How to detect a rectangle and square in an image using OpenCV Python?

To detect rectangles and squares in an image using OpenCV Python, we analyze contours and calculate aspect ratios. A square has an aspect ratio close to 1.0, while rectangles have ratios significantly different from 1.0.

Algorithm Overview

The detection process follows these key steps:

for cnt in contours:
    approx = cv2.approxPolyDP(cnt, epsilon, True)
    if len(approx) == 4:
        x, y, w, h = cv2.boundingRect(cnt)
        ratio = float(w)/h
        if ratio >= 0.9 and ratio <= 1.1:
            # Square detected
            cv2.putText(img, 'Square', (x, y), font, 0.6, color, 2)
        else:
            # Rectangle detected
            cv2.putText(img, 'Rectangle', (x, y), font, 0.6, color, 2)

Step-by-Step Implementation

Import Required Libraries

Import OpenCV for image processing operations ?

import cv2
import numpy as np

Load and Preprocess Image

Read the input image and convert it to grayscale for contour detection ?

img = cv2.imread('shapes.jpg')
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

Apply Thresholding

Create a binary image using thresholding to improve contour detection ?

ret, thresh = cv2.threshold(gray, 50, 255, 0)

Find Contours

Detect all contours in the binary image ?

contours, _ = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)

Process Each Contour

For each contour, approximate the polygon and check if it has 4 vertices ?

for cnt in contours:
    approx = cv2.approxPolyDP(cnt, 0.01*cv2.arcLength(cnt, True), True)
    if len(approx) == 4:
        x, y, w, h = cv2.boundingRect(cnt)
        ratio = float(w)/h
        if 0.9 <= ratio <= 1.1:
            # Square detected
            pass
        else:
            # Rectangle detected
            pass

Complete Example

Here's the complete implementation to detect rectangles and squares ?

import cv2
import numpy as np

img = cv2.imread('shapes.jpg')
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

ret, thresh = cv2.threshold(gray, 50, 255, 0)
contours, hierarchy = cv2.findContours(thresh, 1, 2)
print("Number of contours detected:", len(contours))

for cnt in contours:
    x1, y1 = cnt[0][0]
    approx = cv2.approxPolyDP(cnt, 0.01*cv2.arcLength(cnt, True), True)
    if len(approx) == 4:
        x, y, w, h = cv2.boundingRect(cnt)
        ratio = float(w)/h
        if ratio >= 0.9 and ratio <= 1.1:
            img = cv2.drawContours(img, [cnt], -1, (0,255,255), 3)
            cv2.putText(img, 'Square', (x1, y1), cv2.FONT_HERSHEY_SIMPLEX, 0.6, (255, 255, 0), 2)
        else:
            cv2.putText(img, 'Rectangle', (x1, y1), cv2.FONT_HERSHEY_SIMPLEX, 0.6, (0, 255, 0), 2)
            img = cv2.drawContours(img, [cnt], -1, (0,255,0), 3)

cv2.imshow("Shapes", img)
cv2.waitKey(0)
cv2.destroyAllWindows()

Key Parameters

Input and Output

Consider the following input image:

The output shows detected rectangles and squares:

Number of contours detected: 4

In the output image, one rectangle and one square are successfully detected and labeled.

Conclusion

This method effectively detects rectangles and squares by analyzing contour approximations and aspect ratios. Adjust threshold values and epsilon parameters based on your specific image characteristics for optimal results.