Biometric Measurement Accuracy via Selective Image Part Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing techniques for biometric information measurement are prone to accuracy deterioration due to varying imaging conditions, affecting the reliability of pulse wave velocity and other biometric data.
Innovation Solution
An image processing apparatus that detects multiple part images of a subject, determines the detectability of biometric information for each image, and selectively acquires biometric data based on this determination, using a processor to enhance measurement accuracy by choosing the most suitable image for data acquisition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image processing techniques are used to measure biometric information, then the measurement process is simple, but the measurement accuracy deteriorates due to varying imaging conditions
Solution Approach 1:
The patent divides the image processing task into multiple stages: detecting multiple candidate part images, determining detectability for each candidate, and selectively acquiring biometric information from the most suitable candidate. This segmentation allows the system to handle varying imaging conditions by evaluating multiple options rather than relying on a single image, thereby improving measurement accuracy without excessive complexity increase
Solution Approach 2:
The patent performs preliminary detection of multiple part images and evaluation of their detectability before actual biometric information acquisition. By pre-assessing the quality and suitability of different image regions, the system can select the optimal candidate for measurement, ensuring higher accuracy while managing processing complexity through structured preliminary evaluation
2Reliability
If multiple part images are detected and evaluated, then the reliability of biometric data improves, but the processing time increases
Solution Approach 1:
The patent detects multiple part images (excessive action) but only processes and acquires biometric information from the single most suitable candidate (partial action). By evaluating multiple candidates through detectability determination and then selecting only the best one for actual measurement, the system enhances data reliability while limiting the time-consuming processing to just the selected candidate rather than all detected images
3Stability of the object's composition
If selective acquisition based on detectability is implemented, then the consistency of obtained data improves, but the system complexity increases
Solution Approach 1:
The patent introduces detectability as a key parameter for evaluating and selecting part images. By using detectability determination to assess each candidate image's suitability and selecting based on this parameter, the system ensures consistent data quality across different imaging conditions. The complexity is managed by focusing the selection process on this single critical parameter rather than multiple complex criteria
Data Source
AI summary
An image processing apparatus includes an image analysis unit, a biometric information measurement unit, and a determination selection unit. The image analysis unit detects a plurality of part images of a subject from the image including the subject captured by the imaging unit. The biometric information measurement unit obtains biometric information from at least one of the plurality of part images of the subject detected by the image analysis unit.The determination selection unit determines the detectability of biometric information on each of the plurality of part images of the subject detected by the image analysis unit. The determination selection unit makes a selection regarding acquisition of the biometric information of the subject on the basis of the detectability of the biometric information determined by the determination selection unit.