使用YOLOv2进行图像检测

• 基本配置信息

tensorflow (1.4.0)

tensorflow-tensorboard (0.4.0)

Keras (2.1.5)

Python (3.6.0)

Anaconda 4.3.1 (64-bit)

Windows 7

• darknet链接

https://github.com/pjreddie/darknet

下载后在cfg文件夹下找到yolov2的配置文件yolov2.cfg

• yolov2权重文件链接

https://pjreddie.com/darknet/yolov2/

在页面中选择YOLOV2 weights下载

• yad2k 链接

https://github.com/allanzelener/YAD2K

下载完成后将之前下载好的yolov2.cfg文件，YOLOV2 weights文件拷贝到yad2k目录下

• 使用spyder 运行yad2k目录下的yad2k.py文件

在运行配置里设置运行时所需的参数信息

或使用命令行运行yad2k.py

python yad2k.py yolov2.cfg yolov2.weights model_data/yolo.h5

运行结果如图所示

生成的yolo.h5文件在model_data文件夹内

• 利用生成的权重信息，进行图像检测

使用opencv调用电脑摄像头，进行视频图像信息的检测

opencv版本

opencv-python (3.2.0)

在yad2k目录下创建自己的demo，参考https://www.jianshu.com/p/3e77cefeb49b

 import cv2
 import os
 import time
 import numpy as np
 from keras import backend as K
 from keras.models import load_model

 from yad2k.models.keras_yolo import yolo_eval, yolo_head

 class YOLO(object):
     def __init__(self):
         self.model_path = 'model_data/yolo.h5'
         self.anchors_path = 'model_data/yolo_anchors.txt'
         self.classes_path = 'model_data/coco_classes.txt'
         self.score = 0.3
         self.iou = 0.5

         self.class_names = self._get_class()
         self.anchors = self._get_anchors()
         self.sess = K.get_session()
         self.boxes, self.scores, self.classes = self.generate()

     def _get_class(self):
         classes_path = os.path.expanduser(self.classes_path)
         with open(classes_path) as f:
             class_names = f.readlines()
         class_names = [c.strip() for c in class_names]
         return class_names

     def _get_anchors(self):
         anchors_path = os.path.expanduser(self.anchors_path)
         with open(anchors_path) as f:
             anchors = f.readline()
             anchors = [float(x) for x in anchors.split(',')]
             anchors = np.array(anchors).reshape(-1, 2)
         return anchors

     def generate(self):
         model_path = os.path.expanduser(self.model_path)
         assert model_path.endswith('.h5'), 'Keras model must be a .h5 file.'

         self.yolo_model = load_model(model_path)

         # Verify model, anchors, and classes are compatible
         num_classes = len(self.class_names)
         num_anchors = len(self.anchors)
         # TODO: Assumes dim ordering is channel last
         model_output_channels = self.yolo_model.layers[-1].output_shape[-1]
         assert model_output_channels == num_anchors * (num_classes + 5), \
             'Mismatch between model and given anchor and class sizes'
         print('{} model, anchors, and classes loaded.'.format(model_path))

         # Check if model is fully convolutional, assuming channel last order.
         self.model_image_size = self.yolo_model.layers[0].input_shape[1:3]
         self.is_fixed_size = self.model_image_size != (None, None)

         # Generate output tensor targets for filtered bounding boxes.
         # TODO: Wrap these backend operations with Keras layers.
         yolo_outputs = yolo_head(self.yolo_model.output, self.anchors, len(self.class_names))
         self.input_image_shape = K.placeholder(shape=(2, ))
         boxes, scores, classes = yolo_eval(yolo_outputs, self.input_image_shape, score_threshold=self.score, iou_threshold=self.iou)
         return boxes, scores, classes

     def detect_image(self, image):
         start = time.time()
         #image = cv2.imread(image)
         #cv2.imshow('image',image)
         y, x, _ = image.shape

         if self.is_fixed_size:  # TODO: When resizing we can use minibatch input.
             resized_image = cv2.resize(image, tuple(reversed(self.model_image_size)), interpolation=cv2.INTER_CUBIC)
             image_data = np.array(resized_image, dtype='float32')
         else:
             image_data = np.array(image, dtype='float32')

         image_data /= 255.
         image_data = np.expand_dims(image_data, 0)  # Add batch dimension.

         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.shape[0], image.shape[1]],
                 K.learning_phase(): 0
             })
         print('Found {} boxes for {}'.format(len(out_boxes), 'img'))

         for i, c in reversed(list(enumerate(out_classes))):
             predicted_class = self.class_names[c]
             box = out_boxes[i]
             score = out_scores[i]

             label = '{} {:.2f}'.format(predicted_class, score)
             top, left, bottom, right = box
             top = max(0, np.floor(top + 0.5).astype('int32'))
             left = max(0, np.floor(left + 0.5).astype('int32'))
             bottom = min(y, np.floor(bottom + 0.5).astype('int32'))
             right = min(x, np.floor(right + 0.5).astype('int32'))
             print(label, (left, top), (right, bottom))

             cv2.rectangle(image, (left, top), (right, bottom), (255, 0, 0), 2)
             cv2.putText(image, label, (left, int(top - 4)), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 1, cv2.LINE_AA)
         end = time.time()
         print(end - start)
         return image

     def close_session(self):
         self.sess.close()

 def detect_vedio(yolo):
     camera = cv2.VideoCapture(0)

     while True:
         res, frame = camera.read()

         if not res:
             break

         image = yolo.detect_image(frame)
         cv2.imshow("detection", image)

         if cv2.waitKey(1) & 0xFF == ord('q'):
             break
     yolo.close_session()

 def detect_img(img, yolo):
     image = cv2.imread(img)
     r_image = yolo.detect_image(image)
     cv2.namedWindow("detection")
     while True:
         cv2.imshow("detection", r_image)
         if cv2.waitKey(110) & 0xff == 27:
                 break
     yolo.close_session()

 if __name__ == '__main__':
     yolo = YOLO()
     detect_vedio(yolo)