[Python] How to find the center and radius of Earth in grayscale satellite images using OpenCV?

I have thousands of grayscale satellite images of Earth, and I want to automatically find the center (x, y) and radius of the Earth’s edge in each image.

I tried using Hough Circle Transform in OpenCV. It doesn't work because of clouds and noise in the image. I have to tune the parameters (like radius, thresholds) for each image, which is not practical for thousands of images.

# Here’s a sample of what I tried:

import cv2
import numpy as np

img = cv2.imread("earth_image.tif", cv2.IMREAD_GRAYSCALE)
blurred = cv2.bilateralFilter(img, 9, 75, 75)
edges = cv2.Canny(blurred, 50, 150)

circles = cv2.HoughCircles(edges, cv2.HOUGH_GRADIENT, dp=1.2, minDist=100,
param1=100, param2=60, minRadius=400, maxRadius=0)

if circles is not None:
circles = np.uint16(np.around(circles))
for x, y, r in circles[0, :]:
print(f"Center: ({x}, {y}), Radius: {r}")




Continue reading...