Data Recognition Using Segmented Feature Components
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data recognition technologies face challenges in accurately and efficiently identifying users based on input data, such as images or voice signals, due to limitations in handling variations and complexities in feature data distributions.
Innovation Solution
A processor-implemented method that extracts input feature data, calculates a matching score using common and distribution component data, and recognizes the input data by accumulating partial scores across indices, allowing for partial recognition before complete score calculation, thereby enhancing recognition speed and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complete feature data is used for recognition, then recognition accuracy is improved, but computational time and complexity increase
Solution Approach 1:
The patent segments the feature data into common component data (mean feature vectors) and distribution component data (variance feature vectors). This segmentation allows the system to process and recognize data using only the common components for initial matching, reducing computational time while maintaining acceptable accuracy, and only resorting to distribution components when necessary.
Solution Approach 2:
The patent applies partial action by calculating matching scores based on only the common component data (mean feature vectors) rather than using all available feature data. This partial calculation provides sufficient information for recognition decisions in many cases, avoiding the need to process the complete feature data and thus reducing computational time and complexity.
2Measurement precision
If detailed feature analysis is performed, then recognition accuracy is improved, but device complexity increases
Solution Approach 1:
The patent divides the complex feature data into two manageable segments: common component data representing the mean characteristics and distribution component data representing the variance characteristics. This segmentation simplifies the computational process by allowing sequential processing - first matching common components, then only必要时 evaluating distribution components, thereby reducing overall device complexity.
Solution Approach 2:
The patent extracts and utilizes only the essential common component data (mean feature vectors) for the primary recognition task. By taking out and focusing on this essential information, the system achieves recognition functionality with significantly reduced computational complexity, avoiding the need to process all detailed feature data simultaneously.
3Reliability
If all feature data components are processed, then recognition reliability is improved, but processing speed decreases
Solution Approach 1:
The patent segments the processing into two stages: first processing common component data (mean feature vectors) for rapid initial matching, then only必要时 processing distribution component data (variance feature vectors). This segmentation maintains recognition reliability by considering both common and distribution components when needed, while significantly improving processing speed by avoiding unnecessary computation in the first stage.
Solution Approach 2:
The patent applies partial action by performing recognition using only common component data, which provides sufficient reliability for many recognition tasks. The system only resorts to processing distribution component data when the partial match is insufficient, thereby maintaining high processing speed while ensuring adequate recognition reliability through selective comprehensive analysis.
Data Source
AI summary
A processor-implemented method with data recognition includes: extracting input feature data from input data; calculating a matching score between the extracted input feature data and enrolled feature data of an enrolled user, based on the extracted input feature data, common component data of a plurality of enrolled feature data corresponding to the enrolled user, and distribution component data of the plurality of enrolled feature data corresponding to the enrolled user; and recognizing the input data based on the matching score.