Camera Calibration via Ground Plane Detection and Feature Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous agricultural vehicles face challenges in camera calibration without fiducial markers, which can introduce inaccuracies due to improper marker placement, and require efficient methods to map camera coordinates to vehicle coordinates for accurate navigation and operation.
Innovation Solution
The implementation of camera calibration circuitry that detects features in images captured by stereoscopic cameras to determine calibration parameters such as roll angle, pitch angle, and yaw angle without fiducial markers, using ground plane detection and feature tracking, and separates the calibration process into plane fitting and feature tracking procedures to reduce computational load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If fiducial markers are used for camera calibration, then calibration process is simplified, but measurement precision deteriorates due to improper marker placement
Solution Approach 1:
The patent removes fiducial markers from the calibration process entirely. Instead of using external markers, the system extracts calibration information directly from the natural scene by detecting the ground plane and tracking features in video frames captured by the camera. This extraction approach eliminates the source of measurement errors while maintaining calibration functionality.
Solution Approach 2:
The calibration system uses the camera's own video feed and automatic feature detection capabilities to perform calibration without external aids. The ground plane detection and feature tracking algorithms automatically identify calibration information from the environment, making the system self-calibrating without requiring manual marker placement or external calibration objects.
2Measurement precision
If complete calibration process is performed simultaneously, then all calibration parameters are determined, but computational load increases
Solution Approach 1:
The calibration process is divided into distinct sequential stages: ground plane detection and parameter extraction from static frames, followed by feature tracking and additional parameter determination from video sequences. This segmentation allows computationally intensive operations to be distributed over time and separated by complexity, reducing peak computational load while achieving complete calibration.
Solution Approach 2:
The system performs preliminary calibration by determining roll, pitch, and z-axis parameters from static image analysis and ground plane detection before proceeding to more computationally intensive feature tracking for yaw and x-y axis parameters. This preliminary action reduces the overall computational burden by completing simpler calibration tasks first.
Data Source
AI summary
Methods, systems, apparatus, and articles of manufacture for camera calibration are disclosed. An example apparatus includes memory, instructions, and programmable circuitry to execute the instructions to detect, based on video frames captured by a camera positioned on a vehicle, a ground plane of the vehicle, determine, based on the ground plane, a first position parameter of the camera with respect to a coordinate system of the vehicle, track a plurality of features between ones of the video frames, and determine, based on the plurality of features, a second position parameter of the camera with respect to the coordinate system of the vehicle, the first position parameter different from the second position parameter.