yolov3为什么对大目标检测不好_3种方法用yolov3+python+cv2实现视频目标检测

1 说明：

=====

1.1 技术要点：yolov3+python+cv2(OpenCV)实现目标检测，本文由浅入深，代码注释，小白秒懂，值得收藏。

1.2 3种方法：摄像头视频目标检测、小视频cv2目标检测，和视频目标检测生成avi视频。

1.3 yolo的基本介绍。

pic1：单张静态图片目标检测

1.4 单张静态图片目标检测，我已经讲过，可参考文章

《目标检测和识别：Python+OpenCV+Yolov3》，文章中三个文件：yolov3.cfg、coco.names和yolov3.weights.下载已经告知，本文省略。建议先看这篇文章后，再看本文。

2 效果图：

=======

2.1 摄像头实时目标检测

bottle=矿泉水瓶；cell phone=手机；person=人

2.2 读取视频实时目标检测：

person=人；

2.3 目标检测后生成avi视频：

来自网络，仅供学习，如有侵权，请联系，定删除

3 摄像头实时目标检测代码：

#代码来源，进行注释和分析#https://blog.csdn.net/qq_39567427/article/details/105451962?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.add_param_isCf&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.add_param_isCf#摄像头实时目标检测#第1步：导入模块import cv2import numpy as np#第2步：加载yolov3的模型和三个文件#注意路径net = cv2.dnn.readNet("/home/xgj/Desktop/yolov3/yolov3.weights", "/home/xgj/Desktop/yolov3/yolov3.cfg")#名字分类的空的列表classes = []with open("/home/xgj/Desktop/yolov3/coco.names", "r") as f:    classes = [line.strip() for line in f.readlines()]layer_names = net.getLayerNames()output_layers = [layer_names[i[0] - 1] for i in net.getUnconnectedOutLayers()]#第3步：初始化相关参数#颜色随机产生colors = np.random.uniform(0, 255, size=(len(classes), 3))# Initialize frame rate calculationframe_rate_calc = 1#获取频率freq = cv2.getTickFrequency()# 0提示读取摄像头# 如果格式是'xxx/xxx/xxx.mp4'就是读取本地视频文件cap = cv2.VideoCapture(0)#第4步：循环while True:    # Start timer (for calculating frame rate)    t1 = cv2.getTickCount()    #获取读取摄像头资料    ret, frame = cap.read()    #以下与单张静态图片类似    #获取高、宽和通道    height, width, channels = frame.shape    # Detecting objects，侦测对象相关设置，与静态图片类似    blob = cv2.dnn.blobFromImage(frame, 0.00392, (416, 416), (0, 0, 0), True, crop=False)    net.setInput(blob)    outs = net.forward(output_layers)    # Showing informations on the screen    #屏幕展示信息设置    class_ids = []    confidences = []    boxes = []    #outs是上面获取的    for out in outs:        for detection in out:            scores = detection[5:]            class_id = np.argmax(scores)            confidence = scores[class_id]            if confidence > 0.5:                # Object detected                center_x = int(detection[0] * width)                center_y = int(detection[1] * height)                w = int(detection[2] * width)                h = int(detection[3] * height)                # Rectangle coordinates                x = int(center_x - w / 2)                y = int(center_y - h / 2)                boxes.append([x, y, w, h])                confidences.append(float(confidence))                class_ids.append(class_id)    indexes = cv2.dnn.NMSBoxes(boxes, confidences, 0.5, 0.4)    font = cv2.FONT_HERSHEY_SIMPLEX    for i in range(len(boxes)):        if i in indexes:            x, y, w, h = boxes[i]            label = str(classes[class_ids[i]])            color = colors[i]            cv2.rectangle(frame, (x, y), (x + w, y + h), color, 1)            cv2.putText(frame, label, (x, y - 20), font, 0.7, color, 1)        cv2.putText(frame,'FPS: {0:.2f}'.format(frame_rate_calc), (30,50),font, 0.7, (255,255,0), 1)    cv2.namedWindow("Image", cv2.WINDOW_NORMAL)    cv2.resizeWindow("Image", 960, 540)    cv2.imshow("Image", frame)    #以上与单张静态图片类似    # Calculate framerate    t2 = cv2.getTickCount()    time1 = (t2-t1)/freq    frame_rate_calc= 1/time1        # Press 'q' to quit，默认为1    if cv2.waitKey(1) == ord('q'):        break#第5步：结束关闭cap.release()cv2.destroyAllWindows()

4 读取本地视频进行目标检测：

代码基本与上面相同，仅需给一下这一行代码为

cap = cv2.VideoCapture('/home/xgj/Desktop/yolov3/333.mp4')   #上面注释里有介绍

5 高级一点的视频标准目标检测，并合成新的视频：

======================================

5.1 本机环境：python3.8+OpenCV4.2.0+华为笔记本电脑+deepin-linux操作系统。

5.2 代码说明：

#参考文章，对代码进行注释和适当修改#https://my.oschina.net/u/1464083/blog/2906860#本代码：sp-3.py

5.3 代码：

#第1步：导入模块import numpy as npimport argparseimport imutilsimport timeimport cv2import os#第2步：参数构建#实例化ap = argparse.ArgumentParser()#当前本代码同目录下的input文件夹：原视频位置ap.add_argument("-i", "--input", required=True,    help="path to input video")#当前本代码同目录下的output文件夹：目标检测标定后的合成视频位置ap.add_argument("-o", "--output", required=True,    help="path to output video")#当前本代码同目录下的yolov3文件夹ap.add_argument("-y", "--yolov3", required=True,    help="base path to YOLO directory")#参数定义，默认值ap.add_argument("-c", "--confidence", type=float, default=0.5,    help="minimum probability to filter weak detections")ap.add_argument("-t", "--threshold", type=float, default=0.3,    help="threshold when applyong non-maxima suppression")args = vars(ap.parse_args())#第3步：加载yolov3的3个文件#注意：三个文件在yolov3文件下weightsPath = os.path.sep.join([args["yolov3"], "yolov3.weights"])configPath = os.path.sep.join([args["yolov3"], "yolov3.cfg"])labelsPath = os.path.sep.join([args["yolov3"], "coco.names"])#读取标签名文件LABELS = open(labelsPath).read().strip().split("") # initialize a list of colors to represent each possible class labelnp.random.seed(42)#颜色随机设置COLORS = np.random.randint(0, 255, size=(len(LABELS), 3),dtype="uint8")print("[INFO] loading YOLO from disk...")net = cv2.dnn.readNetFromDarknet(configPath, weightsPath)ln = net.getLayerNames()ln = [ln[i[0] - 1] for i in net.getUnconnectedOutLayers()]#第4步：获取输入的原视频文件vs = cv2.VideoCapture(args["input"])writer = None(W, H) = (None, None)# try to determine the total number of frames in the video filetry:    prop = cv2.cv.CV_CAP_PROP_FRAME_COUNT if imutils.is_cv2()         else cv2.CAP_PROP_FRAME_COUNT    total = int(vs.get(prop))    #测算本视频内的目标检测识别后的总框架    print("[INFO] {} total frames in video".format(total)) # an error occurred while trying to determine the total# number of frames in the video fileexcept:    print("[INFO] could not determine # of frames in video")    print("[INFO] no approx. completion time can be provided")    total = -1#第5步：循环# loop over frames from the video file streamwhile True:    # read the next frame from the file    (grabbed, frame) = vs.read()     # if the frame was not grabbed, then we have reached the end    # of the stream    if not grabbed:        break     # if the frame dimensions are empty, grab them    if W is None or H is None:        (H, W) = frame.shape[:2]    # construct a blob from the input frame and then perform a forward    # pass of the YOLO object detector, giving us our bounding boxes    # and associated probabilities    blob = cv2.dnn.blobFromImage(frame, 1 / 255.0, (416, 416),        swapRB=True, crop=False)    net.setInput(blob)    start = time.time()    layerOutputs = net.forward(ln)    end = time.time()    # initialize our lists of detected bounding boxes, confidences,    # and class IDs, respectively    boxes = []    confidences = []    classIDs = []    # loop over each of the layer outputs    for output in layerOutputs:        # loop over each of the detections        for detection in output:            scores = detection[5:]            classID = np.argmax(scores)            confidence = scores[classID]             if confidence > args["confidence"]:                box = detection[0:4] * np.array([W, H, W, H])                (centerX, centerY, width, height) = box.astype("int")                x = int(centerX - (width / 2))                y = int(centerY - (height / 2))                 boxes.append([x, y, int(width), int(height)])                confidences.append(float(confidence))                classIDs.append(classID)    #绑定box    idxs = cv2.dnn.NMSBoxes(boxes, confidences, args["confidence"],        args["threshold"])     # ensure at least one detection exists    #确保一个侦测存在    if len(idxs) > 0:        # loop over the indexes we are keeping        for i in idxs.flatten():            # extract the bounding box coordinates            (x, y) = (boxes[i][0], boxes[i][1])            (w, h) = (boxes[i][2], boxes[i][3])             # draw a bounding box rectangle and label on the frame            color = [int(c) for c in COLORS[classIDs[i]]]            cv2.rectangle(frame, (x, y), (x + w, y + h), color, 2)            text = "{}: {:.4f}".format(LABELS[classIDs[i]],                confidences[i])            cv2.putText(frame, text, (x, y - 5),                cv2.FONT_HERSHEY_SIMPLEX, 0.5, color, 2)    # check if the video writer is None    #写入新的视频文件    if writer is None:        # initialize our video writer        fourcc = cv2.VideoWriter_fourcc(*"MJPG")        writer = cv2.VideoWriter(args["output"], fourcc, 30,            (frame.shape[1], frame.shape[0]), True)         # some information on processing single frame        if total > 0:            elap = (end - start)            print("[INFO] single frame took {:.4f} seconds".format(elap))            print("[INFO] estimated total time to finish: {:.4f} seconds".format(                elap * total))     # write the output frame to disk    writer.write(frame)#第6步：结束和收尾# release the file pointersprint("[INFO] cleaning up...")writer.release()vs.release()

5.4 运行：

#第7步：使用方法#在本代码sp-3.py的目录下打开终端#本机输入#python3.8 sp-3.py --input videos/222.mp4 --output output/2222.avi --yolov3 yolov3

5.5 图：

视频是一首完整的MV，有点大，时间有一点久

6 小结：

=====

6.1 摄像头视频，有点卡，小bug。

6.2 合成视频，有点慢，xiaobug。

7 yolo介绍：

=========

7.1 yolo是目前比较流行的目标检测算法，速度快结构简单。

7.2 YOLO 是 2016 年提出来的目标检测算法，当前较为火热。

7.3 yolo=You Only Look Once:

Unified, Real-Time Object Detection，

即：You Only Look Once说的是只需要一次CNN运算，

Unified指的是这是一个统一的框架，提供end-to-end的预测，

而Real-Time体现是Yolo算法速度快，达到实时。

7.4 yolov3比较成熟，但目前已经出现yolov4和yolov5(被认为是v4.5版本)。