def pred_by_drawing(img_path, save_path):
img = cv2.imread(img_path)
*_, img_name = os.path.split(img_path)
rclasses, rscores, rbboxes = process_image(img)
visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes,
visualization.colors_plasma)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
def main():
# Test on some demo image and visualize output.
# detect from image
path = 'demo/'
image_names = sorted(os.listdir(path))
for img_name in image_names:
img = mpimg.imread(path + img_name)
rclasses, rscores, rbboxes = process_image(img)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
def print_ssd_result(img_path):
"""打印ssd 检测结果"""
img = mpimg.imread(img_path)
ckpt_filename = r'/Users/zhuangshui/PycharmProjects/github/others/SSD/checkpoint/ssd_300_vgg.ckpt'
model_params = get_ssd_model_params(ckpt_filename)
rclasses, rscores, rbboxes = process_image(img, model_params=model_params)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
def main():
# TensorFlow session: grow memory when needed. TF, DO NOT USE ALL MY GPU MEMORY!!!
gpu_options = tf.GPUOptions(allow_growth=True)
config = tf.ConfigProto(log_device_placement=False,
gpu_options=gpu_options)
sess = tf.Session(config=config)
# isess = tf.InteractiveSession(config=config)
# Input placeholder.
net_shape = (300, 300)
data_format = 'NHWC'
img_input = tf.placeholder(tf.uint8, shape=(None, None, 3))
# Evaluation pre-processing: resize to SSD net shape.
image_pre, labels_pre, bboxes_pre, bbox_img = ssd_vgg_preprocessing.preprocess_for_eval(
img_input,
None,
None,
net_shape,
data_format,
resize=ssd_vgg_preprocessing.Resize.WARP_RESIZE)
image_4d = tf.expand_dims(image_pre, 0)
# Define the SSD model.
reuse = True if 'ssd_net' in locals() else None
ssd_net = ssd_vgg_300.SSDNet()
with slim.arg_scope(ssd_net.arg_scope(data_format=data_format)):
predictions, localisations, _, _ = ssd_net.net(image_4d,
is_training=False,
reuse=reuse)
# Restore SSD model.
ckpt_filename = '../checkpoints/ssd_300_vgg.ckpt'
# ckpt_filename = '../checkpoints/VGG_VOC0712_SSD_300x300_ft_iter_120000.ckpt'
sess.run(tf.global_variables_initializer())
saver = tf.train.Saver()
saver.restore(sess, ckpt_filename)
# SSD default anchor boxes.
ssd_anchors = ssd_net.anchors(net_shape)
# Test on some demo image and visualize output.
path = '../demo/'
image_names = sorted(os.listdir(path))
img = mpimg.imread(path + image_names[-1])
rclasses, rscores, rbboxes = process_image(sess, img, image_4d,
predictions, localisations,
bbox_img, img_input,
ssd_anchors)
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
def run():
cap = get_video()
while(True):
# Capture frame-by-frame
img = get_video()
# Our operations on the frame come here
# gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
rclasses, rscores, rbboxes = process_image(img)
fig = visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
frame = fig2data(fig)
# Display the resulting frame
cv2.imshow('frame',frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
# When everything done, release the capture
cap.release()
cv2.destroyAllWindows()
exit(0)
while KEEP_RUNNING:
# cv2.imshow('Depth', get_depth())
img = get_video()
rclasses, rscores, rbboxes = process_image(img)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
# cv2.imshow('Video', get_video())
# if cv2.waitKey(10) == 27:
# break
# print('Press ESC in window to stop')
# freenect.runloop(depth=display_depth,
# video=display_rgb,
# body=body)
return
rclasses, rscores, rbboxes = np_methods.ssd_bboxes_select(
rpredictions, rlocalisations, ssd_anchors,
select_threshold=select_threshold, img_shape=net_shape, num_classes=21, decode=True)
rbboxes = np_methods.bboxes_clip(rbbox_img, rbboxes)
rclasses, rscores, rbboxes = np_methods.bboxes_sort(rclasses, rscores, rbboxes, top_k=400)
rclasses, rscores, rbboxes = np_methods.bboxes_nms(rclasses, rscores, rbboxes, nms_threshold=nms_threshold)
# Resize bboxes to original image shape. Note: useless for Resize.WARP!
rbboxes = np_methods.bboxes_resize(rbbox_img, rbboxes)
return rclasses, rscores, rbboxes
# Test on some demo image and visualize output.
path = '../VOC2007/demo/'
image_names = sorted(os.listdir(path))
'''
print(image_names[-5])
img = mpimg.imread(path + image_names[1])
rclasses, rscores, rbboxes = process_image(img)
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
'''
for i in range(len(image_names)):
img = Image.open(path + image_names[i])
out = img.convert("RGB")
imgg=np.array(out)
rclasses, rscores, rbboxes = process_image(imgg)
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
visualization.plt_bboxes(imgg, rclasses, rscores, rbboxes)
def preprocess_img(img, shape=(300, 300, 3)):
# VGG mean parameters.
_R_MEAN = 123.
_G_MEAN = 117.
_B_MEAN = 104.
means = [_R_MEAN, _G_MEAN, _B_MEAN]
img = misc.imresize(img, shape, interp='bilinear')
return img - means
mtime.start()
img = misc.imread('dog.jpg')
img2 = misc.imread('person.jpg')
p_img = preprocess_img(img, shape=(300, 300, 3))
p_img2 = preprocess_img(img2, shape=(300, 300, 3))
mtime.consume('preprocess_pic')
bat_p_img = [p_img, p_img2]
bat_normal_img = [img, img2]
bat_rclasses, bat_rscores, bat_rbboxes = process_image(bat_p_img)
for i in range(batch_size):
visualization.plt_bboxes(bat_normal_img[i], bat_rclasses[i],
bat_rscores[i], bat_rbboxes[i])
### 预处理和后续处理已分开，可得到与源代码一样的结果
### 下一步，多进程预处理
def inference(input=0,inputType=1):
slim = tf.contrib.slim
sys.path.append('../')
from nets import ssd_vgg_300, ssd_common, np_methods
from preprocessing import ssd_vgg_preprocessing
from notebooks import visualization
# TensorFlow session: grow memory when needed. TF, DO NOT USE ALL MY GPU MEMORY!!!
gpu_options = tf.GPUOptions(allow_growth=True)
config = tf.ConfigProto(log_device_placement=False, gpu_options=gpu_options)
isess = tf.InteractiveSession(config=config)
# Input placeholder.
net_shape = (300, 300)
data_format = 'NHWC'
img_input = tf.placeholder(tf.uint8, shape=(None, None, 3))
# Evaluation pre-processing: resize to SSD net shape.
image_pre, labels_pre, bboxes_pre, bbox_img = ssd_vgg_preprocessing.preprocess_for_eval(
img_input, None, None, net_shape, data_format, resize=ssd_vgg_preprocessing.Resize.WARP_RESIZE)
image_4d = tf.expand_dims(image_pre, 0)
# Define the SSD model.
reuse = True if 'ssd_net' in locals() else None
ssd_net = ssd_vgg_300.SSDNet()
with slim.arg_scope(ssd_net.arg_scope(data_format=data_format)):
predictions, localisations, _, _ = ssd_net.net(image_4d, is_training=False, reuse=reuse)
# Restore SSD model.
ckpt_filename = '../checkpoints/ssd_300_vgg.ckpt'
# ckpt_filename = '../checkpoints/VGG_VOC0712_SSD_300x300_ft_iter_120000.ckpt'
isess.run(tf.global_variables_initializer())
saver = tf.train.Saver()
saver.restore(isess, ckpt_filename)
# SSD default anchor boxes.
ssd_anchors = ssd_net.anchors(net_shape)
# Main image processing routine.
def process_image(img, select_threshold=0.5, nms_threshold=.45, net_shape=(300, 300)):
# Run SSD network.
rimg, rpredictions, rlocalisations, rbbox_img = isess.run([image_4d, predictions, localisations, bbox_img],
feed_dict={img_input: img})
# Get classes and bboxes from the net outputs.
rclasses, rscores, rbboxes = np_methods.ssd_bboxes_select(
rpredictions, rlocalisations, ssd_anchors,
select_threshold=select_threshold, img_shape=net_shape, num_classes=21, decode=True)
rbboxes = np_methods.bboxes_clip(rbbox_img, rbboxes)
rclasses, rscores, rbboxes = np_methods.bboxes_sort(rclasses, rscores, rbboxes, top_k=400)
rclasses, rscores, rbboxes = np_methods.bboxes_nms(rclasses, rscores, rbboxes, nms_threshold=nms_threshold)
# Resize bboxes to original image shape. Note: useless for Resize.WARP!
rbboxes = np_methods.bboxes_resize(rbbox_img, rbboxes)
return rclasses, rscores, rbboxes
# input is a image
inputType = int(inputType)
if inputType is 1:
if input == 0:
print("At least indicate 1 input video")
exit(-1)
# Test on some demo image and visualize output.
img = mpimg.imread(input)
rclasses, rscores, rbboxes = process_image(img)
# Find the name of the category num
print(list(map(lambda i:"{}:{}".format(i,category[i]),list(rclasses))))
rclasses = np.array(list(map(lambda i:"{}:{}".format(i,category[i]),list(rclasses))))
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
# plot the image directly
visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
elif inputType == 2:
# input is the video
# plot the boxes into the image
cap = cv2.VideoCapture(input)
fps = cap.get(cv2.CAP_PROP_FPS)
size = (int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)), int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)))
fourcc = cap.get(cv2.CAP_PROP_FOURCC)
#fourcc = cv2.CAP_PROP_FOURCC(*'CVID')
print('fps=%d,size=%r,fourcc=%r'%(fps,size,fourcc))
delay=10/int(fps)
print(delay)
if delay <= 1:
delay = 1
while (cap.isOpened()):
ret, frame = cap.read()
print(ret)
if ret == True:
image = frame
# the array based representation of the image will be used later in order to prepare the
# result image with boxes and labels on it.
image_np = image
# Expand dimensions since the model expects images to have shape: [1, None, None, 3]
image_np_expanded = np.expand_dims(image_np, axis=0)
# Actual detection.
rclasses, rscores, rbboxes = process_image(image_np)
#print(list(map(lambda i: "{}:{}".format(i, category[i]), list(rclasses))))
rclasses = np.array(list(map(lambda i: "{}:{}".format(i, category[i]), list(rclasses))))
# Visualization of the results of a detection.
visualization.bboxes_draw_on_img(image_np, rclasses, rscores, rbboxes)
cv2.imshow('frame', image_np)
#cv2.waitKey(np.uint(delay))
if cv2.waitKey(delay) & 0xFF == ord('q'):
break
print('Ongoing...')
else:
break
cap.release()
cv2.destroyAllWindows()
elif inputType ==3:
print("save video")
if input == 0:
print("At least indicate 1 input video")
exit(-1)
def save_image(image_np):
rclasses, rscores, rbboxes = process_image(image_np)
# print(list(map(lambda i: "{}:{}".format(i, category[i]), list(rclasses))))
rclasses = np.array(list(map(lambda i: "{}:{}".format(i, category[i]), list(rclasses))))
visualization.bboxes_draw_on_img(image_np, rclasses, rscores, rbboxes)
return image_np
from moviepy.editor import VideoFileClip
cap = cv2.VideoCapture(input)
fps = cap.get(cv2.CAP_PROP_FPS)
cap.release()
cv2.destroyAllWindows()
video = VideoFileClip(input)
result = video.fl_image(save_image)
output = os.path.join("./videos/output_{}".format(input.split("/")[-1]))
result.write_videofile(output, fps=fps)
else:
cap = cv2.VideoCapture(0)
while (True):
# Capture frame-by-frame
ret, frame = cap.read()
#cv2.imshow('frame', frame)
# Expand dimensions since the model expects images to have shape: [1, None, None, 3]
image_np_expanded = np.expand_dims(frame, axis=0)
# Actual detection.
rclasses, rscores, rbboxes = process_image(frame)
# print(list(map(lambda i: "{}:{}".format(i, category[i]), list(rclasses))))
rclasses = np.array(list(map(lambda i: "{}:{}".format(i, category[i]), list(rclasses))))
# Visualization of the results of a detection.
visualization.bboxes_draw_on_img(frame, rclasses, rscores, rbboxes)
cv2.imshow('frame', frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
# When everything done, release the capture
cap.release()
cv2.destroyAllWindows()
if cls_id >= 0:
score = scores[i]
print(score)
if cls_id not in colors :
colors[cls_id] = (random.randint(0,255), random.randint(0,255), random.randint(0,255))
print(colors[cls_id])
ymin = int(bboxes[i, 0] * height)
xmin = int(bboxes[i, 1] * width)
ymax = int(bboxes[i, 2] * height)
xmax = int(bboxes[i, 3] * width)
cv2.rectangle(img_painted, (xmin,ymin), (xmax, ymax), colors[cls_id], 1)
class_name = str(cls_id)
cv2.putText(img_painted, '{:s} | {:.3f}'.format(class_name, score), (xmin,ymin), cv2.FONT_HERSHEY_PLAIN, 1.5, colors[cls_id], 2, 1)
return img_painted
'''
# Test on some demo image and visualize output.
# 测试的文件夹
path = '../demo/'
image_names = sorted(os.listdir(path))
# 文件夹中的第几张图，-1代表最后一张
img = mpimg.imread(path + image_names[-2])
rclasses, rscores, rbboxes = process_image(img)
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
'''
'''
class A():
aa = ""
def bndbox_image(image):
img = mpimg.imread(path + image)
rclasses, rscores, rbboxes = process_image(img)
plt_obj = visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
plt_obj.savefig(output_dir + image)
rclasses, rscores, rbboxes = np_methods.bboxes_nms(
rclasses, rscores, rbboxes, nms_threshold=nms_threshold)
# Resize bboxes to original image shape. Note: useless for Resize.WARP!
rbboxes = np_methods.bboxes_resize(rbbox_img, rbboxes)
return rclasses, rscores, rbboxes
dataset_dir = os.path.join(os.getcwd(), 'datasets')
labels_to_names = None
if dataset_utils.has_labels(dataset_dir):
labels_to_names = dataset_utils.read_label_file2(dataset_dir,
filename='labels_voc.txt')
# Test on some demo image and visualize output.
path = '/home/ace19/dl_data/MOT/MOT17/test/sample-test/'
# path = './detection_image/coco/'
image_names = sorted(os.listdir(path))
for image in image_names:
img = mpimg.imread(path + image)
# img = mpimg.imread(path + image_names[-5])
start_time = time.time()
rclasses, rscores, rbboxes = process_image(img)
print('Speed %.3f sec' % (time.time() - start_time))
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes, labels_to_names)
rclasses, rscores, rbboxes = np_methods.ssd_bboxes_select(
rpredictions,
rlocalisations,
ssd_anchors,
select_threshold=select_threshold,
img_shape=net_shape,
num_classes=21,
decode=True)
rbboxes = np_methods.bboxes_clip(rbbox_img, rbboxes)
rclasses, rscores, rbboxes = np_methods.bboxes_sort(rclasses,
rscores,
rbboxes,
top_k=400)
rclasses, rscores, rbboxes = np_methods.bboxes_nms(
rclasses, rscores, rbboxes, nms_threshold=nms_threshold)
# Resize bboxes to original image shape. Note: useless for Resize.WARP!
rbboxes = np_methods.bboxes_resize(rbbox_img, rbboxes)
return rclasses, rscores, rbboxes
# Test on some demo image and visualize output.
path = './demo/'
image_names = sorted(os.listdir(path))
img = mpimg.imread(path + image_names[-5])
rclasses, rscores, rbboxes = process_image(img)
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes, name='pp')
rclasses, rscores, rbboxes, nms_threshold=nms_threshold)
# 将box的坐标重新映射到原图上（上文所有的坐标都进行了归一化，所以要逆操作一次）
rbboxes = np_methods.bboxes_resize(rbbox_img, rbboxes)
if case == 1:
bboxes_draw_on_img(img,
rclasses,
rscores,
rbboxes,
colors_plasma,
thickness=8)
return img
else:
return rclasses, rscores, rbboxes
# 做目标定位，同时做预测分析
case = 2
path = '../demo/12.jpg'
# 读取图片
img = mpimg.imread(path)
# 执行主流程函数
rclasses, rscores, rbboxes = process_image(img, case)
#isualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
# 显示分类结果图
visualization.plt_bboxes(img,
rclasses,
rscores,
rbboxes,
figsize=(10, 10),
linewidth=2)
rbboxes = np_methods.bboxes_clip(rbbox_img, rbboxes)
rclasses, rscores, rbboxes = np_methods.bboxes_sort(rclasses,
rscores,
rbboxes,
top_k=400)
rclasses, rscores, rbboxes = np_methods.bboxes_nms(
rclasses, rscores, rbboxes, nms_threshold=nms_threshold)
# Resize bboxes to original image shape. Note: useless for Resize.WARP!
rbboxes = np_methods.bboxes_resize(rbbox_img, rbboxes)
return rclasses, rscores, rbboxes
print('loading frame')
frame = cv2.imread('E:\\challenge2_val\\test\\01\\rgb\\0001.png')
print(frame.shape)
print('getting boxes')
rclasses, rscores, rbboxes = process_image(frame,
select_threshold=0.6,
nms_threshold=.9,
net_shape=(100, 100))
print('drawing boxes')
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
img = visualization.plt_bboxes(frame, rclasses, rscores, rbboxes)
print(img.shape)
# Our operations on the frame come here
#color = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
# Display the resulting frame
cv2.imshow('frame', img)
cv2.waitKey(0)
top_k=400)
# 去重，将重复检测到的目标去掉
rclasses, rscores, rbboxes = np_methods.bboxes_nms(
rclasses, rscores, rbboxes, nms_threshold=nms_threshold)
# 将box的坐标重新映射到原图上（上文所有的坐标都进行了归一化，所以要逆操作一次）
rbboxes = np_methods.bboxes_resize(rbbox_img, rbboxes)
if case is not None:
bboxes_draw_on_img(img,
rclasses,
rscores,
rbboxes,
colors_plasma,
thickness=8)
return img
else:
return rclasses, rscores, rbboxes
if __name__ == '__main__':
# 做目标定位，同时做预测分析
path = 'd:/input_data/threepeople.jpg'
# 读取图片
img = mpimg.imread(path)
# 执行主流程函数
rclasses, rscores, rbboxes = process_image(img)
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes,colors_plasma)
# 显示分类结果图
visualization.plt_bboxes(img, rclasses, rscores, rbboxes, label_to_class)
