煤岩图像边界的K-means识别算法  

Coal-Rock Image Boundary Using K-means Recognition Algorithm

作　　者：江静[1,2] 张雪松[3,4] 

机构地区：[1]北京联合大学信息学院,北京100101 [2]华北科技学院机电学院,河北三河065201 [3]中国科学院光电研究院,北京100094 [4]中国电子科技集团光电研究院光电信息控制和安全技术重点实验室,河北三河065201

出　　处：《煤炭工程》2015年第8期106-109,共4页Coal Engineering

基　　金：国家自然科学基金重点项目(51134024);国家高技术研究发展计划(863)项目(2012AA062203)

摘　　要：提出了一种基于K-means的煤岩边界提取算法。运用小波变换提取出煤岩图像中大尺度特征,以剔除其杂散纹理和噪声对后续聚类过程的影响;采用K-means算法完成煤岩边界分布的聚类;并利用Canny算子提取出二值聚类图像的边缘,引入图像形态学中的腐蚀与膨胀运算,关联相邻分段边界并平滑边界。仿真图像与真实煤岩边界图像的实验结果表明,与直接K-means和Mean shift等图像分割算法相比,该算法能够更为精确完整地提取出真实的煤岩分界。A K - means based algorithm was proposed to identify the coal - rock image boundary. Wavelet transform was used to extract large - scale features in coal - rock images, eliminate spurious textures and imaging noise and facilitate the subsequent clustering process. The K - means algorithm was applied to complete the clustering of coal - rock image boundary distribution. Finally, image edges were extracted from the clustered binary image using Canny operator, and two image morphological operators, erosion and dilation, were used to connect adjacent segments and smooth the boundaries. The experimental results of simulated and real images show that, the algorithm is accurate to extract the true coal - rock boundaries, compared with the direct K - means and Mean - shift image segmentation algorithms. The proposed algorithm is more promising to the autonomous long arm mining applications.

关 键 词：煤岩界面识别 K-MEANS CANNY边缘检测 腐蚀与膨胀 

分 类 号：TD679[矿业工程—矿山机电]