Human face recognition method based on supervision isometric projection
A face recognition and isometric technology, applied in the field of image processing, can solve the problems of high computational complexity of test data points, affecting the recognition rate of algorithms, and the inability of manifold learning algorithms to effectively eliminate redundant information, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0044] Hereinafter, the present invention will be described in more detail with examples in conjunction with the accompanying drawings:
[0045] The implementation steps are as follows:
[0046] (1) Face sample training process
[0047] ① First preprocess the face training image to obtain the original training sample matrix X in the high-dimensional space; the processing here is to crop each training image, set its resolution to 64×64, and then perform downsampling. To achieve a resolution of 32×32. Finally, each image is normalized with a mean value of 0 and a variance of 1.
[0048] ②Gabor wavelet filters the image. If I(x,y) represents the original image, then the new image feature x i ′ = I ( x , y ) ⊗ Φ u , v ( x , y ) , Represents convolution, stacking new features into an M-dimensional long vector x i , Which constitutes an original training sample. Then all training images form an original training sample matrix X=[...
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.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap