Python+OpenCV：图像Harris角点检测(Harris Corner Detection)

最新推荐文章于 2025-03-05 22:01:12 发布

机器视觉001

最新推荐文章于 2025-03-05 22:01:12 发布

阅读量2.1k

点赞数

分类专栏： Python OpenCV 文章标签： opencv python 计算机视觉

本文链接：https://blog.csdn.net/liubing8609/article/details/109966883

版权

Python 同时被 2 个专栏收录

104 篇文章

订阅专栏

OpenCV

82 篇文章

订阅专栏

本文介绍使用Python和OpenCV实现Harris角点检测的方法，包括理论基础与实践代码。通过对不同图像应用Harris角点检测算法，展示了如何定位图像中的关键角点，并进一步进行亚像素级别的精确检测。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

Python+OpenCV：图像Harris角点检测(Harris Corner Detection)

理论

corners are regions in the image with large variation in intensity in all the directions.

It basically finds the difference in intensity for a displacement of (u,v) in all directions. This is expressed as below:

The window function is either a rectangular window or a Gaussian window which gives weights to pixels underneath.

A score, basically an equation, which determines if a window can contain a corner or not.

where

det(M)=λ1λ2
trace(M)=λ1+λ2
λ1 and λ2 are the eigenvalues of M

So the magnitudes of these eigenvalues decide whether a region is a corner, an edge, or flat.

When |R| is small, which happens when λ1 and λ2 are small, the region is flat.
When R<0, which happens when λ1>>λ2 or vice versa, the region is edge.
When R is large, which happens when λ1 and λ2 are large and λ1∼λ2, the region is a corner.

It can be represented in a nice picture as follows:

Harris Corner Detector in OpenCV

####################################################################################################
# 图像Harris角点检测(Harris Corner Detection)
def lmc_cv_image_harris_corner_detection():
    """
        函数功能: 图像Harris角点检测(Harris Corner Detection)。
    """

    stacking_images = []
    image_file_name = ['D:/99-Research/Python/Image/CheckerBoard.jpg',
                       'D:/99-Research/Python/Image/CheckerBoard01.jpg',
                       'D:/99-Research/Python/Image/CheckerBoard02.jpg',
                       'D:/99-Research/Python/Image/CheckerBoard07.jpg']
    for i in range(len(image_file_name)):
        # 读取图像
        image = lmc_cv.imread(image_file_name[i])
        image = lmc_cv.cvtColor(image, lmc_cv.COLOR_BGR2RGB)
        result_image = image.copy()
        gray_image = lmc_cv.cvtColor(image, lmc_cv.COLOR_BGR2GRAY)

        gray_image = np.float32(gray_image)
        corner_harris_image = lmc_cv.cornerHarris(gray_image, 2, 3, 0.04)
        # result is dilated for marking the corners, not important
        corner_harris_image = lmc_cv.dilate(corner_harris_image, None)
        # Threshold for an optimal value, it may vary depending on the image.
        result_image[corner_harris_image > 0.01 * corner_harris_image.max()] = [255, 0, 0]

        # stacking images side-by-side
        stacking_image = np.hstack((image, result_image))
        stacking_images.append(stacking_image)

    # 显示图像
    for i in range(len(stacking_images)):
        pyplot.figure('Harris Corner Detection %d' % (i+1))
        pyplot.subplot(1, 1, 1)
        pyplot.imshow(stacking_images[i], 'gray')
        pyplot.xticks([])
        pyplot.yticks([])
    pyplot.show()

    # 根据用户输入保存图像
    if ord("q") == (lmc_cv.waitKey(0) & 0xFF):
        # 销毁窗口
        pyplot.close('all')
    return

Corner with SubPixel Accuracy

####################################################################################################
# 图像亚像素Harris角点检测(Harris Corner Detection with SubPixel Accuracy)
def lmc_cv_image_harris_corner_detection_subpixel():
    """
        函数功能: 图像亚像素Harris角点检测(Harris Corner Detection with SubPixel Accuracy)。
    """

    stacking_images = []
    image_file_name = ['D:/99-Research/Python/Image/CheckerBoard.jpg',
                       'D:/99-Research/Python/Image/CheckerBoard02.jpg',
                       'D:/99-Research/Python/Image/CheckerBoard04.jpg',
                       'D:/99-Research/Python/Image/CheckerBoard07.jpg']
    for i in range(len(image_file_name)):
        # 读取图像
        image = lmc_cv.imread(image_file_name[i])
        image = lmc_cv.cvtColor(image, lmc_cv.COLOR_BGR2RGB)
        result_image = image.copy()
        gray_image = lmc_cv.cvtColor(image, lmc_cv.COLOR_BGR2GRAY)

        # find Harris corners
        gray_image = np.float32(gray_image)
        corner_harris_image = lmc_cv.cornerHarris(gray_image, 2, 3, 0.04)
        corner_harris_image = lmc_cv.dilate(corner_harris_image, None)
        ret, corner_harris_image = lmc_cv.threshold(corner_harris_image, 0.01 * corner_harris_image.max(), 255, 0)
        corner_harris_image = np.uint8(corner_harris_image)
        # find centroids
        ret, labels, stats, centroids = lmc_cv.connectedComponentsWithStats(corner_harris_image)
        # define the criteria to stop and refine the corners
        criteria = (lmc_cv.TERM_CRITERIA_EPS + lmc_cv.TERM_CRITERIA_MAX_ITER, 100, 0.001)
        corners = lmc_cv.cornerSubPix(gray_image, np.float32(centroids), (5, 5), (-1, -1), criteria)
        # Now draw them
        res = np.hstack((centroids, corners))
        res = np.int0(res)
        result_image[res[:, 1], res[:, 0]] = [255, 0, 0]
        result_image[res[:, 3], res[:, 2]] = [0, 255, 0]

        # stacking images side-by-side
        stacking_image = np.hstack((image, result_image))
        stacking_images.append(stacking_image)

    # 显示图像
    for i in range(len(stacking_images)):
        pyplot.figure('Harris Corner Detection  with SubPixel Accuracy %d' % (i+1))
        pyplot.subplot(1, 1, 1)
        pyplot.imshow(stacking_images[i], 'gray')
        pyplot.xticks([])
        pyplot.yticks([])
    pyplot.show()

    # 根据用户输入保存图像
    if ord("q") == (lmc_cv.waitKey(0) & 0xFF):
        # 销毁窗口
        pyplot.close('all')
    return