A Deep Learning-Based Method and System for Identifying Individual Tibetan Macaques in the Wild

By combining YOLOv8s and Vision Transformer, the problem of identifying Tibetan macaques in small samples and complex backgrounds was solved, achieving high-precision and robust individual identification, which is suitable for monitoring primates in the wild.

CN122090485APending Publication Date: 2026-05-26ANHUI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610301393.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-12
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing technologies struggle to achieve high-precision and robust individual identification of Tibetan macaques under conditions of small samples, complex backgrounds, and class imbalance. Traditional methods are time-consuming, labor-intensive, and limited by researchers' experience. Traditional convolutional neural networks struggle to capture key discriminative features when processing facial features of Tibetan macaques, and field data collection is difficult with poor model generalization ability.

Method used

A YOLOv8s-based face detection model is used for preprocessing, combined with a Vision Transformer individual recognition model. A joint loss function and adaptive hybrid enhancement strategies, including Focal Loss, Supervisor contrastive loss, hierarchical Dropout regularization, and weighted random sampling, are used to construct an efficient individual recognition system.

Benefits of technology

It achieves high-precision individual identification of Tibetan macaques in complex field environments, improves the model's generalization ability and robustness, solves the class imbalance problem, and provides an efficient means of monitoring primates in the wild.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122090485A_ABST
    Figure CN122090485A_ABST
Patent Text Reader

Abstract

This disclosure relates to the fields of computer vision and wildlife monitoring technology, and specifically to a method and system for identifying Tibetan macaques in the wild based on deep learning. First, a face detection model is used to detect and crop faces from raw wild images. The detection model is initialized with pre-trained weights and fine-tuned by freezing the backbone network. Then, an individual identification model is constructed to output classification results. During training, a joint loss function is used, combining Focal Loss to suppress easily classified samples and guide attention to difficult-to-classify samples. Supervised contrastive loss is used to enhance the discriminative power of the feature space. Simultaneously, an adaptive hybrid enhancement strategy and a hierarchical Dropout regularization mechanism are introduced to effectively address the problems of small sample sizes, class imbalance, and overfitting in complex wild environments. This disclosure enables high-precision identification of Tibetan macaques, providing an efficient technical means for long-term ecological monitoring of wild primates.
Need to check novelty before this filing date? Find Prior Art