Image data classification system based on Universum with combination of matrix Ho-Kashyap algorithm

Abstract

The invention provides an image data classification system based on Universum with combination of a matrix Ho-Kashyap algorithm. Firstly, a certain number of a third kind of sample points located between two kinds of samples, namely Universum samples, are generated by employing an In-Between-based generation strategy; the Universum sample points are substituted into a regularization item Runi; the regularization item is introduced into a HK classification model after matrixing to form a complete matrixing HK model with combination of Universum; and finally, the model is trained, optimal parameters of the model for a current training data set are obtained, and an optimal classification decision surface is generated. In the test stage, the test sample points are substituted into a decision surface function for determination, and classification labels are output. Compared with the conventional classification technology, according to the system, the Universum samples are introduced, the comparison of the original two kinds of samples is more obvious, and the accuracy is further improved.

Description

technical field [0001] The invention relates to the technical field of pattern classification, in particular to a Universum combined matrix Ho-Kashyap algorithm and system for identifying and processing image data sets. Background technique [0002] Pattern recognition is the study of using computers to imitate or realize the recognition ability of humans or other animals in order to complete the task of automatic recognition of research objects. In recent years, pattern recognition technology has been widely used in many important fields such as artificial intelligence, machine learning, computer engineering, robotics, neurobiology, medicine, detective science, archaeology, geological exploration, aerospace science and weapon technology. One of the classic problems that pattern recognition needs to deal with is the processing of two-dimensional data, that is, data represented by a matrix. In practical applications, data represented by matrices is often used in image recogn...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the complex structure, low efficiency and low precision of the existing technology, which cannot meet the image problems of accuracy, real-time, or lack of prior knowledge, the present invention provides a classification method based on the Universum combination matrix Ho-Kashyap algorithm, which is used for binary classification The problem is to first generate inter-class Universum samples through the classic In-Between technology, then design a two-dimensional Ho-Kashyap (HK) algorithm model, and then design a regularization item that characterizes the relationship between Universum samples and original samples and Substitute it into the module designed in the second step, and finally use the gradient descent method to solve the optimal parameters for the entire model. The obtained decision boundary can improve the efficiency of model design and model operation while ensuring the classification accuracy of the image data set

Images