Attack resisting method for multi-target tracking

By designing a target generation module (TGM) and optimizing image perturbation using Kalman filtering, the problem of low false alarm generation efficiency in multi-target tracking is solved, enabling efficient and flexible adversarial attacks and improving attack success rate and stealth.

CN121884040APending Publication Date: 2026-04-17BEIJING UNIV OF POSTS & TELECOMM
0 Cites 0 Cited by

Patent Information

Application Number
CN202610019429.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-01-08
Publication Date
2026-04-17

AI Technical Summary

Technical Problem

Existing multi-target tracking adversarial attack techniques are inefficient in generating false alarms, have poor scalability, and are time-consuming due to repeated execution of correlation steps. Furthermore, they are ineffective on mainstream high-performance trackers, exhibit redundancy in the number of false alarms and abnormal detection rates, and lack flexibility and stealth in their attacks.

Method used

By designing a target generation module (TGM) that combines historical trajectory information and Kalman filtering, the system automatically calculates the location and size of false alarms. It uses the PGD algorithm and a targeted loss function to generate image perturbations, optimizes adversarial examples, and performs the association operation only once per frame, thereby improving attack efficiency and applicability.

Benefits of technology

It achieves adaptive generation of false alarms in multi-target tracking, improving attack efficiency and applicability. The simple and efficient attack process enhances the stealth and success rate of the attack, and is suitable for various scenarios and trackers.

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

Abstract

The invention discloses a multi-target tracking-oriented anti-attack method, which relates to the technical field of deep learning model security, and comprises the following steps: S1, extracting a to-be-attacked sample from an image sequence of an MOT17, MOT20 or BDD100K data set; s2, obtaining a target detection result by adopting YOLOX or the Internet; s3, matching the historical trajectory through intersection-to-union ratio and a Hungary algorithm, and determining a trajectory to be attacked; s4, an interference false alarm is generated in the reverse direction of target movement by using a false alarm generation module in combination with Kalman filtering; s5, calculating image disturbance based on a PGD algorithm and a targeted loss function, and generating an adversarial sample; s6, inputting the adversarial sample into trackers such as SORT, ByteTrack and the like to evaluate an attack effect; and S7, optimizing a disturbance generation parameter according to an evaluation result. According to the method, adaptive false alarm generation is realized through the TGM module, only one association operation is executed for each frame, attack focuses on an MOT association stage, verification is valid on multiple data sets and trackers, the attack efficiency, the concealment and the success rate can be improved, and the method is suitable for security test and vulnerability verification of the MOT model.
Need to check novelty before this filing date? Find Prior Art