Road Data Extraction Using Inpainted Ground Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face challenges in accurately extracting and analyzing road data, particularly in automated driving systems, where detailed road information is crucial for navigation and self-driving cars, and existing methods lack efficient methods for filling missing data and identifying road damage.
Innovation Solution
A method and system that involves obtaining ground data from environment detection data, building a ground model, filling in missing regions, identifying road attributes, and storing road data, which includes using digital inpainting to optimize the ground model and perform road damage identification, enabling the generation of precise road data for self-driving car control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If ground model is built from environment detection data, then road data extraction is enabled, but missing regions in the ground model cause incomplete road information
Solution Approach 1:
The patent applies preliminary action by performing digital inpainting to fill missing regions in the ground model before extracting road attributes. This pre-processing step ensures that road data extraction is performed on a complete ground model, preventing information loss. The system proactively identifies and fills gaps in the ground model using surrounding terrain data and statistical methods, so that when road attribute extraction occurs, all necessary information is already available in the completed ground model.
2Measurement precision
If traditional road data extraction methods are used, then processing is simpler, but road damage identification accuracy is insufficient
Solution Approach 1:
The patent applies segmentation by dividing the road data extraction process into distinct stages: ground model construction, missing region filling, road attribute extraction, and road damage identification. Each stage processes specific aspects of road data independently, allowing for specialized algorithms at each step. This segmented approach enables sophisticated damage identification through multiple analysis steps while keeping each individual processing stage manageable and modular.
Solution Approach 2:
The patent introduces the completed ground model as an intermediary between raw environment detection data and final road damage identification. This intermediate ground model serves as a refined representation that bridges the gap between noisy sensor data and accurate road attribute extraction. The ground model acts as a mediator that structures and cleans the data before damage identification algorithms are applied, improving accuracy while managing complexity through staged processing.
3Measurement precision
If comprehensive road data is extracted including damage identification, then navigation accuracy improves, but data processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-processing environment detection data into a completed ground model with filled missing regions before road attribute extraction. This pre-completed ground model serves as refined input for subsequent navigation and damage identification tasks, reducing the computational burden during real-time operation. By performing data completion and initial processing in advance, the system reduces real-time processing time while maintaining high navigation accuracy.
Data Source
AI summary
A method for extracting road data comprises obtaining ground data from first environment detection data of a detected region. A first ground model may be obtained using the ground data. A missing region of the first ground model may be filled to obtain a second ground model. Road attributes of the detected region may be obtained according to the second ground model. Road damage recognition may be performed using second environment detection data of the detected region and the second ground model. The method may include storing the road attributes and the result of the road damage recognition as pieces of road data of the detected region.