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

Image retrieval method and system based on memory image cluster

An image retrieval and memory technology, applied in the direction of still image data retrieval, etc., can solve the problem of low retrieval performance, achieve the effect of life cycle management and improve performance

Active Publication Date: 2019-07-16
ZHEJIANG UNIVIEW TECH
View PDF11 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage of this solution is that Spark, as a general framework for distributed computing, solves the problem of distributed system development, but the whole process needs to start from data loading, coupled with the overhead of the framework itself, the retrieval performance is low

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
  • Image retrieval method and system based on memory image cluster
  • Image retrieval method and system based on memory image cluster
  • Image retrieval method and system based on memory image cluster

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] figure 1 It is a flowchart of an image retrieval method based on memory image clusters provided by Embodiment 1 of the present invention.

[0059] This embodiment provides an image retrieval method based on memory image clusters. Through the steps of data storage, data life cycle management, balanced distribution of memory clusters, and image search, it is possible to search for massive feature data without loss of accuracy. Realize fast image search by image. refer to figure 1 , the image retrieval method based on memory image clusters includes the following steps:

[0060] Step S110, using HBase and Parquet files on ElasticSearch and HDFS (Hadoop Distributed File System, distributed file system), and based on the memory cluster, the image data is stored in a distributed manner to obtain persistent image data;

[0061] Specifically, image data is divided into attribute data and feature data; among them, massive attribute data is stored through HBase and ElasticSearc...

Embodiment 2

[0072] figure 2 It is a flow chart of the method for obtaining image thermal data provided by Embodiment 2 of the present invention.

[0073] This embodiment will describe each step in the previous embodiment in detail.

[0074] In step S110 of the image retrieval method based on memory image clusters, the image data includes attribute data and feature data, and the data storage method provided in this step specifically includes: first, storing attribute data through HBase and ElasticSearch, and storing feature data through Parquet files are stored, and the attribute field data provided for retrieval are stored through Parquet files to obtain persistent image data; second, memory clusters are built, and persistent image data are stored in a distributed manner through memory clusters.

[0075] That is to say, image data is divided into attribute data and feature data, attribute data such as gender, clothes color, etc., feature data is a feature vector. Massive attribute data...

Embodiment 3

[0096] Figure 4 It is a schematic diagram of an image retrieval system based on memory image clusters provided by Embodiment 3 of the present invention.

[0097] The embodiment of the present invention also provides an image retrieval system based on memory image clusters, which is used to realize the above image retrieval method based on memory image clusters. refer to Figure 4 , the image retrieval system based on memory image cluster includes:

[0098] The data storage unit 100 is used to adopt HBase and Parquet files on ElasticSearch and HDFS, and based on the memory cluster, the image data is stored in a distributed manner to obtain persistent image data;

[0099] The data management unit 200 is configured to manage the persistent image data according to preset conditions to obtain image thermal data;

[0100] A node allocation unit 300, configured to allocate image thermal data to nodes according to the principle of balanced allocation to obtain thermal data nodes; ...

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 provides an image retrieval method and system based on a memory image cluster, and relates to the technical field of image retrieval. The image retrieval method comprises the steps of adopting the HBase and Parquet files on the ElasticSearch and HDFS , and based on a memory cluster, carrying out distributed storage on the image data, and obtaining the persistent image data; managingthe persistent image data according to a preset number to obtain the image thermal data; performing node distribution on the image hot data according to a balanced distribution principle to obtain thehot data nodes; and distributing the retrieval requests to each hot data node in parallel through the first hot data node, so that the hot data nodes retrieve the image hot data to obtain the first target image data. According to the method and the system, the image searching performance with images can be improved on the condition of ensuring the accuracy.

Description

technical field [0001] The invention relates to the technical field of image retrieval, in particular to an image retrieval method and system based on memory image clusters. Background technique [0002] With the high-definition of video surveillance cameras, the application of new video codec algorithms, and the maturity of video structuring algorithms, video surveillance systems are no longer limited to regular functions such as real-time video browsing, historical video retrieval and playback. More and more AI (Artificial Intelligence, artificial intelligence) technology is used to extract image information into various attributes and feature information, and store them in the background system for subsequent rapid retrieval and data mining. [0003] The number of cameras in a city is tens of thousands at every turn, and the amount of extracted data is quite astonishing. Generally, it can be kept at the level of tens of billions or even hundreds of billions of records. Th...

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): G06F16/50
CPCG06F16/50G06F16/51
Inventor 周后取刘清炼吴镁叶建云
Owner ZHEJIANG UNIVIEW TECH
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