python - 如何通过 Python 检测图像的 Miss-Moving Pattern

Question

我有一个图像，是规则的图案。我希望可以有一个 Python 代码来帮助指示“Miss-Moving”位置的边界框，是目标，例如：

目标

我曾尝试在侵蚀后再次进行灰色，侵蚀，阈值，侵蚀，结果为：

加里

加里侵蚀

阈值灰蚀

侵蚀阈值-灰色-侵蚀

我不知道下一步该怎么做，或者欢迎对这项工作提出任何意见。谢谢。

这是我的 Python 代码：

# Define Display & Save image function
def displayIMG(img, windowName):
    cv2.namedWindow(windowName, cv2.WINDOW_NORMAL)
    cv2.imshow(windowName, img)
    cv2.imwrite(windowName + '.jpg', img)
    
    #1.Read image file & Show out
    image = cv2.imread(image_filename)
    displayIMG(image, "Original")
    
    #2. Gray
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    name = "Gray"
    draw = gray
    displayIMG(draw, name)
    
    #3. Erosion
    kernel = np.ones((3, 3), np.uint8)
    erosion = cv2.erode(gray, kernel, iterations=2)
    name = "Gary_Erosion"
    draw = erosion
    hist = plotHistogram(draw, name)
    max_gray = max(hist)
    print('Max Gray = ', max_gray)
    displayIMG(draw, name)
    
    #4. Threshold
    (T, thresh) = cv2.threshold(gray, 120, 255, cv2.THRESH_TOZERO)
    name = "Threshold_Gray"
    draw = thresh
    displayIMG(draw, name)
    plotHistogram(draw, name, max_gray)
    
    #5. Erosion again 
    kernel = np.ones((3, 3), np.uint8)
    erosion2 = cv2.erode(draw, kernel, iterations=5)
    name = "Gary_Threshold_Erosion"
    draw = erosion2
    hist = plotHistogram(draw, name, max_gray)
    displayIMG(draw, name)

标签： pythonimage-processingconvexity-defects