Mitigating Catastrophic Forgetting in Pest Detection Through Adaptive Response Distillation

Hongjun Zhang, Zhendong Yin, Dasen Li () and Yanlong Zhao
Additional contact information
Hongjun Zhang: School of Electronics and Information Engineering, Harbin Institute of Technology, 92 Xidazhi Street, Nangang District, Harbin 150001, China
Zhendong Yin: School of Electronics and Information Engineering, Harbin Institute of Technology, 92 Xidazhi Street, Nangang District, Harbin 150001, China
Dasen Li: School of Electronics and Information Engineering, Harbin Institute of Technology, 92 Xidazhi Street, Nangang District, Harbin 150001, China
Yanlong Zhao: School of Electronics and Information Engineering, Harbin Institute of Technology, 92 Xidazhi Street, Nangang District, Harbin 150001, China

Agriculture, 2025, vol. 15, issue 9, 1-18

Abstract: Pest detection in agriculture faces the challenge of adapting to new pest species while preserving the ability to recognize previously learned ones. Traditional model fine-tuning approaches often result in catastrophic forgetting, where the acquisition of new classes significantly impairs the recognition performance of existing ones. Although knowledge distillation has been shown to effectively mitigate catastrophic forgetting, current research predominantly focuses on feature imitation, neglecting the extraction of potentially valuable information from responses. To address this issue, we introduce a response-based distillation method, called adaptive response distillation (ARD). ARD incorporates an adaptive response filtering strategy that dynamically adjusts the weights of classification and regression responses based on the significance of the information. This approach selectively filters and transfers valuable response data, ensuring efficient propagation of category and localization information. Our method effectively reduces catastrophic forgetting during incremental learning, enabling the student detector to maintain memory of old classes while assimilating new pest categories. Experimental evaluations on the large-scale IP102 pest dataset demonstrate that the proposed ARD method consistently outperforms existing state-of-the-art algorithms across various class-incremental learning scenarios, significantly narrowing the performance gap compared to fully trained models.

Keywords: class-incremental learning; pest detection; catastrophic forgetting; knowledge distillation; adaptive response distillation; adaptive response filtering (search for similar items in EconPapers)
JEL-codes: Q1 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2077-0472/15/9/1006/pdf (application/pdf)
https://www.mdpi.com/2077-0472/15/9/1006/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jagris:v:15:y:2025:i:9:p:1006-:d:1650141

Access Statistics for this article

Agriculture is currently edited by Ms. Leda Xuan

More articles in Agriculture from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().