Around View Camera Tolerance Correction via Ground Feature Masking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Around view monitoring systems (AVMS) in vehicles face challenges in maintaining accurate image matching due to changes in camera tolerances caused by environmental factors like vehicle vibration and road slopes, leading to distorted images that are difficult to correct without visiting a service center.
Innovation Solution
An apparatus and method that captures images from multiple directions, sets a mask region around the vehicle to extract ground feature points, estimates camera attitude angles, and rotationally converts images to a top-view format using a generated rotation matrix, effectively correcting tolerance errors even on sloped surfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional tolerance correction methods are used (extracting ground feature points from entire captured images), then correction can be performed without visiting service center, but image distortion increases due to road slopes
Solution Approach 1:
The captured image is divided into multiple regions, and feature points are selectively extracted only from the flat ground region (mask region) rather than the entire image. This segmentation approach isolates the reliable feature points from areas affected by road slopes, enabling accurate tolerance correction without requiring service center visits.
Solution Approach 2:
Different regions of the image are treated differently based on their reliability for feature extraction. The flat ground region is identified and designated as the mask region where feature points are extracted, while other regions (affected by slopes) are excluded. This local quality approach ensures that only high-quality feature points contribute to the tolerance correction calculation.
2Productivity
If feature points are extracted from entire captured images during driving, then tolerance correction can be performed on-the-go, but extraction accuracy decreases due to non-flat ground surfaces
Solution Approach 1:
The image processing is segmented into two stages: first identifying the mask region corresponding to flat ground, then extracting feature points only within that region. This segmentation enables rapid on-the-go correction while maintaining high extraction accuracy by excluding distorted regions caused by road slopes.
Solution Approach 2:
Before extracting feature points, the system performs preliminary identification and masking of the flat ground region. This preliminary action ensures that only reliable feature points are selected, maintaining high extraction accuracy while enabling real-time correction during vehicle operation.
3Manufacturing precision
If cameras are assembled with initial tolerance correction at release, then around view image matching is satisfied initially, but tolerance changes occur due to environmental factors during vehicle use
Solution Approach 1:
The system continuously monitors the around view image quality during vehicle operation and automatically detects when tolerance drift occurs. By extracting feature points from masked flat ground regions and recalculating camera attitudes, the system provides real-time feedback-based correction, compensating for tolerance changes caused by vibrations, temperature, and other environmental factors.
Solution Approach 2:
The tolerance correction system transitions from a static initial correction at release to a dynamic on-the-go correction mechanism. The system can continuously or periodically update camera attitude angles based on current operating conditions, adapting to environmental changes and maintaining image matching accuracy throughout the vehicle's operational life.
Data Source
AI summary
Provided is an apparatus for generating an around view. The apparatus includes a capture unit configured to capture images in front of, behind, to the left of, and to the right of a vehicle using cameras, a mask generation unit configured to set a region ranging a predetermined distance from the vehicle in the captured image as a mask region, a feature point extraction unit configured to extract ground feature points from the mask region of each of the captured images; a camera attitude angle estimation unit configured to generate a rotation matrix including a rotation angle of the camera using the extracted feature points; and an around view generation unit configured to rotationally convert the captured images to a top-view image using the rotation matrix.