Track Boundary Approximation Using Segmented Image Curve Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for estimating road markings for vehicle travel control are prone to errors and blurring due to restrictions in curve fitting and variations in image extraction, leading to unstable road marking estimation for vehicle control.
Innovation Solution
An image processing device and method that divides an image into areas to generate element functions approximating track boundaries based on probability values, combining these functions with a weighting system to create a composite function for accurate road marking estimation and driving control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a curved line is fitted to a plurality of positions extracted from an image to estimate road markings, then the estimation process is simple, but the error in the estimated road markings is large and blurring occurs due to variations in extracted positions
Solution Approach 1:
The image is divided into multiple divided areas, and road marking estimation is performed separately for each area using element functions. This segmentation allows localized accurate fitting while maintaining overall simplicity, resolving the contradiction between simple processing and high accuracy.
Solution Approach 2:
Different element functions are used for different divided areas based on local characteristics. Each area's road marking is estimated with a function tailored to its specific properties, improving local accuracy while keeping the overall system simple through modular processing.
2Device complexity
If a single composite function is used to approximate the entire track boundary, then the model is simple, but it cannot capture local variations and produces blurring
Solution Approach 1:
The track boundary approximation is segmented into multiple element functions, each handling a specific divided area. This creates a composite model that is more complex than a single function but remains manageable through modular structure, while significantly improving reliability by capturing local variations.
Solution Approach 2:
The approximation model uses a composite structure combining multiple element functions (analogous to composite materials). Each element function contributes to the overall approximation, creating a robust model that maintains stability while adapting to local characteristics of different road marking sections.
3Productivity
If curve fitting is performed with restricted degrees of freedom to maintain simplicity, then the model is easy to compute, but the estimation error increases
Solution Approach 1:
The computational task is segmented into multiple element function fittings, each with restricted degrees of freedom for efficiency. The overall accuracy is maintained by combining these simple local fittings into a composite model, achieving both computational efficiency and high precision.
Solution Approach 2:
Instead of performing a single complex curve fitting with high degrees of freedom, the system performs multiple simpler partial fittings (element functions) across divided areas. This partial action approach maintains computational efficiency while achieving superior overall accuracy through the composite model.
Data Source
AI summary
An image processing device divides an image representing an area in front of a mobile object, which is captured by a camera mounted on the mobile object, at predetermined intervals to generate an element function that approximates a track boundary in each of divided areas on the basis of: a probability value indicating an existence probability of the track boundary for each of coordinates in each of the areas; and the coordinates corresponding to the probability value, generates a composite function that approximates a track boundary in the front area by combining the element functions generated for each of the areas, and executes driving control or driving assistance of the mobile object on the basis of the track boundary approximated by the generated composite function.