Image fusion method and device, electronic equipment and storage medium
By identifying and correcting the occlusion state in the image data collected by roadside perception cameras, and using the corrected covariance matrix for image data fusion, the problem of decreased 3D position accuracy caused by object occlusion in roadside perception is solved, and the efficiency and accuracy of image fusion are improved.
CN118644403BActive Publication Date: 2026-07-24BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF 2 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING BAIDU NETCOM SCI & TECH CO LTD
- Filing Date
- 2024-06-19
- Publication Date
- 2026-07-24
AI Technical Summary
Technical Problem
In roadside perception multi-camera fusion technology, complex factors such as traffic flow at intersections can cause object occlusion, which affects the accuracy of converting the two-dimensional position of an object into its three-dimensional position.
Method used
Image data acquired by the first and second cameras at the same time are obtained, occlusion status is identified, the covariance matrix associated with objects in the occlusion status is corrected, and image data fusion is performed using the corrected covariance matrix and the covariance matrix of the unoccluded state.
Benefits of technology
It improves the efficiency and accuracy of image fusion, and enhances the effect of image fusion.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN118644403B_ABST
Abstract
The disclosure provides an image fusion method and device, electronic equipment and storage medium, and relates to the technical field of computers, in particular to the artificial intelligence technical field of image processing, automatic driving, intelligent transportation, intelligent logistics and the like. The specific scheme is: first, the first image data and the second image data are obtained, and the first image data and the second image data are identified respectively to determine the occlusion state of each object in the first image data and the second image data; then the first covariance matrix associated with the first object in the occlusion state in the first image data and the second image data is corrected to obtain a corrected covariance matrix; finally, based on the corrected covariance matrix and the second covariance matrix associated with the second object not in the occlusion state, the first image data and the second image data are fused.
Need to check novelty before this filing date? Find Prior Art