Machine learning recognition method based on embedded coding and contrast learning

Abstract

The invention discloses a machine learning recognition method based on embedded coding and contrast learning, and the method comprises the steps: carrying out the different learning training of a machine learning model f1 for many times through a certain number of multimedia data samples of a known type and different contrast sample input arrangement sequences, so as to carry out the type recognition of the multimedia data. The machine learning model f1 is designed to be a combined model framework of a coding function model and a convolutional neural network model or a full-connection neural network model, thereby greatly reducing the dependence on the mass training samples. Moreover, the method can be conveniently extended for the type recognition of the multimedia data which is not trained, solves problems that the actual application and universality are limited because a conventional multimedia data classification machine learning recognition method depends on the mass training samples and cannot directly achieve the classification recognition of the types which are not trained, and can be applied to more specific multimedia data classification application occasions more widelyand effectively.

Description

technical field [0001] The invention relates to the fields of multimedia data processing technology and machine learning technology, in particular to a machine learning identification method based on embedded coding and comparative learning. Background technique [0002] Multimedia (Multimedia) is a combination of multiple media. In computer systems, multimedia refers to a human-computer interactive information exchange and dissemination media that combines two or more media. The media used include text, pictures, photos, and sound. , animations and videos, and the interactive features provided by the program. [0003] With the advent of the era of big data, the classification and mining technology of massive multimedia data is particularly important. In massive data mining, how to use the information classified and mined from existing data to guide the classification and mining of new data has become a new research hotspot. Especially when the number of samples for certai...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the above-mentioned deficiencies in the prior art, the technical problem to be solved by the present invention is how to provide a machine learning recognition method based on embedded coding and comparative learning to solve the problem that the existing multimedia data classification machine learning recognition method needs to rely on a large number of Training samples lead to the problem of limited practical application, and further solve the problem that the existing multimedia data classification machine learning recognition methods cannot directly classify and recognize categories that have not been trained, resulting in limited versatility

Images