OpenCv 入门 -- 答题卡的数字识别_opencv答题卡识别去掉数字

OpenCv 入门

OpenCv 入门 -- 答题卡的数字识别 @ Fu Xianjun. All Rights Reserved.

---

---

前言

OpenCV是一个跨平台计算机视觉库，用C++语言编写，用于图像处理、分析。本文将讲解如何使用OpenCv对答题卡的数字识别。

---

一、导入库

日常导包：

import cv2
import numpy as np
1
2

预处理：

def cv_show(name,img):
    cv2.imshow(name,img)
    cv2.waitKey(0)
    cv2.destroyAllWindows()

ANSWER_KEY = {0:1,1:4,2:0,3:3,4:1}
img = cv2.imread("test_01.png")
contours_Img = img.copy()
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)#灰度图
blur = cv2.GaussianBlur(gray,(5,5),0)#高斯（平滑处理）
edge = cv2.Canny(blur,75,200)#边缘检测
#轮廓检测
cnts,h = cv2.findContours(edge,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)#外边缘
cv2.drawContours(img,cnts,-1,(0,255,0),2)#绘制轮廓
cv_show("img",img)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

二、透视变换、寻找轮廓

def cv_show(name,img):
    cv2.imshow(name,img)
    cv2.waitKey(0)
    cv2.destroyAllWindows()
def order_points(pts):
    rect = np.zeros((4,2),dtype="float32")
    s = pts.sum(axis=1)
    rect[0]=pts[np.argmin(s)]
    rect[2]=pts[np.argmax(s)]
    d = np.diff(pts,axis=1)
    rect[1]=pts[np.argmin(d)]
    rect[3]=pts[np.argmax(d)]
    return rect
def four_point_transform(image,pts):
    rect = order_points(pts)
    (tl,tr,br,bl)=rect
    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))
    print(rect)
    dst = np.array([[0,0],[maxWidth-1,0],[maxWidth-1,maxHeight-1],[0,maxHeight-1]],dtype="float32")
    M = cv2.getPerspectiveTransform(rect,dst)
    warp = cv2.warpPerspective(image,M(maxWidth,maxHeight))
    return warp
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
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
27
28
29
30
31
32
33
34
35
36
37
38

三、输出结果

ANSWER_KEY = {0:1,1:4,2:0,3:3,4:1}
image = cv2.imread("test_05.png")
contour_Img = image.copy()
gray = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
blur = cv2.GaussianBlur(gray,(5,5),0)
edge = cv2.Canny(blur,75,200)
cnt = cv2.findContours(edge, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)[0]
cv2.drawContours(contour_Img, cnt, -1,(0,255,0),2)
dotCnt = None
if len(cnt)>0:
    cnt = sorted(cnt, key = cv2.contourArea,reverse=True)
    for c in cnt:
        peri = cv2.arcLength(c,True)
        approx = cv2.approxPolyDP(c, 0.02*peri,True)
        if len(approx)==4:
            dotCnt=approx
warp = four_point_transform(gray, dotCnt.reshape(4,2))

thresh = cv2.threshold(warp,0,255,cv2.THRESH_BINARY_INV|cv2.THRESH_OTSU)[1]
thresh_contours = thresh.copy()
cnts = cv2.findContours(thresh_contours,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)[0]
cv2.drawContours(thresh_contours, cnts, -1,(0,255,255),2)

questionCnts = []
#遍历
for c 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)
questionCnts = sort_contours(questionCnts, method="top-to-bottom")[0]
final = cv2.cvtColor(warp,cv2.COLOR_GRAY2BGR)
correct=0
for (q,i) in enumerate(np.arange(0,len(questionCnts),5)):
    cnts = sort_contours(questionCnts[i:i+5])[0]
    bubbled = None
    for (j,c)in enumerate(cnts):
        mask = np.zeros(thresh.shape,dtype="uint8")
        cv2.drawContours(mask, [c], -1,255,-1)
        
        mask = cv2.bitwise_and(thresh,thresh,mask=mask)
        total = cv2.countNonZero(mask)
        if bubbled is None or total >bubbled[0]:
            bubbled = (total,j)
    color = (0,0,255)
    k = ANSWER_KEY[q]
    if k==bubbled[1]:
        color = (0,255,0)
        correct+=1
    
cv2.drawContours(final, cnts[k], -1,color,2)

score = (correct/5.0)*100
cv2.putText(final,"Total:{:.2f}".format(score),(10,30),cv2.FONT_HERSHEY_SIMPLEX,0.9,(0,0,0),2)
cv2.imshow("final",final)
cv2.waitKey(0)
cv2.destroyAllWindows()
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
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57

四、结果展示

总结

以上就是今天要讲的内容，本文仅仅简单介绍了如何使用OpenCv对答题卡进行数字识别，及OpenCv的基础应用。