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

Construction method of dynamic k-nearest neighbor graph and rapid image retrieval method based on dynamic k-nearest neighbor graph

A technology of k-nearest neighbor graph and construction method, applied in the field of information retrieval, can solve the problem of inability to construct high-quality k-nearest neighbor graph, achieve the effect of fast image retrieval and improve retrieval efficiency

Inactive Publication Date: 2021-03-16
XIAMEN UNIV
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in actual scenarios, the data set sometimes changes. In this case, it is impossible to construct a high-quality k-nearest neighbor graph by using the nearest neighbor descent method.

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
  • Construction method of dynamic k-nearest neighbor graph and rapid image retrieval method based on dynamic k-nearest neighbor graph
  • Construction method of dynamic k-nearest neighbor graph and rapid image retrieval method based on dynamic k-nearest neighbor graph

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The invention discloses a method for constructing a dynamic k-nearest neighbor graph, which comprises the following steps:

[0036] Step 1. When a new image is added to the image data set, feature extraction is performed on the newly added image to obtain a feature vector q, and the feature vector q is added to the image feature set S; the image feature set S is in the d-dimensional space A collection of vectors obtained by feature extraction from images in the image dataset. In this embodiment, SIFT / SURF or methods such as deep learning are used for feature extraction.

[0037] Step 2. When the scale of the image feature set S is small, that is, when the number of feature vectors in the image feature set S is less than M, the distance between the vector q and all the vectors in the image feature set S is calculated by exhaustive comparison, to The calculated distance updates the k-nearest neighbor graph G. The distance here can be Euclidean distance, Hamming distance...

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 a construction method of a dynamic k-nearest neighbor graph and a rapid image retrieval method based on the dynamic k-nearest neighbor graph, which realize online updating ofan approximate k-nearest neighbor graph and realize rapid image retrieval based on the dynamic k-nearest neighbor graph, and because distance measurement between vectors is not assumed, the method hasgood generalization, and moreover, the mapping efficiency exceeding that of a nearest neighbor descent method is displayed on most of data sets while the quality of the k-nearest neighbor graph is ensured, and the retrieval efficiency on a plurality of image feature data sets is better than that of widely accepted methods such as HNSW and the like.

Description

technical field [0001] The invention relates to the technical field of information retrieval, in particular to a method for constructing a dynamic k-nearest neighbor graph and a fast image retrieval method based on the dynamic k-nearest neighbor graph, which can be applied to scenarios such as e-commerce, search engines, and installation monitoring. Background technique [0002] The k-nearest neighbor graph is a directed graph. For a given vector set S={x|x∈R d}, each vertex of the directed graph represents a vector in the vector set, and each vertex has k edges, pointing to the vertices (k nearest neighbors) to which the k closest vectors belong to, and the similarity is determined by the distance between the vectors Decision, commonly used are Euclidean distance, Hamming distance, cosine distance and so on. The k-nearest neighbor graph is an important data structure in the fields of manifold learning, computer vision, machine learning and multimedia information retrieval....

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): G06F16/51G06F16/583
CPCG06F16/51G06F16/583
Inventor 赵万磊王辉雷蕴奇
Owner XIAMEN UNIV
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