Camera Localization via Visual Target GPS Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Infrastructure sensing elements, such as cameras, face challenges in calibration to an alternative frame of reference, requiring significant processing resources and overhead to provide object locations in a usable format for vehicles, especially in complex environments like busy intersections, where precise GNSS locations of these elements are difficult to determine and maintain.
Innovation Solution
A system comprising a camera, a visual target with a GPS receiver, and a processor that images the target at multiple locations to derive 3D locations and calculate the camera's GPS location, allowing for transformation of object positions from the camera reference frame to a GPS reference frame, thereby reducing errors and facilitating efficient data transfer to vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If infrastructure elements are calibrated to provide object locations in camera reference frame, then object detection capability is improved, but processing resources and overhead increase significantly
Solution Approach 1:
The patent introduces a GPS reference frame as an intermediary coordinate system between the camera and the vehicle. Instead of directly transforming all detected objects from camera coordinates to vehicle coordinates, the system first establishes a GPS reference frame that both the infrastructure camera and the vehicle can independently reference. This mediator frame reduces the computational burden by providing a common reference that requires less processing than direct coordinate transformation of every detected object.
2Measurement precision
If precise GNSS locations of infrastructure elements are determined and maintained, then common transformation accuracy is improved, but deployment difficulty and cost increase
Solution Approach 1:
The patent enables infrastructure elements to self-determine their GPS coordinates and transformation parameters through automated calibration processes. The system uses visual targets with embedded GPS receivers that allow cameras to automatically calculate their own transformation matrices without requiring manual surveying or professional calibration services. This self-service approach dramatically simplifies deployment while maintaining precision.
3Device complexity
If infrastructure elements are installed at similar heights and poses to facilitate common transformation, then transformation simplicity is improved, but installation flexibility and adaptability decrease
Solution Approach 1:
The patent uses parameter changes in the transformation model to accommodate diverse installation configurations. Instead of requiring all cameras to be installed at identical heights and orientations, the system dynamically adjusts transformation parameters (translation vectors, rotation matrices) based on each camera's actual installation pose. This allows infrastructure elements to be installed flexibly anywhere in the environment while maintaining accurate coordinate transformations through adaptive parameter calculation.
Data Source
AI summary
A system includes a camera, a visual target including a GPS receiver, and a processor. The camera images the visual target at a plurality of locations to obtain a plurality of images. The processor receives GPS coordinates for each respective location. Also, the processor determines a 3D location of the visual target in a camera reference frame for each of the images. The processor derives a plurality of estimated GPS camera locations for each of the plurality of images and mitigates an error associated with each of the derived estimated GPS camera locations to derive an average GPS camera location. This location is used to determine and store a transformation usable to describe objects viewed by the camera in the camera reference frame in terms of the GPS reference frame.