Fisheye Sensor Calibration for Real-World Object Position Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fisheye lenses used in image sensors provide a wider field of view but introduce distortion, requiring additional processing to determine object locations in real-world coordinates, which can lead to discrepancies and increased computing resources.
Innovation Solution
The method involves calibrating sensors with a fiducial marker to determine a reference plane and using object feature points to locate objects in real-world coordinates without transforming fisheye images into rectilinear images, employing techniques like scale-invariant feature transform and correcting for fisheye lens distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If fisheye lenses are used in image sensors to provide a wider field of view, then the field of view is improved, but image distortion increases requiring additional processing
Solution Approach 1:
The patent applies preliminary action by pre-calibrating the fisheye lens distortion parameters and establishing a reference plane before actual object location measurements. The calibration process creates lookup tables and transformation matrices in advance, so that during operation, the system can directly apply these pre-computed corrections without real-time complex processing, thus maintaining measurement precision while preserving the wide field of view advantage
2Measurement precision
If fisheye images are transformed into rectilinear images to correct distortion, then measurement accuracy is improved, but computing resources and processing time increase
Solution Approach 1:
The patent extracts only the essential correction information needed for object location from the full image transformation process. Instead of transforming the entire fisheye image into rectilinear coordinates, the system extracts distortion parameters and applies selective corrections only to the regions containing objects of interest, using feature point detection and reference plane calibration to achieve accurate location measurements with minimal computing resources
Solution Approach 2:
The calibration process performs preliminary computation of distortion correction parameters and stores them for later use. During actual object location, the system retrieves pre-computed transformation matrices and applies them directly, avoiding the need for real-time complex image transformation while maintaining measurement accuracy
Data Source
AI summary
A system is disclosed that includes a computer that includes a processor and a memory, the memory including instructions executable by the processor to acquire a first image with a sensor, wherein the sensor is calibrated with a fiducial marker to determine a real world location of a reference plane. An image of an object is acquired with the sensor to determine that the object is located on the reference plane by determining object feature points. A location of the object is determined in real world coordinates including depth based on the object feature points. The system is operated based on the location of the object.