vline_detector() is a function for finding the pixel location of vertical lines in images of tables. hline_detector() is an analogous function for horizontal lines. These could be useful for cropping cells of tables before running OCR.

import cv2

import matplotlib.pyplot as plt

import numpy as np

import statistics

image = cv2.imread(image_path)

coords, plot = vline_detector(image)

The x-axis of plot denotes the width of the image. Moving along the width of the image, we can count the number of pixels in each vertical slice that is recognised as a vertical edge. The y-axis of plot shows this count. Therefore the spikes in plot are the horizontal coordinates of detected vertical lines. coords is a list of the positions of these spikes (i.e. the horizontal coordinates of detected vertical lines).