How to separate images using watershed algorithm in Python

Question

How to separate indiviual images among multiple images after image segmentaion using watershed algorithm in Python The attached image is consists of 4 images , from which we need to apply image segmentation and separate individual image from those 4 images

Answer 1

We will flood fill it first

import cv2;
import numpy as np;

# Read image
im_in = cv2.imread("2SNAT.jpg", cv2.IMREAD_GRAYSCALE);

# Threshold.
# Set values equal to or above 220 to 0.
# Set values below 220 to 255.

th, im_th = cv2.threshold(im_in, 220, 255, cv2.THRESH_BINARY_INV);

# Copy the thresholded image.
im_floodfill = im_th.copy()

# Mask used to flood filling.
# Notice the size needs to be 2 pixels than the image.
h, w = im_th.shape[:2]
mask = np.zeros((h+2, w+2), np.uint8)

# Floodfill from point (0, 0)
cv2.floodFill(im_floodfill, mask, (0,0), 255);

# Invert floodfilled image
im_floodfill_inv = cv2.bitwise_not(im_floodfill)

# Combine the two images to get the foreground.
im_out = im_th | im_floodfill_inv

Then find contour and crop out

im, contours, hierarchy = cv2.findContours(im_out.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)

final_contours = []
for contour in contours:
    area = cv2.contourArea(contour)
    if area > 1000:
        final_contours.append(contour)

Crop out step, also drawing rectangle on original image

counter = 0        
for c in final_contours:
    counter = counter + 1
# for c in [final_contours[0]]:
    peri = cv2.arcLength(c, True)
    approx = cv2.approxPolyDP(c, 0.01 * peri, True)
    x,y,w,h = cv2.boundingRect(approx)
    print(x, y, w, h)
    aspect_ratio = w / float(h)

    if (aspect_ratio >= 0.8 and aspect_ratio <= 4):
        cv2.rectangle(im_in,(x,y),(x+w,y+h),(0,255,0),2)
        cv2.imwrite('splitted_{}.jpg'.format(counter), im_in[y:y+h, x:x+w])
cv2.imwrite('rectangled_split.jpg', im_in)

Answer 2

Instead of using watershed, here's a simple approach using thresholding + morphological operations. The idea is to obtain a binary image then perform morph close to combine each object as a single contour. We then find contours and extract/save each ROI using Numpy slicing.

Here's each individual object highlighted in green

Individual saved object

Code

import cv2

# Load image, grayscale, Otsu's threshold
image = cv2.imread('1.jpg')
original = image.copy()
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]

# Morph close
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (7,7))
close = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel, iterations=3)

# Find contours and extract ROI
cnts = cv2.findContours(close, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
num = 0
for c in cnts:
    x,y,w,h = cv2.boundingRect(c)
    cv2.rectangle(image, (x, y), (x + w, y + h), (36,255,12), 2)
    ROI = original[y:y+h, x:x+w]
    cv2.imwrite('ROI_{}.png'.format(num), ROI)
    num += 1

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