Image data dimension reduction method and system based on discriminant regularization local reservation projection
A technology that locally preserves projection and image data, applied in the field of image recognition, can solve problems such as insufficient consideration of differences in distribution characteristics, avoid distortion, prevent the introduction of too much non-local information, and improve local essential structural features.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment approach
[0068] As an implementation manner, step 108 specifically includes:
[0069] Obtain the image to be reduced in dimension;
[0070] Cutting the image to be dimensionally reduced into a sample vector x;
[0071] According to Y=W 'T x performs dimensionality reduction on the image to be reduced in dimensionality, wherein W' is a target projection matrix, and Y is data after dimensionality reduction of the image to be reduced in dimensionality.
[0072] As an implementation manner, the method provided in this embodiment also includes:
[0073] Get the test sample image;
[0074] Cutting the test sample image into test sample vectors;
[0075] performing dimensionality reduction on the test sample vector by using the target projection matrix;
[0076] A classifier is used to identify the test sample vector after dimensionality reduction;
[0077] Evaluate the pros and cons of the projection matrix according to the recognition result.
[0078] In this embodiment, the dimensio...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com