Image retrieval method and system

Abstract

Embodiments of the invention provide an image retrieval method and system. The method comprises the steps of obtaining a target picture serving as a retrieval basis and input by a user; screening out a sample picture meeting a first predetermined condition from sample pictures stored in a picture database by utilizing a first image feature of the target picture, and outputting the screened-out sample picture as a retrieval result; monitoring whether a reference picture serving as a re-retrieval basis and input by the user based on the current retrieval result is obtained or not, and if the reference picture is obtained, screening out a sample picture meeting a second predetermined condition from the sample pictures stored in the picture database by utilizing a second image feature of the reference picture, outputting the screened-out sample picture as a retrieval result, and continuing to execute a monitoring action; and after a picture storage instruction sent by the user based on the current retrieval result is obtained, storing a retrieval result which the picture storage instruction aims at. Through the scheme, the comprehensiveness of image retrieval can be improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an image retrieval method and system. Background technique [0002] The existing image retrieval method is as follows: by extracting a target picture as a query standard to the image retrieval system, the image retrieval system extracts the predetermined image features of the target picture, and then calculates the image database according to the extracted predetermined image features. The similarity between each sample picture and the target picture, and output the sample pictures whose similarity meets the predetermined condition as the retrieval result, where the predetermined condition can be that the similarity is greater than a certain threshold, or the arrangement position of the similarity is at a predetermined and so on before the digits. [0003] However, in practical applications, usually a target image provided by the user as a query standard only reflects t...

AI Technical Summary

Problems solved by technology

[0003] However, in practical applications, usually a target image provided by the user as a query standard only reflects the surface characteristics of a viewing angle of the target object, and is easily affected by many other factors such as background, illumination, and imaging quality. Therefore, it cannot Comprehensively describe the user's retrieval needs, and the retrieved results usually cannot meet the user's expectations

Images