Determine division line in text using openCV

Question

I want to detect the division line in the scanned document, and instead of using cv2.Canny, I used cv2.threshold to get a pretty clean pre-processed image. However my parameter for cv2.HoughLines may be improper and I got a chaos in the final output. And the lines didn`t present as the set color.

My code is:

import cv2
import numpy as np
from matplotlib import pyplot as plt

## (1) read
img = cv2.imread("q11.png")
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
plt.figure(figsize=(17,17))
plt.imshow(gray,cmap='gray')

## (2) threshold
th, threshed = cv2.threshold(gray, 200, 20, cv2.THRESH_BINARY_INV|cv2.THRESH_OTSU)
plt.figure(figsize=(15,15))
plt.imshow(threshed,cmap='gray')

## (3) HoughLines
lines = cv2.HoughLines(threshed,rho=1,theta=np.pi/180,threshold = 800)
for i in range(len(lines)):
    for rho,theta in lines[i]:
        a = np.cos(theta)
        b = np.sin(theta)
        x0 = a*rho
        y0 = b*rho
        x1 = int(x0 + 1000*(-b))
        y1 = int(y0 + 1000*(a))
        x2 = int(x0 - 1000*(-b))
        y2 = int(y0 - 1000*(a))
    cv2.line(threshed,(x1,y1),(x2,y2),(0,0,255),2)

plt.figure(figsize=(10, 10))
plt.subplot(111),plt.imshow(threshed)
plt.title('hough'), plt.xticks([]), plt.yticks([])

The original pic is:

and after cv2.threshold with cv2.THRESH_BINARY_INV|cv2.THRESH_OTSU I got:

I want to have:

What I actually got:

https://stackoverflow.com/questions/45322630/how-to-detect-lines-in-opencv — gameon67, Jan 10 '19 at 00:11
As far as your yellow and purple output is concerned, you can refer to this [answer](https://stackoverflow.com/questions/3823752/display-image-as-grayscale-using-matplotlib) — ZdaR, Jan 10 '19 at 06:23

score 0 · Answer 1 · answered Jul 25 '19 at 02:51

Here's a simple approach

Convert image to grayscale and gaussian blur
Threshold image
Dilate to enhance contour
Detect lines

Threshold

Next we dilate to enhance contours then use cv2.HoughLinesP() to detect lines. You mention

The lines don`t show as the set color

This is because you tried to draw the line on a binary image (the thresholded image). Since it only has one channel, pixels are only white or black. Thus it will not display the color. You have to draw it on a colored image

import cv2
import numpy as np

image = cv2.imread('1.jpg')
gray = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
blur = cv2.GaussianBlur(gray, (5,5), 0)

thresh = cv2.threshold(blur,190, 255,cv2.THRESH_BINARY_INV)[1]
kernel = cv2.getStructuringElement(cv2.MORPH_CROSS, (5,5))
dilate = cv2.dilate(thresh, kernel, iterations=1)

minLineLength = 10
maxLineGap = 200
lines = cv2.HoughLinesP(dilate,1,np.pi/180,100,minLineLength,maxLineGap)
for line in lines:
    for x1,y1,x2,y2 in line:
        cv2.line(image,(x1,y1),(x2,y2),(0,0,255),3)

cv2.imshow('image', image)
cv2.imshow('thresh', thresh)
cv2.waitKey()

Determine division line in text using openCV

1 Answers1