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

An Image Retrieval Method Based on Deep Hash Learning Optimization

An image retrieval and depth technology, applied in the field of computer vision, achieves the effects of discrete orthogonal optimization, robustness, accuracy improvement, and strong semantic learning ability

Active Publication Date: 2020-12-01
ZHEJIANG UNIV OF TECH
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to overcome the optimization problems existing in existing hash learning methods, the present invention provides an image retrieval method based on deep hash learning optimization with high precision and high retrieval efficiency, using a multi-layer neural network to construct a nonlinear hash function, The objective function under the discrete orthogonal constraint that maintains semantic similarity is constructed, and then the objective function is optimized by ADMM (Alternating Direction Method of Multipliers) optimization method and stochastic gradient descent method, thus obtaining Ha Hive function, using a hash function to convert the target image features into a hash code, and using the hash code for image retrieval when retrieving

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
  • An Image Retrieval Method Based on Deep Hash Learning Optimization
  • An Image Retrieval Method Based on Deep Hash Learning Optimization
  • An Image Retrieval Method Based on Deep Hash Learning Optimization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0031] refer to figure 1 , an image retrieval method based on deep hash learning optimization, including five processes of hash function construction, objective function construction, objective function optimization, hash function training, and image retrieval and accuracy testing.

[0032] The images in this implementation case are divided into 10 categories, and each category has 60,000 images.

[0033] The image retrieval method optimized based on depth hash learning, comprises the following steps:

[0034] Step 1: Hash function construction, the hash function is obtained by a multi-layer neural network, the process is as follows:

[0035] Step 1.1: Construct a three-layer fully connected layer with the number of ne...

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 image retrieval method based on deep Hash learning optimization comprises the following steps of 1, constructing a multi-layer full connection network firstly, connecting a panh function behind each layer of full connection, and finally conducting the sign operation on the network output; 2, constructing a semantic retention loss function obtained by a classification loss function and a weighted semantic similarity matrix, and a target function with discrete orthogonal constraint obtained by a quantitative loss function and a regular item; 3, optimizing an objective function; 4, dividing the obtained feature data set into a query set Q and a to-be-searched set D, taking a part of data in the to-be-searched set D to form a training data set P, inputting depth characteristics and label information of the training data set P, initializing a weight coefficient and a binary code, performing iterative optimization on the step 3 in sequence to obtain an optimal network weight coefficient,and obtaining a depth hash function by the step 2; and step 4, carrying out the image retrieval and precision testing. The method is relatively higher in precision and relatively higher in retrieval efficiency.

Description

technical field [0001] The invention relates to image big data processing and analysis in the field of computer vision, in particular to a deep hash learning optimization algorithm and an image retrieval method. Background technique [0002] With the development of network sharing technology, the promotion of big data technology and the generation of massive images, traditional image retrieval technology has been unable to adapt to large-scale image search, and hash learning has been widely used due to its high efficiency and easy storage. in large-scale image retrieval. Hash learning obtains the hash function from the data through the method of machine learning, thereby mapping the data into binary codes, and retaining the neighbor relationship in the original space as much as possible, that is, maintaining similarity. Because binary discrete optimization is an NP problem, many methods use mixed integer optimization methods to relax binary codes to continuous values ​​for ...

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): G06F16/583G06N3/04G06N3/08
Inventor 马青白琮陈胜勇
Owner ZHEJIANG UNIV OF TECH
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