1、https://www.mathworks.com/help/images/ref/graycomatrix.html
【灰度共生矩阵原理,步距、方向原理等说明,及MATLAB代码】
2、https://prism.ucalgary.ca/handle/1880/51900
【一本入门教程,可以下载为pdf小册子,里面有详细的范例《GLCM Texture: A Tutorial v. 3.0 March 2017》】
3、Haralick, R.M., K. Shanmugan, and I. Dinstein, "Textural Features for Image Classification", IEEE Transactions on Systems, Man, and Cybernetics, Vol. SMC-3, 1973, pp. 610-621. 【方法提出者】
C++版:GLCM Library - A GLCM algorithms's achieved by OpenCV 2.4.9
https://github.com/upcAutoLang/GLCM-OpenCV
Python版: Gray Level Co-occurrence Matrix (GLCM) with python library
https://github.com/LendisFabri/Gray-Level-Cooccurrence-Matrix
OpenCV:以下是OpenCV自带函数的计算, 头文件:
只需要调用下面俩个函数完成计算后调用get函数就能获得相应的值了。
CV_IMPL CvGLCM*
cvCreateGLCM( const IplImage* srcImage,
int stepMagnitude,
const int* srcStepDirections,/* should be static array..
or if not the user should handle de-allocation */
int numStepDirections,
int optimizationType )
从图像到灰度共生矩阵。
static void
icvCreateGLCMDescriptors_AllowDoubleNest( CvGLCM* destGLCM, int matrixIndex )
灰度共生矩阵大部分值的计算。(不包括能量能量方差,这是cvGetGLCMDescriptorStatistics这个函数给定的,另外能量均值CV_IMPL double cvGetGLCMDescriptor和上面那个函数都有提供。#define 定义的灰度特征都由CV_IMPL double cvGetGLCMDescriptor获得。要注意和创建时候设定的步长一致)。
另外,后面自己改造原函数,支持标记区域的函数。
参考链接:
Allocation error with 'cvCreateGLCM' method
https://answers.opencv.org/question/41/allocation-error-with-cvcreateglcm-method/
GLCM in opencv2 (Gray-Level Co-occurrence Matrices)
https://stackoverflow.com/questions/24875929/glcm-in-opencv2-gray-level-co-occurrence-matrices
// your input image
cv::Mat grayIm = [in_image CVGrayscaleMat];
// create a legacy image
IplImage pGray = grayIm;
// call function
CvGLCM* glcm = cvCreateGLCM(&pGray, 1, NULL, 4, CV_GLCM_OPTIMIZATION_LUT);
Opencv 自带cvtexture.cpp中灰度共生矩阵函数和参数
https://bbs.csdn.net/topics/360141907
OpenCV 1.0 cvtexture.cpp bug 修正 内存错误
https://blog.csdn.net/frankyhit/article/details/6154187
纹理特征提取方法:LBP, 灰度共生矩阵
https://blog.csdn.net/ajianyingxiaoqinghan/article/details/71552744
灰度共生矩阵的原理及实现(特征提取)-OpenCV
https://blog.csdn.net/qq_37059483/article/details/78292869
图像纹理——灰度共生矩阵
https://blog.csdn.net/guanyuqiu/article/details/53117507