Map Landmark Comparison Using F1 Score Sub-Area Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for comparing maps with landmarks are inadequate for assessing the quality of maps used for navigation, particularly in automated vehicle systems, as they fail to effectively evaluate the similarity and accuracy of landmarks between different maps.
Innovation Solution
A method and device for comparing two maps with landmarks by determining the similarity between sub-areas of the maps based on landmark matches, using a similarity value calculated from precision and recall metrics, specifically the F1 score, to assess map quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing methods for comparing maps are used, then the comparison process is simple, but the accuracy of map quality assessment is insufficient
Solution Approach 1:
The patent divides the map comparison process into multiple sub-area comparisons. Each sub-area is independently analyzed for landmark matches, and results are aggregated to produce an overall similarity value. This segmentation enables precise local assessment while maintaining systematic overall evaluation, resolving the contradiction between assessment accuracy and method complexity.
Solution Approach 2:
The patent introduces multiple quantitative parameters including precision, recall, and F1 score to evaluate landmark matches. By changing from simple presence/absence checks to multi-parameter statistical evaluation, the system achieves higher map quality assessment accuracy while the parameters provide a structured framework that manages complexity.
2Reliability
If landmark matching is performed across entire maps, then comprehensive quality assessment is achieved, but computational complexity increases
Solution Approach 1:
The patent segments the map into multiple sub-areas and performs landmark matching independently in each sub-area. This approach maintains comprehensive assessment reliability by covering the entire map while reducing computational power requirements through localized processing and parallelizable operations.
Solution Approach 2:
The patent performs landmark matching on sub-areas rather than requiring complete simultaneous processing of entire maps. This partial action approach achieves sufficient reliability through aggregated sub-area results while significantly reducing the computational power needed at any given moment.
3Measurement precision
If multiple similarity metrics are calculated, then assessment accuracy improves, but processing time increases
Solution Approach 1:
The patent calculates precision and recall metrics for each sub-area before combining them into the final F1 score. This preliminary calculation of component metrics enables accurate overall assessment while the modular structure allows efficient computation that minimizes total processing time.
Solution Approach 2:
By segmenting the similarity measurement into separate precision and recall calculations performed on sub-areas, the patent achieves accurate multi-metric assessment while enabling parallel processing and efficient aggregation, thereby reducing overall processing time compared to computing a single complex metric.
Data Source
Figure 1
Figure 2
Figure 3a~3b
AI summary
The invention relates to a method for comparing two maps (10, 20) with landmarks (A, B) embedded therein, comprising the following steps: comparing at least one sub-area (11) of the first map (10) with a corresponding sub-area (21) in the second map (20), wherein a similarity between the sub-area (11) of the first map (10) and the corresponding sub-area (21) in the second map (20) is determined based on a correspondence between the landmarks (A, B) embedded in each sub-area (11, 21), and wherein the similarity of the compared sub-area (11, 21) is expressed in the form of a derived similarity value (5). The invention further relates to an associated device (1), an associated computer program with program code means, and an associated computer program product with program code means.