{"title":"Infrared and Visible Image Fusion Based on Multi-scale Decomposition and Texture Preservation Model","authors":"Yingmei Zhang, H. Lee","doi":"10.1109/ICCEAI52939.2021.00067","DOIUrl":null,"url":null,"abstract":"The Infrared and visible image fusion technique is to generate an integrated image that can simultaneously preserve more texture information and thermal target from the raw images. To achieve this goal, a new infrared and visible fusion based on a multi-scale decomposition and texture preservation model is proposed. First, the base layers and detail layers images are obtained through a novel multi-scale decomposition method. Then, an adaptive saliency weighting rule is designed to obtain the fused base image. To maintain important image information from the raw images as much as possible, a texture preservation model is present. Specifically, we first apply a “max-absolute” rule to obtain pre-fused images and then calculate a Frobenius norm operator between pre-fused images and the target detail fusion image. Finally, the merged image can be obtained through an add operator. Experimental results show that compared with other state-of-the-art fusion methods, our method can preserve the texture details and infrared targets from the original images in the fusion image in terms of subjective effects and objective indicators.","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/ICCEAI52939.2021.00067","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The Infrared and visible image fusion technique is to generate an integrated image that can simultaneously preserve more texture information and thermal target from the raw images. To achieve this goal, a new infrared and visible fusion based on a multi-scale decomposition and texture preservation model is proposed. First, the base layers and detail layers images are obtained through a novel multi-scale decomposition method. Then, an adaptive saliency weighting rule is designed to obtain the fused base image. To maintain important image information from the raw images as much as possible, a texture preservation model is present. Specifically, we first apply a “max-absolute” rule to obtain pre-fused images and then calculate a Frobenius norm operator between pre-fused images and the target detail fusion image. Finally, the merged image can be obtained through an add operator. Experimental results show that compared with other state-of-the-art fusion methods, our method can preserve the texture details and infrared targets from the original images in the fusion image in terms of subjective effects and objective indicators.