Road Region Correction Using Map-Linked Partial Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques fail to accurately extract road regions from aerial images, leading to incomplete or incorrect representation of roads, which hinders the generation of high-precision-map information for automated driving.
Innovation Solution
An apparatus that segments road regions into partial regions using skeletal lines or Voronoi edges, associates these regions with real-space road sections, and corrects them to accurately cover corresponding roads using morphological operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional road extraction techniques are used from aerial images, then the extraction process is simple and fast, but the extracted road regions are inaccurate and incomplete
Solution Approach 1:
The patent segments the road extraction process into multiple stages: initial road region extraction, division into partial road regions, association with road sections, and correction. This segmentation allows each stage to focus on specific tasks, improving overall accuracy while managing complexity through modular processing
Solution Approach 2:
The patent introduces an intermediary association stage that links partial road regions with road sections from map data. This intermediary step acts as a mediator between image-based extraction and ground truth, enabling correction of extraction errors through comparison and adjustment
2Reliability
If road regions are extracted directly from aerial images, then the process is quick, but the extracted regions do not correctly cover the actual roads
Solution Approach 1:
The patent performs preliminary division of road regions into partial regions before association and correction. This preliminary organization structures the data in advance, making subsequent correction operations more efficient and targeted, reducing overall processing time while improving reliability
Solution Approach 2:
The patent implements a feedback mechanism where extracted road regions are compared with associated road sections, and corrections are applied based on this comparison. The corrected regions can be used to refine future extractions, creating a continuous improvement loop that enhances reliability over time
3Manufacturing precision
If detailed correction processes are applied to road regions, then the accuracy of road representation improves, but the processing complexity and time increase
Solution Approach 1:
The correction system is segmented into distinct functional modules: division unit, association unit, and correction unit. Each module handles specific aspects of the correction process, allowing for targeted improvements in precision while managing overall system complexity through clear separation of concerns
Solution Approach 2:
The patent applies correction operations locally to partial road regions rather than uniformly to entire road networks. This local approach allows precision to be improved where needed while minimizing unnecessary processing elsewhere, balancing precision gains with system complexity management
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An apparatus for correcting a road region includes a segmenting unit (12) that segments a road region extracted from an aerial image and representing a road into a plurality of partial road regions; an associating unit (13) that associates, for each of the partial road regions, the partial road region with a road section existing at the location of the partial road region, the road section being indicated by map information indicating locations of respective road sections; and a correcting unit (14) that corrects the road region so as to cover, for each of the partial road regions, the road section corresponding to the partial road region.