A semi-supervised image classification method and system

Abstract

The invention discloses a semi-supervised image classification method and system, belonging to the field of computer technology, including: collecting original data and converting the original data into a sample set; assuming L={1, 2, ..., c, ..., l} Represents the set of label values, c=1, 2,..., l, establishes a state transition matrix n=1, 2,..., l, T m，n Indicates the transition probability from state m to state n; according to the smallest element on the diagonal in the state transition matrix, set the labeling interval; mark the samples according to the labeling interval, and obtain the labeled sample set and the unlabeled sample set; use the The labeled sample set and the unlabeled sample set are used to train the constructed support vector machine model, and a trained support vector machine is obtained. The sample classification of the invention well reflects the actual situation of data classification, and the model is trained by using the classified data, which improves the accuracy of semi-supervised classification.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a semi-supervised image classification method and system. Background technique [0002] How to utilize massive amounts of data is an important task facing current machine learning. The traditional support vector machine is a supervised learning method that requires a large number of labeled samples for training. However, in practical applications, since most of the sample data that can be used is unlabeled, and there are fewer labeled sample points, if only these few labeled samples are used, it will result in information that exists in a large number of location-labeled samples. was lost. Therefore, some scholars have proposed a method of semi-supervised learning, which is to use both unlabeled sample data knowledge and a small amount of labeled sample data knowledge in the semi-supervised learning process. However, existing semi-supervised learning methods do not fully expl...

AI Technical Summary

Problems solved by technology

However, the existing semi-supervised learning methods do not fully exploit the spatial smoothness, resulting in a large number of misclassifications.

Images