Unmanned aerial vehicle inspection method and system based on cross-modal data class incremental learning

By employing a method of cross-modal feature fusion and virtual category construction, the forgetting problem in cross-modal data learning of UAV inspection models was solved, enabling efficient identification and adaptation to both new and old tasks, and improving the robustness and recognition accuracy of UAV inspection.

CN119131471BActive Publication Date: 2026-05-08TIANJIN UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202411156650.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2026-05-08
Estimated Expiration
2044-08-22

AI Technical Summary

Technical Problem

Existing UAV inspection models suffer from catastrophic forgetting problems in cross-modal data learning, making it difficult to effectively identify and adapt to cross-modal targets in new and old tasks, resulting in a significant decline in the model's ability to identify old tasks.

Method used

A cross-modal data class incremental learning method is adopted. By constructing a cross-modal feature fusion module and virtual categories, combined with adaptive average pooling and loss function, a UAV inspection method based on cross-modal data class incremental learning is designed to incrementally introduce new categories while maintaining the recognition ability of old categories.

Benefits of technology

This improved the cross-modal incremental learning capability of the UAV inspection model, enhanced its robustness and generalization, reduced computational resource consumption and data labeling requirements, and improved the model's adaptability and recognition accuracy.

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

Abstract

The application discloses a kind of based on cross-modal data class increment learning's unmanned plane inspection method and system, comprising: by cross-modal feature mixed fusion module, the features of each mode are spliced and reorganized;Each group of processed input feature map is assigned a set of independent convolution kernel, and grouping convolution is carried out, and finally complete feature map is generated;For each pooling window in complete feature map, the average value of all elements in the window is calculated according to adaptive average pooling, the average value is used as the element of the corresponding position of the output feature map, the process is repeated for all regions, and the final output feature map is obtained;The output feature map obtained after being processed by the adaptive average pooling module is flattened and input into the fully connected layer to extract and combine features, generate the final classification output, then update the model parameters according to the gradient of the loss function;The trained model is deployed on the unmanned plane to collect visible light and thermal imaging images in real time, extract features and identify and predict through the model.
Need to check novelty before this filing date? Find Prior Art