How to extract two circular areas from a 9000x7000 pixel image using Python and OpenCV?

Python and OpenCV efficiently extract two circular areas in 9000x7000 pixel images

Processing ultra-high resolution images (such as 9000x7000 pixels) and extracting specific shapes (such as circles) from them is a common challenge in image processing and computer vision. This article provides a solution using Python and OpenCV libraries to efficiently and accurately extract target circular areas.

The problem with the existing code is that there are too many circles detected and it is impossible to accurately select the two circle areas required. For improvement, we will adopt the following strategies:

1. Image Preprocessing: Scaling and Noise Reduction : First, to improve processing efficiency, we reduce the original image to the right size. At the same time, a Gaussian blur filter is applied to reduce image noise, thereby improving the accuracy of circular detection.

 import cv2
import numpy as np

image_path = r"c:\users\17607\desktop\smls pictures\pic_20231122151507973.bmp"

# Read image img = cv2.imread(image_path)

# Zoom the image (adjust the zoom ratio according to the actual situation)
scale_percent = 10 # Scale to 1/10 of the original image
width = int(img.shape[1] / scale_percent)
height = int(img.shape[0] / scale_percent)
dim = (width, height)
resized_img = cv2.resize(img, dim, interpolation=cv2.INTER_AREA)

# grayscale conversion gray = cv2.cvtColor(resized_img, cv2.COLOR_BGR2GRAY)

# GaussianBlurred = cv2.GaussianBlur(gray, (5, 5), 0)

2. Edge detection: Canny algorithm : Use the Canny edge detection algorithm to extract image edge information and prepare for subsequent circular detection.

 # Canny edge detection edges = cv2.Canny(blurred, 50, 150)

3. Circle detection: Hough Transform : Use Hough Circle transformation to detect circles in images. The key is parameter adjustments to ensure that only the two circles we need are detected. Here we filter according to the radius of the circle and select the two largest circles.

 # HoughCircle Transform Circles = cv2.HoughCircles(edges, cv2.HOUGH_GRADIENT, 1, 40, param1=50, param2=30, minRadius=0, maxRadius=0)

If circles is not None:
    circles = np.uint16(np.around(circles))
    # Select two largest circles = circles[0, :]
    circles = circles[np.argsort(circles[:, 2])[::-1][:2]] # Select two circles with the largest radius for i in circles:
        center_x, center_y, radius = i
        # Draw circle cv2.circle(resized_img, (center_x, center_y), radius, (0, 0, 255), 2)
        cv2.circle(resized_img, (center_x, center_y), 2, (255, 0, 0), 3)

    cv2.imshow("Detected Circles", resized_img)
    cv2.waitKey(0)
    cv2.destroyAllWindows()

Through the above steps, we can effectively extract the two largest circular regions from high-resolution images and verify them by visualization results. It should be noted that the parameters of scale_percent and Hough transformation need to be adjusted according to the actual image to achieve the best detection effect. If two circles are of similar size, a more refined choice may be required based on the center coordinates or other features.

The above is the detailed content of How to extract two circular areas from a 9000x7000 pixel image using Python and OpenCV?. For more information, please follow other related articles on the PHP Chinese website!