Deep neural networks for single-pixel compressive video reconstruction

Antonio Lorente Mur, B. Montcel, F. Peyrin, N. Ducros
{"title":"Deep neural networks for single-pixel compressive video reconstruction","authors":"Antonio Lorente Mur, B. Montcel, F. Peyrin, N. Ducros","doi":"10.1117/12.2553326","DOIUrl":null,"url":null,"abstract":"Single-pixel imaging is a paradigm that enables the capture of an image from a single point detector using a spatial light modulator. This approach is particularly interesting for optical set-ups where pixelated arrays of detectors are either too expensive or too cumbersome (e.g., multispectral, infrared imaging). It acquires the inner product between the image of the scene and a set of user-defined patterns that are sequentially uploaded onto the spatial light modulator. Compressed data acquisition reduces the acquisition time, although it leads to an ill-posed reconstruction problem, which is very challenging for real-time applications. Recently, neural networks have emerged as competitive alternatives to traditional reconstruction methods. Neural networks are parametric models that are trained by exploiting large datasets. Their noniterative nature allows for fast reconstructions, which opens the door to real-time image reconstruction from compressed acquisition. In this study, we evaluate the different networks for static and dynamic imaging. In particular, we introduce a recurrent neural network that is designed to exploit the spatiotemporal redundancy in videos via a memory state. We validate our algorithms on simulated data from the UCF-101 dataset, with a resolution of 128x128 pixels and a compression ratio of 98%. We also show experimentally that we can resolve small spectral differences in the spectrum of human skin measured in vivo.","PeriodicalId":234467,"journal":{"name":"Unconventional Optical Imaging II","volume":"87 1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Unconventional Optical Imaging II","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1117/12.2553326","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6

Abstract

Single-pixel imaging is a paradigm that enables the capture of an image from a single point detector using a spatial light modulator. This approach is particularly interesting for optical set-ups where pixelated arrays of detectors are either too expensive or too cumbersome (e.g., multispectral, infrared imaging). It acquires the inner product between the image of the scene and a set of user-defined patterns that are sequentially uploaded onto the spatial light modulator. Compressed data acquisition reduces the acquisition time, although it leads to an ill-posed reconstruction problem, which is very challenging for real-time applications. Recently, neural networks have emerged as competitive alternatives to traditional reconstruction methods. Neural networks are parametric models that are trained by exploiting large datasets. Their noniterative nature allows for fast reconstructions, which opens the door to real-time image reconstruction from compressed acquisition. In this study, we evaluate the different networks for static and dynamic imaging. In particular, we introduce a recurrent neural network that is designed to exploit the spatiotemporal redundancy in videos via a memory state. We validate our algorithms on simulated data from the UCF-101 dataset, with a resolution of 128x128 pixels and a compression ratio of 98%. We also show experimentally that we can resolve small spectral differences in the spectrum of human skin measured in vivo.
基于深度神经网络的单像素压缩视频重构
单像素成像是一种范例,可以使用空间光调制器从单点探测器捕获图像。这种方法对于像素化探测器阵列过于昂贵或过于笨重的光学装置(例如,多光谱、红外成像)来说特别有趣。它获取场景图像和一组用户自定义模式之间的内积,这些模式依次上传到空间光调制器上。压缩数据采集减少了采集时间,但会导致不适定重构问题,这对实时应用来说是一个很大的挑战。最近,神经网络已经成为传统重建方法的竞争替代品。神经网络是通过利用大型数据集来训练的参数化模型。它们的非迭代特性允许快速重建,这为压缩采集的实时图像重建打开了大门。在这项研究中,我们评估了不同的网络用于静态和动态成像。特别是,我们引入了一个循环神经网络,旨在通过记忆状态利用视频中的时空冗余。我们在UCF-101数据集的模拟数据上验证了我们的算法,该数据集的分辨率为128 × 128像素,压缩比为98%。我们还通过实验表明,我们可以解决体内测量的人体皮肤光谱中的微小光谱差异。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
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:604180095
Book学术官方微信
小红书