One-Click Background Replacement for ID Photos Using Python OpenCV

In daily life we often need ID photos with different background colors (red, white, blue). This guide shows how to replace the background of an ID photo automatically without manual cropping, using Python and OpenCV.

Knowledge Points

1. Image processing 2. OpenCV 3. NumPy 4. Basic Python knowledge

Environment

Windows, PyCharm, Python 3

Steps

1. Import libraries

<code>import numpy as np
import cv2</code>

2. Install OpenCV

<code>pip install OpenCV-python</code>

3. Load image

<code>img = cv2.imread('timg.jpg')</code>

4. Resize image

<code>rows, cols, channels = img.shape
print(rows, cols, channels)
img = cv2.resize(img, None, fx=0.5, fy=0.5)
rows, cols, channels = img.shape
print(rows, cols, channels)</code>

5. Display and convert to HSV, define color range

<code>cv2.imshow('img', img)
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
cv2.imshow('hsv', hsv)
lower_blue = np.array([90, 70, 90])
upper_blue = np.array([110, 255, 255])</code>

6. Create binary mask

<code>mask = cv2.inRange(hsv, lower_blue, upper_blue)
cv2.imshow('Mask', mask)</code>

7. Erosion

<code>erosion = cv2.erode(mask, None, iterations=1)
cv2.imshow('erosion', erosion)</code>

8. Dilation

<code>dilation = cv2.dilate(mask, None, iterations=1)
cv2.imshow('dilation', dilation)</code>

9. Replace background color where mask is white

<code>for i in range(rows):
    for j in range(cols):
        if dilation[i, j] == 255:
            img[i, j] = (0, 0, 255)  # BGR (red)
cv2.imshow('res', img)</code>

10. Save result and clean up

<code>cv2.imwrite('ting.png', img)
cv2.waitKey(0)  # wait indefinitely
cv2.destroyAllWindows()  # close all windows</code>

The script processes the input image, isolates the desired background color, replaces it with red (or any chosen color), and saves the final image, providing a quick solution for generating ID photos with custom backgrounds.