Research on mung bean variety identification algorithm based on machine learning and hyperspectral technology

LinTao Yu; Bin Liu; Shuoou Qu; ShaoZhong Song

doi:10.1117/12.2685660

28 July 2023 Research on mung bean variety identification algorithm based on machine learning and hyperspectral technology

LinTao Yu, Bin Liu, Shuoou Qu, ShaoZhong Song

Author Affiliations +

Proceedings Volume 12716, Third International Conference on Digital Signal and Computer Communications (DSCC 2023); 127160E (2023) https://doi.org/10.1117/12.2685660
Event: Third International Conference on Digital Signal and Computer Communications (DSCC 2023), 2023, Xi'an, China

Abstract

Mung bean seed variety identification is extremely important for ensuring agricultural productivity, safeguarding farmers' interests, and controlling the seed market. Hyperspectral techniques have the advantages of easy operation, non-destruction and rapid detection. However, it usually generates a large number of data, causing a low identification efficiency.This paper adopts the hyperspectral method and machine learning model to conduct experiments and analysis in the feature extraction which is the key to the identification of mung bean seek varieties. The experiments input the original spectral data into the RF classification model and use the SPA, CARS and LAR algorithms for spectral feature extraction, and the constructed SPA-RF, CARS-RF and LAR-RF discrimination models all have a substantial increase in running speed, of which the CARS-RF accuracy and recall are 0.988, with no decrease compared with the original RF model; the RMSE is 0.391, which is 0.029 greater than the RF model; the shortest running time is only 29% of the original RF model, and the best modelability. The results of the experiments demonstrated that the CARS-RF model chosen in this study successfully identifies different types of mung bean seeds while also reducing equipment costs, facilitating the production and fabrication of specialized equipment, and significantly enhancing identification efficiency.

Citation Download Citation

LinTao Yu, Bin Liu, Shuoou Qu, and ShaoZhong Song "Research on mung bean variety identification algorithm based on machine learning and hyperspectral technology", Proc. SPIE 12716, Third International Conference on Digital Signal and Computer Communications (DSCC 2023), 127160E (28 July 2023); https://doi.org/10.1117/12.2685660

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