Weakly supervised sar image target detection method based on constant false alarm rate and cam

By combining constant false alarm detection with CAM, and utilizing the statistical characteristics and class activation maps of SAR images, image-level labels are updated, solving the problems of false alarms and redundancy in ship target detection in spaceborne SAR images, and achieving efficient weakly supervised target detection.

CN118247653BActive Publication Date: 2026-07-24NORTHWESTERN POLYTECHNICAL UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202410312301.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-03-19
Publication Date
2026-07-24
Estimated Expiration
2044-03-19

AI Technical Summary

Technical Problem

In spaceborne SAR images, ship target detection is difficult due to the fact that the number of target pixels is much lower than the number of background pixels and is affected by sea clutter. Existing methods require a lot of manual annotation, making it difficult to effectively perform ship target detection under weak supervision.

Method used

A weakly supervised SAR image target detection method using constant false alarm rate (CFAR) and CAM is proposed. Candidate regions are obtained through a feature extraction network, and image-level labels are updated by combining class activation maps and deep learning networks to reduce the false alarm rate and improve detection accuracy.

Benefits of technology

It effectively avoids candidate region redundancy, improves detection accuracy and recall, reduces false alarm rate, and enhances the ship target detection effect under image-level annotation.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application discloses a weakly supervised SAR image target detection method based on constant false alarm rate and CAM, which is realized by using a weakly supervised target detection network, and the weakly supervised target detection network comprises a feature extraction network; deep features of a to-be-detected SAR image are extracted by using the feature extraction network, and candidate regions in the to-be-detected SAR image are generated by using a constant false alarm rate detector; local deep features corresponding to the candidate regions are extracted in the deep features; classification detection results and classification contribution degrees of the local deep features are respectively generated; and image-level labels of the to-be-detected SAR image are obtained by fusing the classification detection results and the classification contribution degrees; features of the image are extracted by using the weakly supervised target detection, target candidate regions are obtained in combination with the constant false alarm rate detector, and subsequent target detection procedures are performed after the combination of the two, so that the candidate regions containing only target parts can be classified as targets when the candidate regions are classified, and the accuracy of candidate region detection can be increased.
Need to check novelty before this filing date? Find Prior Art