Journal article 36 views 13 downloads
A Generic Cryptographic Deep-Learning Inference Platform for Remote Sensing Scenes
,
Yulin Wu ,
Xuan Wang ,
Zoe L. Jiang ,
Weizhe Zhang ,
Yang Liu ,
Mamoun Alazab
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Volume: 16, Pages: 3309 - 3321
Swansea University Author:
Yang Liu
-
PDF | Version of Record
This work is licensed under a Creative Commons Attribution 4.0 License.
Download (3.31MB)
DOI (Published version): 10.1109/jstars.2023.3260867
Abstract
Deep learning plays an essential role in multidisciplinary research of remote sensing. We will encounter security problems during the data acquisition, processing, and result generation stages. Therefore, secure deep-learning inference services are one of the most important links. Some theoretical p...
| Published in: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
|---|---|
| ISSN: | 1939-1404 2151-1535 |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2024
|
| Online Access: |
Check full text
|
| URI: | https://cronfa.swan.ac.uk/Record/cronfa67393 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract: |
Deep learning plays an essential role in multidisciplinary research of remote sensing. We will encounter security problems during the data acquisition, processing, and result generation stages. Therefore, secure deep-learning inference services are one of the most important links. Some theoretical progress has been made in cryptographic deep-learning inference, but it lacks a general platform that can be realized in reality. Constantly modifying the corresponding models to approximate the plaintext results reveal the model information to a certain extent. This article proposes a generic post-quantum platform named the PyHENet, which perfectly combines cryptography with plaintext deep learning libraries. Second, we optimize the convolution, activation, and pooling functions and complete the ciphertext operation under floating point numbers for the first time. Moreover, the computation process is accelerated by single instruction multiple data streams and GPU parallel computing. The experimental results show that the PyHENet is closer to the plaintext inference platform than any other cryptographic model and has satisfactory robustness. The optimized PyHENet obtained a better accuracy of 95.05% in the high-resolution NaSC-TG2 database, which the Tiangong-2 space station received. |
|---|---|
| Keywords: |
Convolutional neural network (CNN), deep learning inference, fully homomorphic encryption (HE), privacy preserving, remote sensing scenes |
| College: |
Faculty of Science and Engineering |
| Funders: |
. This work was supported in part by the Basic Research
Project of Shenzhen, China under Grant JCYJ20200109113405927 and Grant
JCYJ20200109113427092, in part by the National Natural Science Foundation of China under Grant 61872109, Grant 62272131, and 62203134, in
part by the National Science and Technology Major Project Carried on by
Shenzhen under Grant CJGJZD20200617103000001, in part by the Guangdong
Provincial Key Laboratory of Novel Security Intelligence Technologies under
Grant 2022B1212010005, in part by the Science and Technology Project of
Guangzhou under Grant 2020A1515010652, in part by the Key Fields R&D
Project of Guangdong Province under Grant 2020B0101380001, and in part
by the PINGAN-HITsz Intelligence Finance Research Center. (Corresponding
authors: Xuan Wang; Zoe L. Jiang.) |
| Start Page: |
3309 |
| End Page: |
3321 |