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

Near-duplicate image detection method and device and electronic equipment

A technology of approximate repetition and image detection, which is applied in the computer field, can solve the problem of low detection accuracy of approximate repetition images, and achieve the effect of improving comprehensiveness and accuracy, and improving accuracy

Active Publication Date: 2018-10-16
BEIJING SANKUAI ONLINE TECH CO LTD
View PDF8 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present application provides a method for detecting near-duplicate images, which at least solves the problem of low detection accuracy of near-duplicate images in the prior art

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
  • Near-duplicate image detection method and device and electronic equipment
  • Near-duplicate image detection method and device and electronic equipment
  • Near-duplicate image detection method and device and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] An approximate duplicate image detection method disclosed in this embodiment, such as figure 1 As shown, the method includes: Step 110 to Step 130.

[0030] Step 110, respectively determine the first feature and the second feature of the target image in the input image pair through the multi-task network model.

[0031] Wherein, the first features include image features reflecting inter-class differences and intra-class differences, and the second features are image features reflecting intra-class differences.

[0032] During the specific implementation of this application, the multi-task network model is trained in advance. The multi-task network model includes a plurality of sub-networks, such as a classification network, a similarity measurement network, etc., and the classification network and the similarity measurement network share a basic network. Wherein, the basic network is used to extract the first feature of the input image, and the similarity measurement ...

Embodiment 2

[0041] An approximate duplicate image detection method disclosed in this embodiment, such as figure 2 As shown, the method includes: Step 210 to Step 260.

[0042] Step 210, training a classification model based on image samples including approximately repeated images.

[0043] The approximately repeated images described in the embodiments of the present application may be composed of different images of the same object, for example, images of the same object taken under different lighting conditions, or the original image of a certain object and the cropping of the original image Image composition obtained after cutting, rotating, and brightness adjustment processing. The non-approximately repeated images described in the embodiments of the present application refer to image pairs composed of different images.

[0044] In the specific implementation, first collect similar images of different categories on the platform as image samples to train the classification model. The ...

Embodiment 3

[0076] An approximate duplicate image detection device disclosed in this embodiment, such as Figure 7 As shown, the device includes:

[0077] The feature extraction module 710 is used to respectively determine the first feature and the second feature of the target image in the input image pair through the multi-task network model, wherein the first feature includes image features reflecting inter-class differences and intra-class differences, so The second feature is an image feature that reflects intra-class differences;

[0078] A feature fusion module 720, configured to construct a fusion feature of the target image according to the first feature and the second feature of the target image extracted by the feature extraction module 710;

[0079] The approximate duplicate image detection module 730 is configured to determine whether the target images are approximate duplicate images according to the fusion features determined by the feature fusion module 720 .

[0080] Opt...

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

An embodiment of the invention discloses a near-duplicate image detection method and belongs to the computer technical field and solves the problem of low accuracy of near-duplicate image detection inthe prior art. The method comprises the steps of: determining a first feature and a second feature of a target image in an input image pair by a multi-task network model respectively, wherein the first feature comprises image features reflecting inter-class differences and intra-class differences, and the second feature comprises image features reflecting the intra-class differences; constructinga fusion feature of the target image according to the first feature and the second feature of the target image; and according to the fusion feature, determining whether the target image is a near-duplicate image. The near-duplicate image detection method disclosed in the present application further improves the comprehensiveness and accuracy of the image fusion feature by combining the intra-class information and inter-class information of the image to construct the fusion feature of the image, thereby improving the accuracy of the near-duplicate image detection.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a method and device for detecting approximately repeated images, and electronic equipment. Background technique [0002] Approximate duplicate images refer to images that only differ in color, saturation, cropping, shooting angle, watermark, etc. There are many types of approximate duplicate images in UGC (User Generated Content, User Generated Content) data. The appearance of these near-duplicate images will have a negative impact on the training of search and recommendation models and the display of search recommendation results. For example, it will reduce the accuracy of the model, reduce the utilization rate of the platform, and it is not conducive to the merchants to output more valuable information. At the same time, it will also affect the user experience. Therefore, there is an urgent need for an approximate duplicate image detection method to detect approxi...

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): G06T7/00G06V10/40G06V10/75G06V10/764
CPCG06T7/0002G06T2207/20221G06T2207/20084G06T2207/20081G06V10/40G06V10/82G06V10/75G06V10/764G06F18/22G06N3/04G06F18/254
Inventor 康丽萍魏晓明
Owner BEIJING SANKUAI ONLINE 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

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