Vehicle Assist Super-Resolution Road Structure Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing road sign recognition technologies using camera images struggle with low image resolution, particularly in distant regions, leading to inaccurate recognition of road structures.
Innovation Solution
A vehicle assist method that extracts stationary object regions from multiple images, aligns them based on movement, performs super-resolution processing to enhance image resolution, and recognizes road structures using the resulting high-resolution image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If template matching is used for road sign recognition, then the recognition process is simple and fast, but the recognition accuracy deteriorates in regions with low image resolution
Solution Approach 1:
The patent combines template matching with super-resolution processing to create a hybrid recognition system. Multiple low-resolution images are merged through alignment and super-resolution techniques to generate a high-resolution composite image, which is then used for accurate road structure recognition while maintaining processing efficiency
Solution Approach 2:
The patent transitions from 2D template matching in the spatial domain to a multi-dimensional approach by capturing images at different times, aligning them based on vehicle movement, and reconstructing a high-resolution image in the frequency domain through super-resolution processing
2Measurement precision
If multiple images are captured and processed to improve resolution, then the recognition accuracy improves, but the processing time and computational complexity increase
Solution Approach 1:
The patent performs preliminary alignment of multiple images based on vehicle movement data before super-resolution processing. By pre-aligning images using movement amounts calculated from vehicle behavior, the system reduces the computational burden during the super-resolution stage and enables real-time processing
Solution Approach 2:
The patent segments the image processing task into distinct stages: capturing multiple images, calculating movement amounts, aligning images based on movement, performing super-resolution processing, and recognizing road structures. This segmentation allows each stage to be optimized independently, reducing overall processing time
Data Source
AI summary
A vehicle assist device includes a camera that captures the surroundings of a host vehicle at different times, and a controller that processes the plurality of images captured at different times by the camera. The controller extracts a stationary object region, which is a region corresponding to a stationary object, from each of a plurality of images captured at different times, aligns the plurality of images based on the movement amount of the stationary object in the image in the stationary object region, performs super-resolution processing using the plurality of aligned images to generate a super-resolution image that exceeds the resolution of the image captured by the camera, and recognizes a road structure based on the super-resolution image.