Road Network Extraction Using Trajectory and Satellite Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of modern traffic networks due to rapid urbanization makes it challenging to accurately extract and update road data in navigation systems, requiring efficient methods to combine user trajectory data and satellite aerial images for accurate road network extraction.
Innovation Solution
A method that extracts a road network by combining user trajectories, satellite aerial images, and a pre-trained road recognition model, utilizing region heat and trajectory density peak profiles to refine the extraction process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If road network extraction is performed using traditional single-source methods, then the extraction process is simple, but the accuracy and timeliness of road data updates deteriorate due to inability to capture complex urban traffic changes
Solution Approach 1:
The patent merges multiple data sources including user trajectories, satellite aerial images, and pre-trained road recognition models into a unified extraction framework. This combination allows the system to leverage complementary information from different sources, improving extraction accuracy while managing complexity through integrated processing
2Productivity
If multiple data sources are combined for road network extraction, then the accuracy and timeliness improve, but the processing complexity and computational requirements increase
Solution Approach 1:
The patent employs pre-trained road recognition models that have been previously trained on large datasets. This preliminary action allows the model to have pre-acquired knowledge about road patterns and features, enabling faster and more accurate extraction from new data sources without requiring extensive real-time computation
Solution Approach 2:
The patent introduces region heat maps and trajectory density peak profiles as intermediary representations that bridge raw user trajectory data and final road network extraction. These intermediaries simplify the processing by providing structured summaries of mobility patterns, reducing computational complexity while maintaining accuracy
3Reliability
If user trajectory data is used for road network extraction, then the timeliness of road updates improves, but the accuracy may deteriorate due to noise and irregularities in trajectory data
Solution Approach 1:
The patent uses region heat maps and trajectory density peak profiles as intermediary representations that filter and structure raw user trajectory data. These intermediaries aggregate individual trajectories into meaningful patterns, reducing noise and irregularities while preserving the timeliness information from user mobility data
Solution Approach 2:
The patent combines user trajectory data with satellite aerial images and pre-trained models, allowing the strengths of each source to compensate for the weaknesses of others. The satellite images provide accurate spatial references that help correct inaccuracies in trajectory-based extraction
Data Source
AI summary
Provided are a road network extraction method, a device, and a storage medium, which relate to the technical field of artificial intelligence and, in particular, to the fields of image processing, computer vision, and the like and are specifically applicable to scenarios such as intelligent transportation and a smart city. A specific implementation scheme includes: extracting a first road network of a target region according to user trajectories of the target region; extracting a second road network of the target region according to a satellite aerial image of the target region; and extract a target road network of the target region according to the first road network, the second road network, and the user trajectories. Efficient and accurate road network extraction can be achieved through techniques in embodiments of the present disclosure.