import numpy as np 
import cv2 

def get_vvList(list_data):
    #取出list中像素存在的區(qū)間
    vv_list=list()
    v_list=list()
    for index,i in enumerate(list_data):
        if i>0:
            v_list.append(index)
        else:
            if v_list:
                vv_list.append(v_list)
                #list的clear與[]有區(qū)別
                v_list=[]
    return vv_list

if __name__=='__main__':
    img_bgr=cv2.imread('./testImg/1.png',1)
    if not img_bgr is None:
        img=img_bgr.copy()
        img_gray=cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
        #二值化
        t,binary=cv2.threshold(img_gray,0,255,cv2.THRESH_OTSU+cv2.THRESH_BINARY)
        '''
        水平投影從左向右投射，計(jì)算每一行的黑色像素總數(shù)
        '''
        rows,cols=binary.shape
        hor_list=[0]*rows
        for i in range(rows):
            for j in range(cols):
                #統(tǒng)計(jì)每一行的黑色像素總數(shù)
                if binary.item(i,j)==0:
                    hor_list[i]=hor_list[i]+1
        '''
        對(duì)hor_list中的元素進(jìn)行篩選，可以去除一些噪點(diǎn)
        '''
        hor_arr=np.array(hor_list)
        hor_arr[np.where(hor_arr<5)]=0
        hor_list=hor_arr.tolist()

        #繪制水平投影
        img_white=np.ones(shape=(rows,cols),dtype=np.uint8)*255
        for i in range(rows):
            pt1=(cols-1,i)
            pt2=(cols-1-hor_list[i],i)
            cv2.line(img_white,pt1,pt2,(0,),1)
        cv2.imshow('水平投影',img_white)
        cv2.waitKey(0)
        
        #取出各個(gè)文字區(qū)間
        vv_list=get_vvList(hor_list)
        for i in vv_list:
            img_hor=img_bgr[i[0]:i[-1],:,:]
            cv2.imshow('文本行',img_hor)
            cv2.waitKey(0)