Dictionary learning method, visual word bag characteristic extracting method and retrieval system

Abstract

The invention provides a dictionary learning method. The dictionary learning method includes 1), dividing local characteristic vector of images into first segments and second segments on the basis of dimensionality; 2) establishing a first data matrix by the first segments of a plurality of local characteristic vectors, and establishing a second data matrix by the second segments of a plurality of local characteristic vectors; 3) subjecting the first data matrix to sparse non-negative matrix factorization to obtain a first dictionary sparsely coding the first segments of the local characteristic vectors; subjecting the second data matrix to sparse non-negative matrix factorization to obtain a second dictionary sparsely coding the second segments of the local characteristic vectors. The invention further provides a visual word bag characteristic extracting method for sparsely indicating the local characteristic vectors of the images segment by segment on the basis of the dictionaries and provides a corresponding retrieval system. Memory usage can be greatly reduced, wordlist training time and characteristic extraction time are shortened, and the dictionary learning method is particularly suitable for mobile terminals.

Description

Technical field [0001] The present invention relates to the technical field of multimedia content analysis and retrieval. Specifically, the present invention relates to a dictionary learning, a feature extraction method of visual bag of words, and a retrieval system. Background technique [0002] Simply put, visual search is "search for pictures with pictures". In order to realize visual search, it is first necessary to build an index library for extracting features from a large-scale image library. When a user searches, extract features from the query image, quickly search in the feature index library, sort by relevance (ie similarity), and return results. The result is a sorted list of images in the library, where each result image may contain related information related to the query image, combined with user characteristics and search scenarios. At present, traditional PC-oriented visual search has accumulated a large number of algorithms and technical solutions to choose fro...

AI Technical Summary

Problems solved by technology

[0006] (3) The processor performance, memory and power resources of the mobile terminal are limited, so it is necessary to study the feature extraction and representation algorithm suitable for the mobile terminal to meet the requirements of the memory occupation, processing speed and accuracy of the mobile terminal in practical applications

At the same time, because the K-Means clustering tends to focus on areas with high data density, some areas with significant features but less data are merged, which leads to a great decrease in the discrimination of descriptors.

[0014] To sum up, although the large vocabulary BoW has achieved great success in the field of PC visual retrieval, all current large vocabulary BoW methods cannot be applied to mobile terminals with limited computing resources. The biggest obstacle is the presence of words. The problem that the table occupies too much memory

The existing large vocabulary BoW method, a 1M-dimensional vocabulary (1M×128) occupies up to 512M of memory, even if the mobile phone memory can store such a large vocabulary, memory loading and calculation time are still a big problem

Images