{"title":"局部低秩混合正则化的高光谱图像融合","authors":"Zhaoyang Liu, Mingxi Ma, Zhaoming Wu","doi":"10.1109/ICCEAI52939.2021.00022","DOIUrl":null,"url":null,"abstract":"Hyperspectral images usually have higher spectral resolution but lower spatial resolution, compared with the multispectral images. Low spatial resolution brings difficulties to the practical applications of hyperspectral images. Therefore, to get high spatial resolution hyperspectral image, it is very important to fuse low spatial resolution hyperspectral image with the high spatial resolution multispectral image in the same scene. In this paper, we propose a hybrid regularization model by integrating sparse prior, local low-rank regularization and total variation based on l2 norm to reconstruct high spatial resolution hyperspectral images. In addition, we design an alternating direction method of multipliers (ADMM) to solve it. The experimental results show the superiority and competitiveness of our method over the state-of-the-art methods.","PeriodicalId":331409,"journal":{"name":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Hyperspectral image fusion by hybrid regularizations with local low-rank\",\"authors\":\"Zhaoyang Liu, Mingxi Ma, Zhaoming Wu\",\"doi\":\"10.1109/ICCEAI52939.2021.00022\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Hyperspectral images usually have higher spectral resolution but lower spatial resolution, compared with the multispectral images. Low spatial resolution brings difficulties to the practical applications of hyperspectral images. Therefore, to get high spatial resolution hyperspectral image, it is very important to fuse low spatial resolution hyperspectral image with the high spatial resolution multispectral image in the same scene. In this paper, we propose a hybrid regularization model by integrating sparse prior, local low-rank regularization and total variation based on l2 norm to reconstruct high spatial resolution hyperspectral images. In addition, we design an alternating direction method of multipliers (ADMM) to solve it. The experimental results show the superiority and competitiveness of our method over the state-of-the-art methods.\",\"PeriodicalId\":331409,\"journal\":{\"name\":\"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-08-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICCEAI52939.2021.00022\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCEAI52939.2021.00022","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Hyperspectral image fusion by hybrid regularizations with local low-rank
Hyperspectral images usually have higher spectral resolution but lower spatial resolution, compared with the multispectral images. Low spatial resolution brings difficulties to the practical applications of hyperspectral images. Therefore, to get high spatial resolution hyperspectral image, it is very important to fuse low spatial resolution hyperspectral image with the high spatial resolution multispectral image in the same scene. In this paper, we propose a hybrid regularization model by integrating sparse prior, local low-rank regularization and total variation based on l2 norm to reconstruct high spatial resolution hyperspectral images. In addition, we design an alternating direction method of multipliers (ADMM) to solve it. The experimental results show the superiority and competitiveness of our method over the state-of-the-art methods.