利用cv2的findContours 把黑白图片上的文字部分裁剪下来保存

findContours对于二值化图片的物体找轮廓真的挺好用的。

项目所需，效果如下:

每个模块有用的部分都在注释里了，希望小白的自己以后有用的话，能找得到。

import cv2
import numpy as np
import os
import shutil##获得一张图片的文字框信息，左上角和右下角 并写入文件中
def get_condition(img,path,title):#img 图片 path :txt文件的路径+名字 title:这一行的标注名gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)ret, binary = cv2.threshold(gray, 127, 255, cv2.THRESH_BINARY)binary = 255-binarycontours, hierarchy = cv2.findContours(binary, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)# x0,y0 左上角 x1,y1右下角 chushihuax0,y0,x1,y1 = 255,255,0,0for c in contours:##不断比较，得到框的边界x, y, w, h = cv2.boundingRect(c)x0 = min(x,x0)y0 = min(y,y0)x1 = max(x1,x+w)y1 = max(y1,y+h)##enlarge box 把框搞的大一丢丢x0=int(x0-5)y0=int(y0-5)x1=int(x1+5)y1=int(y1+5)# draw box#h=cv2.rectangle(img, (x0,y0),(x1,y1), (0, 0, 255), 2)#cv2.imshow('h',h)#cv2.waitKey(10000)condition = title+' ' +str(x0)+' '+str(y0)+ ' '+str(x1)+" "+str(y1)+" "+"\n"#print(condition)with open(path, 'a') as f:  # 'a'表示append,即在原来文件内容后继续写数据（不清楚原有数据）f.write(condition)f.close()#根据文件和坐标，将图片进行裁剪，保存
def cut_img(img,condition,save_path):#防止有些左上角越界if condition[0]<0:condition[0]=0if condition[1]<0:condition[1]=0crop = img[condition[1]:condition[3],condition[0]:condition[2]]  #y0:y1, x0:x1print(condition)cv2.imwrite(save_path,crop)#读取txt文件的每一行，并对每一行用空格进行分开，再根据每一行,对图片进行裁剪,保存到对应字母的文件夹下
def read_image(img_path,img_path_new,txt_path):aa = []with open(txt_path, "r") as f:for line in f.readlines():data = line.split('\n\t')for str in data:sub_str = str.split(' ')if sub_str:aa.append(sub_str)#print(aa)for j in range(9500): #txt文件中的数据行数condition=[]for i in range(4):condition.append(int(aa[j][i+1]))#save_path='/home/cpy/PycharmProjects/SingleGAN-master/image_new/'+'a'+'/'+'a1'+'.jpg'img = cv2.imread(img_path+aa[j][0]+'.jpg')cut_img(img,condition,save_path = img_path_new+aa[j][0][0]+'/'+aa[j][0]+'.jpg')#新建 a-z文件夹
def file():for i in range(25):os.mkdir('./image_new/'+chr(97+i))#对文件夹中包含‘0to1’的文件进行改名，把文字位置保存下来，把不包含0to1的文件删除#注意 os.listdir()返回的文件名不一定是顺序的，可能是随机的
def process_img(img_path,txt_path):for i in os.listdir(img_path):if '0to1' in i:img = cv2.imread(img_path+i)get_condition(img,txt_path,i[0:-13])os.rename(img_path+i,img_path+i[0:-13]+'.jpg')else:os.remove(img_path+i)img_path = '/home/cpy/PycharmProjects/SingleGAN-master/images/'
txt_path = './condition.txt'
img_path_new = '/home/cpy/PycharmProjects/SingleGAN-master/image_new/'
#process_img(img_path,txt_path)
read_image(img_path,img_path_new,txt_path)