Deep hash-based medical image distributed retrieval method

Abstract

The invention provides a deep hash-based medical image distributed retrieval method. The method comprises deep hash extraction image features and Hadoop-based batch image feature matching parallel computing; in the deep hash extraction image features, similar or dissimilar image pairs are used as training input through a convolutional neural network model, the gradient of an objective function relative to multilayer network weights is calculated by use of a back propagation algorithm, and finally multiple output values of each image are guided to be approximate to discrete 0 or 1; and in the Hadoop-based batch image feature matching parallel computing, a feature file of batch images is divided into a plurality of blocks, the blocks are independent from each other, the blocks are distributed to different nodes for execution through an Apache Hadoop YARN (Yet Another Resource Negotiator) resource manager, and finally results executed by all Mappers are combined into a Reducer. According to the method provided by the invention, the gap between image representation and semanteme can be reduced, the retrieval precision is improved, the retrieval process is accelerated through parallel feature matching, and the retrieval efficiency of batch medical images is enhanced.

Description

technical field [0001] The invention relates to the technical field of medical image retrieval, in particular to a distributed retrieval method for medical images based on deep hashing. Background technique [0002] At present, medical imaging systems have produced more and more digitized images in various medical fields, such as X-ray images, nuclear magnetic resonance images, and CT images, and most of these images are stored in databases. Based on medical image retrieval, it is a new mode of medical diagnosis to efficiently organize and manage these images to provide clinical diagnosis services. Content-based medical image retrieval mainly includes two stages, medical image feature extraction and feature matching. [0003] Image grayscale histogram feature extraction is a common method for medical image feature extraction. The methods used by this method to reflect the histogram feature of the image mainly include: [0004] (1) Mean value: it reflects the average gray v...

AI Technical Summary

Problems solved by technology

[0013] Aiming at the technical problems existing in the artificial image feature extraction method and single-node image retrieval in the prior art, the present invention provides a distributed retrieval method for medical images based on deep hashing, which can better align image features with visual or The semantics are similar, thereby improving the accuracy of retrieval, and realizing distributed storage and retrieval through the Hadoop framework

Images