Multi-camera extrinsic parameter calibration using object detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge is to perform real-time extrinsic parameter calibration for multiple cameras on a movable device, such as a vehicle, to account for slight changes in camera positions due to factors like installation position movement, vibration, and collisions, ensuring accurate performance in automatic driving tasks.
Innovation Solution
A multi-camera extrinsic parameter calibration method and device that involves acquiring multi-frame environmental images from different view angles, detecting a predetermined object type, mapping detection information to a pre-set coordinate system, dividing cameras into groups based on spatial layout, constructing cross-image matching information, and performing calibration using this information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If cameras are installed on a movable device, then the device can perform automatic driving tasks, but camera positions change due to vibration and movement causing extrinsic parameter drift
Solution Approach 1:
The patent implements a feedback mechanism by continuously acquiring multi-frame environmental images from multiple cameras, detecting objects in each frame, and using the detection results to calculate and update extrinsic parameters in real-time, correcting drift caused by vehicle movement and vibration
Solution Approach 2:
The system performs self-calibration by using its own camera images and object detection capabilities to automatically compute and correct extrinsic parameters without requiring external calibration equipment or manual intervention, enabling the system to self-correct for position changes
2Measurement precision
If real-time calibration is performed using multi-frame images and object detection, then extrinsic parameter accuracy is maintained, but computational complexity and processing time increase
Solution Approach 1:
The patent extracts only the necessary information for calibration by focusing on detecting specific predetermined objects (such as traffic signs, road markings, or other reference objects) rather than processing all image data, reducing computational load while maintaining calibration accuracy
Solution Approach 2:
The calibration process is segmented into distinct modules: image acquisition, object detection, coordinate transformation, and parameter calculation. This modular segmentation allows each component to be optimized independently and facilitates parallel processing to reduce overall complexity
Data Source
AI summary
A multi-camera extrinsic parameter calibration method, a storage medium and an electronic apparatus are disclosed. The method includes: acquiring multi-frame environmental images from different view angles collected by a plurality of cameras provided at different orientations of a movable device; performing detection of a predetermined type of object respectively on the multi-frame environmental images to obtain initial detection information respectively corresponding to the multi-frame environmental images; mapping the initial detection information to a pre-set coordinate system to obtain corresponding transformed detection information; dividing the plurality of cameras into at least one camera group based on the spatial layout of the plurality of cameras; constructing, for each camera group, cross-image matching information of a predetermined type of object based on the transformed detection information respectively corresponding to each camera included in the camera group; and performing extrinsic parameter calibration on the plurality of cameras based on the cross-image matching information.