Visual Surveillance for Big Data
Chairs: John See (Multimedia University, Malaysia), Shahrel Azmin Suandi (USM, Malaysia) and Atiqur Rahman Ahad (Dhaka University, Bangladesh)
Over the past two decades, visual surveillance research has made considerable progress in the areas of signal processing and computer vision. In this era of Big Data, visual surveillance tasks are confronted with new challenges and the need for intelligent applications as a massive amount of video data is now available through various surveillance infrastructure such as camera networks, autonomous drones, and vehicle dash-cams. This special session welcomes papers discussing contemporary aspects of visual surveillance, including large-scale search and retrieval, the use of deep learning techniques, event detection and recognition, long-term video analytics, parallel and distributed computing.
Deep Learning for Biomedical and Medical Applications
Chairs: Fabrice Meriaudeau, Tong Boon Tang, Aznul Qalid Md Sabri
Recently, there has been a revolutionary step forward in machine learning with the advent of deep learning architectures where a many layered neural network is trained on a huge set of data to extract the former “hand crafted features”. Specifically, the development of convolutional neural network (CNN) allowed for impressive gains in the ability to classify images and detect objects in a picture. Recently, deep architectures of various forms have been used for medical applications: computer aided detection , lung disease classification, hemorrhages detection on fundus images , brain segmentation, cerebral microbleeds , cytology and very recently, some pioneer works on AMD detection on SD-OCT images. As biomedical and medical data fully annotated are not available in large quantity, researchers are developing new strategies on fine tuning and transfer learning from pre-trained networks with promising results.
Intelligent Signals for Instrumentation, Control and Automation Systems
Chairs: Rosdiazli Bin Ibrahim, Vijanth S. Asirvadam, Nordin Bin Saad
Signal analysis and manipulation is an integral part of instrumentation, control and automation systems. Signals obtained from sensors and transducers have to be analysed, processed and manipulated to achieve the desired objective. This results in an increasing research interest in the area of signal filtering, monitoring, analysis and conditioning for instrumentation, control and automation systems. This session will focus on topics that encompass the application of signal monitoring, processing, filtering, analysis and conditioning in control systems, instrumentation systems and automation systems.
Signal Processing for Humanitarian Activities
Chairs: Mohammad Faizal Ahmad Fauzi
This special session aims to bring together individuals conducting project and research to innovate signal and image processing technologies for humanitarian applications. It is intended to inspire professionals and engineers to become socially responsible and be engaged in developing technological solutions which are cost effective and sustainable; improve living conditions and livelihood of people thus making a difference in peoples’ lives. The session is introduced to demonstrate the conference committee’s support of the IEEE Board-endorsed vision for IEEE volunteers around the world carrying out and/or supporting impactful humanitarian activities at the local level - that is, "feet on the ground."
Big Data Image and Signal Processing
Chairs: Rajasvaran Logeswaran
As the volume, variety, velocity and veracity (the 4V's) of data has grown at a tremendous pace in recent years, and will continue to do so in the foreseeable future, the handling, processing and analysis of Big Data is fundamental to data science activities in making sense of and taking the relevant action on the available data. As this requirement permeates all domains, hence this special session welcomes papers discussing various efforts taken with regards to Big Data in relation to the signal and image analysis applications.
Challenges in Biosignal Analysis
Chairs: Norliza Mohd Noor
Biological systems are highly interconnected subsystems and exhibit characteristics that are nolinear, stochastic, deterministic, and nonstationary. The signal is normally has low SNR and the measurement is highly variable and difficult. Biosignal analysis has a wide application area and is an important topic in eHealth and biomedical engineering. This special session welcomes papers discussing the challenges in biosignal analysis that are ranged from the development of sensor devices, the algorithms, improving the low SNR, and improving the accuracy, sensitivity and specificity.
Simulation Studies in Signal Processing
Chairs: Syed Khaleel Ahmed
Many modern signal processing applications deal with systems that are often nonlinear, stochastic, and nonstationary. Exact analytical solutions based on probability theory and statistics are difficult to obtain for even moderately complex systems. In many situations, analytical solutions become intractable. However, the growth of computational power and availability of low cost computing has reopened avenues for analysis using simulation techniques. In this special session¸ authors are invited to share their findings in developing and exploring methodologies and practices dealing with simulation approaches for signal processing applications.
Education in Signal Analysis
Chairs: Nor’aini Abd Jalil
Signal Analysis describes the field of study whose objective is to gather, understand, and interpret information and intelligence from various signal sources. It has very wide applications such as in biomedical, biometrics, identification of foreign signals for security, communication and etcetera. These types of signal analysis require a strong and extensive background in Engineering, Computer Science, and Mathematics. Knowledge in Signal Analysis is very demandable in producing precise interpretation of the acquired signals for exact identification or decision. The importance of this knowledge not only created awareness in embedding the course on Signal Analysis as a core in the engineering education curriculum but also in most technical related programs.
Signal and Imaging Neuroinformatics
Chairs: Vijanth S. Asirvadam, Naufal Saad
This special session focuses on advancement in the field of neuro signal and imaging using multi-modality acquisition tools such as MRI, EEG, SPECT, FNiR PET, real-time MEG and EEG with fundamental contribution contains of both signal and image processing methods. The special session also accepts novel findings related to signaling and imaging pathways associated to both neuro vascular or nervous system and also in the area of neuroinformatics when dealing with large amount of data with different modalities.
Human Action Recognition
Chairs: Syed Abd Rahman
Lately, research on Human Action Recognition has received a great deal of attention from reseachers from various disciplines. This is because this field has a wide spectrum of applications such as intelligent video surveillance, man-machine interaction, video indexing and health care monitoring. Since then, many interesting and innovative methods have been reported and published. Nevertheless the challenges in developing fully functional system have not been fully resolved. These challenges include complex and cluttered background, multiple persons with multiple actions scenes, real time video implementation algorithms and adequate compilations of realistic video footage (actions in the wild) with ground truth. Hence, this special session would like to invite prospective authors to submit papers that adress novel solutions to these challenges.
Digital Pathology (Histopathological Image Analysis)
Chairs: Mohammad Faizal Ahmad Fauzi
Digital pathology is the management and interpretation of pathology information in a digital environment that enables better, faster and cheaper diagnosis, prognosis and prediction of cancer and other important diseases. With the advent of whole-slide imaging, the field of digital pathology has exploded, and is currently regarded as one of the most promising avenues of diagnostic medicine. Histopathological analysis of tissues has also been gaining a lot of interests as a result; developing computer algorithms to assist pathologists from cell detection and counting, to tissue classification and cancer grading. Research in pathological image analysis however has yet to take off in Malaysia, and this special session is intended to bring digital pathology experts around the world to showcase their latest research to local researchers.
Geo-Signal and Seismic Processing
Chairs: Maxwell Meju