一种面向毛发图像的数据处理方法及系统
By constructing a hair data processing model and utilizing principal component analysis and adaptive threshold adjustment, the problems of poor real-time performance and low accuracy in hair image processing were solved, achieving efficient and accurate extraction and classification of hair endpoints.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI PANGYINGCE TECH CO LTD
- Filing Date
- 2023-01-11
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for hair image processing suffer from poor real-time computation, low accuracy, and loss of hair length, especially at high resolutions where effective hair endpoint extraction and classification are difficult.
By constructing a hair data processing model, principal component analysis is used to calculate the intersection of the main direction and the contour. The average nearest distance is then used to extract the hair endpoints. Adaptive threshold adjustment is used for classification to reduce the number of iterations and avoid hair length loss.
It achieves efficient and accurate extraction and classification of hair endpoints, reduces hair length loss, improves computational accuracy, and adapts to hair classification needs in different scenarios.
Smart Images
Figure CN116310314B_ABST