Jian Lin;Qiurong Yan;Quan Zou;Shida Sun;Zhen Wei;Hua Du
{"title":"用于单像素成像的变量多尺度误差补偿网络","authors":"Jian Lin;Qiurong Yan;Quan Zou;Shida Sun;Zhen Wei;Hua Du","doi":"10.1109/JPHOT.2024.3421574","DOIUrl":null,"url":null,"abstract":"Single-pixel imaging is an advanced computational imaging technique based on compressive sensing that offers higher signal-to-noise ratio and broader application scope compared to traditional imaging techniques. However, conventional reconstruction algorithms suffer from issues such as long processing time and low reconstruction accuracy during the sampling and reconstruction processes. Deep learning-based compressed reconstruction networks can circumvent the complex iterative computations of traditional algorithms and achieve fast, high-quality reconstruction. In this paper, we propose a Variational Multi-Scale Error Compensation Network (VMSE) based on variational autoencoders. VMSE designs an error compensation network to enhance the feature representation capability of the sampling reconstruction network. We employ multiple latent variables to generate error features at different scales in the intermediate layers of the error compensation network, compensating the reconstructed image. Additionally, we design a module that simultaneously learns in the spatial and frequency domains, which is used for upsampling and complementing the missing high-frequency information in the frequency domain. On the MNIST dataset, when the sampling rate is 0.025, VMSE achieved higher Peak Signal-to-Noise Ratio(PSNR) and Structural Similarity Index(SSIM) scores, especially with an SSIM score of 0.963, significantly surpassing Reconnet and DR2Net's scores of 0.930 and 0.920, respectively. This was further corroborated by practical experiments, where at low sampling rates, VMSE could reconstruct outlines more clearly compared to TVAL3.","PeriodicalId":13204,"journal":{"name":"IEEE Photonics Journal","volume":null,"pages":null},"PeriodicalIF":2.1000,"publicationDate":"2024-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10582416","citationCount":"0","resultStr":"{\"title\":\"A Variational Multi-Scale Error Compensation Network for Single-Pixel Imaging\",\"authors\":\"Jian Lin;Qiurong Yan;Quan Zou;Shida Sun;Zhen Wei;Hua Du\",\"doi\":\"10.1109/JPHOT.2024.3421574\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Single-pixel imaging is an advanced computational imaging technique based on compressive sensing that offers higher signal-to-noise ratio and broader application scope compared to traditional imaging techniques. However, conventional reconstruction algorithms suffer from issues such as long processing time and low reconstruction accuracy during the sampling and reconstruction processes. Deep learning-based compressed reconstruction networks can circumvent the complex iterative computations of traditional algorithms and achieve fast, high-quality reconstruction. In this paper, we propose a Variational Multi-Scale Error Compensation Network (VMSE) based on variational autoencoders. VMSE designs an error compensation network to enhance the feature representation capability of the sampling reconstruction network. We employ multiple latent variables to generate error features at different scales in the intermediate layers of the error compensation network, compensating the reconstructed image. Additionally, we design a module that simultaneously learns in the spatial and frequency domains, which is used for upsampling and complementing the missing high-frequency information in the frequency domain. On the MNIST dataset, when the sampling rate is 0.025, VMSE achieved higher Peak Signal-to-Noise Ratio(PSNR) and Structural Similarity Index(SSIM) scores, especially with an SSIM score of 0.963, significantly surpassing Reconnet and DR2Net's scores of 0.930 and 0.920, respectively. This was further corroborated by practical experiments, where at low sampling rates, VMSE could reconstruct outlines more clearly compared to TVAL3.\",\"PeriodicalId\":13204,\"journal\":{\"name\":\"IEEE Photonics Journal\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":2.1000,\"publicationDate\":\"2024-07-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10582416\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE Photonics Journal\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/10582416/\",\"RegionNum\":4,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Photonics Journal","FirstCategoryId":"5","ListUrlMain":"https://ieeexplore.ieee.org/document/10582416/","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
A Variational Multi-Scale Error Compensation Network for Single-Pixel Imaging
Single-pixel imaging is an advanced computational imaging technique based on compressive sensing that offers higher signal-to-noise ratio and broader application scope compared to traditional imaging techniques. However, conventional reconstruction algorithms suffer from issues such as long processing time and low reconstruction accuracy during the sampling and reconstruction processes. Deep learning-based compressed reconstruction networks can circumvent the complex iterative computations of traditional algorithms and achieve fast, high-quality reconstruction. In this paper, we propose a Variational Multi-Scale Error Compensation Network (VMSE) based on variational autoencoders. VMSE designs an error compensation network to enhance the feature representation capability of the sampling reconstruction network. We employ multiple latent variables to generate error features at different scales in the intermediate layers of the error compensation network, compensating the reconstructed image. Additionally, we design a module that simultaneously learns in the spatial and frequency domains, which is used for upsampling and complementing the missing high-frequency information in the frequency domain. On the MNIST dataset, when the sampling rate is 0.025, VMSE achieved higher Peak Signal-to-Noise Ratio(PSNR) and Structural Similarity Index(SSIM) scores, especially with an SSIM score of 0.963, significantly surpassing Reconnet and DR2Net's scores of 0.930 and 0.920, respectively. This was further corroborated by practical experiments, where at low sampling rates, VMSE could reconstruct outlines more clearly compared to TVAL3.
期刊介绍:
Breakthroughs in the generation of light and in its control and utilization have given rise to the field of Photonics, a rapidly expanding area of science and technology with major technological and economic impact. Photonics integrates quantum electronics and optics to accelerate progress in the generation of novel photon sources and in their utilization in emerging applications at the micro and nano scales spanning from the far-infrared/THz to the x-ray region of the electromagnetic spectrum. IEEE Photonics Journal is an online-only journal dedicated to the rapid disclosure of top-quality peer-reviewed research at the forefront of all areas of photonics. Contributions addressing issues ranging from fundamental understanding to emerging technologies and applications are within the scope of the Journal. The Journal includes topics in: Photon sources from far infrared to X-rays, Photonics materials and engineered photonic structures, Integrated optics and optoelectronic, Ultrafast, attosecond, high field and short wavelength photonics, Biophotonics, including DNA photonics, Nanophotonics, Magnetophotonics, Fundamentals of light propagation and interaction; nonlinear effects, Optical data storage, Fiber optics and optical communications devices, systems, and technologies, Micro Opto Electro Mechanical Systems (MOEMS), Microwave photonics, Optical Sensors.