D pcode privacy preserving frequent visual patterns publication on cloud
0 likes20 views
The document discusses the necessity of privacy-preserving techniques for mining frequent visual patterns in cloud-stored graphical data, addressing concerns over sensitivity leaks from video and image datasets. It introduces the 'dpcode' algorithm, which is the first differentially private approach for this purpose, combining visual word conversion with noise allocation strategies to safeguard privacy while minimizing data distortion. Experimental results indicate that this method effectively balances high utility with stringent privacy requirements.
1 of 1
Download to read offline
More Related Content
What's hot (20)
Ad
Viewers also liked (12)
Ad
Similar to D pcode privacy preserving frequent visual patterns publication on cloud (11)
More from ieeepondy (20)
D pcode privacy preserving frequent visual patterns publication on cloud
- 1. DPcode: Privacy-Preserving Frequent Visual Patterns Publication on Cloud Abstract: Nowadays, cloud has become a promising multimedia data processing and sharing platform. Many institutes and companies plan to outsource and share their large-scale video and image datasets oncloud for scientific research and public interest. Among various video applications, the discovery of frequent visual patterns over graphical data is an exploratory and important technique. However, the privacy concerns over the leakage of sensitive information contained in the videos/images impedes the further implementation. Although the frequent visual patterns mining (FVPM) algorithm aggregates summary over individual frames and seems not to pose privacy threat, the private information contained in individual frames still may be leaked from the statistical result. In this paper, we study the problem of privacy-preserving publishing of graphical data FVPM on cloud. We propose the first differentially private frequent visual patterns mining algorithm for graphical data, named DPcode. We propose a novel mechanism that integrates the privacy-preserving visual word conversion with the differentially private mechanism under the noise allocation strategy of the sparse vector technique. The optimized algorithms properly allocate the privacy budgets among different phases in FPM algorithm over images and reduce the corresponding data distortion. Extensive experiments are conducted based on datasets commonly used in visual mining algorithms. The results show that our approach achieves high utility while satisfying a practical privacy requirement.