# Research on Sika Deer Behavior Recognition Based on YOLOv11 Lightweight SDB-YOLO Model for Small Sample Learning

## TL;DR

This paper introduces a lightweight model for accurately recognizing sika deer behavior using limited data and challenging conditions.

## Contribution

A novel lightweight SDB-YOLO model is proposed for improved small-sample behavior recognition in sika deer.

## Key findings

- SDB-YOLO achieves 90.2% recognition accuracy with low computational requirements.
- The model outperforms existing baseline models in small-sample environments.
- The architecture includes modules like FPSC, C3_GDConv, and CBAM for enhanced feature extraction.

## Abstract

In practical breeding scenarios, automatic behavior recognition for sika deer often lacks accuracy due to limited behavioral samples collected, coupled with factors like lighting variations and occlusions. This study proposes a more compact new model, SDB-YOLO, to address this issue. The model effectively utilizes limited data to capture more representative behavioral information from images while reducing computational overhead. Simultaneously, the newly designed recognition architecture enhances training stability and facilitates higher accuracy in small-sample environments. Experimental results demonstrate that SDB-YOLO achieves a 90.2% recognition accuracy while maintaining extremely low computational requirements, outperforming existing baseline models.

In the breeding scene, limited by the small number of samples and environmental interference such as illumination occlusion, sika deer behavior recognition still faces challenges such as insufficient feature representation and weak cross-scale modeling ability. To this end, this study builds a lightweight improved model SDB-YOLO based on YOLOv11n. Firstly, the FPSC module is proposed to enhance the correlation between multi-scale features through the shared convolution mechanism, so as to significantly improve the quality of feature fusion under the condition of small samples. Secondly, the Ghost feature generation and dynamic convolution strategy are introduced into the C3k2 module to construct the C3_GDConv structure, so as to strengthen the fine-grained behavior pattern modeling ability and reduce redundant calculations. In addition, the CBAM attention mechanism is added to the neck of the network to further improve the ability of key information extraction and enhance the discrimination of feature expression. Finally, the EfficientHead was used to replace the original detection head to obtain a more robust training process and higher detection accuracy in small-sample scenarios. Experimental results show that SDB-YOLO achieves 90.2% detection accuracy with only 4.3 GFLOPs of calculation, which achieves significant performance improvement compared with YOLOv11n, and verifies the effectiveness and lightweight advantages of the proposed method in small-sample special animal behavior recognition tasks.