An event feature extraction method based on spatial gradient variance maximization

Through the method based on spatial gradient variance maximization, features are extracted from the data of event cameras, which solves the problem of insufficiently accurate and real-time feature extraction in the prior art, and achieves more efficient and accurate feature extraction, improving system performance.

CN119380042BActive Publication Date: 2025-03-25TIANMUSHAN LABORATORY +1
2 Cites 0 Cited by

Patent Information

Application Number
CN202411974945.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-12-31
Publication Date
2025-03-25
Estimated Expiration
2044-12-31

AI Technical Summary

Technical Problem

The prior art is difficult to extract useful features from data captured by event cameras, especially while maintaining high temporal resolution, and failing to provide sufficient real-time and accuracy.

Method used

An event feature extraction method based on spatial gradient variance maximization is adopted. By obtaining event stream data of event cameras, a binary event frame is generated, event patch is determined, its spatial gradient variance is calculated, and the patch with the largest variance is selected for de-overlapping processing to generate a feature map.

Benefits of technology

It improves the stability and accuracy of feature extraction, can capture key details in the image more accurately, provides richer and meaningful feature representations for subsequent identification, classification and tracking, and enhances the performance and reliability of the system.

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

Abstract

The present invention relates to the field of image processing and computer vision technology, and in particular to an event feature extraction method based on maximizing spatial gradient variance, comprising: acquiring event stream data shot by an event camera, processing the event stream data, and generating a binary event frame; determining an event patch corresponding to each pixel in the binary event frame; acquiring multiple spatial gradients of each event patch, and calculating the spatial gradient variance of the event patch based on the multiple spatial gradients; for the spatial gradient variances of all the event patches, selecting the first m event patches with the largest spatial gradient variance for de-overlapping processing, and obtaining multiple event patches without any overlapping areas as feature maps of the binary event frame; the present invention can improve the accuracy and effectiveness of event camera data feature extraction.
Need to check novelty before this filing date? Find Prior Art