Unsupervised multi-view feature selection method and system based on low-rank tensor learning
A feature selection method and feature selection technology, applied in the field of image processing, can solve the problems that affect the performance of feature selection, do not consider high-level information of different views, etc., and achieve the effect of high discriminant
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0031] In one or more embodiments, an unsupervised multi-view feature selection method based on low-rank tensor learning is disclosed, combining figure 1 and 2 , including:
[0032] Step 1: Obtain a real original data set; extract several view features for each image data to obtain a multi-view feature data set;
[0033] In this embodiment, the original data set is composed of multiple image data; feature extraction is performed on each image data to obtain a multi-view feature data set; view features such as LBP, SIFT, and GIST.
[0034] Step 2: For the multi-view data set, the pseudo-label matrix of each view data is obtained based on multi-view spectral clustering and low-rank tensor learning, and the feature selection matrix is learned through a sparse regression model to construct an unsupervised model based on low-rank tensor learning. Multi-view feature selection objective function;
[0035] Step 2.1: Pseudo-label learning.
[0036] Since no label information is a...
Embodiment 2
[0083] In one or more embodiments, an unsupervised multi-view feature selection system based on low-rank tensor learning is disclosed, including:
[0084] The data set obtaining module is used to obtain the image data set, extracts several view features for each image data, and obtains the multi-view feature data set;
[0085] The objective function construction module is used to obtain the pseudo-label matrix of each view data based on multi-view spectral clustering and low-rank tensor learning, learn the feature selection matrix through the sparse regression model, and construct unsupervised multi-view based on low-rank tensor learning Feature selection objective function;
[0086] A solving module is used to solve the objective function by an iterative optimization method to obtain a feature selection matrix;
[0087] The feature selection module is used to calculate the importance of each feature according to the feature selection matrix, sort the features according to th...
Embodiment 3
[0090] In one or more embodiments, a terminal device is disclosed, including a server, the server includes a memory, a processor, and a computer program stored on the memory and operable on the processor, and the processor executes the The program implements the unsupervised multi-view feature selection method based on low-rank tensor learning in Embodiment 1. For the sake of brevity, details are not repeated here.
[0091] It should be understood that in this embodiment, the processor can be a central processing unit CPU, and the processor can also be other general-purpose processors, digital signal processors DSP, application specific integrated circuits ASIC, off-the-shelf programmable gate array FPGA or other programmable logic devices , discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
[0092] The memory may include read-only...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More