Neural network image matching method for spacecraft rendezvous and docking optical auxiliary means

By introducing the Class-Tuned Invariant FeatureTransformer (CtIFT) network algorithm in the spacecraft rendezvous and docking mission, the improved symmetric multi-layer cross-attention mechanism and feature classification module are used to solve the feature matching problem in low-texture and complex optical scenarios, and the feature matching effect with high accuracy, robustness and real-time performance is achieved.

CN119832282AInactive Publication Date: 2025-04-15BEIJING INST OF TECH
1 Cites 0 Cited by

Patent Information

Application Number
CN202411890857.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-20
Publication Date
2025-04-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The prior art is difficult to achieve high-precision feature matching in low-texture or complex optical scenarios in spacecraft rendezvous and docking tasks, and traditional methods and some deep learning methods perform poorly in complex situations such as viewing angles and lighting changes.

Method used

A Class-Tuned Invariant FeatureTransformer (CtIFT) network algorithm based on deep learning is proposed, and an improved symmetric multi-layer cross-attention mechanism and feature classification module is adopted to enhance the robustness and directionality of feature matching, reduce the calculation load, and improve real-time.

Benefits of technology

CtIFT significantly improves the feature matching accuracy and robustness in spacecraft rendezvous and docking missions, and can achieve high-precision matching in complex lighting, viewing angles and low texture scenarios, with good generalization and real-time processing capabilities.

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

Abstract

The invention discloses a neural network image matching method for spacecraft rendezvous and docking optical auxiliary means, and the method comprises the following steps: S1, carrying out data collection and image preprocessing, and obtaining collected data and a preprocessed image; s2, performing feature matching on the acquired data and the preprocessed image according to a CtIFT algorithm to obtain a pixel-level high-precision matching point set, and outputting the pixel-level high-precision matching point set; s3, calculating the spatial relationship between the images through the high-precision matching point set, and further estimating and outputting the relative pose of the spacecraft; the problems that a traditional method and a part of deep learning methods based on an attention mechanism are limited in feature matching capacity for a low-texture region and cannot adapt to complex conditions such as view angle and illumination change in the spacecraft docking process are solved.
Need to check novelty before this filing date? Find Prior Art