Image Calibration Using Road Feature Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving assistance technologies face challenges in performing proper calibration of captured images due to misalignment of imaging devices on mobile objects or product variations, leading to potential errors in image recognition and assistance.
Innovation Solution
An image processing device that includes an acquirer, extractor, detector, feature point extractors, and calibrator to enhance image calibration by using feature points from both moving and cross road areas, allowing for coordinate system conversion from camera to bird's-eye view, thereby improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature points are extracted from general image areas for calibration, then calibration can be performed, but calibration accuracy is insufficient due to misalignment and product variations
Solution Approach 1:
The patent applies local quality by selecting feature points from specific road areas (moving road area and cross road area) rather than uniformly from the entire image. This localized selection ensures that feature points are extracted from regions with stable geometric characteristics, improving calibration accuracy while compensating for misalignment and product variations in the imaging device.
Solution Approach 2:
The patent segments the image into different functional areas (moving road area and cross road area) and extracts feature points from each segment separately. This segmentation allows the system to utilize the distinct geometric properties of different road regions, enhancing the overall calibration accuracy through diversified feature point sources.
2Productivity
If imaging devices are mounted on mobile objects for driving assistance, then real-time image capture is enabled, but misalignment and product variations cause calibration errors
Solution Approach 1:
The patent implements self-service by enabling the imaging system to perform self-calibration using feature points extracted from road areas visible in the captured images. The system automatically detects and utilizes geometric features of the road environment to correct calibration errors, eliminating the need for external calibration equipment or manual adjustment, thus maintaining driving assistance functionality while improving calibration accuracy.
3Area of stationary object
If multiple camera views are used to capture peripheral images, then coverage is improved, but calibration complexity increases
Solution Approach 1:
The patent applies universality by using the same calibration approach (extracting feature points from moving road area and cross road area) for all camera views. This unified method allows the system to handle multiple camera perspectives consistently, maintaining wide coverage while avoiding the need for view-specific calibration procedures, thus reducing overall calibration complexity.
Data Source
AI summary
An image processing device of an embodiment includes an acquirer configured to acquire an image in a periphery of a mobile object, an extractor configured to extract feature points from an image acquired, a first detector configured to detect a moving road area in which the mobile object moves based on the image, a second detector configured to detect a cross road area that crosses the moving road area based on the image, a first feature point extractor configured to extract a feature point of a moving road area detected by the first detector as a first feature point among feature points extracted, a second feature point extractor configured to extract a feature point of a cross road area detected by the second detector as a second feature point among the feature points, and a calibrator configured to calibrate the image based on the first and the second feature point.