DESSERT

10th International IEEE Conference
Dependable Systems, Services and Technologies
Leeds, United Kingdom, June 5-7, 2019

IEEE
  • Information

  • Deadlines

    (23:59 UTC+0)

    Paper submission: March 4, 2018 March 14, 2018 March 19, 2018 (Extra deadline for update only)
    Notification of paper acceptance: April 25, 2018
    Final manuscript: April 30, 2018
    Registration and payment: May 3, 2018

    Program draft publication: May 7, 2018
    Conference date: May 24-27, 2018

  • Contacts

    Department 503, DESSERT’2018 Organizing Committee,
    National Aerospace University n. a. N. E. Zhukovsky “KhAI”,
    Chkalov str., 17, Kharkiv, 61070, Ukraine
    Phone: +38 (095) 564 76 69
    (contact person – Anastasiia Strielkina)
    e-mail: dessert@csn.khai.edu

Volodymyr Ponomaryov

Sparse Learning Approach in Restoration and Filtering of Multidimensional Signals with Rapid Hardware Implementation (CPU multicore, GPU, DSP).

Abstract:

Satellite, Radar, Medical, Digital Photographs, HDTV, Virtual Reality, etc. are some of the 2D/3D signal applications where the restoration and filtering procedures are obligatory. Usually, these signals are corrupted by sensors non ideality, during transmission, or by noise contamination.
The main objective of this paper is to discuss state-of-the-art methods and to suggest justified novel approach in restoration of 2D/3D data that can be used in mentioned applications. In applications, it is necessary to have several efficient restoration schemes, which depend on type of distortion and other priory information.
Several original frameworks in processing of 2D/3D signals will be exposed in this paper comparing them with known techniques justifying the efficiency of novel approach via different criteria: objective as well as subjective ones.
The principal idea of our approach consists of usage of several instruments: sparse learning techniques, order statistics, and fuzzy set theory. Modern theoretical methods in restoration are generally based on a possibility to gather more samples for similar parches into 2D/3D data during learning stage. Then, the restoration procedures use sophisticated statistical methods, which depend on data distortion model. The difficulty here is in selection, measuring and employing the similarity of group of objects for their best restoration.
Several designed and better existing algorithms have been implemented on the CPU multicore and GPU platforms performing restoration of 2D/3D data in a real time environment.
Keywords: Digital 2D/3D Processing, Video sequence, Vector Order Statistics, Fuzzy logic, additive, impulsive, speckle noise; CPU multicore, GPU.

Volodymyr I. Ponomaryov attained his Ph.D. (1974), D. Tech.Sci. (1981), and has been a Full Professor (1984). During more than 20 years, he is working as Professor at Instituto Politecnico Nacional (Mexico-city).
He is a referee for more than 20 international scientific journals (editorials: IEEE, Elsevier, Springer, SPIE, etc.), and a member of the editorial board of international journals and conferences. He was the lead guest editor of a special issue on Image and Video Quality Improvement Techniques for Emerging Applications in EURASIP Journal on Advances in Signal Processing (2011).
Dr. Ponomaryov served and currently serves as Associate Editor in Journal of Real Time Image Processing (Springer). Over the years, he has also been a promoter of 42 Ph.D.s (ex-USSR, Ukraine, Mexico) on his areas of interests.
His current research interests include 2D/3D signal and video processing, restoration, denoising, intelligent pattern recognition, real-time filtering, 3D reconstruction, super-resolution, digital watermarking, medical sensors, remote sensing, etc.
He has authored or co-authored of about 180 international scientific pear referred papers, 300 international prestige conference papers, 20 author certificates (Mexico), and also 23 patents of ex USSR, Russia and Mexico, and five scientific books in international editorials.
Dr. Ponomaryov is academician of National Academy of Sciences of Mexico, and member of National System of Investigators (highest III level). He is a Member of scientific societies: IEEE, SPIE, IEICE (Japan).