Image retriever combined with significant features, database and retrieval method

An image retrieval and feature map technology, applied in the fields of paleontology and computer vision, can solve the problem of low accuracy of fossil image retrieval

Pending Publication Date: 2021-06-04
NORTHWEST UNIV
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies in the prior art, the object of the present invention is to provide an image retriever, database and retri

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 retriever combined with significant features, database and retrieval method
  • Image retriever combined with significant features, database and retrieval method
  • Image retriever combined with significant features, database and retrieval method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0102] This embodiment provides a kind of image retriever, database and retrieval method combined with salient features, such as figure 1 shown, follow the steps below:

[0103] Step 1, obtaining the 3D model of the paleontology, performing mapping processing on the 3D model of the paleontology, and obtaining a view set of the 3D model of the paleontology and a mask image set;

[0104] In this embodiment, the selected three-dimensional model of paleontology is a three-dimensional model of trilobites, which will be as follows figure 2 The three-dimensional model of the trilobite shown in the mapping process is obtained as follows image 3 The three-dimensional model view of the trilobite paleontology is shown, and the three-dimensional model view of the trilobite paleontology is binarized using an adaptive threshold algorithm to obtain the mask image of the trilobite as shown in Figure 4.

[0105] Step 2, obtain the real fossil image set, and preprocess the real fossil image...

Embodiment 2

[0165] This embodiment presents an image database combined with salient features. The real fossil image training set is input into the final image retrieval network combined with salient features to obtain a feature vector set, which is stored in the database.

Embodiment 3

[0167] This embodiment provides an image retrieval method combined with salient features, selects a real fossil image in the test set of real fossil images, inputs it into the final image retrieval network combined with salient features, extracts the feature vector of the real fossil image, and compares it with The similarity comparison is performed on the feature vector set in the database.

[0168] Specifically, the method is carried out according to the following steps:

[0169] Step 1: Select a real fossil image in the test set of real fossil images, input the real fossil image into the image retrieval network that finally combines salient features, and obtain the feature vector P of the real fossil image K ;

[0170] Step 2: Calculate the eigenvector P of the real fossil image and each eigenvector Q in the eigenvector set through the Euclidean distance formula 8 m The distance value D m , select the first 10 eigenvectors Q with the smallest distance value in the eigenv...

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 retriever combined with significant features, a database and a retrieval method. The retrieval method comprises the steps: obtaining a three-dimensional model of a paleontology, and carrying out the mapping processing of the three-dimensional model of the paleontology, and obtaining a three-dimensional model view set and a mask image set of the paleontology; collecting a real fossil image set, and preprocessing the real fossil image set to obtain a simulated fossil image set GX; taking the simulated fossil image set GX and the mask image set as input, and training a saliency detection network to obtain a trained saliency detection network; according to the method, the saliency features extracted by the saliency detection network encoder are fused with the global features to serve as the description features of the final image, the description features serve as the final features of fossil image retrieval, feature description of the main body part in the fossil image is enhanced, the influence of noise such as rock on retrieval is reduced, and the retrieval efficiency is improved. The technical problem of low fossil image retrieval precision in the prior art is solved.

Description

technical field [0001] The invention belongs to the fields of paleontology informatics and computer vision, and relates to a retrieval method, in particular to an image retrieval device combined with salient features, a database and a retrieval method. Background technique [0002] With the development of digital media technology and the improvement of deep learning network performance, image retrieval has a very wide range of applications in many fields such as education, culture, and life sciences. However, in terms of paleontology, the retrieval of fossil images still mainly relies on human identification to give specific categories. If an unknown fossil image can be given, its similar image can be found from the existing fossil image database. Good practical application value; this not only contributes to the development of paleontology, but also contributes to the dissemination of paleontological science knowledge. [0003] Traditional image retrieval methods extract a...

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): G06F16/583G06T17/00G06T7/136G06N3/08G06N3/04
CPCG06F16/583G06T17/00G06T7/136G06N3/08G06T2207/10012G06N3/045Y02D10/00
Inventor 冯宏伟侯刚冯筠刘建妮韩健
Owner NORTHWEST UNIV
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