Data Generator Validity Scoring Using Spatio-Temporal Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for an effective method to evaluate the validity of virtual data generated through synthesis, ensuring it meets specific demand scenarios and requirements.
Innovation Solution
A test method and system for evaluating the validity of a data generator, involving the formation of matching and mismatching combinations of targets and backgrounds based on predefined rules, followed by data generation and comparison to determine sub-scores and a total score for the generator's validity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If virtual synthesis is used to generate new data, then data quantity and diversity are improved, but data validity and reliability deteriorate
Solution Approach 1:
The patent implements a feedback mechanism by comparing generated virtual data with real reference data across multiple dimensions (spatial distribution, temporal characteristics, feature correlations). This feedback loop enables continuous validation and adjustment of the data generator, ensuring that synthesized data maintains high reliability while achieving data augmentation goals.
Solution Approach 2:
The patent transforms the abstract concept of data validity into measurable parameters including spatial distribution consistency, temporal characteristic alignment, and feature correlation accuracy. By changing validation from qualitative to quantitative parameter-based assessment, the system can objectively evaluate and ensure data reliability.
2Measurement precision
If multiple test combinations are created to evaluate data generator validity, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the validity assessment into three independent modules: spatial distribution validation, temporal characteristic validation, and feature correlation validation. Each module focuses on specific aspects of data quality, allowing comprehensive measurement precision without requiring a monolithic complex system.
Solution Approach 2:
The test system is designed with multi-functional components that can perform multiple validation tasks. For example, the comparison module can simultaneously evaluate spatial distribution, temporal characteristics, and feature correlations across different data types, reducing overall system complexity while maintaining high measurement precision.
Data Source
AI summary
A validity testing method and system for a data generator, and an electronic device. The method includes: obtaining a plurality of test targets and test backgrounds and a plurality of matching rules; obtaining physical fourth spatio-temporal feature data of a matching combination; forming a first mismatching combination and a second mismatching combination in both of which a background does not conform to the corresponding matching rule; respectively obtaining corresponding first spatio-temporal feature data and second spatio-temporal feature data; obtaining third spatio-temporal feature data by means of a data generator; comparing the third spatio-temporal feature data with the fourth spatio-temporal feature data, and obtaining a sub-score of the validity of the data generator; obtaining a plurality of sub-scores on the basis of the plurality of matching rules; and forming a total score of the validity of the data generator. The method assist evaluating data generator can for a desired data augmentation scenario.