def val(valloader, yolo):
"""Validation wrapper
Args
valloader: (Dataloader) validation data loader
yolo: (nn.Module) YOLOv3 model
"""
class_names = config.datasets[args.dataset]['class_names']
tbar = tqdm(valloader, ncols=80, ascii=True)
tbar.set_description("evaluation")
for batch_idx, (names, inputs, targets) in enumerate(tbar):
inputs = inputs.cuda()
detections = yolo(inputs)
for idx, name in enumerate(names):
img_path = opj(config.datasets[args.dataset]['val_imgs'], name)
img_name = img_path.split('/')[-1]
try:
detection = detections[detections[:, 0] == 0]
except Exception:
detection = torch.Tensor([])
save_path = opj(config.evaluate['result_dir'], img_name)
draw_detection(img_path, detection, yolo.reso, class_names,
save_path)
def predict():
input_size = (416, 416)
tf.app.flags.DEFINE_string('image_file', './images/person.jpg', 'image_path')
FLAGS = tf.app.flags.FLAGS
image_file = FLAGS.image_file
image = cv2.imread(image_file)
image_shape = image.shape[:2]
image_cp = preprocess_image(image, input_size)
images = tf.placeholder(tf.float32, [1, input_size[0], input_size[1], 3])
detection_feat = darknet(images)
feat_sizes = input_size[0] // 32, input_size[1] // 32
detection_results = decode(detection_feat, feat_sizes, len(class_names), anchors)
checkpoint_path = "./checkpoint_dir/yolo2_coco.ckpt"
#checkpoint_path = "/Users/xiang/Downloads/DeepLearning_tutorials-master/ObjectDetections/yolo2/checkpoint_dir/yolo2_coco.ckpt"
saver = tf.train.Saver()
with tf.Session() as sess:
saver.restore(sess, checkpoint_path)
bboxes, obj_probs, class_probs = sess.run(detection_results, feed_dict={images: image_cp})
bboxes, scores, class_inds = postprocess(bboxes, obj_probs, class_probs,
image_shape=image_shape)
img_detection = draw_detection(image, bboxes, scores, class_inds, class_names)
cv2.imwrite("./res/detection.jpg", img_detection)
cv2.imshow("detection results", img_detection)
print('*****Click the window,and press any key to close!*****')
cv2.waitKey(0)
cv2.destroyAllWindows()
def main():
input_size = (416,416)
image_file = './yolo2_data/car.jpg'
image = cv2.imread(image_file)
image_shape = image.shape[:2] #只取wh，channel=3不取
# copy、resize416*416、归一化、在第0维增加存放batchsize维度
image_cp = preprocess_image(image,input_size)
# 【1】输入图片进入darknet19网络得到特征图，并进行解码得到：xmin xmax表示的边界框、置信度、类别概率
tf_image = tf.placeholder(tf.float32,[1,input_size[0],input_size[1],3])
model_output = darknet_slim(tf_image) # darknet19网络输出的特征图
output_sizes = input_size[0]//32, input_size[1]//32 # 特征图尺寸是图片下采样32倍
output_decoded = decode(model_output=model_output,output_sizes=output_sizes,
num_class=len(class_names),anchors=anchors) # 解码
model_path = "models/model.ckpt"
init_fn = slim.assign_from_checkpoint_fn(model_path,slim.get_variables())
with tf.Session() as sess:
init_fn(sess)
bboxes,obj_probs,class_probs = sess.run(output_decoded,feed_dict={tf_image:image_cp})
# 【2】筛选解码后的回归边界框——NMS(post process后期处理)
bboxes,scores,class_max_index = postprocess(bboxes,obj_probs,class_probs,image_shape=image_shape)
# 【3】绘制筛选后的边界框
img_detection = draw_detection(image, bboxes, scores, class_max_index, class_names)
cv2.imwrite("./yolo2_data/detection.jpg", img_detection)
print('YOLO_v2 detection has done!')
cv2.imshow("detection_results", img_detection)
cv2.waitKey(0)
def val(valloader, yolo, save_img=True):
"""Validation wrapper
@Args
valloader: (Dataloader) validation data loader
yolo: (nn.Module) YOLOv3 model
save_img: (bool) whether to save images during validation
"""
mAPs = []
tbar = tqdm(valloader, ncols=80)
for batch_idx, (names, inputs, targets) in enumerate(tbar):
inputs = inputs.cuda()
detections = yolo(inputs)
mAP_batch = mAP(detections, targets, args.reso)
mAPs += mAP_batch
tbar.set_description("mAP=%.2f" % (np.mean(mAPs) * 100))
if save_img == True and batch_idx % 4 == 0:
img_path = opj(config.datasets[args.dataset]['val_imgs'], names[0])
img_name = img_path.split('/')[-1]
try:
detection = detections[detections[:, 0] == 0]
except Exception:
img = Image.open(img_path)
else:
img = draw_detection(img_path, detection, yolo.reso, type='pred')
img.save(opj(config.evaluate['result_dir'], img_name))
return mAPs
def main():
input_size = (416, 416)
# image_file = "/home/zdq/darknet/data/1.jpg"
# image = cv2.imread(image_file)
image_shape = image.shape[:2]
image_cp = preprocess_image(image, input_size)
images = tf.placeholder(tf.float32, [1, input_size[0], input_size[1], 3])
detection_feat = darknet(images)
feat_sizes = input_size[0] // 32, input_size[1] // 32
detection_results = decode(detection_feat, feat_sizes, len(class_names),
anchors)
checkpoint_path = "/home/zdq/YOLO/checkpoint_dir/yolo2_coco.ckpt"
saver = tf.train.Saver()
with tf.Session() as sess:
saver.restore(sess, checkpoint_path)
bboxes, obj_probs, class_probs = sess.run(detection_results,
feed_dict={images: image_cp})
bboxes, scores, class_inds = postprocess(bboxes,
obj_probs,
class_probs,
image_shape=image_shape)
img_detection = draw_detection(image, bboxes, scores, class_inds,
class_names) #回归框，得到矩阵框的X左右，Y下像素坐标
print('\n')
# #检测车道线的像素，并放入字典中
# lane_cor = {}
# #手动选取ROI，待修改
# vertices = np.array([[(110, 194), (110, 0), (150, 0), (150, 194)]], dtype=np.int32)
# roi = region_of_interest(line_img, vertices)
# # cv2.imshow("roi", roi)
# for i in range(0, (roi.shape)[0]):
# for j in range(0, (roi.shape)[1]):
# if roi[i, j, 2] == 255: #roi[i,j,num]这里num代表着BGR第几通道
# lane_cor[i] = j
# print("The coodinate of the detected_lane y：x")
#
# print(lane_cor)
#
# global box
# if (utils.box[0] + m * (utils.box[2] - utils.box[0])) <= lane_cor[utils.box[3]] <= (utils.box[2] - m * (utils.box[2] - utils.box[0])):
# print("The car is on the solid line!!!")
# else:
# print("The car is permitted~")
# mix1 = weight_add(img_detection, line_img, alpha=0.7, belta=1, gamma=0.)
# mixed = weight_add(img_detection,roi , alpha=0.7, belta=1, gamma=0.)
# cv2.imshow("mix1", mix1)
# cv2.imshow("mixed",mixed)
#
cv2.imshow("detection results", img_detection)
cv2.imwrite("/home/zdq/PycharmProjects/YOLOv2/detection.jpg",
img_detection)
cv2.waitKey(0)
return img_detection
def camera_detect():
tf.app.flags.DEFINE_string('video', False, 'Whether to output video file')
FLAGS = tf.app.flags.FLAGS
input_size = (416, 416)
cv2.namedWindow("camera")
capture = cv2.VideoCapture(0) #开启摄像头
success, image = capture.read()
images = tf.placeholder(tf.float32, [1, input_size[0], input_size[1], 3])
detection_feat = darknet(images)
if FLAGS.video:
fps = 1
size = (int(capture.get(cv2.CAP_PROP_FRAME_WIDTH)),
int(capture.get(cv2.CAP_PROP_FRAME_HEIGHT)))
videoWriter = cv2.VideoWriter('./test/result.avi',
cv2.VideoWriter_fourcc('M','J', 'P', 'G'), fps, size)
num = 1
while success and cv2.waitKey(1) == -1:
cv2.imshow('camera', image)
image_shape = image.shape[:2]
image_cp = preprocess_image(image, input_size)
feat_sizes = input_size[0] // 32, input_size[1] // 32
detection_results = decode(detection_feat, feat_sizes, len(class_names), anchors)
checkpoint_path = "./checkpoint_dir/yolo2_coco.ckpt"
#checkpoint_path = "/Users/xiang/Downloads/DeepLearning_tutorials-master/ObjectDetections/yolo2/checkpoint_dir/yolo2_coco.ckpt"
saver = tf.train.Saver()
with tf.Session() as sess:
saver.restore(sess, checkpoint_path)
bboxes, obj_probs, class_probs = sess.run(detection_results, feed_dict={images: image_cp})
bboxes, scores, class_inds = postprocess(bboxes, obj_probs, class_probs,
image_shape=image_shape)
img_detection = draw_detection(image, bboxes, scores, class_inds, class_names)
if FLAGS.video:
videoWriter.write(img_detection)
else:
cv2.imwrite("./test/"+str(num)+"test.jpg", img_detection)
success, image = capture.read()
num += 1
cv2.destroyWindow("camera")
capture.release()
def main():
anchors = cfg.ANCHORS
class_names = cfg.COCONAME
input_size = (416,416)
image_file = './yolo2_data/train.jpg'
tf_image = tf.placeholder(tf.float32,[1,input_size[0],input_size[1],3])
net = YOLONet(False)
output_sizes = input_size[0]//32, input_size[1]//32 # 特征图尺寸是图片下采样32倍
output_decoded = decode(model_output=net.logits,output_sizes=output_sizes,
num_class=len(class_names),anchors=anchors) # 解码
model_path = "./model.ckpt"
saver = tf.train.Saver()
cap = cv2.VideoCapture('./video.mp4')
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
#saver.restore(sess,model_path)
#bboxes,obj_probs,class_probs = sess.run(output_decoded,feed_dict={tf_image:image_cp})
ret, _ = cap.read()
while ret:
ret, frame = cap.read()
image = frame
image_shape = image.shape[:2] #只取wh，channel=3不取
# copy、resize416*416、归一化、在第0维增加存放batchsize维度
image_cp = preprocess_image(image,input_size)
bboxes,obj_probs,class_probs = sess.run(output_decoded,feed_dict={net.images: image_cp})
#img1 = sess.run(net.img1,feed_dict={net.images: image_cp})
bboxes,scores,class_max_index = postprocess(bboxes,obj_probs,class_probs,image_shape=image_shape)
print(scores)
print(class_max_index)
img_detection = draw_detection(image, bboxes, scores, class_max_index, class_names)
#cv2.imwrite("./yolo2_data/detection.jpg", img_detection)
#print('YOLO_v2 detection has done!')
cv2.imshow("detection_results", img_detection)
cv2.waitKey(10)
ret, frame = cap.read()
def main():
s = time.time()
input_size = (416, 416)
image_file = '/Users/aaron/Study/机器学习/深度学习/YOLOv3/darknet/data/dog.jpg'
image = cv2.imread(image_file)
image_shape = image.shape[:2] # 只取wh，channel=3不取
# copy、resize416*416、归一化、在第0维增加存放batchsize维度
image_cp = preprocess_image(image, input_size)
# 【1】输入图片进入darknet19网络得到特征图，并进行解码得到：xmin xmax表示的边界框、置信度、类别概率
tf_image = tf.placeholder(tf.float32, [1, input_size[0], input_size[1], 3])
model_output = darknet(tf_image) # darknet19网络输出的特征图
output_sizes = input_size[0] // 32, input_size[1] // 32 # 特征图尺寸是图片下采样32倍
output_decoded = decode(model_output=model_output,
output_sizes=output_sizes,
num_class=len(class_names),
anchors=anchors) # 解码
model_path = "./yolo2_model/yolo2_coco.ckpt"
saver = tf.train.Saver()
with tf.Session() as sess:
saver.restore(sess, model_path)
bboxes, obj_probs, class_probs = sess.run(
output_decoded, feed_dict={tf_image: image_cp})
# 【2】筛选解码后的回归边界框——NMS(post process后期处理)
bboxes, scores, class_max_index = postprocess(bboxes,
obj_probs,
class_probs,
image_shape=image_shape)
# 【3】绘制筛选后的边界框
img_detection = draw_detection(image, bboxes, scores, class_max_index,
class_names)
cv2.imwrite("./yolo2_data/detection.jpg", img_detection)
print('YOLO_v2 detection has done!')
e = time.time()
print(e - s)
cv2.imshow("detection_results", img_detection)
cv2.waitKey(0)
def main():
anchors = cfg.ANCHORS
class_names = cfg.COCONAME
input_size = (416, 416)
image_file = './yolo2_data/car.jpg'
image = cv2.imread(image_file)
image_shape = image.shape[:2]
tf_image = tf.placeholder(tf.float32, [1, input_size[0], input_size[1], 3])
net = YOLONet(False)
output_sizes = input_size[0] // 32, input_size[1] // 32
output_decoded = decode(model_output=net.logits,
output_sizes=output_sizes,
num_class=len(class_names),
anchors=anchors)
image_cp = preprocess_image(image, input_size)
model_path = "./model.ckpt"
saver = tf.train.Saver()
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
saver.restore(sess, model_path)
#bboxes,obj_probs,class_probs = sess.run(output_decoded,feed_dict={tf_image:image_cp})
bboxes, obj_probs, class_probs = sess.run(
output_decoded, feed_dict={net.images: image_cp})
#img1= sess.run([net.img1],feed_dict={net.images: image_cp})
#print(img1)
#print(img2)
#print(obj_probs)
#print(sess.run(tf.reduce_max(obj_probs),feed_dict={net.images: image_cp}))
bboxes, scores, class_max_index = postprocess(bboxes,
obj_probs,
class_probs,
image_shape=image_shape)
print(scores)
print(class_max_index)
img_detection = draw_detection(image, bboxes, scores, class_max_index,
class_names)
#cv2.imwrite("./yolo2_data/detection.jpg", img_detection)
#print('YOLO_v2 detection has done!')
cv2.imshow("detection_results", img_detection)
cv2.waitKey(-1)
image_cp = np.expand_dims(image_cp, 0)
#print(image_cp)
"""
images = tf.placeholder(tf.float32, [1, input_size[0], input_size[1], 3])
detection_feat = darknet(images)
feat_sizes = input_size[0] // 32, input_size[1] // 32
start = time.clock()
detection_results = decode(detection_feat, feat_sizes, len(class_names),
anchors)
checkpoint_path = "./checkpoint_dir/yolo2_coco.ckpt"
saver = tf.train.Saver()
with tf.Session() as sess:
saver.restore(sess, checkpoint_path)
bboxes, obj_probs, class_probs = sess.run(detection_results,
feed_dict={images: image_cp})
bboxes, scores, class_inds = postprocess(bboxes,
obj_probs,
class_probs,
image_shape=image_shape)
#plt_bboxes(image,class_inds,scores,bboxes)
img_detection = draw_detection(image, bboxes, scores, class_inds, class_names)
end = time.clock()
cv2.imwrite("detection.jpg", img_detection)
cv2.startWindowThread()
cv2.imshow("detection results", img_detection)
print(end - start)
cv2.waitKey(0) # waitkey代表读取键盘的输入，括号里的数字代表等待多长时间，单位ms。 0代表一直等待
"""
image = Image.open(image_file)
image_cp = image.resize(input_size, Image.BICUBIC)
image_cp = np.array(image_cp, dtype=np.float32)/255.0
image_cp = np.expand_dims(image_cp, 0)
#print(image_cp)
"""
images = tf.placeholder(tf.float32, [1, input_size[0], input_size[1], 3])
detection_feat = darknet(images)
feat_sizes = input_size[0] // 32, input_size[1] // 32
detection_results = decode(detection_feat, feat_sizes, len(class_names), anchors)
checkpoint_path = "./checkpoint_dir/yolo2_coco.ckpt"
saver = tf.train.Saver()
with tf.Session() as sess:
saver.restore(sess, checkpoint_path)
bboxes, obj_probs, class_probs = sess.run(detection_results, feed_dict={images: image_cp})
bboxes, scores, class_inds = postprocess(bboxes, obj_probs, class_probs,
image_shape=image_shape)
img_detection = draw_detection(image, bboxes, scores, class_inds, class_names)
cv2.imwrite("detection.jpg", img_detection)
cv2.imshow("detection results", img_detection)
cv2.waitKey(0)
image_cp = preprocess_image(image) #图像预处理，resize image, normalization归一化， 增加一个在第0的维度--batch_size
tf_image = tf.placeholder(tf.float32,[1,input_size[0],input_size[1],3]) #定义placeholder
model_output = darknet(tf_image) #网络的输出
output_sizes = input_size[0]//32, input_size[1]//32 # 特征图尺寸是图片下采样32倍
#这个函数返回框的坐标（左上角-右下角），目标置信度，类别置信度
output_decoded = decode(model_output=model_output,output_sizes=output_sizes, num_class=len(class_names),anchors=anchors)
with tf.Session() as sess:
sess.run(tf.global_variables_initializer()) #初始化tensorflow全局变量
saver = tf.train.Saver()
saver.restore(sess, model_path) #把模型加载到当前session中
bboxes, obj_probs, class_probs = sess.run(output_decoded, feed_dict={tf_image: image_cp}) #这个函数返回框的坐标，目标置信度，类别置信度
bboxes,scores,class_max_index = postprocess(bboxes,obj_probs,class_probs,image_shape=image_shape) #得到候选框之后的处理，先留下阈值大于0.5的框，然后再放入非极大值抑制中去
colors = generate_colors(class_names)
img_detection = draw_detection(image, bboxes, scores, class_max_index, class_names, colors) #得到图片
cv2.imwrite(image_detection, img_detection)
cv2.imshow("detection_results", img_detection) #显示图片
cv2.waitKey(0) #等待按任意键退出
print(colored("\n==>", 'blue'), emojify("Prepare data ... :coffee:\n"))
img_datasets, dataloader = prepare_demo_dataset(config.demo['images_dir'], args.reso)
print("Number of demo images:", len(img_datasets))
print(colored("\n==>", 'blue'), emojify("Loading network ... :hourglass:\n"))
yolo = YOLOv3(cfg, args.reso).cuda()
start_epoch, start_iteration = args.checkpoint.split('.')
start_epoch, start_iteration, state_dict = load_checkpoint(
opj(config.CKPT_ROOT, args.dataset),
int(start_epoch),
int(start_iteration)
)
yolo.load_state_dict(state_dict)
print("Model starts training from epoch %d iteration %d" % (start_epoch, start_iteration))
print(colored("\n==>", 'blue'), emojify("Evaluation ...\n"))
yolo.eval()
for batch_idx, (img_path, inputs) in enumerate(tqdm(dataloader, ncols=80)):
inputs = inputs.cuda()
detections = yolo(inputs)
# take idx 0
detections = detections[detections[:, 0] == 0]
path = img_path[0]
img_name = path.split('/')[-1]
img = draw_detection(path, detections.data, args.reso, type='pred')
img.save(opj(config.demo['result_dir'], img_name))
print(emojify("Done! :+1:\n"))
