识别银行卡号

1.读取模板，获取模板数据，便于图像比较

2.读取图像，处理图像，获取数字区域

3.将每个数字区域与模板进行匹配，获取最优结果

4.输出结果

#coding:gbk
from imutils import contours
import cv2
import numpy as np
import imutils


def cvshow(img):
    cv2.imshow('',img)
    cv2.waitKey(0)
    cv2.destroyAllWindows()

#读取输入图像
img = cv2.imread('szjc\\k2.jpg')
img = cv2.resize(img,(331,203))
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
cvshow(img)
#读取模板
tem = cv2.imread('szjc\\mb.jpg')
gray0 = cv2.cvtColor(tem,cv2.COLOR_BGR2GRAY)#灰度图
cvshow(gray0)
#二值图像
gray0 = cv2.threshold(gray0,10,255,cv2.THRESH_BINARY_INV)[1]
#gray0 = cv2.threshold(gray0,210,255,cv2.THRESH_BINARY)[1]
cvshow(gray0)
#获取轮廓
contours,hierarchy = cv2.findContours(gray0,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)
#cv2.drawContours(tem,contours,-1,(0,0,255),2)
#cvshow(tem)
#排序
contours = imutils.contours.sort_contours(contours,method='Left-to-right')[0]
digits = {}
#遍历每一个轮廓
for (i, c) in enumerate(contours):
    # 计算外接矩形并且resize成合适大小
    (x,y,w,h) = cv2.boundingRect(c)
    if w>3 and h>3:
        a = gray0[y-5:y+h+5,x-5:x+w+5]
        a = cv2.resize(a,(57,88))
        digits[i] = a
        #cvshow(a)
#处理输入图像
#初始化卷积核
rectKernel = cv2.getStructuringElement(cv2.MORPH_RECT,(9,3))
sqKernel = cv2.getStructuringElement(cv2.MORPH_RECT,(5,5))
#礼帽操作，突出更明亮的区域
tophat = cv2.morphologyEx(gray,cv2.MORPH_TOPHAT,rectKernel)
#cvshow(tophat)
gradX = cv2.Sobel(tophat,ddepth=cv2.CV_32F,dx=1,dy=0,ksize=-1)  #ksize=-1相当于3*3
gradX = np.absolute(gradX)
minVal,maxVal = np.min(gradX),np.max(gradX)
gradX = 255*((gradX-minVal)/(maxVal-minVal))
gradX = gradX.astype("uint8")
#print(np.array(gradX).shape)
#cvshow(gradX)
#通过闭操作（先膨胀,再腐蚀）将数字连在一起
gradX = cv2.morphologyEx(gradX,cv2.MORPH_CLOSE,rectKernel)
cvshow(gradX)
#THRESH_OTSU会自动寻找合适的阈值，适合双峰，需把阈值参数设置为0
thresh = cv2.threshold(gradX,0,255,cv2.THRESH_BINARY|cv2.THRESH_OTSU)[1]
cvshow(thresh)
#再来一个闭操作
thresh = cv2.morphologyEx(thresh,cv2.MORPH_CLOSE,sqKernel)
cvshow(thresh)
#计算轮廓
cnts,hierarchy = cv2.findContours(thresh,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)
img1 = img.copy()
cv2.drawContours(img1,cnts,-1,(0,0,255),2)
cvshow(img1)
locs = []
#遍历轮廓
for (i, c) in enumerate(cnts):
    #计算矩形
    (x, y, w, h) = cv2.boundingRect(c)
    ar = w/float(h)
    #选择合适的区域，根据实际任务来，这里的基本都是四个数字一组
    if ar>2.5 and ar<4.0:
        if (w>40 and w<70) and (h>10 and h<30):
            locs.append((x,y,w,h))
# 将符合的轮廓从左到右排序
locs = sorted(locs,key=lambda x:x[0])
#print(locs)
output = []
#遍历每一个轮廓中的数字
for (i,(gX,gY,gW,gH)) in enumerate(locs):
    groupOutput = []
    #根据坐标提取每一个组
    group = gray[gY-5:gY+gH+5,gX-5:gX+gW+5]
    #cvshow(group)
    #二值化
    group = cv2.threshold(group,0,255,cv2.THRESH_BINARY|cv2.THRESH_OTSU)[1]
    cvshow(group)
    #计算每一组的轮廓
    digitCnts,hierarchy = cv2.findContours(group,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)
    digitCnts = imutils.contours.sort_contours(digitCnts,method='Left-to-right')[0]
    #计算每一组中的每一个数值
    for c in digitCnts:
        #找到当前数值的轮廓，resize成合适的的大小
        (x,y,w,h) = cv2.boundingRect(c)
        roi = group[y:y+h,x:x+w]
        roi = cv2.resize(roi,(57,88))
        cvshow(roi)
        #计算匹配得分
        scores = []
        #在模板中计算每一个得分
        for (digit,digitROI) in digits.items():
            # 模板匹配
            result = cv2.matchTemplate(roi,digitROI,cv2.TM_CCOEFF)
            score = cv2.minMaxLoc(result)[1]
            scores.append(score)
        #得到最合适的数字
        groupOutput.append(str(np.argmax(scores)))
    #得到结果
    cv2.rectangle(img,(gX-5,gY-5),(gX+gW+5,gY+gH+5),(0,0,255),1)
    cv2.putText(img,"".join(groupOutput),(gX,gY-15),cv2.FONT_HERSHEY_SIMPLEX,0.65,(0,0,255),2)
    output.extend(groupOutput)
# 打印结果
print("Credit Card #: {}".format("".join(output)))
cvshow(img)

模板

 

 结果：