An approach of occlusion perception based on knowledge distillation in complex vineyard environment

Jingxian Zhao; Jinhai Wang; Xuemin Lin; Mingyou Chen; Huiling Wei; Lufeng Luo

doi:10.1117/12.3051785

18 November 2024 An approach of occlusion perception based on knowledge distillation in complex vineyard environment

Jingxian Zhao, Jinhai Wang, Xuemin Lin, Mingyou Chen, Huiling Wei, Lufeng Luo

Proceedings Volume 13403, International Conference on Algorithms, High Performance Computing, and Artificial Intelligence (AHPCAI 2024) ; 134031E (2024) https://doi.org/10.1117/12.3051785
Event: International Conference on Algorithms, High Performance Computing, and Artificial Intelligence, 2024, Zhengzhou, China

Abstract

To address the challenges of occlusion in intelligent picking within unstructured environments, this paper proposes an occlusion perception algorithm utilizing knowledge distillation. A grape occlusion dataset was developed to support the study. The proposed method employs a lightweight MobileNetV3 student model, trained under the supervision of a high-accuracy ResNet50 teacher model, to achieve comparable performance with significantly fewer network parameters. Experimental results demonstrate that the knowledge-distilled student model operates with only 1.50% of the teacher model's operations while achieving an accuracy of 99.4%, representing a 2.4% improvement over the baseline model. These findings underscore the method's effectiveness in developing efficient, high-performance models suitable for deployment in resource-constrained environments.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Jingxian Zhao, Jinhai Wang, Xuemin Lin, Mingyou Chen, Huiling Wei, and Lufeng Luo "An approach of occlusion perception based on knowledge distillation in complex vineyard environment", Proc. SPIE 13403, International Conference on Algorithms, High Performance Computing, and Artificial Intelligence (AHPCAI 2024) , 134031E (18 November 2024); https://doi.org/10.1117/12.3051785

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