Image representation method and server

Abstract

The embodiments of the present application disclose an image representation method and server. The image characterization method includes: obtaining an image matrix set corresponding to a target image in a predetermined color space; performing dimensionality reduction processing on the image matrix set to obtain at least one dimensionality reduction matrix set; according to the first dimensionality reduction matrix set of the dimensionality reduction matrix set , the second dimensionality reduction matrix, and elements with the same two-dimensional coordinates in the third dimensionality reduction matrix are operated according to a preset algorithm, and the screening matrix corresponding to the dimensionality reduction matrix set is output; based on the screening corresponding to the dimensionality reduction matrix set matrix, performing data screening on the dimensionality reduction matrix set to obtain at least one representation matrix set of the target image; wherein, the representation matrix set includes a first representation matrix, a second representation matrix and a third representation matrix. The image characterization method and server provided in the present application can reduce the computational workload generated when acquiring the characterization data of the target image.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to an image representation method and server. Background technique [0002] With the development of computer technology, images have become an important form of information bearing. Image characterization generally refers to the process of obtaining characterization data of the image and using the characterization data to characterize the image, wherein the characterization data is generally data that can express information such as color, texture, and shape contained in the image. Image representation has a wide range of applications in image retrieval, image stitching, object detection and recognition, robot scene positioning, and video content analysis. For example, in the field of image search, after receiving a query request containing a target image, it is usually necessary to obtain the characterization data of the target image, and compare the characterization d...

AI Technical Summary

Problems solved by technology

In particular, when the above-mentioned prior art method is applied to the field of image search, it may be necessary to obtain the characterization data of a very large number of target images at the same time, which will generate a huge computing workload and bring a burden to the server performing the image search. to a huge load, and may even bring down the server performing the image search

Images