Study on denoising method of hyperspectral data with multi-standard reflectivity correction

Hui Li; Jianlun He; Xin Tan; Yuhang Li; Hongzhang Ning

doi:10.1117/12.2606347

24 November 2021 Study on denoising method of hyperspectral data with multi-standard reflectivity correction

Hui Li, Jianlun He, Xin Tan, Yuhang Li, Hongzhang Ning

Proceedings Volume 12064, AOPC 2021: Optical Spectroscopy and Imaging; 120640F (2021) https://doi.org/10.1117/12.2606347
Event: Applied Optics and Photonics China 2021, 2021, Beijing, China

Abstract

Hyperspectral imager is easily affected by noise in the process of imaging, so it is difficult to evaluate the noise in it. It is difficult to obtain accurate image noise values by existing technical methods, and it is also a problem to correct different environmental light sources.In this paper, by comparing the reflectivity inversion results obtained from the inversion of different number of reflectance plates.Inversion method based on the double reflectance plate is the best choice finally. The mean square error of the reflectance spectrum and the calibration spectrum obtained from the inversion in the band of 500~900nm is less than 0.0001, which effectively corrects the radiation error in the image.Based on PCA and SVM algorithm to build the grain heavy hyperspectral classification model.The improvement of the total accuracy and test accuracy of the inversion model based on double reflectance plate is better than several other inversion methods.Compared with single reflectivity inversion reflectance spectra, inversion method based on the double reflectance plate build hyperspectral database after training to obtain a set of test precision of 4.409%, total classification accuracy of 3.104%.

Citation Download Citation

Hui Li, Jianlun He, Xin Tan, Yuhang Li, and Hongzhang Ning "Study on denoising method of hyperspectral data with multi-standard reflectivity correction", Proc. SPIE 12064, AOPC 2021: Optical Spectroscopy and Imaging, 120640F (24 November 2021); https://doi.org/10.1117/12.2606347

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available