Question

我目前正在尝试使用背景减法器从检测器中消除误报。 Eack链接读取一个mjpeg视频，并将减法器应用于每个视频。该代码有效，结果如下：

MoG2背景分隔符的代码为：

    for index, link in enumerate(onlyfiles):
        print(link)
        subtractor = cv2.createBackgroundSubtractorMOG2(history=20, varThreshold=100, detectShadows=True)
        count=0
        count2=0
        # Create a VideoCapture object and read from input file
        # If the input is the camera, pass 0 instead of the video file name

        cap = cv2.VideoCapture(link)
        # Check if camera opened successfully
        if (cap.isOpened() == False):
            print("Error opening video stream or file")

        # Read until video is completed
        while (cap.isOpened()):
            # Capture frame-by-frame
            ret, frame = cap.read()
            if ret == True:
                print("Frame detected")
                frame1 = frame.copy()  
                gray = cv2.cvtColor(frame1, cv2.COLOR_BGR2GRAY)
                blurred = cv2.bilateralFilter(gray, 9, 9, 9)
                mask = subtractor.apply(blurred)     
                cv2.imshow("mask1", mask)
                # Copy the thresholded image.
                im_floodfill = mask.copy()               
                # Mask used to flood filling.
                # Notice the size needs to be 2 pixels than the image.
                h, w = mask.shape[:2]
                mask1 = np.zeros((h+2, w+2), np.uint8)                 
                # Floodfill from point (0, 0)
                cv2.floodFill(im_floodfill, mask1, (0,0), 255);                
                # Invert floodfilled image
                im_floodfill_inv = cv2.bitwise_not(im_floodfill)                 
                # Combine the two images to get the foreground.
                im_out = mask | im_floodfill_inv
                cv2.imshow("Foreground", im_out)
                cv2.imshow('Video', frame1)
                cv2.waitKey(25)                 

                # Press Q on keyboard to  exit
                if cv2.waitKey(25) & 0xFF == ord('q'):
                    break

            # Break the loop

                cv2.imshow('Video', frame)  
                cv2.waitKey(25)
            else:
                break

        # When everything done, release the video capture object
        cap.release()
        # Closes all the frames
        cv2.destroyAllWindows()

在某些情况下，会输出一批灰色图像，如下所示：

为什么会这样，是由于历史原因？如何纠正？

为了确定这种情况的发生，我输出了黑色像素的数量。零表示图像是灰度的，像素值为127。

Frame detected
zero pixels:  0
Frame detected
zero pixels:  414628
Frame detected
zero pixels:  414615
Frame detected
zero pixels:  41465

Frame detected
zero pixels:  0
Frame detected
zero pixels:  413462
Frame detected
zero pixels:  414719
Frame detected
zero pixels:  414720
Frame detected
zero pixels:  414720
Frame detected
zero pixels:  414592
Frame detected
zero pixels:  413932
Frame detected
zero pixels:  412518
Frame detected
zero pixels:  412495
Frame detected
zero pixels:  414221

Frame detected
zero pixels:  0
Frame detected
zero pixels:  412651
Frame detected
zero pixels:  414290
Frame detected
zero pixels:  414490
Frame detected
zero pixels:  414707
Frame detected
zero pixels:  414687
Frame detected
zero pixels:  414689
Frame detected
zero pixels:  414665
Frame detected
zero pixels:  414704
Frame detected
zero pixels:  414583

Frame detected
zero pixels:  0

如果该项目链接到历史参数，那么如何编辑减法器以设置何时可以收集图像进行捕获？

Opencv Mog2背景减法器：输出为灰色图像

0 个答案: