Small target detection model and method based on anchor frame focusing loss distribution correction

By using the AFL-DETR model, the performance lag problem of small target detection is solved by adjusting the contribution weight of object area in the loss function and fusing image information, thereby improving the accuracy of small target detection and enhancing the detection effect of medium and large targets.

CN120852744APending Publication Date: 2025-10-28CHONGQING UNIV OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510915629.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-03
Publication Date
2025-10-28

AI Technical Summary

Technical Problem

Existing Transformer-based object detection models suffer from performance lag when detecting small objects, especially in complex scenes where they struggle to effectively identify small objects. Existing methods rely on large amounts of labeled data and have limited local attention enhancement effects, resulting in a lack of information about small objects.

Method used

A small object detection model with anchor focal loss distribution correction (AFL-DETR) is introduced. The contribution weight of object area in the loss function is adjusted by Anchor Focal Loss (AFL), and the Context Aggregation Pyramid (CAP) module is combined to fuse image information and shallow features to enhance the feature representation and contextual semantic information of small objects.

Benefits of technology

It significantly improves the detection accuracy of small targets and enhances the model's ability to recognize small objects. At the same time, it improves the detection performance of medium and large objects without increasing computational costs. Experimental results show excellent robustness and scalability on multiple datasets.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120852744A_ABST
    Figure CN120852744A_ABST
Patent Text Reader

Abstract

The invention discloses a small target detection model and method based on anchor frame focusing loss distribution correction, and relates to the technical field of target detection. According to the method, an overall framework of an AFL-DETR model for end-to-end small target detection is provided, a novel loss function specially customized for small targets is provided in the model building process, the area can be deduced by extracting position information from an object query, the area is incorporated into the loss function, contribution of the size of each target is adjusted, and the detection accuracy of the AFL-DETR model is improved. Therefore, the accuracy of small target detection is improved. In addition, an information enhancement module is also introduced, and the module fuses the preprocessed image details and the shallow feature information. And then, the features are further extracted and fused to obtain cross-fusion information, so that context semantic information suitable for the small target is generated.
Need to check novelty before this filing date? Find Prior Art