Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

An image retrieval method based on tree cluster vector quantization

An image retrieval and vector quantization technology, which is applied in still image data retrieval, vector format still image data, still image data indexing, etc., to achieve the effects of improving recall, reducing query space, and improving speed and accuracy

Inactive Publication Date: 2019-06-21
CHONGQING UNIV OF POSTS & TELECOMM
View PDF4 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] The present invention aims to solve the problems of image retrieval speed and accuracy

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
  • An image retrieval method based on tree cluster vector quantization
  • An image retrieval method based on tree cluster vector quantization
  • An image retrieval method based on tree cluster vector quantization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] see image 3 , an image retrieval method based on tree clustering vector quantization, including two processes of training and retrieval, wherein the training process includes the following steps:

[0053] S11. First preprocess the image, scale the image size to 224*224, use the ResNet-50 CNN model to extract 2048-dimensional image vector features from the image, and save the vector features of all images;

[0054] S12. Use the k-means++ clustering algorithm to cluster the pictures, the number of clusters is 5, and save the clustering model to the current root node of the tree model;

[0055] S13. For recursive clustering of the data in these 5 classes, the number of clusters is still 5, so as to divide all the data into leaf nodes;

[0056] S14. Stop clustering when the number in the subclass is less than N, or the depth of the number reaches H;

[0057] S15. Save the tree model, calculate the position of the leaf node for the existing picture, the path passed is the...

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

The invention claims an image retrieval method based on tree cluster vector quantization. The method comprises the following steps: S1, preprocessing a picture and extracting vector characteristics ofthe picture; S2, clustering the vectors of the images; S3, dividing the data space by using recursive clustering; S4, stopping clustering according to the formulated rules; And S5, storing the path fingerprint of the picture and the tree model. According to the method, the high-dimensional feature vector of the image is extracted by using the feature extraction capability of the deep learning model based on the tree clustering vector quantization algorithm, the content and semantics of the image are fully expressed, and the search capability of the tree clustering algorithm is combined, so that the image retrieval accuracy and the retrieval speed are improved.

Description

technical field [0001] The invention belongs to the field of image retrieval, in particular to a vector quantization algorithm based on tree clustering, a feature extraction method of deep learning and the combination of the two methods. Background technique [0002] With the rapid development of mobile Internet technology, heterogeneous data such as images, videos, audios, and texts are growing at an alarming rate every day. For example, Facebook has more than 1 billion registered users and uploads more than 1 billion pictures every month; the number of pictures uploaded by users of the Flickr photo social networking site reached 728 million in 2015, with an average of about 2 million pictures per day; Taobao, the largest e-commerce system in China More than 28.6 billion images are stored on the backend system of . For these massive pictures containing rich visual information, how to conveniently, quickly and accurately query and retrieve the images that users need or are ...

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
IPC IPC(8): G06F16/583G06F16/56G06F16/51G06K9/62
Inventor 丰江帆付雪君夏英周耀韩思祺
Owner CHONGQING UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com