OpenCV python 获取轮廓的点位置

OpenCV python 获取轮廓的点位置

处理图片:[cc.jpg]

import cv2
import numpy as np


def get_contour(img):
    """获取连通域

    :param img: 输入图片
    :return: 最大连通域
    """
    # 灰度化, 二值化, 连通域分析
    img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

    ret, img_bin = cv2.threshold(img_gray, 127, 255, cv2.THRESH_BINARY)

    img_contour, contours, hierarchy = cv2.findContours(img_bin, cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)

    return img_gray, contours[0]


def main():

    # 1.导入图片
    img_src = cv2.imread("cc.jpg")

    # 2.获取连通域
    img_gray, contour = get_contour(img_src)

    # 3.轮廓外边缘点
    mask_out_edge = np.zeros(img_gray.shape, np.uint8)
    cv2.drawContours(mask_out_edge, [contour], 0, 255, 2)

    pixel_point1 = cv2.findNonZero(mask_out_edge)
    print("pixel point shape:", pixel_point1.shape)
    print("pixel point:\n", pixel_point1)

    # 4.轮廓内部
    mask_inside = np.zeros(img_gray.shape, np.uint8)
    cv2.drawContours(mask_inside, [contour], 0, 255, -1)
    pixel_point2 = cv2.findNonZero(mask_inside)

    print("pixel point2 shape:", pixel_point2.shape)
    print("pixel point2:\n", pixel_point2)

    # 5.显示图片
    cv2.imshow("img_src", img_src)
    cv2.imshow("mask_out_edge", mask_out_edge)
    cv2.imshow("mask_inside", mask_inside)

    cv2.waitKey()
    cv2.destroyAllWindows()


if __name__ == '__main__':
    main()

处理结果图片:[mask_out_edge.jpg]

处理结果图片:[mask_inside.jpg]