Extending Effective Dynamic Range of Hyperspectral Line Cameras for Short Wave Infrared Imaging.

Muhammad Saad Shaikh,Keyvan Jaferzadeh,Benny Thörnberg

doi:10.3390/s22051817

Muhammad Saad Shaikh, Keyvan Jaferzadeh + Show 1 more

Open Access

https://doi.org/10.3390/s22051817

Copy DOI

Abstract

In this work, a multi-exposure method is proposed to increase the dynamic range (DR) of hyperspectral imaging using an InGaAs-based short-wave infrared (SWIR) hyperspectral line camera. Spectral signatures of materials were captured for scenarios in which the DR of a scene was greater than the DR of a line camera. To demonstrate the problem and test the proposed multi-exposure method, plastic detection in food waste and polymer sorting were chosen as the test application cases. The DR of the hyperspectral camera and the test samples were calculated experimentally. A multi-exposure method is proposed to create high-dynamic-range (HDR) images of food waste and plastic samples. Using the proposed method, the DR of SWIR imaging was increased from 43 dB to 73 dB, with the lowest allowable signal-to-noise ratio (SNR) set to 20 dB. Principal Component Analysis (PCA) was performed on both HDR and non-HDR image data from each test case to prepare the training and testing data sets. Finally, two support vector machine (SVM) classifiers were trained for each test case to compare the classification performance of the proposed multi-exposure HDR method against the single-exposure non-HDR method. The HDR method was found to outperform the non-HDR method in both test cases, with the classification accuracies of 98% and 90% respectively, for the food waste classification, and with 95% and 35% for the polymer classification.

Highlights

These abstractly defined categories are known as Multi-Spectral Imaging (MSI) and Hyper-Spectral Imaging (HSI)
ReFsourlttsesatncdasAen1a,lPyCsiAs of food and plastic measurements was performed for both HDR and nFoonr-tHesDt Rcadseat1a,.PFFCooArr tohfeefPoCodAasnhdowplnnasitnicFmigeuarseu1re1m, tehnetsblwacaks plearsftoircmaendd fborrigbhotthplHasDtiRc calnadssnesonw-eHrDe tRredaatetad. sFeopratrhaetePlyCfAorsvhioswuanliizninFgigduartea 1cl1u,sttheersb. lTahcke plastic caonldlecbtreidghfrtopmlasthtiec wclasstesfowresraemtrpelaintegdhsaedpasrmataellytrfaocrevsiosuf adleizcionmgpdoasteadclfuosotde,rsa.nTdhtehpelfaosotidc csoamllepcltesdmfraoymatlhsoe hwaavsetehfaodr ssaommpeliinmgphuuarrdiittiiseemss,asslolot5r5a%%ceoosffooofuudttellyycioinnmggpddoaastteaadwwfoeeorrede,rraeemnmdoovtvheeeddffofrrooodmmsaamllllptthlherrseeeme ccallyaassassleessos
Many applications of spectral imaging require a spectral camera in short-wave infrared (SWIR) with a high dynamic range (DR) to capture all the meaningful spectral details in a scene

Summary

Introduction

Spectral imaging allows spectral information to be acquired across the electromagnetic spectrum [1]. Spectral imaging is divided into different categories depending upon the number of spectral bands used, the width of the spectral bands and the gaps between them, and the spectral resolution. These abstractly defined categories are known as Multi-Spectral Imaging (MSI) and Hyper-Spectral Imaging (HSI). The use of spectral imaging was first proposed in 1985 by Goetz [2] for remote sensing of Earth. Satellite-based remote sensing [3], agriculture [4], the defense industry [5], medical diagnostics [6], and food inspection [7] are just a few examples of situations in which the use of spectral imaging is very popular. In other applications, reflected signatures are captured by a camera and fed into a Machine Learning (ML) algorithm to detect and classify the objects present in the scene [9]

Methods

Results

Discussion

Conclusion