python+opencv进行表格识别并写入excel中

效果图如下：


对于任意图标都不需要自定义模板，直接程序生成，不过需要注意，图中的表格必须是水平的，无法适配倾斜的表格。

直接上代码：

import cv2
import numpy as np
import math
import xlwt
src='图片路径'

raw = cv2.imread(src, 1)
# 灰度图片
gray = cv2.cvtColor(raw, cv2.COLOR_BGR2GRAY)
binary = cv2.adaptiveThreshold(~gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 35, -5)
# 展示图片

rows, cols = binary.shape
scale2=15
scale = 20
# 自适应获取核值
# 识别横线:
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (cols // scale, 1))
kernel1 = cv2.getStructuringElement(cv2.MORPH_RECT, (cols // scale2, 1))

eroded = cv2.erode(binary, kernel, iterations=1)
dilated_col = cv2.dilate(eroded, kernel1, iterations=1)
# cv2.imwrite("横线图.jpg", dilated_col)

# 识别竖线：
# scale = 40#scale越大，越能检测出不存在的线
kernel2 = cv2.getStructuringElement(cv2.MORPH_RECT, (1, rows // scale2))

kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (1, rows // scale))
eroded = cv2.erode(binary, kernel, iterations=1)
dilated_row = cv2.dilate(eroded, kernel2, iterations=1)
# cv2.imwrite("竖线图.jpg", dilated_row)

# cv2.imwrite("3.png", dilated_row)

# 将识别出来的横竖线合起来
bitwise_and = cv2.bitwise_and(dilated_col, dilated_row)#对二值图进行与操作
# cv2.imwrite("交点二值图.jpg", bitwise_and)

# 标识表格轮廓
merge = cv2.add(dilated_col, dilated_row)
ret,binary = cv2.threshold(merge, 127, 255, cv2.THRESH_BINARY)
_,contours, hierarchy = cv2.findContours(binary, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
area=[]
for k in range(len(contours)):
	area.append(cv2.contourArea(contours[k]))
max_idx = np.argmax(np.array(area))
m_d_r=[]
m_u_l=[]
max_p=0
min_p=1e6
for  l1 in contours[max_idx]:
	for l2 in l1:
		if sum(l2)>max_p:
			max_p=sum(l2)
			d_r=l2
		if sum(l2)<min_p:
			min_p=sum(l2)
			u_l=l2
m_d_r=d_r
m_u_l=u_l
padding=5
x0=max(m_u_l[0]-padding,0)
x1=min(m_d_r[0]+padding,raw.shape[