Scenery Image Database for Mobile Object Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing positioning systems face challenges in accurately identifying the location of mobile objects, especially in environments where radio wave reception is limited, and require high computational effort due to the need to handle feature points from objects that change over time, such as vehicles and plants.
Innovation Solution
A scenery image database is created by correlating feature points from scenery images with their locations, excluding areas likely to change, such as vehicles and plants, to reduce the number of feature points stored, thereby reducing the computational effort required for location identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature points from all objects including vehicles and plants are stored in the database, then the positioning accuracy is improved, but the computational effort and processing time increase significantly
Solution Approach 1:
The patent extracts and removes feature points corresponding to objects that change over time (vehicles, plants, pedestrians) from the scenery image database. By separating these dynamic objects from static background features, the system reduces the number of feature points that need to be processed during positioning, thereby decreasing computational effort and processing time while maintaining positioning accuracy through the use of stable, permanent features.
Solution Approach 2:
The patent performs preliminary processing to identify and remove dynamic object feature points during database creation, before the actual positioning operation. This advance preparation ensures that only stable, permanent features are stored in the database, eliminating the need for complex real-time filtering during positioning operations and reducing processing time.
2Measurement precision
If feature points from all objects including vehicles and plants are stored in the database, then the positioning accuracy is improved, but the computational resources and processor cost increase
Solution Approach 1:
The patent extracts and removes feature points corresponding to objects that change over time (vehicles, plants, pedestrians) from the scenery image database. By separating these dynamic objects from static background features, the system reduces the number of feature points that need to be processed during positioning, thereby decreasing computational effort and processing time while maintaining positioning accuracy through the use of stable, permanent features.
3Reliability
If trial-and-error methods like RANSAC are used for location identification, then the robustness against erroneous feature points is improved, but the computational effort increases
Solution Approach 1:
The patent extracts and removes feature points corresponding to objects that change over time (vehicles, plants, pedestrians) from the scenery image database. By separating these dynamic objects from static background features, the system reduces the number of feature points that need to be processed during positioning, thereby decreasing computational effort and processing time while maintaining positioning accuracy through the use of stable, permanent features.
Solution Approach 2:
The patent performs preliminary processing to identify and remove dynamic object feature points during database creation, before the actual positioning operation. This advance preparation ensures that only stable, permanent features are stored in the database, eliminating the need for complex real-time filtering during positioning operations and reducing processing time.
Data Source
AI summary
An object of the present invention is to provide a positioning system which makes it possible to perform positioning processing in a positioning target mobile object with a smaller calculation amount.A scenery image database according to an aspect of the present invention is characterized by comprising a plurality of scenery image data and image acquisition locations at which the scenery image data have been acquired are correlated with each other and stored, wherein each of the plurality of scenery image data include a feature amount of a feature point corresponding to a thing other than a thing of which a location or a shape is unlikely to be kept as it is on the real world during a period larger than or equal to a predetermined period.