数字图像处理_冈萨雷顿纹理分割

YandexOj

已于 2022-06-24 12:26:44 修改

阅读量126

点赞数

文章标签： opencv python

于 2022-06-10 19:22:10 首次发布

本文链接：https://blog.csdn.net/weixin_47989730/article/details/125226933

版权

数字图像处理_第三版_冈萨雷斯

9.34：

效果示意：

# -*- coding=GBK -*-
import cv2 as cv
def boundary(image):
    kernel = cv.getStructuringElement(cv.MORPH_RECT, (4, 4))
    img_grad1 = cv.morphologyEx(image, cv.MORPH_GRADIENT, kernel)
    img_grad2 = cv.subtract(cv.dilate(image, kernel), cv.erode(image, kernel))
    cv.imshow("boundary",img_grad2)
    return img_grad1


def open_image(image):
    kernel = cv.getStructuringElement(cv.MORPH_ELLIPSE, (45, 45))
    # shape: MORPH_RECT(矩阵)，MORPH_CORSS(交叉形状)，MORPH_ELLIPSE(椭圆) ;
    binary = cv.morphologyEx(image, cv.MORPH_OPEN, kernel)
    cv.imshow("Open", binary)
    return binary


def close_image(image):
    kernel = cv.getStructuringElement(cv.MORPH_ELLIPSE, (25, 25))
    binary = cv.morphologyEx(image, cv.MORPH_CLOSE, kernel)
    cv.imshow("Close", binary)
    return binary


src = cv.imread("point.jpg")
cv.imshow("before", src)
gray = cv.cvtColor(src, cv.COLOR_BGR2GRAY)
ret, binary = cv.threshold(gray, 0, 255, cv.THRESH_BINARY | cv.THRESH_OTSU)
tmp1 = close_image(binary)
tmp2 = open_image(tmp1)
tmp3 = boundary(tmp2)
bound=cv.cvtColor(tmp3, cv.COLOR_GRAY2BGR)
res = cv.add(src , bound)
cv.imshow("res",res)
cv.waitKey(0)
cv.destroyAllWindows()

参考：

图像开闭操作：python实现opencv学习二十：图像的开闭操作_云散月明晴的博客-CSDN博客_开操作python

getStructuringElement()，morphologyEx()参考：OpenCV 中 getStructuringElement() 与 morphologyEx() 函数用法_程序员大飞1的博客-CSDN博客_getstructuringelement()函数作用

图像加法：OpenCV的图像加法运算_半濠春水的博客-CSDN博客_opencv图像相加

图像的形态梯度：【千律】OpenCV基础：图像的形态学梯度_奇恪的博客-CSDN博客_opencv形态学梯度