On-Vehicle Camera Overlap Processing for Motion-Blurred 3D Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems using on-vehicle cameras face challenges in accurately recognizing targets due to image blurring from vehicle movement during exposure, leading to increased calculation costs and decreased parallax distance measurement accuracy.
Innovation Solution
An image processing system utilizing a camera group with overlapping imaging regions, including an image acquisition unit, representative distance calculation unit, and three-dimensional information acquisition unit to reduce blurring effects by calculating posture changes and feature point movements, enabling accurate three-dimensional information acquisition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If exposure time is uniformly changed according to vehicle speed to display real-time images, then real-time image display is achieved, but clear images cannot be acquired at high-speed movement
Solution Approach 1:
The patent applies dynamics by making the exposure time adjustment dynamic and adaptive rather than uniform. The system calculates blurring amounts based on actual vehicle speed, distance to objects, and other factors, then adjusts exposure time dynamically to maintain image clarity across varying driving conditions. This resolves the contradiction by enabling real-time imaging at high speeds while preserving image quality through adaptive parameter adjustment.
Solution Approach 2:
The patent changes multiple parameters simultaneously (exposure time, matching window size, calculation precision) based on vehicle speed and imaging conditions. At high speeds, the system increases exposure time and adjusts stereo matching parameters to compensate for motion blur, rather than simply reducing exposure time for real-time display. This multi-parameter adjustment resolves the contradiction between real-time display and image clarity.
2Reliability
If matching window is widened to handle blurred images, then target recognition becomes possible, but calculation cost increases and parallax distance measurement accuracy decreases
Solution Approach 1:
The patent dynamically adjusts the matching window size based on the calculated blurring amount and vehicle speed. Rather than using a fixed wide matching window that degrades precision, the system optimizes the window size adaptively - using wider windows only when necessary for target recognition while maintaining tighter windows for accurate parallax measurement when conditions permit. This dynamic adjustment resolves the contradiction between recognition reliability and measurement precision.
Solution Approach 2:
The patent applies different processing qualities to different regions and objects in the image. For distant objects or objects of interest, the system uses larger matching windows to ensure recognition, while for nearby objects where precision is critical, it uses smaller windows to maintain measurement accuracy. This local differentiation resolves the contradiction by applying appropriate processing intensity where needed rather than uniformly across the entire image.
3Measurement precision
If exposure time is extended to capture clear images, then image clarity improves, but real-time display capability deteriorates
Solution Approach 1:
The patent dynamically adjusts exposure time based on real-time vehicle speed, distance to objects, and imaging conditions. At high speeds, the system extends exposure time and compensates through algorithmic adjustments to maintain clarity, while at lower speeds it uses shorter exposure times for real-time display. This dynamic adaptation resolves the contradiction by optimizing the trade-off between clarity and real-time capability based on actual driving conditions.
Solution Approach 2:
The patent changes multiple parameters in coordination - extending exposure time when needed for clarity while simultaneously adjusting stereo matching parameters, matching window sizes, and processing algorithms to compensate for the increased exposure duration. This coordinated parameter changes approach allows the system to maintain both image clarity and real-time performance by distributing the computational load and optimizing the overall processing pipeline.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An image processing system includes: an image acquisition unit that acquires a plurality of images captured by a plurality of cameras arranged so that there is an overlapping imaging region where imaging regions at least partially overlap; a representative distance calculation unit that calculates a posture change of each camera on a basis of a change in a posture of a host vehicle; a movement amount computation unit that finds an amount of change in a position of a feature point in the overlapping imaging region of the plurality of images captured by the plurality of cameras; and a three-dimensional information acquisition unit that reduces the images on a basis of the amount of change in the position of the feature point and uses the reduced images to acquire three-dimensional information of the overlapping imaging region.