2021005期：OpenCV识别答题卡

1
2
3
4
5
6
7
8
9

import imutils
import cv2

# 读入图片
image = cv2.imread("/Users/nestor/Downloads/09142111_jB7C.jpg")

# 转换为灰阶，轻度模糊
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(gray, (5, 5), 0)

1

edged = imutils.auto_canny(blurred)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

from imutils.perspective import four_point_transform

# 从边缘图中寻找轮廓，然后初始化答题卡对应的轮廓
cnts = cv2.findContours(edged.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)

docCnt = None
# 确保至少有一个轮廓被找到
if len(cnts) > 0:
    # 将轮廓按大小降序排序
    cnts = sorted(cnts, key=cv2.contourArea, reverse=True)

    # 对排序后的轮廓循环处理
    for c in cnts:
        # 获取近似的轮廓
        peri = cv2.arcLength(c, True)
        approx = cv2.approxPolyDP(c, 0.02 * peri, True)

        # 如果我们的近似轮廓有四个顶点，那么就认为找到了答题卡
        if len(approx) == 4:
            docCnt = approx
            break
            
# 对原始图像和灰度图都进行四点透视变换
paper = four_point_transform(image, docCnt.reshape(4, 2))
warped = four_point_transform(gray, docCnt.reshape(4, 2))

1
2
3
4
5
6
7

# 固定大小
paper = imutils.resize(paper, height=1000)
warped = imutils.resize(warped, height=1000)

# 对灰度图应用大津二值化算法
thresh = cv2.threshold(warped, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1]
cv2.imshow('thresh', thresh)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

# 在二值图像中查找轮廓，然后初始化题目对应的轮廓列表
cnts = cv2.findContours(thresh.copy(), cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
questionCnts = []

# 对每一个轮廓进行循环处理
for c in cnts:
    # 计算轮廓的边界框，然后利用边界框数据计算宽高比
    (x, y, w, h) = cv2.boundingRect(c)
    ar = w / float(h)

    # 辨别答案
    if 20 <= w <= 50 and 10 <= h <= 40 and 1.5 <= ar <= 3:
        M = cv2.moments(c)
        cX = int(M["m10"] / M["m00"])
        cY = int(M["m01"] / M["m00"])
        # 绘制中心及其轮廓
        cv2.drawContours(paper, c, -1, (0, 0, 255), 5, lineType=0)
        cv2.circle(paper, (cX, cY), 7, (255, 255, 255), -1)
        # 保存题目坐标信息
        questionCnts.append((cX, cY))
        
cv2.imshow('paper', paper)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

def judge(x, y):
    begin = (70, 350) # 第一题选项[A]的坐标
    x_param = 47.5    # 每个选项x轴的间隔
    y_param = 22      # 每个选项y轴的间隔
    options = {
        0: 'unknown',
        1: 'A',
        2: 'B',
        3: 'C',
        4: 'D'
    }
    column = round((x - begin[0]) / x_param)
    row = round((y - begin[1]) / y_param)
    option = options[(column + 1) % 5]
    line = (row + 1) // 6 * 20 + (1 + row) % 6 + 5 * (column // 5)

    return line, option
    
# 识别题号和答案
result = []
for x, y in questionCnts:
    line, option = judge(x, y)
    result.append({'id': line, 'option': option, 'c': (x, y)})
result.sort(key=(lambda item: item['id']))
print(json.dumps(result))

1

[{"id": 1, "option": "B", "c": [120, 349]}, {"id": 2, "option": "C", "c": [170, 374]}, {"id": 3, "option": "A", "c": [70, 398]}, {"id": 4, "option": "C", "c": [169, 423]}, {"id": 5, "option": "D", "c": [219, 447]}, {"id": 6, "option": "A", "c": [309, 349]}, {"id": 7, "option": "A", "c": [310, 374]}, {"id": 8, "option": "A", "c": [308, 398]}, {"id": 9, "option": "C", "c": [409, 423]}, {"id": 10, "option": "C", "c": [408, 449]}, {"id": 11, "option": "B", "c": [597, 348]}, {"id": 12, "option": "D", "c": [697, 373]}, {"id": 13, "option": "A", "c": [546, 399]}, {"id": 14, "option": "C", "c": [647, 422]}, {"id": 15, "option": "B", "c": [597, 449]}, {"id": 16, "option": "D", "c": [938, 348]}, {"id": 17, "option": "C", "c": [887, 373]}, {"id": 18, "option": "B", "c": [836, 399]}, {"id": 19, "option": "B", "c": [834, 424]}, {"id": 20, "option": "A", "c": [785, 447]}, {"id": 21, "option": "B", "c": [121, 480]}, {"id": 22, "option": "C", "c": [169, 505]}, {"id": 23, "option": "D", "c": [220, 530]}, {"id": 24, "option": "A", "c": [69, 554]}, {"id": 25, "option": "C", "c": [170, 580]}, {"id": 26, "option": "B", "c": [358, 481]}, {"id": 27, "option": "D", "c": [458, 505]}, {"id": 28, "option": "C", "c": [405, 530]}, {"id": 29, "option": "D", "c": [458, 554]}, {"id": 30, "option": "A", "c": [309, 579]}, {"id": 31, "option": "C", "c": [646, 482]}, {"id": 32, "option": "D", "c": [698, 506]}, {"id": 33, "option": "B", "c": [595, 530]}, {"id": 34, "option": "C", "c": [646, 555]}, {"id": 35, "option": "A", "c": [546, 580]}, {"id": 36, "option": "A", "c": [782, 482]}, {"id": 37, "option": "B", "c": [835, 506]}, {"id": 38, "option": "D", "c": [936, 531]}, {"id": 39, "option": "C", "c": [888, 556]}, {"id": 41, "option": "B", "c": [120, 612]}, {"id": 42, "option": "D", "c": [219, 638]}, {"id": 43, "option": "B", "c": [120, 662]}, {"id": 44, "option": "D", "c": [218, 687]}, {"id": 45, "option": "A", "c": [70, 711]}, {"id": 46, "option": "C", "c": [408, 612]}, {"id": 47, "option": "D", "c": [458, 638]}, {"id": 48, "option": "B", "c": [358, 662]}, {"id": 49, "option": "D", "c": [458, 687]}, {"id": 50, "option": "A", "c": [308, 712]}, {"id": 55, "option": "C", "c": [885, 582]}]