A mouse automatic detection model and method combining YOLOV11 and residual graph, and a storage medium

By combining YOLOv11 with residual maps to create an automatic mouse detection model, dynamic targets are extracted between video frames using multi-scale feature maps and residual map technology. This solves the problems of low efficiency and poor adaptability to complex environments in automatic mouse detection technology, and achieves high-precision automated identification and real-time monitoring.

CN122116227APending Publication Date: 2026-05-29NANJING AGRICULTURAL UNIVERSITY +2
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NANJING AGRICULTURAL UNIVERSITY
Filing Date
2026-01-29
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing automatic mouse detection technologies suffer from low efficiency, poor adaptability to complex environments, high computational costs, and challenges in achieving automated identification.

Method used

An automatic mouse detection model combining YOLOv11 and residual maps is used to perceive targets of different sizes through multi-scale feature maps and multi-target detection. Dynamic targets are extracted between video frames using residual map technology, and SAHI technology is used for feature exchange and enhancement to achieve the recognition of small and distant targets.

Benefits of technology

It significantly improves the accuracy of mouse target detection in complex backgrounds and dynamic scenes, reduces false detections and missed detections, can automatically identify mouse targets without human intervention, monitors in real time and generates detection logs, adapts to complex backgrounds and dynamic scenes, and improves the level of intelligence and automation of detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122116227A_ABST
    Figure CN122116227A_ABST
Patent Text Reader

Abstract

The application discloses a mouse automatic detection model combined with YOLOV11 and residual graphs, a method and a storage medium, relates to the technical field of computer vision and artificial intelligence, and comprises information collection and processing, collection of image data, denoising and deduplication preprocessing operation on the image, labeling of the image, and generation of a label file; a target detection model is built, a mouse automatic detection model combined with YOLOV11 and residual graphs is built; real-time detection and application, the results after post-processing and optimization, continuous monitoring and dynamic adjustment strategy and model parameters. The method significantly improves the detection accuracy and real-time performance of the mouse detection system in a complex dynamic background, greatly reduces the false detection, missed detection and background interference problems existing in the traditional target detection method in a complex environment, so that the system can accurately and efficiently detect the target in various environmental conditions, especially in a dynamic change and complex background.
Need to check novelty before this filing date? Find Prior Art