基于灰度共生矩阵的迷彩图像相似度研究

Research on Similarity of Camouflage Images Based on Gray Level Co-occurrence Matrix

针对迷彩伪装中图案斑点与颜色在相似度效果评估时的权重问题,通过迷彩设计软件生成多种林地和海洋背景四色迷彩,运用灰度共生矩阵计算图像相似度,采用控制变量法研究斑点图案和背景颜色对图像相似度的影响规律。研究结果表明:灰度共生矩阵包含图像的亮度(颜色)信息和分布(纹理)信息,可以综合评价迷彩图像的相似度;在迷彩设计中当图像中的斑点形状尺寸相同时,改变图像的颜色,图像相似度为0.5左右,当颜色相同,改变图像中的斑点形状尺寸,图像相似度达到0.9以上,说明迷彩伪装中颜色权重高于斑点图案权重,可为变形迷彩设计提供依据和指导。

Aiming at the weight problem of pattern spot and color in the evaluation of similarity effect in camouflage,this paper uses camouflage design software to generate a variety of forest land and ocean background four color camouflage,uses the gray level co-occurrence matrix(GLCM)to calculate the image similarity,and uses the control variates method to study the influence of pattern spot and background color on image similarity.The research results indicate that the GLCM contains both brightness(color)and distribution(texture)information of the image,which can comprehensively evaluate the similarity of camouflage images;In camouflage design,when the shape and size of the spots in the image are the same,the color of the image is changed,and the image similarity is about 0.5.When the color is the same,the shape and size of the spots in the image are changed,and the image similarity reaches above 0.9.This indicates that the color weight in camouflage is higher than the weight in the spot pattern,which can provide basis and guidance for deformation camouflage design.