摘要
针对煤矿井下图像对比度小、纹理不清晰和数据量大等问题,根据各向异性扩散在图像处理中具有良好的边缘保持与增强的作用,提出一种基于各向异性扩散的图像分割算法.首先在图像分割前对原图像进行各向异性扩散运算,在消除原图像噪声的同时,更好地划分了图像的边缘和纹理区域;然后提取图像的纹理特性运用到聚类算法中,从而对图像进行分割.实验证明:与未经扩散处理的分割算法相比,基于各向异性扩散的图像分割算法不仅改善了分割效果,而且提高了计算速度.
According to the characteristics of image including low contrast, unclear texture, large quantity of data etc in coal mine underground, based on the good effects of anisotropic diffusion on preservation and enhancement of edge in image processing, an image segmentation algorithm based on anisotropic diffusion is presented. Firstly, the anisotropic dif- fusion algorithm is implemented on the original image, to eliminate the noise and determine the edge and texture district better. And then clustering algorithm is applied to the texture characteristic of image to segment the image. The experimental results show that, compared with the algorithm without diffusion, the algorithm based on anisotropic diffusion improves the segmentation effect and computation speed.
出处
《天津师范大学学报(自然科学版)》
CAS
2013年第1期35-37,共3页
Journal of Tianjin Normal University:Natural Science Edition
基金
安徽省高等学校省级自然科学研究项目(KJ2011Z354)
淮南市科技计划资助项目(2011A08017)