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

Image retrieval method, device, computing device and medium based on small-world network

A small-world network and image retrieval technology, applied in computing, computing models, computer parts, etc., can solve problems such as consuming a lot of manpower and time, prone to misjudgment, etc., achieve high accuracy, improve speed, and increase business The effect of throughput

Active Publication Date: 2021-07-02
GUANGZHOU HUIRUI SITONG INFORMATION SCI & TECH CO LTD
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] For a long time, the classification and search of evidence images by the public security, procuratorate and law departments have still remained at the stage of manual processing. However, due to the fact that most of the evidence images have text and the differences are small, when the amount of data in the image database is large, manual processing not only consumes a lot of manpower and time, and it is prone to misjudgment

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, device, computing device and medium based on small-world network
  • Image retrieval method, device, computing device and medium based on small-world network
  • Image retrieval method, device, computing device and medium based on small-world network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0076] Such as figure 1 As shown, the present embodiment provides a small-world network-based image retrieval method, which includes the following steps:

[0077] S101. Create an image library.

[0078] Creating an image library can directly acquire a large number of images and then create them. However, the image library to be created in this embodiment is an evidence image library, so the images that need to be acquired are evidence images. Evidence images are converted through evidence files. Evidence files are in various forms. Quality is the priority, and all evidence materials need to be converted into a unified image format and named according to the content, so this step S101 is as follows: figure 2 shown, including:

[0079] S1011. Acquire multiple images.

[0080] The image acquired in this embodiment is an evidence image, which can be obtained through collection, for example, by scanning the paper evidence documents uniformly into an electronic version of the im...

Embodiment 2

[0114] Such as Image 6 As shown, this embodiment provides an image retrieval device based on a small-world network, which includes an image library creation module 601, a first feature extraction module 602, a learning module 603, a small-world network construction module 604, and a second feature extraction module. Module 605, feature matching module 606 and similarity judgment module 607, the specific functions of each module are as follows:

[0115] The image library creation module 601 is configured to create an image library.

[0116] Further, the image library creation module 601 is as Figure 7 shown, including:

[0117] An acquisition unit 6011, configured to acquire multiple images.

[0118] A detection unit 6012, configured to detect the resolution of the acquired image.

[0119] The resolution reduction unit 6013 is configured to reduce the resolution of the image when there is an image with a resolution greater than a predetermined resolution value, so as to o...

Embodiment 3

[0136] This embodiment provides a computing device, which may be a computer, such as Figure 10 As shown, it includes a processor 1002 , a memory, a display 1003 and a network interface 1004 connected through a system bus 1001 . Wherein, the processor 1002 is used to provide calculation and control capabilities, and the memory includes a non-volatile storage medium 1005 and an internal memory 1006, the non-volatile storage medium 1005 stores an operating system, a computer program and a database, and the internal memory 1006 is The operating system in the non-volatile storage medium and the operation of the computer program provide an environment. When the computer program is executed by the processor 1002, the image retrieval method of the above-mentioned embodiment 1 is realized, specifically: creating an image library; extracting each image in the image library The local features of the image; learn the local features of each image in the image library to obtain a learning ...

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 discloses an image retrieval method, device, computing device and medium based on a small world network. The method includes: creating an image library; extracting local features of each image in the image library; The local features are learned to obtain the learning dictionary; according to the learning dictionary, the feature vector of each image in the image library is generated, and the small-world network of the feature vector is constructed; the local features of the target image are extracted; the local features of the target image are combined with the small-world network The image features in the image are matched; according to the matching result, the images in the image library that meet the similarity threshold condition are screened out and / or the images in the image library are sorted according to the similarity measure from high to low. The invention uses machine learning to obtain a learning dictionary, and constructs a small-world network with image features according to the learning dictionary, and uses the small-world network as an image retrieval model, which greatly improves the processing efficiency and accuracy of image retrieval, and increases business throughput .

Description

technical field [0001] The invention relates to an image retrieval method, in particular to an image retrieval method, device, computing device and medium based on a small world network, belonging to the field of image retrieval. Background technique [0002] For a long time, the classification and search of evidence images by the public security, procuratorate and law departments have still remained at the stage of manual processing. However, due to the fact that most of the evidence images have text and the differences are small, when the amount of data in the image database is large, manual processing not only consumes a lot of manpower and time, and it is prone to misjudgment. [0003] With the development of artificial intelligence technology and the improvement of hardware computing power, machine learning algorithms have been widely used in the field of image processing. Evidence images are usually scanned documents, mostly text, and there are noises and offsets. At t...

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 Patents(China)
IPC IPC(8): G06F16/583G06K9/62G06N20/00
CPCG06F18/28
Inventor 卢溜
Owner GUANGZHOU HUIRUI SITONG INFORMATION SCI & TECH CO LTD
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