Vehicle Distance Estimation Using Selective Super-Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately calculating the distance between vehicles on a road with high processing load and speed, especially in varying environmental conditions, which affects the safety and efficiency of advanced driver-assistance systems and automated driving.
Innovation Solution
An information processing apparatus that extracts a target area from an environment image for super-resolution processing, rather than the entire image, and adjusts the processing based on environmental factors like weather and illuminance, allowing for accurate distance calculation while minimizing processing load and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If super-resolution processing is performed on the entire environment image, then measurement precision of object size is improved, but productivity (processing speed) deteriorates
Solution Approach 1:
The patent divides the environment image into multiple regions and identifies candidate areas containing specific objects (vehicles, pedestrians, cyclists). Super-resolution processing is applied only to these candidate areas rather than the entire image, thereby maintaining measurement precision while reducing processing load and improving processing speed.
Solution Approach 2:
The patent extracts target areas (candidate areas containing specific objects) from the environment image and performs super-resolution processing only on these extracted regions. This selective extraction approach ensures accurate size measurement of objects while significantly reducing the processing burden compared to processing the entire image.
2Measurement precision
If super-resolution processing is performed on the entire environment image, then measurement precision of object size is improved, but processing load increases
Solution Approach 1:
The environment image is segmented into multiple regions, and super-resolution processing is applied only to candidate areas containing specific objects. This segmentation strategy maintains measurement precision while reducing processing load by avoiding unnecessary processing of empty or irrelevant image regions.
Solution Approach 2:
Candidate areas containing specific objects are extracted from the environment image, and super-resolution processing is performed only on these extracted regions. This approach reduces processing load while ensuring accurate size measurement of the objects of interest.
3Measurement precision
If a lower threshold value is set for target area extraction, then measurement precision is improved, but productivity deteriorates
Solution Approach 1:
The patent dynamically adjusts the threshold value for distance determination based on environmental conditions (weather, illuminance, luminance). In low-image-quality environments, a lower threshold triggers super-resolution processing for closer objects, improving measurement precision. In high-image-quality environments, a higher threshold reduces processing by only processing distant objects, thereby maintaining productivity.
Solution Approach 2:
The threshold value parameter is changed based on environmental conditions. When image quality is low (poor weather, low illuminance), the threshold is lowered to include more objects for super-resolution processing, improving measurement precision. When image quality is high, the threshold is raised to reduce processing load, maintaining productivity.
Data Source
AI summary
[Object] To calculate a distance between vehicles traveling on a road accurately and with a low load.[Solving Means] An information processing apparatus includes: a candidate area detection unit that detects a candidate area including a specific object from an environment image acquired by an imaging apparatus; a target area extraction unit that extracts a target area from the candidate area, the target area being a target of super-resolution processing; and a super-resolution processing unit that generates a super-resolution image by performing super-resolution processing on the target area.