Road Boundary Image Processing for Hazard-Aware Vehicle Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In autonomous driving systems, there are disconnections between the sensing module and subsequent control signals, affecting control accuracy and credibility, leading to potential safety issues due to incomplete or inaccurate road boundary detection.
Innovation Solution
An image processing method and apparatus that acquires road images from vehicles, detects multiple road boundaries, and identifies a target road boundary dangerous to the vehicle, using convolutional neural networks and deep learning techniques to enhance detection accuracy and credibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional road boundary detection methods are used, then the system complexity is low, but the measurement precision and reliability of road boundary detection are insufficient
Solution Approach 1:
The patent replaces traditional mechanical or rule-based road boundary detection methods with deep learning-based image processing. The system uses convolutional neural networks to automatically learn and extract road boundary features from images, achieving higher detection precision while the computational complexity is managed through optimized network architectures and training approaches.
2Measurement precision
If deep learning-based road boundary detection is implemented, then the measurement precision improves, but the loss of time in processing increases
Solution Approach 1:
The patent implements preliminary action by pre-training deep learning models offline with large datasets before deployment. The models are prepared in advance with optimized parameters and feature extractors, allowing real-time or near-real-time inference during actual road boundary detection without requiring extensive processing time during operation.
Solution Approach 2:
The patent segments the image processing task into multiple stages: preprocessing (image enhancement, normalization), feature extraction (using pre-trained networks), and post-processing (boundary refinement, validation). This segmentation allows parallel processing and optimization of each stage independently, reducing overall processing time while maintaining high precision.
3Reliability
If comprehensive road boundary detection is performed, then the reliability of control signals improves, but the device complexity increases
Solution Approach 1:
The patent introduces an intermediary layer between image capture and control signal generation. This intermediary consists of multiple processing modules including image preprocessing, deep learning-based boundary detection, result validation, and control signal generation. Each module performs a specific function and passes results to the next, improving reliability through systematic processing while managing complexity through modular architecture.
Solution Approach 2:
The patent implements feedback mechanisms where detection results are validated against multiple criteria (geometric consistency, physical plausibility, temporal coherence). The system continuously monitors detection quality and adjusts processing parameters accordingly, ensuring high reliability of control signals while using feedback loops to avoid unnecessary complexity in the forward processing chain.
Data Source
AI summary
An image processing method and apparatus, a device, and a storage medium. The method includes: acquiring a road image acquired by an image acquisition apparatus mounted on a vehicle; detecting multiple road boundaries in the road image on the basis of the road image; and determining a target road boundary that is dangerous to the vehicle among the multiple road boundaries. In this case, the driving of the vehicle can be controlled more accurately on the basis of the target road boundary.