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

Convolutional neural network image classification method based on homomorphic encryption

A convolutional neural network and homomorphic encryption technology, applied in the field of image processing, can solve the problems of being unable to resist collusion attacks and easy leakage of private information, so as to reduce the probability of private information leakage, improve computing efficiency, and resist collusion attacks Effect

Active Publication Date: 2021-06-11
XIDIAN UNIV
View PDF9 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to address the defects and insufficiencies of the above-mentioned prior art, and propose a convolutional neural network image classification method based on homomorphic encryption, which is used to solve the problems of easy disclosure of private information and inability to resist collusion attacks in the prior art technical issues

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
  • Convolutional neural network image classification method based on homomorphic encryption
  • Convolutional neural network image classification method based on homomorphic encryption
  • Convolutional neural network image classification method based on homomorphic encryption

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] Below in conjunction with accompanying drawing and specific embodiment, the present invention is described in further detail:

[0055] refer to figure 1 , the present invention comprises the following steps:

[0056] Step 1) Build a multi-party deep learning scene model:

[0057] Build includes parameter server, auxiliary server and 10 users P = {P i |1≤i≤10} multi-party deep learning scene model, where, P i Indicates the i-th user.

[0058] Step 2) The parameter server initializes the encrypted parameters:

[0059] The parameter server selects security parameters and selects k=1024, and then generates a 1024-bit long prime number p as a security modulus through the randprime() function in the SymPy library of Python, and then constructs the remaining class ring according to p in Constructed as a set of integers from 0 to p-1, and then in Randomly select the original root g of p.

[0060] Step 3) Each user generates their own public and private keys:

[0061...

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 provides a convolutional neural network image classification method based on homomorphic encryption, which is used for solving the technical problems that privacy information is easy to leak and collusion attacks cannot be resisted in the prior art, and comprises the following steps: constructing a multi-party deep learning scene model; initializing encryption parameters by the parameter server; enabling each user to generate a public key and a private key of the user; enabling the parameter server to generate a public key and a private key of the parameter server; enabling the auxiliary server to generate a public key and a private key of the auxiliary server and a joint public key; enabling each user to obtain a training image sample set and a test image sample set; enabling the parameter server to construct a convolutional neural network model and initialize training parameters; enabling the user P to obtain and upload a gradient vector ciphertext; enabling the parameter server to aggregate the gradient ciphertext vectors; enabling the parameter server and the auxiliary server to perform homomorphic re-encryption on the aggregation gradient vector ciphertext; enabling the user P to obtain a training result of the convolutional neural network model; and enabling each user to obtain an image classification result.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to an image classification method, in particular to a convolutional neural network image classification method based on homomorphic encryption. Background technique [0002] The image classification problem is a basic problem in the field of image processing technology. Its purpose is to distinguish different types of images according to the semantic information of the image to achieve the smallest classification error. Now the main image classification method is to use convolutional neural network, but to train a good convolutional neural network model for image classification requires a large amount of image data. However, the user's image data contains a large amount of private data, so the interaction of massive image data during the training of the convolutional neural network model will inevitably bring about privacy and security issues. The image classification service...

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): G06K9/62G06N3/04G06N3/08H04L9/00H04L29/06
CPCG06N3/08H04L9/008H04L63/0442G06N3/048G06N3/045G06F18/241
Inventor 王保仓何苏豫段普张本宇
Owner XIDIAN 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

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