Exploring Parallel Computation Techniques for Real-Time Video Processing and Analysis
In today's digital environment, the demand for efficient and real-time video processing and analysis is growing rapidly. From surveillance to autonomous vehicles and augmented reality applications the need for accurate and rapid video analysis is important. To meet these demands researchers have turned to parallel computation techniques. Here, by harnessing the power of parallel processing the speed and accuracy of video processing tasks could be enhanced.
Parallel computation involves breaking down complex tasks into smaller subtasks that can be executed simultaneously on multiple processing units leading to faster processing and enabling real-time video analysis. Recent trends in parallel computation for video processing include the use of graphics processing units (GPUs) and specialized hardware accelerators like field-programmable gate arrays (FPGAs). In this, GPUs provide high computational power while FPGAs offer flexibility in designing custom processing pipelines. These advancements have revolutionized the field by enabling real-time object detection, tracking, recognition and content analysis. Additionally, the application of deep learning techniques, specifically convolutional neural networks (CNNs) has further improved parallel video processing. Also, parallelizing CNN computations across multiple units has made real-time object detection and tracking in videos possible leading to advancements in surveillance systems, autonomous vehicles and augmented reality applications. Moreover, the future of parallel computation for real-time video processing holds exciting possibilities. Researchers can explore novel parallelization techniques, such as data parallelism, task parallelism, and hybrid approaches, to optimize video processing algorithms further.
While parallel computation techniques offer tremendous potential, researchers face several challenges in this field. This special issue focuses on using parallel computing methods to improve the speed and accuracy of video processing tasks, such as object detection, tracking, recognition and content analysis. Researchers can utilize the power of parallel processing to overcome the constraints of sequential algorithms and enable real-time video analysis in various applications such as surveillance systems, autonomous vehicles, augmented reality and video editing tools.
Possible topics include, but are not limited to:
- Design and optimization strategies for parallel architectures in real-time video processing.
- Load balancing techniques to ensure efficient parallel video analysis.
- Task scheduling algorithms for effective parallel video processing.
- Utilizing deep learning for parallelization in video analysis.
- Real-time object detection and tracking using parallel computation methods.
- Parallel video analysis for surveillance and security applications.
- Parallel processing techniques for real-time video analytics in autonomous vehicles.
- Efficient algorithms for content-based video retrieval and indexing using parallel computation.
- Energy-efficient parallel computation techniques for video processing.
- Distributed and parallel video processing in cloud and edge computing environments.
- Parallel video analysis for augmented reality applications.
- Real-time video editing and post-processing using parallel computation.
- Integration of GPUs and FPGAs for parallel video processing.
Guest Editor Credentials & Biographies:
Dr. Mohammad Nishat Akhtar
School of Aerospace Engineering,
Engineering Campus, Universiti Sains Malaysia,
14300 Nibong Tebal, Pulau Pinang, Malaysia
E-mail: nishat@usm.my, iresearchertech2023@gmail.com
ResearcherID: S-7313-2018
ORCID ID: https://orcid.org/0000-0001-8592-5966
Scopus Author ID: 55065002500
Google Scholar: https://scholar.google.com/citations?user=ie7z5hQAAAAJ&hl=en
Research Background:
Dr. Mohammad Nishat Akhtar received his B.E in Computer Science from VTU, India during the year 2010, MSc in Electrical and Electronics from Universiti Sains Malaysia during the year 2013 and PhD in the field of Parallel Image Processing from Universiti Sains Malaysia during the year 2018. Currently, He is a Lecturer at School of Aerospace Engineering in Universiti Sains Malaysia. His research interests include High Performance Computation, Image Processing, Control & Embedded Systems, System-on-Chip. Currently, He has several International Journals and International Conferences under his banner. He has been actively involved as a lead researcher and co-researcher in USM’s research grant project. He has imparted several programming-based trainings to internship students and technicians. In this regard, Dr. Mohammad Nishat Akhtar also serves as Liaison Officer for School of Aerospace Business Unit whereby He coordinates in conducting training to the employees of multinational companies. He has also been designated as one of the lead trainers for a professional certification program conducted by School of Aerospace Business Unit for the session 2022/23.
Dr. Muhammad Rafiq Khan Kakar
Department of Architecture, Wood and Civil Engineering,
Bern University of Applied Sciences (BFH),
Pestalozzistrasse 20, 3400 Burgdorf, Switzerland
E-mail: muhammad.kakar@bfh.ch
ORCID ID: https://orcid.org/0000-0001-8669-897X
Scopus Author ID: 55994095000
Google Scholar: https://scholar.google.com/citations?user=0YKlSqEAAAAJ&hl=en
Research Background:
Dr. Muhammad Rafiq Khan Kakar completed his Bachelor of Science (Civil Engineering) from University of Engineering & Technology, Taxila, Pakistan in 2006, Master of Engineering (Transportation Engineering) from NED University of Engineering & Technology, Karachi, Pakistan in 2010 and Doctor of Philosophy (Ph.D.) in Civil Engineering (Asphalt Technology) from Universiti Sains Malaysia during the year 2015. Currently, He is a Scientific Collaborator in Department of Architecture, Bern University of Applied Sciences (BFH), Switzerland. His research interests include Asphalt Technology, Pavement Engineering, and Highway Engineering Material. His other professional skills include Road Maintenance execution, Budgeting and Tendering process, Government policies for road infrastructure. His current responsibilities include Leading Science & Technology funded Projects, Supervision of MSE student’s thesis, conducting research, and Establishing and developing test procedures.
Dr. Asha Crastaa
Department of Mathematics, Centre for advanced learning,
Bejai, Mangalore, India
E-mail: ashacrasta81@gmail.com
GoogleScholar: https://scholar.google.co.in/citations?user=xiyg5XIAAAAJ&hl=en
University Link: https://mite.ac.in/member/dr-asha-crasta/
Research Background:
Dr. Asha Crastaa completed his B.Sc in Physics, Computer Science & Mathematics, St. Aloysius College, Mangalore in 2004. M.Sc in Mathematics Mangalore University, Mangalore in 2002. M.Phil in Mathematics Vinayaka Missions University, Salem in 2008. Ph.D in applied Mathematics Jain University, Bangalore in 2014. His research interests include Engineering Mathematics, Advanced Mathematics I and II, Discrete Mathematical Structures, Management Information systems, Statistics for management, Business research methods, Probability Statistics and Queuing theory.
Important Dates:
Submission Deadline: 10 November, 2024
Authors Notification: 20 January, 2025
Revised Version Submission: 25 March, 2025
Final Decision Notification: 05 June, 2025