Efficient image retrieval method based on discrete local linear imbedding Hash

Abstract

The invention discloses an efficient image retrieval method based on discrete local linear imbedding Hash and relates to image retrieval. For images in an image library, a part of images are randomlyselected as a training set, and corresponding image characteristics are extracted; a principal component analysis method is adopted, and dimensions of original image characteristics are lowered till the dimensions are equal to the lengths of Hash codes; a similarity relation matrix of training samples and an optimal reconstruction weight of each data point are established; a corresponding Hash function is learned through iterative optimization; the corresponding Hash function is output, and the Hash codes of the whole image library are calculated; for query images, firstly, corresponding characteristics are extracted, then the image characteristics are subjected to Hash coding by adopting the same method according to a trained Hash coding function, the Hamming distance between the Hash codes of the query images and image characteristic codes of the image library is calculated, the similarity between the query images and images to be retrieved in the image library is measured by using the Hamming distance, and images with high similarity are returned.

Description

technical field [0001] The invention relates to image retrieval, in particular to an efficient image retrieval method based on discrete local linear embedding hash. Background technique [0002] With the rapid development of computer technology, the Internet has penetrated into all aspects of our lives, so the data on it is increasing all the time. At present, the existing capacity and growth rate of the data have far exceeded the processing capacity of current technology. Faced with such a rapid growth rate of data volume, how to make good use of these data faces two problems that need to be solved, that is, how to use storage space more effectively and how to accurately and quickly find the required information in the massive content. Compared with how to effectively use storage space, how to accurately and quickly find the information users need in the massive content is more difficult under the existing technical conditions. When retrieving and searching audio, pictures...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the traditional unsupervised hash learning algorithm, in the process of training the model, the spectral relaxation of the original problem brings precision loss, that is, the model is usually model learning and optimization in the real number space. Scale image search problem, to overcome various problems in large-scale image retrieval, improve the scope of use of the model, can deal with image search problems in different feature metric spaces, and provide an efficient image retrieval method based on discrete local linear embedding hash

Images