python - 我正在尝试分析 Scantron 表单并为 OMR 项目创建 key

Question

因此，我尝试使用 OpenCV 扫描此表格，并从本质上为标记的问题制定一个 key 。现在，我尝试关闭一些我在网上找到的示例，并将其转换为二进制图像，但是在检测问题的标记时遇到了问题。我是通过我在网上找到的教程完成的，但他们使用了不同格式的表单，并且此表单上的 Material 比示例中显示的要多得多，我可以使用更熟悉 OpenCV 的人的一些输入。帮助不一定是对我的代码甚至工作代码的改进，它可能是对更有用的文档、 Material 或教程的链接和引用。

def analyzeKey(self):
    keypix = self.doc.getPagePixmap(0, alpha=False)
    keyim = self.pixel2np(keypix)
    cv2.imwrite("keyimage.jpg",keyim)
    key = cv2.imread("keyimage.jpg")

    grayscale = cv2.cvtColor(key, cv2.COLOR_BGR2GRAY)
    blurred = cv2.GaussianBlur(grayscale, (5, 5), 0)
    edged = cv2.Canny(blurred, 75, 200)
    cv2.imshow("Key", edged)


    cnts = cv2.findContours(edged.copy(), cv2.RETR_EXTERNAL,
                            cv2.CHAIN_APPROX_SIMPLE)
    cnts = cnts[0] if imutils.is_cv2() else cnts[1]
    docCnt = None

    # ensure that at least one contour was found
    if len(cnts) > 0:
        # sort the contours according to their size in
        # descending order
        cnts = sorted(cnts, key=cv2.contourArea, reverse=True)

        # loop over the sorted contours
        for c in cnts:
            # approximate the contour
            peri = cv2.arcLength(c, True)
            approx = cv2.approxPolyDP(c, 0.02 * peri, True)
    # if our approximated contour has four points,
            if len(approx) == 4:
                docCnt = approx
                break

    #originalkey = four_point_transform(key, docCnt.reshape(4, 2))
    #newkey = four_point_transform(grayscale, docCnt.reshape(4, 2))
    keyim[:,:,2] = 0
    cv2.imshow("Split",keyim)
    thresh = cv2.threshold(grayscale, 0,255,
                           cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1]
    cv2.imshow("Otsu", thresh)
    cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,
                            cv2.CHAIN_APPROX_SIMPLE)
    cnts = cnts[0] if imutils.is_cv2() else cnts[1]
    questionCnts = []

    for cntrs in cnts:
        (x, y, w, h) = cv2.boundingRect(c)
        ar = w / float(h)

        if w>= 20 and h>=20 and ar>=0.9 and ar<=1.1:
            questionCnts.append(c)

    cv2.imshow("cnts", questionCnts[0])

Answer 1

为了获得铅笔标记的轮廓，而不是阈值，您可以使用 cv2.inRange 区分灰色/黑色铅笔标记和橙色文本。 .您将为允许的颜色选择一个下限和上限。这里我简单地选择黑色作为下界，灰色(180,180,180)作为上界，并应用于你给的彩色图像。这些值之间的任何像素都在下面显示的输出掩码中指示。

img = cv2.imread('keyimage.jpg')
lower_bound = (0,0,0)
upper_bound = (180,180,180)
mask = cv2.inRange(img, lower_bound, upper_bound)
plt.imshow(mask, cmap='gray')
plt.show()