Large-scale medical image retrieval method based on random sparse codes

Abstract

Provided is a large-scale medical image retrieval method based on random sparse codes. The method includes the following steps that firstly, images to be retrieved are input; secondly, the random sparse codes of all images are solved; thirdly, an autocorrelation matrix is computed; fourthly, the solution of an optimization problem is carried out, low dimensional embedding of all the images is solved through characteristic spectral factorization; fifthly, in a low dimensional space, the Euclidean distance between the images to be retrieved and database images is computed, then ascending sort is carried out on the images, and the Euclidean distance is in negative correlation with similarity; sixthly, the images subjected to the ascending sort are output. The large-scale medical image retrieval method is high in retrieval accuracy and retrieval speed by the application of relative comparison constraints and the random sparse codes.

Description

technical field [0001] The invention relates to a large-scale medical image retrieval method. Background technique [0002] Image retrieval technology refers to the technology of retrieving a list of images similar to an input image from an image database. Existing technologies include three main branches: text-based image retrieval technology, content-based image retrieval technology, and technology that comprehensively considers text and content. The limitations of text-based techniques are manifested in the subjective tendency and semantic limitations of text annotation. Content-based retrieval technology is the mainstream technology of current research, but there are several challenges: (1) There is no universally applicable method for images in different segments; (2) Feature selection is a key problem that is difficult to solve for a long time; (3) Which similarity measurement method to choose; (4) Whether the image retrieval system can respond quickly to massive ima...

AI Technical Summary

Problems solved by technology

The technology that comprehensively considers text and content has more information, but there is still room for improvement in the degree of problem solving

[0003] The defect of the existing technology is that the feature selection is mainly based on field experience, such as using gray histogram as a feature, using texture as a feature, etc. The similarity measurement method is generally based on experience, it is difficult to find sufficient reasons, and the versatility is poor.

Images