Unlock instant, AI-driven research and patent intelligence for your innovation.

Rapid image retrieval method and device based on semantics and content

An image retrieval and semantic technology, applied in the field of image processing, can solve the problems of slow retrieval speed in large databases, achieve the effect of eliminating linear growth relationship, improving retrieval accuracy, and increasing retrieval speed

Active Publication Date: 2019-07-05
HARBIN INST OF TECH
View PDF7 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problem of slow retrieval speed of large databases in existing image retrieval methods, a fast image retrieval method and device based on semantics and content are provided

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Rapid image retrieval method and device based on semantics and content
  • Rapid image retrieval method and device based on semantics and content
  • Rapid image retrieval method and device based on semantics and content

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0096] Adopt the following examples to verify the beneficial effects of the present invention:

[0097] A kind of fast image retrieval method based on semantics and content described in this embodiment is carried out according to the following steps:

[0098]1. On a certain floor, push a multimedia mobile acquisition platform carrying a rechargeable battery, a notebook computer, and an industrial camera, run the MATLAB program in the computer to realize the call of the industrial camera, and collect images of the environment in the corridor, aiming at the multi-semantic environment Multi-angle image acquisition is carried out and stored in the database. Using the labeling tool that comes with MATLAB, the images in the database are semantically labeled, and the neural network training set is constructed. Input the images in the training set into the neural network framework with the required learning rate and number of iterations to complete the construction of the semantic se...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

In order to solve the problem that an existing image retrieval method is slower in retrieval speed for a large database, the invention provides a rapid image retrieval method and device based on semantics and content, and belongs to the technical field of image processing. The method comprises the following steps of constructing a semantic segmentation network; inputting all the images in the offline database into a semantic segmentation network to obtain one or more semantic tags of each image; classifying all the same images of the semantic tags into one class to obtain a plurality of semantic databases, converting the semantic tag corresponding to each semantic database into a unique digital tag, and marking the corresponding semantic database by adopting the digital tag; inputting theto-be-retrieved image into a semantic segmentation network to obtain a semantic tag of the to-be-retrieved image, and converting the semantic tag into a digital tag; and determining a semantic database having the same digital label as the to-be-retrieved image, and retrieving an image matched with the to-be-retrieved image in the determined semantic database based on a content retrieval mode.

Description

technical field [0001] The invention relates to an image retrieval method in visual positioning, in particular to a fast image retrieval method based on semantics and content, and belongs to the technical field of image processing. Background technique [0002] In the online stage of indoor visual positioning technology, it is necessary to quickly retrieve the image to be positioned provided by the user, and find its matching image in the positioning database, so as to perform precise positioning. Most of the existing image retrieval algorithms extract and compare the features of the images in the database and the user input images to find the matching feature vector with the minimum Euclidean distance. However, with the increase of the database, the number of extracted features gradually increases, and the process of finding matching feature vectors in the online stage takes a lot of time, which cannot meet the real-time performance of online positioning. Therefore, there ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/583
Inventor 马琳戴进谭学治何晨光
Owner HARBIN INST OF TECH