矢量量化的混合学习方法及其在图像压缩中的应用

Noritaka Shigei, H. Miyajima, M. Maeda, S. Fukumoto
{"title":"矢量量化的混合学习方法及其在图像压缩中的应用","authors":"Noritaka Shigei, H. Miyajima, M. Maeda, S. Fukumoto","doi":"10.1109/ISSPA.2003.1224859","DOIUrl":null,"url":null,"abstract":"Neural networks for vector quantization such as K-means, neural-gas (NG) network and Kohonen's self-organizing map (SOM) is proposed. K-means, which is a \"hard-max\" approach, converges very fast, but it easily falls into local minima. On the other hand, the NG and SOM methods, which are \"soft-max\" approaches, are good at the global search ability. Though NG and SOM exhibit better performance in coming close to the optimum than K-means, they converge slower than K-means. In order to the drawbacks that exist when K-means, NG and SOM are used individually, we have developed hybrid methods such as NG-K and SOM-K. This paper investigates the effectiveness of NG- K and SOM-K in an image compression application. Our simulation results show that NG-K and SOM-K have good scalability to the number of weight vectors.","PeriodicalId":264814,"journal":{"name":"Seventh International Symposium on Signal Processing and Its Applications, 2003. Proceedings.","volume":"82 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2003-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Hybrid learning methods for vector quantization and its to image compression\",\"authors\":\"Noritaka Shigei, H. Miyajima, M. Maeda, S. Fukumoto\",\"doi\":\"10.1109/ISSPA.2003.1224859\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Neural networks for vector quantization such as K-means, neural-gas (NG) network and Kohonen's self-organizing map (SOM) is proposed. K-means, which is a \\\"hard-max\\\" approach, converges very fast, but it easily falls into local minima. On the other hand, the NG and SOM methods, which are \\\"soft-max\\\" approaches, are good at the global search ability. Though NG and SOM exhibit better performance in coming close to the optimum than K-means, they converge slower than K-means. In order to the drawbacks that exist when K-means, NG and SOM are used individually, we have developed hybrid methods such as NG-K and SOM-K. This paper investigates the effectiveness of NG- K and SOM-K in an image compression application. Our simulation results show that NG-K and SOM-K have good scalability to the number of weight vectors.\",\"PeriodicalId\":264814,\"journal\":{\"name\":\"Seventh International Symposium on Signal Processing and Its Applications, 2003. Proceedings.\",\"volume\":\"82 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2003-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Seventh International Symposium on Signal Processing and Its Applications, 2003. Proceedings.\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISSPA.2003.1224859\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Seventh International Symposium on Signal Processing and Its Applications, 2003. Proceedings.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISSPA.2003.1224859","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

提出了K-means、Neural -gas (NG)网络和Kohonen’s自组织映射(SOM)等矢量量化神经网络。K-means是一种“硬最大值”方法,收敛速度非常快,但很容易陷入局部最小值。另一方面,作为“软最大”方法的NG和SOM方法具有较好的全局搜索能力。虽然NG和SOM在接近最优方面表现出比K-means更好的性能,但它们的收敛速度比K-means慢。为了克服单独使用K-means、NG和SOM时存在的缺点,我们开发了NG- k和SOM- k等混合方法。本文研究了NG- K和SOM-K在图像压缩应用中的有效性。仿真结果表明,NG-K和SOM-K对权向量的数量具有良好的可扩展性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Hybrid learning methods for vector quantization and its to image compression
Neural networks for vector quantization such as K-means, neural-gas (NG) network and Kohonen's self-organizing map (SOM) is proposed. K-means, which is a "hard-max" approach, converges very fast, but it easily falls into local minima. On the other hand, the NG and SOM methods, which are "soft-max" approaches, are good at the global search ability. Though NG and SOM exhibit better performance in coming close to the optimum than K-means, they converge slower than K-means. In order to the drawbacks that exist when K-means, NG and SOM are used individually, we have developed hybrid methods such as NG-K and SOM-K. This paper investigates the effectiveness of NG- K and SOM-K in an image compression application. Our simulation results show that NG-K and SOM-K have good scalability to the number of weight vectors.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信