cv.py

import sys
import numpy as np
import cv2
 
im = cv2.imread('tt.png')
im3 = im.copy()
 
gray = cv2.cvtColor(im,cv2.COLOR_BGR2GRAY)   #先转换为灰度图才能够使用图像阈值化

 
thresh = cv2.adaptiveThreshold(gray,255,cv2.ADAPTIVE_THRESH_GAUSSIAN_C,cv2.THRESH_BINARY,11,2)  #自适应阈值化
 
##################      Now finding Contours         ###################
# 
contours,hierarchy = cv2.findContours(thresh,cv2.RETR_LIST,cv2.CHAIN_APPROX_SIMPLE)
#边缘查找，找到数字框，但存在误判
 
samples =  np.empty((0,900))    #将每一个识别到的数字所有像素点作为特征，储存到一个30*30的矩阵内
responses = []                  #label
pre={}
for i in range(48,58):
    pre[i]=-1
keys = [i for i in range(48,58)]    #48-58为ASCII码
count =0
for cnt in contours:
    if cv2.contourArea(cnt)>80:     #使用边缘面积过滤较小边缘框
        [x,y,w,h] = cv2.boundingRect(cnt)   
        if  h>25 and h < 30:        #使用高过滤小框和大框
            count+=1
            cv2.rectangle(im,(x,y),(x+w,y+h),(0,0,255),2)
            roi = thresh[y:y+h,x:x+w]
            roismall = cv2.resize(roi,(30,30))
            cv2.imshow('norm',im)
            
            key = cv2.waitKey(0)
            #print(key,end=' ')
            if key == 27:  # (escape to quit)
                sys.exit()
            elif key in keys:
                #print(pre[key])
                responses.append(int(chr(key)))
                sample = roismall.reshape((1,900))
                samples = np.append(samples,sample,0)
                print(sample)
            if count == 100:        #过滤一下过多边缘框，后期可能会尝试极大抑制
                break
responses = np.array(responses,np.float32)
responses = responses.reshape((responses.size,1))
print()
print ("training complete")
#print(samples)
#print(responses)
np.savetxt('generalsamples.data',samples)
np.savetxt('generalresponses.data',responses)
#
cv2.waitKey()
cv2.destroyAllWindows()