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}
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.