Discrimination of wheat unsound grains based on deep learning and terahertz spectral image technology

Fei Wang; Yuying Jiang; Hongyi Ge; Xinyu Chen; Li Li

doi:10.1117/12.2646081

16 December 2022 Discrimination of wheat unsound grains based on deep learning and terahertz spectral image technology

Fei Wang, Yuying Jiang, Hongyi Ge, Xinyu Chen, Li Li

Proceedings Volume 12501, Seventeenth National Conference on Laser Technology and Optoelectronics; 1250105 (2022) https://doi.org/10.1117/12.2646081
Event: Seventeenth National Conference on Laser Technology and Optoelectronics, 2022, Shanghai, China

Abstract

The problem of insignificant image features and poor image quality in the acquisition of terahertz spectral images of unsound wheat. Therefore, this study proposed the FFDNet-VGG terahertz image enhancement algorithm based on the FFDNet denoising algorithm and VGG feature extraction algorithm. The experimental results showed that this algorithm could effectively improve the image quality compared with the traditional algorithms such as BM3D and WNNM, and the PSNR and SSIM of the enhanced images were 39.10 dB and 0.93, respectively. The wheat unsound grain images processed by FFDNet, VGG and FFDNet-VGG algorithms were verified by CNN classification network with the classification accuracy of 93.1% and 93.7%, respectively. FFDNet-VGG effectively enhances the wheat unsound grain terahertz spectral images.

Citation Download Citation

Fei Wang, Yuying Jiang, Hongyi Ge, Xinyu Chen, and Li Li "Discrimination of wheat unsound grains based on deep learning and terahertz spectral image technology", Proc. SPIE 12501, Seventeenth National Conference on Laser Technology and Optoelectronics, 1250105 (16 December 2022); https://doi.org/10.1117/12.2646081

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