Vehicle Image Processing Region Margin Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control technologies do not adequately set margins for image processing regions, leading to potential deterioration in target recognition accuracy when subsequent image processing is performed.
Innovation Solution
An information processing device and system that includes an image acquisition unit, a target detection unit, a classification identification unit, and a region determination unit, which determines an image processing region by adding a margin corresponding to the identified classification of the detected target.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a margin is added to the target region to ensure visibility of the target, then the target visibility is improved, but the target recognition accuracy deteriorates when subsequent image processing is performed
Solution Approach 1:
The patent applies different image processing qualities to different regions of the image. The target region is processed with high quality to maintain recognition accuracy, while other regions are processed with lower quality to reduce bandwidth usage. This local differentiation resolves the contradiction by ensuring the target remains visible and recognizable even when overall image quality is reduced.
Solution Approach 2:
The patent segments the image into multiple regions: a target region that requires high processing quality for accurate recognition, and other regions that can be processed with lower quality. By dividing the image processing task into different quality levels for different segments, the system maintains target recognition accuracy while reducing overall bandwidth requirements.
2Manufacturing precision
If image processing is performed on a target region with margin to improve image quality, then the image quality is improved, but the recognition accuracy deteriorates
Solution Approach 1:
The patent implements local quality control by processing only the target region with high image quality parameters, while other regions receive lower quality processing. This ensures that the target maintains its recognition accuracy without requiring the entire image to be processed at high quality, thus resolving the contradiction between image quality and recognition accuracy.
Solution Approach 2:
The patent segments the image processing task by identifying the target region and applying different processing parameters to it versus the rest of the image. The target region receives processing optimized for recognition accuracy, while other regions are processed with parameters optimized for bandwidth efficiency, resolving the quality-accuracy contradiction.
3Reliability
If a wide margin is set for targets at short distance, then the visibility is ensured, but the image processing region becomes larger increasing bandwidth usage
Solution Approach 1:
The patent applies local quality processing where only the identified target region receives high-quality processing with appropriate margins, while the rest of the image is processed with lower quality and smaller margins. This resolves the contradiction by ensuring visibility of the target without unnecessarily processing the entire image at high quality, thus reducing bandwidth usage.
Solution Approach 2:
The patent segments the image into a target region requiring high quality with appropriate margins for visibility, and non-target regions that can be processed with lower quality and smaller margins. This segmentation allows the system to maintain target visibility while significantly reducing the total bandwidth required for image transmission.
Data Source
AI summary
An information processing device includes: an image acquisition unit configured to acquire an image captured by an imaging unit mounted on a vehicle; a target detection unit configured to detect a target region including a target in the acquired image; a classification identification unit configured to identify classification including a type of the detected target and a size of the detected target region; and a region determination unit configured to determine a region obtained by adding a margin corresponding to the identified classification to the target region as an image processing region.