Local spline embedding-based orthogonal semi-monitoring subspace image classification method

Abstract

The invention discloses a local spline embedding-based orthogonal semi-monitoring subspace image classification method. The method comprises the following steps of: 1) selecting n samples serving as training sets and the balance serving as testing sets from image data sets, wherein the training sets comprise marked data and unmarked data; 2) building an extra-class divergence matrix and an intra-class divergence matrix by using the marked data; (3) training data characteristic space distribution by using a whole and building a Laplacian matrix in a local spline embedding mode; 4) according toa local spline, embedding an orthogonal semi-monitoring subspace model, and searching a projection matrix to perform dimensionality reduction on the original high dimension characteristic; 5) building a classifier for the training samples after the dimensionality reduction by using a support vector machine; and 6) performing the dimensionality reduction on the testing sets by using the projectionmatrix and classifying the testing sets after the dimensionality reduction by using the classifier. In the method, the information, such as image sample marking, characteristic space distribution andthe like, is fully utilized; potential semantic relevance among image data can be found out; and image semantics can be analyzed and expressed better.

Description

technical field [0001] The invention relates to an orthogonal semi-supervised subspace image classification method based on local spline embedding. This method extracts features from image data and expresses them as feature vectors, and finds an effective dimensionality reduction method to project them into a low-dimensional semantic space, so as to classify image data by training a classifier model. Background technique [0002] With the popularization of digital cameras and the development of the Internet, the amount of image data collection, storage and access has exploded, and how to efficiently manage the increasingly large-scale image data has become increasingly important. In order to manage images better, an effective means is to manage image data in categories. Therefore, classifying image data has become a hot research issue in recent years. [0003] In the research of image classification, the biggest challenge is the semantic gap, that is, the underlying featur...

AI Technical Summary

Problems solved by technology

Although supervised algorithms can effectively improve the accuracy of image expression, existing supervised algorithms still have limitations, especially obtaining accurate annotation information takes a lot of time

With the rapid increase in the amount of image data, it is more convenient to obtain unlabeled data, which makes the limitations of supervised learning algorithms that can only process labeled data more obvious.

Images