Method and device for searching for first image through second image

An image and score map technology, applied in the field of image search, can solve problems such as search result errors

Active Publication Date: 2018-01-09
ZHEJIANG DAHUA TECH CO LTD
View PDF3 Cites 24 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] This application provides a method and device for searching images by images, which is used to solve the problem in the prior art that there is a large error in the search results in the method of searching images by images

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
  • Method and device for searching for first image through second image
  • Method and device for searching for first image through second image
  • Method and device for searching for first image through second image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0070] In order to solve the problem of large search result error in the way of image search in the prior art, a method of image search by image is proposed in the embodiment of this application, which can be based on the fully convolutional twin neural network model to achieve the first The local area image of an image is used as a search condition, and a second image similar to it is found from a preset database. And in this application, the area block associated with the local area image is first searched from the score map, then the image feature corresponding to the area block in the second image of the preset database can be regarded as the local area block of the first image. The most similar part of the image features of the regional image, and further determine whether the second image is the same as Images that are similar to the first image, so that the search results of image search by image are more accurate.

[0071] In the following, in order to facilitate 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 relates to the field of image processing technology, in particular to a method and a device for searching for a first image through a second image. The method and the device are used forsolving the problem that a search result error is large when images are searched for through images in the prior art. According to the method for searching for the first image through the second image in the embodiment, image features of a local region image in the first image and image features of the second image can be extracted based on a full-convolution twin neural network model; a score chart is generated based on the extracted image features; region blocks, associated with the local region image, in the score chart are determined according to a sum of scores of points in a designatedquantity in the score chart; and when it is judged that the similarity between the image features, corresponding to the determined region blocks, in the second image and the image features of the local region image is greater than a set similarity threshold value, it is determined that the second image is an image similar to the first image.

Description

technical field [0001] The present application relates to the technical field of image processing, in particular to a method and device for searching images by image. Background technique [0002] As the data information of images on the Internet is becoming larger and larger, users' requirements for image search are also increasing, and the search method of searching images by images has also emerged as the times require. [0003] Image search is a search technology for retrieving similar images by inputting images, which mainly involves image feature extraction and image feature similarity analysis. The inventors found that most of the existing implementations of searching for images by images use a typical convolutional neural network (Convolutional Neural Network, CNN) to extract the overall image features of the target image, and compare it with the overall image features of the images stored in the database. Compare the similarity and find images that meet the similar...

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): G06F17/30G06V10/764
CPCG06F16/53G06F16/5854G06V10/454G06V10/82G06V10/764G06F18/24143G06F16/532G06N3/04G06N3/08G06F18/22G06F18/214
Inventor 陈羽飞
Owner ZHEJIANG DAHUA TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap