Multi- pattern analysis: A case study in image classification

详细信息	查看全文 | 推荐本文 |

• 作者：Ricardo Fabbri^b ; ^{rfabbri@gmail.com} ; Wesley N. Gonç ; alves^a ; ^{wnunes@ursa.ifsc.usp.br} ; Francisco J.P. Lopes^c ; ^{fjplopes@gmail.com} ; Odemir M. Bruno^a ; ^{bruno@ifsc.usp.br}
• 关键词：Image pattern classification ; Texture ; Tsallis entropy ; Non-additive entropy
• 刊名：Physica A
• 出版年：2012
• 期刊代码：138_03784371
• 类别：et
• 出版时间：1 October, 2012
• 卷：391
• 期：19
• 页码：4487-4496
• 文件大小：831 K

摘要

This paper compares the effectiveness of the Tsallis entropy over the classic Boltzmann-Gibbs-Shannon entropy for general pattern recognition, and proposes a multi- approach to improve pattern analysis using entropy. A series of experiments were carried out for the problem of classifying image patterns. Given a dataset of 40 pattern classes, the goal of our image case study is to assess how well the different entropies can be used to determine the class of a newly given image sample. Our experiments show that the Tsallis entropy using the proposed multi- approach has great advantages over the Boltzmann-Gibbs-Shannon entropy for pattern classification, boosting image recognition rates by a factor of 3. We discuss the reasons behind this success, shedding light on the usefulness of the Tsallis entropy and the multi- approach.