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 = "/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 creat_model(classes,
load_pretrained=True,
freeze_body=2,
weights_path='imagenet_models/darknet53.h5'):
K.clear_session()
image_input = Input(shape=(None, None, 3))
darknet53 = darknet_body(image_input)
if load_pretrained:
darknet53.load_weights(weights_path, by_name=True, skip_mismatch=True)
print('load_weights succeed!!!')
darknet_model = darknet(darknet53, classes)
model = Model(image_input, darknet_model.output)
return model
tf.reset_default_graph() #重设图
input_size = (416, 416)
image_file = "car.jpg"
image = cv2.imread(image_file)
image_shape = image.shape[:2]
image_cp = preprocess_image(image, input_size)
"""
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
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,
def get_flops(cfg,img_h,img_w):
flops = 0
in_channles=3
for l in cfg:
if l != 'M':
flops += in_channles*l[0]*l[1]*l[1]*img_h*img_w
in_channles = l[0]
else:
img_h=img_h/2
img_w=img_w/2
return flops
loss = np.random.random(7200)
loss_rank = np.argsort(loss)
cfg=[[32,3],'M',[64,3],'M',[128,3],[64,1],[128,3],'M',[256,3],[128,1],[256,3],'M',[512,3],[256,1],[512,3],[256,1],[512,3],'M',[1024,3],[512,1],[1024,3],[512,1],[1024,3]]
origin_flops=get_flops(cfg,224,224)
pruned_model_cfg,pruned_layer_cfg = generate_pruned_cfg(loss_rank, cfg, int(7200*args.percent))
pruned_flops=get_flops(pruned_model_cfg,224,224)
print pruned_model_cfg
model=torch.load(args.model)
print model
pruned_model = darknet(cfg=pruned_model_cfg)
pruned_model.cuda()
print pruned_model
pruned_model_init(model, pruned_model, pruned_layer_cfg)
print 'origin model flops: %d, pruned_model flops: %d'%(origin_flops,pruned_flops)
torch.save(pruned_model, './models/prune/prune_%.2f_darknet19.pkl'%args.percent)
print('pruned model saved at models/prune')
# image = cv2.imread(image_file)
image_shape = image.shape[:2] #只能取wh，channel=3不取
# copy,resize 416*416,归一化，在第0维增加存放batchsize维度
image_cp = preprocess_image(image, input_size)
"""
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)
"""
# （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
images = tf.placeholder(tf.float32, [1, input_size[0], input_size[1], 3])
detection_feat = darknet(images) #darknet网络输出的特征图
feat_sizes = input_size[0] // 32, input_size[1] // 32 #特征图尺寸是图像下采样32倍
detection_results = decode(detection_feat, feat_sizes, len(class_names),
anchors) #解码
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})
# （2）筛选解码后回归边界框——NMS（post process后期处理）
bboxes, scores, class_inds = postprocess(bboxes,
obj_probs,
class_probs,
def __init__(self, model_path, img_size=(416, 416)):
self.model_path = model_path
self.img_size = img_size
self.darknet = darknet(self.model_path)
input_size = (416, 416)
image_file = "./images/car.jpg"
image = cv2.imread(image_file)
image_shape = image.shape[:2]
image_cp = preprocess_image(image, input_size)
"""
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)
])
train_dataset = torchvision.datasets.ImageFolder(
root='/media/lulugay/PC/CCCV-30/train_set', transform=train_data_transform)
train_loader = torch.utils.data.DataLoader(train_dataset,
batch_size=args.batch_size,
shuffle=True,
num_workers=12)
test_dataset = torchvision.datasets.ImageFolder(
root='/media/lulugay/PC/CCCV-30/test_set', transform=test_data_transform)
test_loader = torch.utils.data.DataLoader(test_dataset,
batch_size=args.test_batch_size,
shuffle=True,
num_workers=12)
model = darknet()
model.cuda()
start_epoch = args.start_epoch
total_epoch = args.epochs
if start_epoch != 0:
model.load_state_dict(torch.load(args.resume))
history_score = np.zeros((total_epoch + 1, 4))
for epoch in range(start_epoch, total_epoch):
start = time.time()
print('epoch%d...' % epoch)
lr = poly(args.lr, 4, total_epoch, epoch)
optimizer = torch.optim.SGD(model.parameters(), lr, momentum=0.9)
