3], results[0][:, 4], results[0][:, 5]
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
#-----------------------------------------------------------#
# 去掉灰条部分
#-----------------------------------------------------------#
if self.letterbox_image:
boxes = ssd_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.input_shape[0], self.input_shape[1]]),
image_shape)
else:
top_xmin = top_xmin * image_shape[1]
top_ymin = top_ymin * image_shape[0]
top_xmax = top_xmax * image_shape[1]
top_ymax = top_ymax * image_shape[0]
boxes = np.concatenate([top_ymin, top_xmin, top_ymax, top_xmax],
axis=-1)
font = ImageFont.truetype(font='model_data/simhei.ttf',
size=np.floor(3e-2 * np.shape(image)[1] +
0.5).astype('int32'))
thickness = max(
(np.shape(image)[0] + np.shape(image)[1]) // self.input_shape[0],
1)
for i, c in enumerate(top_label_indices):
predicted_class = self.class_names[int(c) - 1]
score = top_conf[i]
top, left, bottom, right = boxes[i]
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(
np.shape(image)[0],
np.floor(bottom + 0.5).astype('int32'))
right = min(
np.shape(image)[1],
np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label, top, left, bottom, right)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[int(c) - 1])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[int(c) - 1])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
del draw
return image
def detect_image(self, image):
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
if self.letterbox_image:
boxed_image = letterbox_image(
image, (self.model_image_size[1], self.model_image_size[0]))
else:
boxed_image = image.convert('RGB')
boxed_image = boxed_image.resize(
(self.model_image_size[1], self.model_image_size[0]),
Image.BICUBIC)
image_data = np.array(boxed_image, dtype='float32')
image_data /= 255.
#---------------------------------------------------------#
# 添加上batch_size维度
#---------------------------------------------------------#
image_data = np.expand_dims(image_data, 0)
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
out_boxes, out_scores, out_classes = self.sess.run(
[self.boxes, self.scores, self.classes],
feed_dict={
self.yolo_model.input: image_data,
self.input_image_shape: [image.size[1], image.size[0]],
K.learning_phase(): 0
})
print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
#---------------------------------------------------------#
# 设置字体
#---------------------------------------------------------#
font = ImageFont.truetype(font='font/simhei.ttf',
size=np.floor(3e-2 * image.size[1] +
0.5).astype('int32'))
thickness = max((image.size[0] + image.size[1]) // 300, 1)
for i, c in list(enumerate(out_classes)):
predicted_class = self.class_names[c]
box = out_boxes[i]
score = out_scores[i]
top, left, bottom, right = box
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(image.size[1], np.floor(bottom + 0.5).astype('int32'))
right = min(image.size[0], np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label, top, left, bottom, right)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[c])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[c])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
del draw
return image
def get_FPS(self, image, test_interval):
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
if self.letterbox_image:
boxed_image = letterbox_image(
image, (self.model_image_size[1], self.model_image_size[0]))
else:
boxed_image = image.convert('RGB')
boxed_image = boxed_image.resize(
(self.model_image_size[1], self.model_image_size[0]),
Image.BICUBIC)
image_data = np.array(boxed_image, dtype='float32')
image_data /= 255.
#---------------------------------------------------------#
# 添加上batch_size维度
#---------------------------------------------------------#
image_data = np.expand_dims(image_data, 0) # Add batch dimension.
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
if self.eager:
# 预测结果
input_image_shape = np.expand_dims(
np.array([image.size[1], image.size[0]], dtype='float32'), 0)
out_boxes, out_scores, out_classes = self.get_pred(
image_data, input_image_shape)
else:
# 预测结果
out_boxes, out_scores, out_classes = self.sess.run(
[self.boxes, self.scores, self.classes],
feed_dict={
self.yolo_model.input: image_data,
self.input_image_shape: [image.size[1], image.size[0]],
K.learning_phase(): 0
})
t1 = time.time()
for _ in range(test_interval):
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
if self.eager:
# 预测结果
input_image_shape = np.expand_dims(
np.array([image.size[1], image.size[0]], dtype='float32'),
0)
out_boxes, out_scores, out_classes = self.get_pred(
image_data, input_image_shape)
else:
# 预测结果
out_boxes, out_scores, out_classes = self.sess.run(
[self.boxes, self.scores, self.classes],
feed_dict={
self.yolo_model.input: image_data,
self.input_image_shape: [image.size[1], image.size[0]],
K.learning_phase(): 0
})
t2 = time.time()
tact_time = (t2 - t1) / test_interval
return tact_time
def detect_image(self, image_id, image):
f = open("./input/detection-results/" + image_id + ".txt", "w")
self.confidence = 0.01
self.nms_threhold = 0.5
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
#---------------------------------------------------------#
crop_img = letterbox_image(image,
[self.image_size[0], self.image_size[1]])
#----------------------------------------------------------------------------------#
# 将RGB转化成BGR，这是因为原始的centernet_hourglass权值是使用BGR通道的图片训练的
#----------------------------------------------------------------------------------#
photo = np.array(crop_img, dtype=np.float32)[:, :, ::-1]
#-----------------------------------------------------------#
# 图片预处理，归一化。获得的photo的shape为[1, 512, 512, 3]
#-----------------------------------------------------------#
photo = np.reshape(
np.transpose(preprocess_image(photo), (2, 0, 1)),
[1, self.image_size[2], self.image_size[0], self.image_size[1]])
with torch.no_grad():
images = Variable(
torch.from_numpy(np.asarray(photo)).type(torch.FloatTensor))
if self.cuda:
images = images.cuda()
outputs = self.centernet(images)
if self.backbone == 'hourglass':
outputs = [
outputs[-1]["hm"].sigmoid(), outputs[-1]["wh"],
outputs[-1]["reg"]
]
#-----------------------------------------------------------#
# 利用预测结果进行解码
#-----------------------------------------------------------#
outputs = decode_bbox(outputs[0], outputs[1], outputs[2],
self.image_size, self.confidence, self.cuda)
#-------------------------------------------------------#
# 对于centernet网络来讲，确立中心非常重要。
# 对于大目标而言，会存在许多的局部信息。
# 此时对于同一个大目标，中心点比较难以确定。
# 使用最大池化的非极大抑制方法无法去除局部框
# 所以我还是写了另外一段对框进行非极大抑制的代码
# 实际测试中，hourglass为主干网络时有无额外的nms相差不大，resnet相差较大。
#-------------------------------------------------------#
try:
if self.nms:
outputs = np.array(nms(outputs, self.nms_threhold))
except:
pass
output = outputs[0]
if len(output) <= 0:
return image
batch_boxes, det_conf, det_label = output[:, :
4], output[:,
4], output[:,
5]
det_xmin, det_ymin, det_xmax, det_ymax = batch_boxes[:,
0], batch_boxes[:,
1], batch_boxes[:,
2], batch_boxes[:,
3]
#-----------------------------------------------------------#
# 筛选出其中得分高于confidence的框
#-----------------------------------------------------------#
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
#-----------------------------------------------------------#
# 去掉灰条部分
#-----------------------------------------------------------#
boxes = centernet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.image_size[0], self.image_size[1]]),
image_shape)
for i, c in enumerate(top_label_indices):
predicted_class = self.class_names[int(c)]
score = str(top_conf[i])
top, left, bottom, right = boxes[i]
f.write("%s %s %s %s %s %s\n" %
(predicted_class, score[:6], str(int(left)), str(
int(top)), str(int(right)), str(int(bottom))))
f.close()
return
import numpy as np
import tensorflow as tf
from PIL import Image, ImageFont, ImageDraw
from utils.utils import letterbox_image
from utils.setup_tool import get_classes, get_anchors
import cv2
img = "test_data/london.jpg"
image = Image.open(img)
model_image_size = (416, 416)
image_shape = (image.size[1], image.size[0], 3)
model_image_size[0] % 32 == 0, 'Multiples of 32 required'
model_image_size[1] % 32 == 0, 'Multiples of 32 required'
boxed_image = letterbox_image(image, tuple(reversed(model_image_size)))
image_data = np.array(boxed_image, dtype='float32')
image_data /= 255.
image_data = np.expand_dims(image_data, 0)
#print(image.size)
print(image_shape)
print(image_data.shape)
# Load TFLite model and allocate tensors.
interpreter = tf.lite.Interpreter(
model_path="model_data/small_mobilenet_yolo.tflite")
interpreter.allocate_tensors()
# Get input and output tensors.
def detect_image(self, image, classroom_id):
start = timer()
# 调整图片使其符合输入要求
new_image_size = (image.width - (image.width % 32),
image.height - (image.height % 32))
boxed_image = letterbox_image(image, new_image_size)
image_data = np.array(boxed_image, dtype='float32')
image_data /= 255.
image_data = np.expand_dims(image_data, 0) # Add batch dimension.
# 预测结果
out_boxes, out_scores, out_classes = self.sess.run(
[self.boxes, self.scores, self.classes],
feed_dict={
self.yolo_model.input: image_data,
self.input_image_shape: [image.size[1], image.size[0]],
K.learning_phase(): 0
})
print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
# 设置字体
font = ImageFont.truetype(font='font/simhei.ttf',
size=np.floor(3e-2 * image.size[1] + 0.5).astype('int32'))
thickness = (image.size[0] + image.size[1]) // 300
small_pic = []
for i, c in list(enumerate(out_classes)):
predicted_class = self.class_names[c]
box = out_boxes[i]
score = out_scores[i]
top, left, bottom, right = box
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(image.size[1], np.floor(bottom + 0.5).astype('int32'))
right = min(image.size[0], np.floor(right + 0.5).astype('int32'))
# print(top, left, bottom, right)
# point_x = (right + left) / 2
# point_y = (top + bottom) / 2
# # 判断是否有人
# if predicted_class == 'person':
# result = mysql.seat_select(point_x, point_y, classroom_id)
# if result.__len__() == 1:
# mysql.seat_update(result[0][0])
# elif result.__len__() >= 2:
# distance = 0.00
# r_id = 0
# for r in result:
# pic_x = (r[3] + r[4]) / 2
# pic_y = (r[1] + r[2]) / 2
# aa = round(math.sqrt(math.pow((pic_x - point_x)) + math.pow((pic_y - point_y))), 2)
# if aa > distance:
# r_id = r[0]
# distance = aa
# if r_id != 0:
# mysql.seat_update(r_id)
# print('6' * 60)
if predicted_class == 'person':
# 画框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle(
[left + i, top + i, right - i, bottom - i],
outline=self.colors[c])
draw.rectangle(
[tuple(text_origin), tuple(text_origin + label_size)],
fill=self.colors[c])
draw.text(text_origin, str(label, 'UTF-8'), fill=(0, 0, 0), font=font)
del draw
end = timer()
print(end - start)
return image
def detect_image(self, image):
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
if self.letterbox_image:
crop_img = np.array(letterbox_image(image, (self.model_image_size[1],self.model_image_size[0])))
else:
crop_img = image.convert('RGB')
crop_img = crop_img.resize((self.model_image_size[1],self.model_image_size[0]), Image.BICUBIC)
photo = np.array(crop_img,dtype = np.float32) / 255.0
photo = np.transpose(photo, (2, 0, 1))
#---------------------------------------------------------#
# 添加上batch_size维度
#---------------------------------------------------------#
images = [photo]
with torch.no_grad():
images = torch.from_numpy(np.asarray(images))
if self.cuda:
images = images.cuda()
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
outputs = self.net(images)
output_list = []
for i in range(3):
output_list.append(self.yolo_decodes[i](outputs[i]))
#---------------------------------------------------------#
# 将预测框进行堆叠，然后进行非极大抑制
#---------------------------------------------------------#
output = torch.cat(output_list, 1)
batch_detections = non_max_suppression(output, self.config["yolo"]["classes"],
conf_thres=self.confidence,
nms_thres=self.iou)
#---------------------------------------------------------#
# 如果没有检测出物体，返回原图
#---------------------------------------------------------#
try :
batch_detections = batch_detections[0].cpu().numpy()
except:
return image
#---------------------------------------------------------#
# 对预测框进行得分筛选
#---------------------------------------------------------#
top_index = batch_detections[:,4] * batch_detections[:,5] > self.confidence
top_conf = batch_detections[top_index,4]*batch_detections[top_index,5]
top_label = np.array(batch_detections[top_index,-1],np.int32)
top_bboxes = np.array(batch_detections[top_index,:4])
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(top_bboxes[:,0],-1),np.expand_dims(top_bboxes[:,1],-1),np.expand_dims(top_bboxes[:,2],-1),np.expand_dims(top_bboxes[:,3],-1)
#-----------------------------------------------------------------#
# 在图像传入网络预测前会进行letterbox_image给图像周围添加灰条
# 因此生成的top_bboxes是相对于有灰条的图像的
# 我们需要对其进行修改，去除灰条的部分。
#-----------------------------------------------------------------#
if self.letterbox_image:
boxes = yolo_correct_boxes(top_ymin,top_xmin,top_ymax,top_xmax,np.array([self.model_image_size[0],self.model_image_size[1]]),image_shape)
else:
top_xmin = top_xmin / self.model_image_size[1] * image_shape[1]
top_ymin = top_ymin / self.model_image_size[0] * image_shape[0]
top_xmax = top_xmax / self.model_image_size[1] * image_shape[1]
top_ymax = top_ymax / self.model_image_size[0] * image_shape[0]
boxes = np.concatenate([top_ymin,top_xmin,top_ymax,top_xmax], axis=-1)
font = ImageFont.truetype(font='model_data/simhei.ttf',size=np.floor(3e-2 * np.shape(image)[1] + 0.5).astype('int32'))
thickness = max((np.shape(image)[0] + np.shape(image)[1]) // self.model_image_size[0], 1)
for i, c in enumerate(top_label):
predicted_class = self.class_names[c]
score = top_conf[i]
top, left, bottom, right = boxes[i]
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(np.shape(image)[0], np.floor(bottom + 0.5).astype('int32'))
right = min(np.shape(image)[1], np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label, top, left, bottom, right)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle(
[left + i, top + i, right - i, bottom - i],
outline=self.colors[self.class_names.index(predicted_class)])
draw.rectangle(
[tuple(text_origin), tuple(text_origin + label_size)],
fill=self.colors[self.class_names.index(predicted_class)])
draw.text(text_origin, str(label,'UTF-8'), fill=(0, 0, 0), font=font)
del draw
return image
def detect_image(self, image):
'''检测图片
Parameters
----------
image: Image,
Returns
-------
image:
'''
start = timer()
if self.model_image_size != (None, None):
assert self.model_image_size[
0] % 32 == 0, 'Multiples of 32 required'
assert self.model_image_size[
1] % 32 == 0, 'Multiples of 32 required'
boxed_image = letterbox_image(image,
tuple(reversed(
self.model_image_size))) # 填充图像
else:
new_image_size = (image.width - (image.width % 32),
image.height - (image.height % 32))
# <PIL.Image.Image image mode=RGB size=416*416 at 0x.. >
boxed_image = letterbox_image(image, new_image_size)
# 调整图片使其符合输入要求
# new_image_size = (image.width - (image.width % 32),
# image.height - (image.height % 32))
# boxed_image = letterbox_image(image, new_image_size)
# image_date: array, shape=(416,416,3)
image_data = np.array(boxed_image, dtype='float32')
image_data /= 255.
# image_data: array, shape=(1, 414, 414, 3)
image_data = np.expand_dims(image_data, 0) # Add batch dimension.
# 预测结果:参数盒子、得分、类别；输入图像0~1，4维；原始图像的尺寸
# out_boxes:array,shape=(n, 4), n为输出图片中方框数量
# out_scores:array, shape=(n,), n同上
# out_classes:array, shape=(n,), n同上
out_boxes, out_scores, out_classes = self.sess.run(
[self.boxes, self.scores, self.classes],
feed_dict={
self.yolo_model.input: image_data,
self.input_image_shape: [image.size[1], image.size[0]],
K.learning_phase(): 0
})
# 输出检测出的框
print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
# 设置字体
font = ImageFont.truetype(font='font/simhei.ttf',
size=np.floor(3e-2 * image.size[1] +
0.5).astype('int32'))
thickness = (image.size[0] + image.size[1]) // 300
# small_pic=[]
# c为类别标号，如c=6,表示为car
# i表示图片中第几个方框，如： i=0,c=6 表示第1个方框为car，predicted_class='car'
for i, c in list(enumerate(out_classes)):
predicted_class = self.class_names[c]
box = out_boxes[i]
score = out_scores[i]
top, left, bottom, right = box
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(image.size[1], np.floor(bottom + 0.5).astype('int32'))
right = min(image.size[0], np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[c])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[c])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
del draw
end = timer()
# 检测执行时间
print("检测执行时间：" + str(end - start))
return image
def detect_image(self, image_id, image):
self.confidence = 0.001
f = open("./input/detection-results/" + image_id + ".txt", "w")
image_shape = np.array(np.shape(image)[0:2])
crop_img = np.array(
letterbox_image(
image, (self.model_image_size[0], self.model_image_size[1])))
photo = np.array(crop_img, dtype=np.float32)
photo /= 255.0
photo = np.transpose(photo, (2, 0, 1))
photo = photo.astype(np.float32)
images = []
images.append(photo)
images = np.asarray(images)
images = torch.from_numpy(images)
if self.cuda:
images = images.cuda()
with torch.no_grad():
outputs = self.net(images)
output_list = []
for i in range(3):
output_list.append(self.yolo_decodes[i](outputs[i]))
output = torch.cat(output_list, 1)
batch_detections = non_max_suppression(
output,
self.config["yolo"]["classes"],
conf_thres=self.confidence,
nms_thres=0.3)
try:
batch_detections = batch_detections[0].cpu().numpy()
except:
return image
top_index = batch_detections[:,
4] * batch_detections[:,
5] > self.confidence
top_conf = batch_detections[top_index, 4] * batch_detections[top_index,
5]
top_label = np.array(batch_detections[top_index, -1], np.int32)
top_bboxes = np.array(batch_detections[top_index, :4])
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
top_bboxes[:, 0],
-1), np.expand_dims(top_bboxes[:, 1], -1), np.expand_dims(
top_bboxes[:, 2], -1), np.expand_dims(top_bboxes[:, 3], -1)
# 去掉灰条
boxes = yolo_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.model_image_size[0], self.model_image_size[1]]),
image_shape)
for i, c in enumerate(top_label):
predicted_class = self.class_names[c]
score = str(top_conf[i])
top, left, bottom, right = boxes[i]
f.write("%s %s %s %s %s %s\n" %
(predicted_class, score[:6], str(int(left)), str(
int(top)), str(int(right)), str(int(bottom))))
f.close()
return
def detect_image(self, image):
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
#---------------------------------------------------------#
crop_img = letterbox_image(image,
[self.input_shape[0], self.input_shape[1]])
#----------------------------------------------------------------------------------#
# 将RGB转化成BGR，这是因为原始的centernet_hourglass权值是使用BGR通道的图片训练的
#----------------------------------------------------------------------------------#
photo = np.array(crop_img, dtype=np.float32)[:, :, ::-1]
photo = np.reshape(
preprocess_image(photo),
[1, self.input_shape[0], self.input_shape[1], self.input_shape[2]])
preds = self.centernet.predict(photo)
#--------------------------------------------------------------------------#
# 对于centernet网络来讲，确立中心非常重要。
# 对于大目标而言，会存在许多的局部信息。
# 此时对于同一个大目标，中心点比较难以确定。
# 使用最大池化的非极大抑制方法无法去除局部框
# 所以我还是写了另外一段对框进行非极大抑制的代码
# 实际测试中，hourglass为主干网络时有无额外的nms相差不大，resnet相差较大。
#---------------------------------------------------------------------------#
if self.nms:
preds = np.array(nms(preds, self.nms_threhold))
if len(preds[0]) <= 0:
return image
#-----------------------------------------------------------#
# 将预测结果转换成小数的形式
#-----------------------------------------------------------#
preds[0][:, 0:4] = preds[0][:, 0:4] / (self.input_shape[0] / 4)
det_label = preds[0][:, -1]
det_conf = preds[0][:, -2]
det_xmin, det_ymin, det_xmax, det_ymax = preds[0][:, 0], preds[
0][:, 1], preds[0][:, 2], preds[0][:, 3]
#-----------------------------------------------------------#
# 筛选出其中得分高于confidence的框
#-----------------------------------------------------------#
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
#-----------------------------------------------------------#
# 去掉灰条部分
#-----------------------------------------------------------#
boxes = centernet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.input_shape[0], self.input_shape[1]]), image_shape)
font = ImageFont.truetype(font='model_data/simhei.ttf',
size=np.floor(3e-2 * np.shape(image)[1] +
0.5).astype('int32'))
thickness = max(
(np.shape(image)[0] + np.shape(image)[1]) // self.input_shape[0],
1)
for i, c in enumerate(top_label_indices):
predicted_class = self.class_names[int(c)]
score = top_conf[i]
top, left, bottom, right = boxes[i]
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(
np.shape(image)[0],
np.floor(bottom + 0.5).astype('int32'))
right = min(
np.shape(image)[1],
np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label, top, left, bottom, right)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[int(c)])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[int(c)])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
del draw
return image
def get_FPS(self, image, test_interval):
# 调整图片使其符合输入要求
image_shape = np.array(np.shape(image)[0:2])
crop_img = letterbox_image(
image, [self.model_image_size[0], self.model_image_size[1]])
photo = np.array(crop_img, dtype=np.float32)
# 图片预处理，归一化
photo = np.reshape(preprocess_input(photo), [
1, self.model_image_size[0], self.model_image_size[1],
self.model_image_size[2]
])
preds = self.get_pred(photo)
preds = [pred.numpy() for pred in preds]
# 将预测结果进行解码
results = self.bbox_util.detection_out(
preds, self.prior, confidence_threshold=self.confidence)
if len(results[0]) > 0:
results = np.array(results)
# 筛选出其中得分高于confidence的框
det_label = results[0][:, 5]
det_conf = results[0][:, 4]
det_xmin, det_ymin, det_xmax, det_ymax = results[0][:, 0], results[
0][:, 1], results[0][:, 2], results[0][:, 3]
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
# 去掉灰条
boxes = efficientdet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.model_image_size[0], self.model_image_size[1]]),
image_shape)
t1 = time.time()
for _ in range(test_interval):
preds = self.get_pred(photo)
preds = [pred.numpy() for pred in preds]
# 将预测结果进行解码
results = self.bbox_util.detection_out(
preds, self.prior, confidence_threshold=self.confidence)
if len(results[0]) > 0:
results = np.array(results)
# 筛选出其中得分高于confidence的框
det_label = results[0][:, 5]
det_conf = results[0][:, 4]
det_xmin, det_ymin, det_xmax, det_ymax = results[
0][:, 0], results[0][:, 1], results[0][:, 2], results[0][:,
3]
top_indices = [
i for i, conf in enumerate(det_conf)
if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices], -1), np.expand_dims(
det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
# 去掉灰条
boxes = efficientdet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array(
[self.model_image_size[0], self.model_image_size[1]]),
image_shape)
t2 = time.time()
tact_time = (t2 - t1) / test_interval
return tact_time
def get_FPS(self, image, test_interval):
# 调整图片使其符合输入要求
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
if self.letterbox_image:
crop_img = np.array(
letterbox_image(image,
(self.input_shape[1], self.input_shape[0])))
else:
crop_img = image.convert('RGB')
crop_img = crop_img.resize(
(self.input_shape[1], self.input_shape[0]), Image.BICUBIC)
photo = np.array(crop_img, dtype=np.float64)
photo = preprocess_input(
np.reshape(photo,
[1, self.input_shape[0], self.input_shape[1], 3]))
preds = self.ssd_model.predict(photo)
results = self.bbox_util.detection_out(
preds, confidence_threshold=self.confidence)
if len(results[0]) > 0:
det_label = results[0][:, 0]
det_conf = results[0][:, 1]
det_xmin, det_ymin, det_xmax, det_ymax = results[0][:, 2], results[
0][:, 3], results[0][:, 4], results[0][:, 5]
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
#-----------------------------------------------------------#
# 去掉灰条部分
#-----------------------------------------------------------#
if self.letterbox_image:
boxes = ssd_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.input_shape[0], self.input_shape[1]]),
image_shape)
else:
top_xmin = top_xmin * image_shape[1]
top_ymin = top_ymin * image_shape[0]
top_xmax = top_xmax * image_shape[1]
top_ymax = top_ymax * image_shape[0]
boxes = np.concatenate(
[top_ymin, top_xmin, top_ymax, top_xmax], axis=-1)
t1 = time.time()
for _ in range(test_interval):
preds = self.ssd_model.predict(photo)
results = self.bbox_util.detection_out(
preds, confidence_threshold=self.confidence)
if len(results[0]) > 0:
det_label = results[0][:, 0]
det_conf = results[0][:, 1]
det_xmin, det_ymin, det_xmax, det_ymax = results[
0][:, 2], results[0][:, 3], results[0][:, 4], results[0][:,
5]
top_indices = [
i for i, conf in enumerate(det_conf)
if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices], -1), np.expand_dims(
det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
#-----------------------------------------------------------#
# 去掉灰条部分
#-----------------------------------------------------------#
if self.letterbox_image:
boxes = ssd_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.input_shape[0], self.input_shape[1]]),
image_shape)
else:
top_xmin = top_xmin * image_shape[1]
top_ymin = top_ymin * image_shape[0]
top_xmax = top_xmax * image_shape[1]
top_ymax = top_ymax * image_shape[0]
boxes = np.concatenate(
[top_ymin, top_xmin, top_ymax, top_xmax], axis=-1)
t2 = time.time()
tact_time = (t2 - t1) / test_interval
return tact_time
def get_FPS(self, image, test_interval):
image = np.array(image, np.float32)
im_height, im_width, _ = np.shape(image)
scale = [
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0]
]
scale_for_landmarks = [
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0],
np.shape(image)[1],
np.shape(image)[0]
]
#---------------------------------------------------------#
# letterbox_image可以给图像增加灰条，实现不失真的resize
#---------------------------------------------------------#
if self.letterbox_image:
image = letterbox_image(image,
[self.input_shape[1], self.input_shape[0]])
else:
self.anchors = Anchors(self.cfg,
image_size=(im_height,
im_width)).get_anchors()
photo = np.expand_dims(preprocess_input(image), 0)
preds = self.retinaface.predict(photo)
results = self.bbox_util.detection_out(
preds, self.anchors, confidence_threshold=self.confidence)
if len(results) > 0:
results = np.array(results)
#---------------------------------------------------------#
# 如果使用了letterbox_image的话，要把灰条的部分去除掉。
#---------------------------------------------------------#
if self.letterbox_image:
results = retinaface_correct_boxes(
results,
np.array([self.input_shape[0], self.input_shape[1]]),
np.array([im_height, im_width]))
results[:, :4] = results[:, :4] * scale
results[:, 5:] = results[:, 5:] * scale_for_landmarks
t1 = time.time()
for _ in range(test_interval):
preds = self.retinaface.predict(photo)
results = self.bbox_util.detection_out(
preds, self.anchors, confidence_threshold=self.confidence)
if len(results) > 0:
results = np.array(results)
#---------------------------------------------------------#
# 如果使用了letterbox_image的话，要把灰条的部分去除掉。
#---------------------------------------------------------#
if self.letterbox_image:
results = retinaface_correct_boxes(
results,
np.array([self.input_shape[0], self.input_shape[1]]),
np.array([im_height, im_width]))
results[:, :4] = results[:, :4] * scale
results[:, 5:] = results[:, 5:] * scale_for_landmarks
t2 = time.time()
tact_time = (t2 - t1) / test_interval
return tact_time
def get_bbox(self, image):
image_shape = np.array(np.shape(image)[0:2])
crop_img = np.array(
letterbox_image(
image, (self.model_image_size[0], self.model_image_size[1])))
photo = np.array(crop_img, dtype=np.float32)
photo /= 255.0
photo = np.transpose(photo, (2, 0, 1))
photo = photo.astype(np.float32)
images = []
images.append(photo)
images = np.asarray(images)
images = torch.from_numpy(images)
if self.cuda:
images = images.cuda()
with torch.no_grad():
outputs = self.net(images)
output_list = []
for i in range(3):
output_list.append(self.yolo_decodes[i](outputs[i]))
output = torch.cat(output_list, 1)
batch_detections = non_max_suppression(
output,
self.config["yolo"]["classes"],
conf_thres=self.confidence,
nms_thres=0.3)
try:
batch_detections = batch_detections[0].cpu().numpy()
except:
return [None]
#print(batch_detections)
top_index = batch_detections[:,
4] * batch_detections[:,
5] > self.confidence
top_conf = batch_detections[top_index, 4] * batch_detections[top_index,
5]
top_label = np.array(batch_detections[top_index, -1], np.int32)
top_bboxes = np.array(batch_detections[top_index, :4])
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
top_bboxes[:, 0],
-1), np.expand_dims(top_bboxes[:, 1], -1), np.expand_dims(
top_bboxes[:, 2], -1), np.expand_dims(top_bboxes[:, 3], -1)
boxes = yolo_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.model_image_size[0], self.model_image_size[1]]),
image_shape)
bboxes = []
for i, c in enumerate(top_label):
top, left, bottom, right = boxes[i]
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(
np.shape(image)[0],
np.floor(bottom + 0.5).astype('int32'))
right = min(
np.shape(image)[1],
np.floor(right + 0.5).astype('int32'))
score = top_conf[i]
box = [left, top, right, bottom, score, c]
bboxes.append(box)
return bboxes
def detect_image(self, image_id, image):
self.confidence = 0.01
self.iou = 0.5
f = open("./input/detection-results/" + image_id + ".txt", "w")
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
#---------------------------------------------------------#
crop_img = np.array(
letterbox_image(image,
(image_sizes[self.phi], image_sizes[self.phi])))
photo = np.array(crop_img, dtype=np.float32)
photo = np.transpose(preprocess_input(photo), (2, 0, 1))
with torch.no_grad():
images = torch.from_numpy(np.asarray([photo]))
if self.cuda:
images = images.cuda()
#---------------------------------------------------------#
# 传入网络当中进行预测
#---------------------------------------------------------#
_, regression, classification, anchors = self.net(images)
#-----------------------------------------------------------#
# 将预测结果进行解码
#-----------------------------------------------------------#
regression = decodebox(regression, anchors, images)
detection = torch.cat([regression, classification], axis=-1)
batch_detections = non_max_suppression(detection,
len(self.class_names),
conf_thres=self.confidence,
nms_thres=self.iou)
#--------------------------------------#
# 如果没有检测到物体，则返回原图
#--------------------------------------#
try:
batch_detections = batch_detections[0].cpu().numpy()
except:
return
#-----------------------------------------------------------#
# 筛选出其中得分高于confidence的框
#-----------------------------------------------------------#
top_index = batch_detections[:, 4] > self.confidence
top_conf = batch_detections[top_index, 4]
top_label = np.array(batch_detections[top_index, -1], np.int32)
top_bboxes = np.array(batch_detections[top_index, :4])
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
top_bboxes[:, 0],
-1), np.expand_dims(top_bboxes[:, 1], -1), np.expand_dims(
top_bboxes[:, 2],
-1), np.expand_dims(top_bboxes[:, 3], -1)
#-----------------------------------------------------------#
# 去掉灰条部分
#-----------------------------------------------------------#
boxes = efficientdet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([image_sizes[self.phi], image_sizes[self.phi]]),
image_shape)
for i, c in enumerate(top_label):
predicted_class = self.class_names[c]
score = str(top_conf[i])
top, left, bottom, right = boxes[i]
f.write("%s %s %s %s %s %s\n" %
(predicted_class, score[:6], str(int(left)), str(
int(top)), str(int(right)), str(int(bottom))))
f.close()
return
def detect_image(self, image_id, image, results):
self.confidence = 0.01
self.nms_threhold = 0.5
image_shape = np.array(np.shape(image)[0:2])
crop_img = letterbox_image(image,
[self.input_shape[0], self.input_shape[1]])
# 将RGB转化成BGR，这是因为原始的centernet_hourglass权值是使用BGR通道的图片训练的
photo = np.array(crop_img, dtype=np.float32)[:, :, ::-1]
# 图片预处理，归一化
photo = np.reshape(
preprocess_image(photo),
[1, self.input_shape[0], self.input_shape[1], self.input_shape[2]])
preds = self.centernet.predict(photo)
if self.nms:
preds = np.array(nms(preds, self.nms_threhold))
if len(preds[0]) <= 0:
return results
preds[0][:, 0:4] = preds[0][:, 0:4] / (self.input_shape[0] / 4)
# 筛选出其中得分高于confidence的框
det_label = preds[0][:, -1]
det_conf = preds[0][:, -2]
det_xmin, det_ymin, det_xmax, det_ymax = preds[0][:, 0], preds[
0][:, 1], preds[0][:, 2], preds[0][:, 3]
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
# 去掉灰条
boxes = centernet_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.input_shape[0], self.input_shape[1]]), image_shape)
for i, c in enumerate(top_label_indices):
result = {}
predicted_class = self.class_names[int(c)]
top, left, bottom, right = boxes[i]
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(image.size[1], np.floor(bottom + 0.5).astype('int32'))
right = min(image.size[0], np.floor(right + 0.5).astype('int32'))
result["image_id"] = int(image_id)
result["category_id"] = clsid2catid[c]
result["bbox"] = [
float(left),
float(top),
float(right - left),
float(bottom - top)
]
result["score"] = float(top_conf[i])
results.append(result)
return results
def detect_image(self, image):
image_shape = np.array(np.shape(image)[0:2])
crop_img, x_offset, y_offset = letterbox_image(
image, [self.model_image_size[0], self.model_image_size[1]])
photo = np.array(crop_img, dtype=np.float64)
# 图片预处理，归一化
photo = preprocess_input(
np.reshape(photo, [
1, self.model_image_size[0], self.model_image_size[1],
self.model_image_size[2]
]))
preds = self.m2det.predict(photo)
# 将预测结果进行解码
results = self.bbox_util.detection_out(
preds, self.prior, confidence_threshold=self.confidence)
if len(results[0]) <= 0:
return image
# 筛选出其中得分高于confidence的框
det_label = results[0][:, 0]
det_conf = results[0][:, 1]
det_xmin, det_ymin, det_xmax, det_ymax = results[0][:, 2], results[
0][:, 3], results[0][:, 4], results[0][:, 5]
top_indices = [
i for i, conf in enumerate(det_conf) if conf >= self.confidence
]
top_conf = det_conf[top_indices]
top_label_indices = det_label[top_indices].tolist()
top_xmin, top_ymin, top_xmax, top_ymax = np.expand_dims(
det_xmin[top_indices],
-1), np.expand_dims(det_ymin[top_indices], -1), np.expand_dims(
det_xmax[top_indices],
-1), np.expand_dims(det_ymax[top_indices], -1)
# 去掉灰条
boxes = m2det_correct_boxes(
top_ymin, top_xmin, top_ymax, top_xmax,
np.array([self.model_image_size[0], self.model_image_size[1]]),
image_shape)
font = ImageFont.truetype(font='model_data/simhei.ttf',
size=np.floor(3e-2 * np.shape(image)[1] +
0.5).astype('int32'))
thickness = (np.shape(image)[0] +
np.shape(image)[1]) // self.model_image_size[0]
for i, c in enumerate(top_label_indices):
predicted_class = self.class_names[int(c - 1)]
score = top_conf[i]
top, left, bottom, right = boxes[i]
top = top - 5
left = left - 5
bottom = bottom + 5
right = right + 5
top = max(0, np.floor(top + 0.5).astype('int32'))
left = max(0, np.floor(left + 0.5).astype('int32'))
bottom = min(
np.shape(image)[0],
np.floor(bottom + 0.5).astype('int32'))
right = min(
np.shape(image)[1],
np.floor(right + 0.5).astype('int32'))
# 画框框
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=self.colors[int(c - 1)])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=self.colors[int(c - 1)])
draw.text(text_origin,
str(label, 'UTF-8'),
fill=(0, 0, 0),
font=font)
del draw
return image
def detect_image(sess, image):
start = timer()
image_shape = (image.size[1], image.size[0], 3)
model_image_size[0] % 32 == 0, 'Multiples of 32 required'
model_image_size[1] % 32 == 0, 'Multiples of 32 required'
boxed_image = letterbox_image(image, tuple(reversed(model_image_size)))
image_data = np.array(boxed_image, dtype='float32')
image_data /= 255.
image_data = np.expand_dims(image_data, 0)
#print(image.size)
#print(image_shape)
#print(image_data.shape)
outs = tf_out(sess, image_data)
out_boxes, out_classes, out_scores = yolo_out(outs, image_shape)
print(model_image_size)
if not out_boxes is None:
print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
#print(out_boxes)
font = ImageFont.truetype(font='font/FiraMono-Medium.otf',
size=np.floor(3e-2 * image.size[1] +
0.5).astype('int32'))
thickness = (image.size[0] + image.size[1]) // 300
for i, c in reversed(list(enumerate(out_classes))):
predicted_class = class_names[c]
box = out_boxes[i]
score = out_scores[i]
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
x, y, w, h = box
top = max(0, np.floor(y + 0.5).astype('int32'))
left = max(0, np.floor(x + 0.5).astype('int32'))
bottom = min(image.size[1], np.floor(y + h + 0.5).astype('int32'))
right = min(image.size[0], np.floor(x + w + 0.5).astype('int32'))
print(label, (left, top), (right, bottom),
((right - left), (bottom - top)))
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
# My kingdom for a good redistributable image drawing library.
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],
outline=colors[c])
draw.rectangle(
[tuple(text_origin),
tuple(text_origin + label_size)],
fill=colors[c])
draw.text(text_origin, label, fill=(0, 0, 0), font=font)
del draw
else:
print('No Boxes')
end = timer()
print(end - start)
return image
