Deep belief network image classification protocol based on bottom layer fusion feature

Abstract

The invention provides a deep belief network image classification protocol based on a bottom layer fusion feature. The deep belief network image classification protocol comprises the steps of firstly extracting colors, textures and shape features in a sample image to form a multi-feature fusion weight matrix; then normalizing the weight matrix; and finally training and testing the deep belief network by taking the normalized weight matrix as original data to obtain a classification result. The deep belief network is a structure formed by connecting a plurality of restricted Boltzmann machine (RBM) models together and a deep structure formed by a BP (Back Propagation) network; connecting RBMs of each layer, and forming the whole deep belief network by taking output of a previous layer RBM as input of a next layer RBM and output of a final layer RBM as input of the BP network; inputting the original data to the first layer RBM, and utilizing output of the BP network as the classification result. The deep belief network image classification protocol has good expandability, the classification correct rate is improved through multi-feature fusion and the deep belief network improves the classification efficiency.

Description

technical field [0001] The invention relates to a classification algorithm of natural images, in particular to a deep belief network image classification protocol based on underlying fusion features, which is applicable to the technical fields of image information, image classification and retrieval. Background technique [0002] With the rapid development of digital technology, information technology and multimedia technology, digital images have become an indispensable part of people's daily life, and the number of images is growing at an alarming rate. Faced with more and more image information, image classification And retrieval has become the focus of research. Traditional classification and retrieval methods based on text and annotation have some disadvantages: time-consuming and labor-intensive; the rapid increase of digital images makes it almost impossible to annotate all images; annotators have a great subjective influence. This limits the development of text-base...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to overcome the problems of low classification accuracy and low efficiency of existing single feature descriptors and shallow structure classification algorithms, and provide a deep belief network image classification protocol with high classification accuracy and high efficiency

Images