A Case Study for Practical Issues of DCT Based Bitmap Selective Encryption Methods

Selective Encryption (SE) methods are mainly used for protecting multimedia contents. One approach, DCT based SE methods, is exploited for image protection especially bitmap protection use case in many previous works. The special focus of our investigation is looking into two main practical issues i...

Full description

Saved in:
Bibliographic Details
Published in2018 Third International Conference on Security of Smart Cities, Industrial Control System and Communications (SSIC) pp. 1 - 7
Main Authors Qiu, Han, Enfrin, Nathalie, Memmi, Gerard
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2018
Subjects
bitmap
DCT
Discrete cosine transforms
Encryption
Frequency-domain analysis
Image recovery
Industrial control
Reliability
Selective encryption
Smart cities
Visualization
Online AccessGet full text

Cover

More Information
Summary:Selective Encryption (SE) methods are mainly used for protecting multimedia contents. One approach, DCT based SE methods, is exploited for image protection especially bitmap protection use case in many previous works. The special focus of our investigation is looking into two main practical issues ignored by existing DCT based SE methods: rounding errors and recovery from non-selected DCT coefficients. In this paper, we re-implement previous work and present the issue or rounding errors with a practical implementation. According to our work, the rounding errors due to truncation operations inside the DCT based SE cannot be ignored although multimedia contents can tolerate some level of noise introduced. Also, we improved recovery methods that can recover visual elements of an image by guessing the DC coefficients from only a small set of known high frequency AC coefficients. It is shown that some published DCT based SE methods are not reliable for a high level of protection purpose even they protect low frequency coefficients representing 99% of the energy in frequency domain.
DOI:10.1109/SSIC.2018.8556828