python OpenCV实现答题卡识别判卷_Python

本文实例为大家分享了python OpenCV实现答题卡识别判卷的具体代码，供大家参考，具体内容如下

完整代码：

				?

									#导入工具包

									import numpy as np

									import argparse

									import imutils

									import cv2

									# 设置参数

									ap = argparse.ArgumentParser()

									ap.add_argument("-i", "--image", default="./images/test_03.png",

									 help="path to the input image")

									args = vars(ap.parse_args())

									# 正确答案

									ANSWER_KEY = {0: 1, 1: 4, 2: 0, 3: 3, 4: 1}

									def order_points(pts):

									 # 一共4个坐标点

									 rect = np.zeros((4, 2), dtype = "float32")

									 # 按顺序找到对应坐标0123分别是 左上，右上，右下，左下

									 # 计算左上，右下

									 s = pts.sum(axis = 1)

									 rect[0] = pts[np.argmin(s)]

									 rect[2] = pts[np.argmax(s)]

									 # 计算右上和左下

									 diff = np.diff(pts, axis = 1)

									 rect[1] = pts[np.argmin(diff)]

									 rect[3] = pts[np.argmax(diff)]

									 return rect

									def four_point_transform(image, pts):

									 # 获取输入坐标点

									 rect = order_points(pts)

									 (tl, tr, br, bl) = rect

									 # 计算输入的w和h值

									 widthA = np.sqrt(((br[0] - bl[0]) ** 2) + ((br[1] - bl[1]) ** 2))

									 widthB = np.sqrt(((tr[0] - tl[0]) ** 2) + ((tr[1] - tl[1]) ** 2))

									 maxWidth = max(int(widthA), int(widthB))

									 heightA = np.sqrt(((tr[0] - br[0]) ** 2) + ((tr[1] - br[1]) ** 2))

									 heightB = np.sqrt(((tl[0] - bl[0]) ** 2) + ((tl[1] - bl[1]) ** 2))

									 maxHeight = max(int(heightA), int(heightB))

									 # 变换后对应坐标位置

									 dst = np.array([

									  [0, 0],

									  [maxWidth - 1, 0],

									  [maxWidth - 1, maxHeight - 1],

									  [0, maxHeight - 1]], dtype = "float32")

									 # 计算变换矩阵

									 M = cv2.getPerspectiveTransform(rect, dst)

									 warped = cv2.warpPerspective(image, M, (maxWidth, maxHeight))

									 # 返回变换后结果

									 return warped

									def sort_contours(cnts, method="left-to-right"):

									    reverse = False

									    i = 0

									    if method == "right-to-left" or method == "bottom-to-top":

									        reverse = True

									    if method == "top-to-bottom" or method == "bottom-to-top":

									        i = 1

									    boundingBoxes = [cv2.boundingRect(c) for c in cnts]

									    (cnts, boundingBoxes) = zip(*sorted(zip(cnts, boundingBoxes),

									                                        key=lambda b: b[1][i], reverse=reverse))

									    return cnts, boundingBoxes

									def cv_show(name,img):

									        cv2.imshow(name, img)

									        cv2.waitKey(0)

									        cv2.destroyAllWindows()  

									# 预处理

									image = cv2.imread(args["image"])

									contours_img = image.copy()

									gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

									blurred = cv2.GaussianBlur(gray, (5, 5), 0)

									cv_show('blurred',blurred)

									edged = cv2.Canny(blurred, 75, 200)

									cv_show('edged',edged)

									# 轮廓检测

									cnts = cv2.findContours(edged.copy(), cv2.RETR_EXTERNAL,

									 cv2.CHAIN_APPROX_SIMPLE)[0]

									cv2.drawContours(contours_img,cnts,-1,(0,0,255),3) 

									cv_show('contours_img',contours_img)

									docCnt = None

									# 确保检测到了

									if len(cnts) > 0:

									 # 根据轮廓大小进行排序

									 cnts = sorted(cnts, key=cv2.contourArea, reverse=True)

									 # 遍历每一个轮廓

									 for c in cnts:

									  # 近似

									  peri = cv2.arcLength(c, True)

									  approx = cv2.approxPolyDP(c, 0.02 * peri, True)

									  # 准备做透视变换

									  if len(approx) == 4:

									   docCnt = approx

									   break

									# 执行透视变换

									warped = four_point_transform(gray, docCnt.reshape(4, 2))

									cv_show('warped',warped)

									# Otsu's 阈值处理

									thresh = cv2.threshold(warped, 0, 255,

									 cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1] 

									cv_show('thresh',thresh)

									thresh_Contours = thresh.copy()

									# 找到每一个圆圈轮廓

									cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,

									 cv2.CHAIN_APPROX_SIMPLE)[0]

									cv2.drawContours(thresh_Contours,cnts,-1,(0,0,255),3) 

									cv_show('thresh_Contours',thresh_Contours)

									questionCnts = []

									# 遍历

									for c in cnts:

									 # 计算比例和大小

									 (x, y, w, h) = cv2.boundingRect(c)

									 ar = w / float(h)

									 # 根据实际情况指定标准

									 if w >= 20 and h >= 20 and ar >= 0.9 and ar <= 1.1:

									  questionCnts.append(c)

									# 按照从上到下进行排序

									questionCnts = sort_contours(questionCnts,

									 method="top-to-bottom")[0]

									correct = 0

									# 每排有5个选项

									for (q, i) in enumerate(np.arange(0, len(questionCnts), 5)):

									 # 排序

									 cnts = sort_contours(questionCnts[i:i + 5])[0]

									 bubbled = None

									 # 遍历每一个结果

									 for (j, c) in enumerate(cnts):

									  # 使用mask来判断结果

									  mask = np.zeros(thresh.shape, dtype="uint8")

									  cv2.drawContours(mask, [c], -1, 255, -1) #-1表示填充

									  cv_show('mask',mask)

									  # 通过计算非零点数量来算是否选择这个答案

									  mask = cv2.bitwise_and(thresh, thresh, mask=mask)

									  total = cv2.countNonZero(mask)

									  # 通过阈值判断

									  if bubbled is None or total > bubbled[0]:

									   bubbled = (total, j)

									 # 对比正确答案

									 color = (0, 0, 255)

									 k = ANSWER_KEY[q]

									 # 判断正确

									 if k == bubbled[1]:

									  color = (0, 255, 0)

									  correct += 1

									 # 绘图

									 cv2.drawContours(warped, [cnts[k]], -1, color, 3)

									score = (correct / 5.0) * 100

									print("[INFO] score: {:.2f}%".format(score))

									cv2.putText(warped, "{:.2f}%".format(score), (10, 30),

									 cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 0, 255), 2)

									cv2.imshow("Original", image)

									cv2.imshow("Exam", warped)

									cv2.waitKey(0)