python traffic detection traffic statistics vehicle count yolov5 deepsort traffic detection
Based on the previous yolo+deepsort, change the person category to the vehicle category, because there are several vehicle categories in the coco data set [car, bus, truck], so they must be saved.
First, let's take a look at the initial effect of yolov5+deepsort's vehicle tracking, look at the dense boxes and ids, and think about how to design these rules for statistics.
labelsPath = "./yolo-coco/coco.names"
LABELS = open(labelsPath).read().strip().split("\n")
np.random.seed(42)
COLORS = np.random.randint(0, 255, size=(200, 3),dtype="uint8")
weightsPath = "./yolo-coco/yolov3.weights"
configPath = "./yolo-coco/yolov3.cfg"
net = cv2.dnn.readNetFromDarknet(configPath, weightsPath)
ln = net.getLayerNames()
ln = [ln[i - 1] for i in net.getUnconnectedOutLayers()]
# ----------------------------------------------------------------------------------------------------------------------
"""
# cv2.VideoCapture() --- 里面参数如果地址，打开视频文件 --- 里面参数是0/1，打开摄像头
# 当参数是0的时候，打开计算机的内置摄像头，当参数为1的时候打开计算机的外置摄像头
# (W, H) = (None, None) --- 视频的宽度和高度，初始化视频编写器（writer）和帧尺寸
"""
# ----------------------------------------------------------------------------------------------------------------------
vs = cv2.VideoCapture('./input/123.mp4')
(W, H) = (None, None)
# ----------------------------------------------------------------------------------------------------------------------
"""
# try to determine the total number of frames in the video file
# 打开一个指向视频文件的文件指针，循环读取帧 --- 尝试确定视频文件中的总帧数（total），以便估计整个视频的处理时间;
# CV_CAP_PROP_FRAME_COUNT --- 视频的帧数
# 这里使用是处理视频的时候固定的过程，不必过度的纠结其使用 ---
# if imutils.is_cv2():
# prop = cv2.cv.CV_CAP_PROP_FRAME_COUNT
# else:
# prop = cv2.CAP_PROP_FRAME_COUNT
#
# vs.get(prop) --- cv2.VideoCapture.get(prop) --- 得到视频的总帧数
# print("[INFO] {} total frames in video".format(total)) --- 输出视频的帧数
"""
# ----------------------------------------------------------------------------------------------------------------------
try:
prop = cv2.cv.CV_CAP_PROP_FRAME_COUNT if imutils.is_cv2() \
else cv2.CAP_PROP_FRAME_COUNT
total = int(vs.get(prop))
print("[INFO] {} total frames in video".format(total))
except:
print("[INFO] could not determine # of frames in video")
print("[INFO] no approx. completion time can be provided")
total = -1
while True:
# ----------------------------------------------------------------------------------------------------------------------
"""
# cv2.VideoCapture.read() ---> 读取视频，在while中循环读取视频的frame
# vs.read() ---> 得到两个参数，其中ret是布尔值，如果读取帧是正确的则返回True，
# 如果文件读取到结尾，它的返回值就为False。frame就是每一帧的图像，是个三维矩阵。
# 第一个参数为False的时候，if not grabbed --- True --- 循环结束，
"""
(grabbed, frame) = vs.read()
if not grabbed:
break
if W is None or H is None:
(H, W) = frame.shape[:2]
python traffic flow detection two-way traffic counting