第十九天分水岭及归一化标准化

  pyrMeanShiftFiltering。这个函数严格来说并不是图像的分割，而是图像在色彩层面的平滑滤波，它可以中和色彩分布相近的颜色，平滑色彩细节，侵蚀掉面积较小的颜色区域

 

 

• 归一化（normalization）：   ，如果对输出结果范围有要求，用归一化
• 标准化（standardization）： 如果数据存在异常值和较多噪音，用标准化，可以间接通过中心化避免异常值和极端值的影响一般来说，我个人建议优先使用标准化。

 

import cv2 as cv
import numpy as np


def watershed_demo():
    # remove noise if any
    print(src.shape)
    blurred = cv.pyrMeanShiftFiltering(src, 10, 100)
    # gray\binary image
    gray = cv.cvtColor(blurred, cv.COLOR_BGR2GRAY)
    ret, binary = cv.threshold(gray, 0, 255, cv.THRESH_BINARY | cv.THRESH_OTSU)
    cv.imshow("binary-image", binary)

    # morphology operation
    kernel = cv.getStructuringElement(cv.MORPH_RECT, (3, 3))##彩色图像均值漂移
    mb = cv.morphologyEx(binary, cv.MORPH_OPEN, kernel, iterations=2)  #闭算法
    sure_bg = cv.dilate(mb, kernel, iterations=3)
    cv.imshow("mor-opt", sure_bg)

    # distance transform
    dist = cv.distanceTransform(mb, cv.DIST_L2, 3)
    dist_output = cv.normalize(dist, 0, 1.0, cv.NORM_MINMAX)  #归一化
    cv.imshow("distance-t", dist_output*50)

    ret, surface = cv.threshold(dist, dist.max()*0.6, 255, cv.THRESH_BINARY)

    surface_fg = np.uint8(surface)
    cv.imshow("surface-bin", surface_fg)
    unknown = cv.subtract(sure_bg, surface_fg)     #未知区域
    ret, markers = cv.connectedComponents(surface_fg)##连通分支

    # watershed transform
    markers = markers + 1
    markers[unknown==255] = 0
    markers = cv.watershed(src, markers=markers)  ##分水岭算法函数
    src[markers==-1] = [0, 0, 255]
    cv.imshow("result", src)



print("--------- Python OpenCV Tutorial ---------")
src = cv.imread("D:/vcprojects/images/coins_001.jpg")
cv.namedWindow("input image", cv.WINDOW_AUTOSIZE)
cv.imshow("input image", src)
watershed_demo()
cv.waitKey(0)

cv.destroyAllWindows()