vlambda博客
学习文章列表

openCV全局阈值分割

全局阈值(又称简单阈值)


顾名思义,整幅图像采用一个阈值,当某一点像素值大于阈值时,赋予该点一个新值,否则就赋予另外一种值。采用函数cv2.threshold()实现。


cv2.threshold(src, thresh, maxval, type)


参数:

  1. src:输入的图像

  2. thresh:图像分割所用的阈值(threshold value)

  3. maxval:当阈值类型(thresholding type)采用cv2.THRESH_BINARY和cv2.THRESH_BINARY_INV时像素点被赋予的新值

  4. type:包括5种类型:

  • cv2.THRESH_BINARY(当图像某点像素值大于thresh(阈值)时赋予maxval,反之为0。注:最常用)

  • cv2.THRESH_BINARY_INV(当图像某点像素值小于thresh时赋予maxval,反之为0)

  • cv2.THRESH_TRUNC(当图像某点像素值大于thresh时赋予thresh,反之不变。注:虽然maxval没用了,但是调用函数不能省略)

  • cv2.THRESH_TOZERO(当图像某点像素值小于thresh时赋予0,反之不变)

  • cv2.THRESH_TOZERO_INV(当图像某点像素值大于thresh时赋予0,反之不变)


返回值:

  • retval:设定的thresh值

  • dst:阈值分割后的图像


下面通过一段代码进行直观认识:

import cv2import matplotlib.pyplot as plt#设定阈值thresh=130#载入原图,并转化为灰度图像img_original=cv2.imread('E:\ShannonT\\notebook workspace\\images\\4.24.tudou.png',0)img_original=cv2.resize(img_original,(0,0),fx=0.3,fy=0.3)#采用5种阈值类型(thresholding type)分割图像retval1,img_binary=cv2.threshold(img_original,thresh,255,cv2.THRESH_BINARY)retval2,img_binary_invertion=cv2.threshold(img_original,thresh,255,cv2.THRESH_BINARY_INV)retval3,img_trunc=cv2.threshold(img_original,thresh,255,cv2.THRESH_TRUNC)retval4,img_tozero=cv2.threshold(img_original,thresh,255,cv2.THRESH_TOZERO)retval5,img_tozero_inversion=cv2.threshold(img_original,thresh,255,cv2.THRESH_TOZERO_INV)#采用plt.imshow()显示图像imgs=[img_original,img_binary,img_binary_invertion,img_trunc,img_tozero,img_tozero_inversion]titles=['original','binary','binary_inv','trunc','tozero','tozero_inv']for i in range(6): plt.subplot(2,3,i+1) plt.imshow(imgs[i],'gray') plt.title(titles[i]) plt.xticks([]) plt.yticks([])plt.show()


结果显示如下:


下面通过加入滑动条动态改变阈值参数,代码如下:

import cv2import numpy as npimport matplotlib.pyplot as plt#载入原图,转化为灰度图像,并通过cv2.resize()等比调整图像大小img_original=cv2.imread('E:\ShannonT\\notebook workspace\\images\\4.24.tudou.png',0)img_original=cv2.resize(img_original,(0,0),fx=0.3,fy=0.3)#初始化阈值,定义全局变量imgsthresh=130imgs=0#创建滑动条回调函数,参数thresh为滑动条对应位置的数值def threshold_segmentation(thresh): #采用5种阈值类型(thresholding type)分割图像 retval1,img_binary=cv2.threshold(img_original,thresh,255,cv2.THRESH_BINARY) retval2,img_binary_invertion=cv2.threshold(img_original,thresh,255,cv2.THRESH_BINARY_INV) retval3,img_trunc=cv2.threshold(img_original,thresh,255,cv2.THRESH_TRUNC) retval4,img_tozero=cv2.threshold(img_original,thresh,255,cv2.THRESH_TOZERO) retval5,img_tozero_inversion=cv2.threshold(img_original,thresh,255,cv2.THRESH_TOZERO_INV) #由于cv2.imshow()显示的是多维数组(ndarray),因此我们通过np.hstack(数组水平拼接) #和np.vstack(竖直拼接)拼接数组,达到同时显示多幅图的目的 img1=np.hstack([img_original,img_binary,img_binary_invertion]) img2=np.hstack([img_trunc,img_tozero,img_tozero_inversion]) global imgs imgs=np.vstack([img1,img2])#新建窗口cv2.namedWindow('Images')#新建滑动条,初始位置为130cv2.createTrackbar('threshold value','Images',130,255,threshold_segmentation)#第一次调用函数threshold_segmentation(thresh)#显示图像while(1): cv2.imshow('Images',imgs) if cv2.waitKey(1)==ord('q'): breakcv2.destroyAllWindows()


视频演示如下: