Conference Paper/Proceeding/Abstract 598 views
Efficient PCA-based image compression via secure outsourcing edge cloud
2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys)
Swansea University Author:
Scott Yang
Full text not available from this repository: check for access using links below.
DOI (Published version): 10.1109/hpcc-dss-smartcity-dependsys57074.2022.00242
Abstract
The increasing of image pixel size in recent years has lifted the expense for the image data storage and transmission drastically. Image compression could be one effectively solution to alleviate this difficulty. However, in some cases, the computational terminal resource is constrained and insuffic...
| Published in: | 2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys) |
|---|---|
| Published: |
IEEE
2022
|
| Online Access: |
http://dx.doi.org/10.1109/hpcc-dss-smartcity-dependsys57074.2022.00242 |
| URI: | https://cronfa.swan.ac.uk/Record/cronfa63210 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract: |
The increasing of image pixel size in recent years has lifted the expense for the image data storage and transmission drastically. Image compression could be one effectively solution to alleviate this difficulty. However, in some cases, the computational terminal resource is constrained and insufficient to effectively perform the image compression. An image compression based on principal component analysis (PCA) outsourcing protocol is proposed in this paper to handle such problems. The abundant computing resources of the edge cloud are used to perform the image compression effectively that the terminal cannot. The image is encrypted and sent to the cloud by terminal. The encrypted matrix is directly calculated by the cloud and the computational result is returned to the terminal. After the result is received by the terminal, the correctness of the result is verified. The result will be decrypted if it passed the verification. Otherwise, the result is returned to the edge cloud for recalculation. According to a series of comprehensive performance analysis, the proposed protocol can substantially improve the efficiency of local computation, also guarantee the privacy of the terminal and correctness of the results. |
|---|---|
| College: |
Faculty of Science and Engineering |
| Funders: |
This research was supported by the National Natural Science
Foundation of China under Grant 61801131, Guangxi Natural Science Foundation under Grant 2022GXNSFAA035632,
research fund of Guangxi Normal University under Grant
2021JC006, and the AI+Education research project of Guangxi
Humanities Society Science Development Research Center
under Grant ZXZJ202205. |