Breast cancer detection using MRF-based probable texture feature and decision-level fusion-based classification using HMM on thermography images
作者：
Highlights：
• We propose a two-stage breast cancer detection algorithm by decision-level fusion.
• We tried to improve false accept of previous algorithms by our proposed algorithm.
• We used Hidden Markov Model as a fusion algorithm to fuse primitive decisions.
• We propose a novel texture feature based on Markov Random Field model.
• To extract color and edge information of images, we modified Local Binary Pattern.
摘要
•We propose a two-stage breast cancer detection algorithm by decision-level fusion.•We tried to improve false accept of previous algorithms by our proposed algorithm.•We used Hidden Markov Model as a fusion algorithm to fuse primitive decisions.•We propose a novel texture feature based on Markov Random Field model.•To extract color and edge information of images, we modified Local Binary Pattern.
论文关键词：Breast cancer,Local Binary Pattern,MRF-based probable texture feature,Hidden Markov Model,Decision-level fusion
论文评审过程：Received 14 July 2014, Revised 16 August 2015, Accepted 15 September 2015, Available online 28 September 2015, Version of Record 27 November 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.09.009
