Deep Learning for Internet of Things

Submission Deadline:  20 September 2021

IEEE Access invites manuscript submissions in the area of Deep Learning for Internet of Things.   

In recent years, the techniques of Internet of Things (IoT) and mobile communications have been developed to detect and collect human and environment information (e.g. geo-information, weather information, bio-information, human behaviors, etc.) for a variety of intelligent services and applications. The three layers in IoT are the sensor, networking, and application layers; several techniques and standards (e.g. oneM2M, Open Connectivity Foundation, etc.) have been proposed and established for these three layers. For the sensor and networking layers, the rise of mobile technology advancements (e.g. wireless sensor networks, LoRaWAN, Sigfox, narrow band-IoT, etc.) has led to a new wave of machine-to-machine (M2M), machine-to-human (M2H), human-to-human (H2H), and human-to-machine (H2M) communications. For the application layer, the IoT techniques in several applications, including energy, enterprise, healthcare, public services, residency, retail, and transportation, have been designed and implemented to detect environmental changes and send instant updates to a cloud computing server farm via mobile communications and middleware for big data analyses. One of the perfect examples is that the vehicle on-board units can instantly detect and share information about the vehicle geolocation, speed, following distance, as well as gaps with other neighboring vehicles. Big data can be collected by IoT techniques and then analyzed by deep learning techniques for a variety of applications and services.

Deep learning techniques, e.g. neural network (NN), convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory (LSTM), etc., have been popularly applied into image recognition and time-series inference for IoT applications. Advanced driver assistance systems and autonomous cars, for instance, have been developed based on machine learning and deep learning techniques, which perform forward collision warning, blind spot monitoring, lane departure warning, traffic sign recognition, traffic safety, infrastructure management and congestion, and so on. Autonomous cars can share their detected information, such as traffic signs, collision events, etc., with other cars via vehicular communication systems, e.g., dedicated short range communications (DSRC), vehicular ad hoc networks (VANETs), long term evolution (LTE), and 5th generation mobile networks for cooperation. However, how to enhance the performance and efficiency of these deep learning techniques is one of the big challenges for implementing these real-time applications.

Furthermore, several optimization techniques, such as stochastic gradient descent algorithm (SGD), adaptive moment estimation algorithm (Adam), and Nesterov-accelerated Adaptive Moment Estimation (Nadam), have been proposed to support deep learning algorithms for faster solution searching; for example, the gradient descent method is a popular optimization technique to quickly seek the optimized weight sets and filters of CNN for image recognition. The IoT applications based on these image recognition techniques (autonomous cars, augmented reality navigation systems, etc.) have gained considerable attention, and the hybrid approaches typical of mathematics for engineering and computer science (deep learning and optimization techniques) can be investigated and developed to support a variety of IoT applications.

The topics of interest include, but are not limited to:

• Deep learning for massive IoT
• Deep learning for critical IoT
• Deep learning for enhancing IoT security
• Deep learning for enhancing IoT privacy
• Preprocessing of IoT data for AI modeling
• Deep learning for IoT applications (smart home, smart agriculture, interactive art, etc.)

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility and downloads of articles.

Associate Editor: Chi-Hua Chen, Fuzhou University, China

Guest Editors:

1. 1. Yi-Bing Lin, National Yang Ming Chiao Tung University, Taiwan
   2. Kuo-Ming Chao, Coventry University, UK

Relevant IEEE Access Special Sections:

1. 1. Intelligent Logistics Based on Big Data
   2. Real-Time Machine Learning Applications In Mobile Robotics
   3. Advances in Machine Learning and Cognitive Computing for Industry Applications

IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

 For inquiries regarding this Special Section, please contact: chihua0826@gmail.com.

Deep Learning Technologies for Internet of Video Things

Submission Deadline:  31 March 2021

IEEE Access invites manuscript submissions in the area of Deep Learning Technologies for Internet of Video Things.   

The past decade has witnessed tremendous advances in Internet of Things (IoT) technologies, protocols, and applications. The goal of IoT is to connect every physical device/sensor (with video, audio, texting capabilities) in a seamless network to allow communication and perform intelligent decisions. Among these, video data devices (things) in IoT are increasingly becoming important since video communications are an essential part of everyone’s experiences and everything that is happening around the world. The technologies of Internet of Video Things (IoVT) are widely expected to bring exciting services and applications for security, healthcare, education, transportation, and so on.

Due to the huge amount of video data that is being generated and consumed nowadays, there are many challenges and problems that are yet to be solved to have practical real-world applications of IoVT, from sensing/capturing to displaying the data. First, efficient video sensing technologies are required to capture high quality videos with low-power. Second, video coding and communication technologies are essential for compressing and transmitting enormous volumes of data. Third, since not all transmitted data are useful for (or targets) human consumption, there are huge challenges in the areas of data learning/understanding to filter and extract high-level information. Finally, video quality enhancement and assessment algorithms are indispensable for improving and evaluating the video quality, respectively.

Recently, data-driven algorithms such as deep neural networks have attracted a lot of attention and have become a popular area of research and development. This interest is driven by several factors, such as recent advances in processing power (cheap and powerful hardware), the availability of large data sets (big data) and several small but important algorithmic advances (e.g., convolutional layers). Nowadays, deep neural networks are the state-of-the-art for several computer vision tasks, such as the ones that require high level understanding of image semantics, e.g. image classification, object segmentation, saliency detection, but also in low level image processing tasks, such as image denoising, inpainting, super-resolution, among others. Deep learning can handle large volumes of video data by making use of its powerful non-linear mapping, and extracting high-level features with very deep networks. Incorporating deep learning into IoVT can provide radical innovations in video sensing, coding, enhancing, understanding, and evaluation areas, to better handle the enormous growth in video data compared to traditional methods. On the other hand, incorporating deep learning into IoVT also brings a lot of challenges such as the long training latency of data training and huge computational cost.

This Special Section in IEEE Access provides a perfect platform for researchers from academia and industry to discuss the prospective developments and innovative ideas in applying deep learning technologies for IoVT.

The topics of interest include, but are not limited to:

• Deep learning technologies in video sensing/capturing systems
• Deep learning technologies in visual communications
• Deep learning algorithms, architectures, and databases for image and video compression/coding
• Deep learning for enhancement and quality assessment of visual data
• Deep learning-based visual attention and saliency detection
• Deep learning-based real-time and low-power video coding technologies
• Deep learning-based algorithms, architectures, and databases for video analysis and understanding
• Deep learning-based 3D visual coding and processing (from 360º degrees to light-fields)
• Technologies for reducing the complexity of deep learning-based IoVT
• Technologies for reducing training latency of deep learning-based IoVT

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility and downloads of articles.

  Associate Editor:  Jinjia Zhou, Hosei University, Japan

  Guest Editors:

1. 1. Joao Ascenso, University of Lisbon, Portugal
   2. Victor Sanchez, University of Warwick, UK
   3. Lu Zhang, INSA Rennes, France
   4. Jianquan Liu, NEC Corporation, Japan
   5. Jiu Xu, Apple, USA

Relevant IEEE Access Special Sections:

1. 1. Mobile Multimedia: Methodology and Applications
   2. Innovation and Application of Internet of Things and Emerging Technologies in Smart Sensing
   3. Deep Learning Algorithms for Internet of Medical Things

 IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

 Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

 For inquiries regarding this Special Section, please contact: jinjia.zhou.35@hosei.ac.jp.

Emerging Deep Learning Theories and Methods for Biomedical Engineering

Submission Deadline: 31 August 2020

IEEE Access invites manuscript submissions in the area of Emerging Deep Learning Theories and Methods for Biomedical Engineering.

The accelerating power of deep learning in diagnosing disease and analyzing medical data will empower physicians and speed up decision-making in clinical environments. Applications of modern medical instruments and digitalization of medical care have generated large amounts of biomedical information in recent years. However, new AI methods and computational models for efficient data processing, analysis, and modeling with the generated data is important for clinical applications and in understanding the underlying biological process.

Deep learning has rapidly developed in recent years, in terms of both methodological development and practical applications. It provides computational models of multiple processing layers to learn and represent data with various levels of abstraction. It can implicitly capture intricate structures of large-scale data and is ideally suited to some of the hardware architectures that are currently available.

The purpose of this Special Section aims to provide a diverse, but complementary, set of contributions to demonstrate new theories, techniques, developments, and applications of Deep learning, and to solve emerging problems in biomedical engineering.

The ultimate goal of this Special Section is to promote research and development of deep learning for multimodal & multidimensional signals in biomedical engineering by publishing high-quality research articles and reviews in this rapidly growing interdisciplinary field.

The topics of interest include, but are not limited to:

• Theoretical understanding of deep learning in biomedical engineering
• Transfer learning and multi-task learning
• Joint Semantic Segmentation, Object Detection and Scene Recognition on biomedical images
• Improvising on the computation of a deep network; exploiting parallel computation techniques and GPU programming
• Multimodal imaging techniques: data acquisition, reconstruction; 2D, 3D, 4D imaging, etc.
• Translational multimodality imaging and biomedical applications (e.g., detection, diagnostic analysis, quantitative measurements, image guidance of ultrasonography)
• Optimization by deep neural networks, multi-dimensional deep learning
• New model or new structure of convolutional neural network
• Visualization and explainable deep neural network
• Missing data imputation for multi-source biomedical data
• Sparse screening, feature screening, feature merging, quality assessment for biomedical data

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility and downloads of articles.

Associate Editor: Yu-Dong Zhang, University of Leicester, United Kingdom

Guest Editors:

1. 1.  Zhengchao Dong, Columbia University, USA
   2. Juan Manuel Gorriz, University of Granada, Spain
   3. Yizhang Jiang, Jiangnan University, China
   4. Ming Yang, Nanjing Medical University, China
   5. Shui-Hua Wang, Loughborough University, UK

Relevant IEEE Access Special Sections:

1. Deep Learning Algorithms for Internet of Medical Things
2. Smart Health Sensing and Computational Intelligence: From Big Data to Big Impacts
3. Data-Enabled Intelligence for Digital Health

IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

For inquiries regarding this Special Section, please contact:  yudongzhang@ieee.org.

Real-Time Machine Learning Applications In Mobile Robotics

Submission Deadline: 31 December 2020

IEEE Access invites manuscript submissions in the area of Real-Time Machine Learning Applications In Mobile Robotics.

In the last ten years, advances in machine learning methods have brought tremendous developments to the field of robotics. The performance in many robotic applications such as robotics grasping, locomotion, human-robot interaction, perception and control of robotic systems, navigation, planning, mapping, and localization has improved since the appearance of modern machine learning methods. In particular, deep learning methods have brought significant improvements in a broad range of robot applications, including drones, mobile robots, robotics manipulators, bipedal robots, and self-driving cars. The availability of big data and more powerful computational resources, such as Graphics Processing Units (GPUs), have made numerous robotics applications feasible that were not possible previously.

Despite recent advances, there are still gaps in applying available machine learning methods to real robots. Directly transferring algorithms that work successfully in the simulation to the real robot and self-learning of robots are among the current challenges. Moreover, there is also a need for new real-time algorithms and more explainable and interpretable models that receive and process data from the sensors such as cameras, Light Detection and Ranging (LIDAR), Inertial Measurement Unit (IMU), and Global Positioning System (GPS), preferably in an unsupervised or semi-supervised fashion.

This Special Section in IEEE Access aims to present the works relating to the design and usage of the real-time machine learning methods on all mobile robots including legged and humanoid platforms, focusing on state-of-the-art methods, such as deep learning, generative adversarial networks, scalable evolutionary algorithms, reinforcement learning, probabilistic graphical models, Bayesian methods, and explainable and interpretable approaches. The Special Section will present original research articles covering the implementations and applications of mobile robots and incorporating up-to-date results, theorems, algorithms, and systems.

The topics of interest include, but are not limited to:

• Robotic learning by simulations
• Learning and adaptive algorithms for robotic mobile manipulators and humanoid robots
• Human-robot interaction, learning from human demonstrations
• Learning for human-robot collaborative tasks
• Autonomous grasping and manipulation by using mobile robots
• Multi-robot systems, networked robots, and robot soccer
• Control, complex action learning, and predictive learning from sensorimotor information for bio-inspired social robots
• Autonomous driving, navigation, planning, mapping, localization, collision avoidance, and exploration
• Robotics and autonomous system design and implementation
• Nonlinear control and visual servoing in robotic systems
• Soft robotics
• Usage of sensors, such as EEG, ECG, and IMU in robotics
• Usage of explainable machine learning and interpretable artificial intelligence in robotics

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility and downloads of articles.

Associate Editor:  Aysegul Ucar, Firat University, Turkey

Guest Editors:

1. 1. Jessy W. Grizzle, University of Michigan, USA
   2. Maani Ghaffari Jadidi, University of Michigan, USA
   3. Mattias Wahde, Chalmers University of Technology, Sweden
   4. Levent Akin, Bogazici University, Turkey
   5. Jacky Baltes, National Taiwan Normal University, Taiwan
   6. Işıl Bozma, Bogazici University, Turkey
   7. Jaime Valls Miro, University of Technology Sydney, Australia

Relevant IEEE Access Special Sections:

1. Advances in Machine Learning and Cognitive Computing for Industry Applications
2. Integrative Computer Vision and Multimedia Analytics
3. Uncertainty Quantification in Robotic Applications

IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

For inquiries regarding this Special Section, please contact: agulucar@firat.edu.tr.

Behavioral Biometrics for eHealth and Well-Being

Submission Deadline: 28 February 2021

IEEE Access invites manuscript submissions in the area of Behavioral Biometrics for eHealth and Well-Being.

Artificial Intelligence (AI) is changing the healthcare industry from many perspectives. A very challenging issue deals with the development of non-intrusive AI technologies that could be integrated into everyday activities, thus allowing continuous health state monitoring and enabling automatic warnings when a dangerous change is predicted. Behavioral biometrics play a crucial role within this challenge. Behavioral biometrics, such as speech, handwriting, gait, etc. can be used to quantify human physiology, pathophysiological mechanisms, and actions. The final acquired signal is a mixture of at least four components:

• The physical one, which enables the user to make the action (e.g. mouth, lips, tongue, etc.);
• The cognitive one, which deals with mental abilities (learning, thinking, reasoning, remembering, problem-solving, decision-making, and attention);
• The learned one, which deals with culture, habits, personalization, etc.;
• The contingent contour one, which deals with the acquisition device, the emotional state, the specific task to be performed, etc.

It is evident that disease at its early stage, as well as during its course, could affect one or more of these components. Behavioral biometrics in eHealth seek solutions to diagnose, assess, and monitor diseases that are measurable just when the patient performs an action. This action could be walking, talking, writing or typing on a touchscreen, and many more. Behavioral biometrics also deal with the way the human being responds to natural and social events around her/him and emotions. The adoption of non-intrusive behavioral biometrics techniques in the set of daily activities would be pervasive: the user would be asked to do what she/he already does normally. The output of these systems could be provided to doctors, thus helping them in a deep disease inspection. At the same time these technologies could be directly adopted by doctors. These aspects are extremely important for the development of Computer Aided Diagnosis (CAD) tools. Nevertheless, specific behavioral biometrics tasks and activities could be planned to support rehabilitation activities.

This Special Section in IEEE Access aims to attract original research articles that advance the state of the art in behavioral biometrics for e-health and well-being. The goal is that it provides an opportunity to gain a significantly better understanding of the field’s current developments and future direction.

The topics of interest include, but are not limited to:

• Signal processing techniques
• Pattern Recognition techniques
• Computer Vision techniques
• Artificial Intelligence techniques
• Continuous learning and recognition
• Acquisition tools, procedures and protocols
• Biometrics data mining
• Wearable and non-intrusive sensors
• Brain signals analysis for disease and emotional states recognition
• Eye movement analysis for disease recognition
• Face analysis for disease and emotional state recognition
• Gait analysis for disease and emotional state recognition
• Handwriting analysis for disease and emotional state recognition
• Keystroke dynamics for disease and emotional state recognition
• Sleep analysis for disease and emotional state recognition
• Speech analysis for disease and emotional state recognition
• Biometric data and clinical data fusion
• Multiple behavioral biometrics
• Development of complete CAD systems
• Real-time health alerts and long-term health trend analytics
• Applications

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility and downloads of articles.

Associate Editor:  Donato Impedovo, University of Bari Aldo Moro, Italy

Guest Editors:

1. 1. Thurmon Lockhart, Arizona State University, United States
   2. Jiri Mekyska, Brno University of Technology, Czech Republic
   3. Bijan Najafi, Baylor College of Medicine, United States
   4. Toshihisa Tanaka, Tokyo University of Agriculture and Technology, Japan

Relevant IEEE Access Special Sections:

1. Data-Enabled Intelligence for Digital Health
2. Smart Health Sensing and Computational Intelligence: From Big Data to Big Impacts
3. Data Analytics and Artificial Intelligence for Prognostics and Health Management (PHM) Using Disparate Data Streams

IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

For inquiries regarding this Special Section, please contact: donato.impedovo@uniba.it.

Intelligent Biometric Systems for Secure Societies

Submission Deadline: 30 September 2020

IEEE Access invites manuscript submissions in the area of Intelligent Biometric Systems for Secure Societies.

Recent discourse in security domains points to the emergence of intelligent biometric systems as some of the most potent authentication systems for government agencies, border surveillance, anomaly detection, and medical data protection. New and emerging biometric research spurred scientific debate and empowered innovations in on-line security, e-banking, healthcare, and collaborative spaces, to name a few.  This Special Section on Intelligent Biometric Systems for Secure Societies in  IEEE  Access  will  focus  on  novel  biometrics  knowledge  representation  theories, methodologies, applications and technological implementations, aimed towards ensuring smarter, safer and more secure societies. Particular emphasis will be put on real-world and industrial applications supported by experimental or empirical studies.

The topics of interest include, but are not limited to:

• Intelligent biometric-based authentication protocols
• Intelligent biometric system architectures
• Biometric-based identity and trust management
• Cloud-based biometric identification
• Cognitive biometric systems
• Biometric intelligence from social media
• Social behavioral biometrics
• Machine learning and deep learning for biometrics
• Multi-modal biometrics
• Information fusion for reliable decision-making
• Biometric-based digital rights management
• Quality of biometric data
• Security of biometric databases
• Biometrics and online security
• Biometric template protection and privacy
• Mobile biometrics
• Biometrics-based healthcare
• Emerging biometrics

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility and downloads of articles.

Associate Editor:  Marina L. Gavrilova, University of Calgary, Canada

Guest Editors:

1. 1. Gee-Sern Jison Hsu, National Taiwan University of Science and Technology, Taiwan
   2. Khalid Saeed, Bialystok University of Technology, Poland
   3. Svetlana Yanushkevich, University of Calgary, Canada

Relevant IEEE Access Special Sections:

1. Advanced Data Mining Methods for Social Computing
2. Distributed Computing Infrastructure for Cyber-Physical Systems
3. Emerging Approaches to Cyber Security

IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

For inquiries regarding this Special Section, please contact: mgavrilo@ucalgary.ca.

Gigapixel Panoramic Video with Virtual Reality

Submission Deadline: 15 August 2020

IEEE Access invites manuscript submissions in the area of Gigapixel Panoramic Video with Virtual Reality.

Panoramic video is also known as a Panoramic Video Loop, in which traditional static photos are replaced by more dynamic representations. As a counterpart of the image stitching, panoramic video can provide more information and improve the quality of digital entertainment. Unlike a typical rectangular video that shows only the front view of a scene, gigapixel panoramic video captures omni-directional lights from the surrounding environment. This allows a viewer to interactively look around the scene, possibly providing a strong sense of presence. This potential change of the viewing paradigm arising from the use of gigapixel panoramic vides has attracted much attention from the industry and the general public. Panoramic video streaming services are now available through companies such as YouTube and Facebook, and head-mounted display devices such as Samsung Gear VR and Oculus Rift, which support 360-degree viewing, are starting to be more commonly used. In the fields of virtual reality and augmented reality, content creators have constructed gigapixel panoramic video in order to deliver stories with more visually immersive experiences than previously possible.

Although constructing image panoramas by assembling multiple photos from a shared viewpoint is a well-studied problem, it is still difficult to construct large-scale panoramic video. Generally, the key step for constructing gigapixel panoramic video is to stitch unsynchronized videos into a large-scale dynamic panorama. The processing of gigapixel panoramic video construction involves several stages including video stabilization, dynamic feature tracking, vignetting correction, gain compensation, loop optimization, color consistency and image blending. At present, all existing panoramic video devices use tiled multiscale image structures to enable viewers to interactively explore the captured image stream. Size, weight, power and cost of the devices are central challenges in gigapixel panoramic video.

This Special Section aims to review the latest results of image panorama techniques and devices in gigapixel panoramic video construction, as well as their applications. We hope that the Special Section will also help researchers exchange the latest technical progress in the field.

The topics of interest include, but are not limited to:

• Gigapixel panoramic video loops
• Gigapixel image stitching
• Video stabilization for gigapixel video panorama
• Gain compensation
• Color consistency for gigapixel panoramic video
• Image blending for gigapixel panoramic video
• Vignetting correction for gigapixel panoramic video
• Loop optimization for gigapixel panoramic video
• Gigapixel panoramic video for virtual reality
• Gigapixel panoramic video for augmented reality
• Novel devices for producing Gigapixel panoramic video
• Novel approaches to gigapixel panoramic video-based content creation
• Super-resolution for Gigapixel image/video

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility and downloads of articles.

Associate Editor:  Zhihan Lv, University of Barcelona, Spain

Guest Editors:

1. 1. Shangfei Wang, School of Computer Science and Technology, China
   2. Rong Shi, Facebook, USA
   3. Neeraj Kumar, Thapar Institute of Engineering and Technology, India

Relevant IEEE Access Special Sections:

1. Recent Advances in Video Coding and Security
2. Advanced Optical Imaging for Extreme Environments
3. Biologically inspired image processing challenges and future directions

IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

For inquiries regarding this Special Section, please contact: lvzhihan@gmail.com.

Trends and Advances in Bio-Inspired Image-Based Deep Learning Methodologies and Applications

Submission Deadline: 31 October 2020

IEEE Access invites manuscript submissions in the area of Trends and Advances in Bio-Inspired Image-Based Deep Learning Methodologies and Applications.

Many of the technological advances we enjoy today have been inspired by biological systems due to their ease of operation and outstanding efficiency. Designing technological solutions based on biological inspiration has become a cornerstone of research in a variety of areas ranging from control theory and optimization to computer vision, machine learning and artificial intelligence. Especially in the latter few areas, biologically relevant solutions are becoming increasingly important as we look for new ways to make artificial systems more efficient, intelligent and overall effective.

It is generally acknowledged that the human brain is a multitude of times more efficient than the best artificial intelligence algorithms and machine learning models available today. This suggests that there is still something fundamental to learn from the way the brain processes information and new (biologically-inspired) ideas are needed to devise a more effective form of computation capable of competing with the efficiency of biological systems.

One of the hottest and most active research topics in the field of machine learning and artificial intelligence right now is deep learning. Deep learning models exhibit a certain kind of biological relevance, but differ significantly from what we see in the human brain in their structure and efficiency, and the way they process information. Deep learning models, such as convolutional neural networks, consist of several processing layers that represent data at multiple levels of abstraction. Such models are able to implicitly capture the intricate structures of large-scale data and are closer in terms of information processing mechanisms to biological systems than earlier so-called shallow machine learning models.

However, despite the recent progress in deep learning methodologies and their success in various fields, such as computer vision, speech technologies, natural language processing, medicine, and the like, it is obvious that current models are still unable to compete with biological intelligence. It is, therefore, natural to believe that the state of the art in this area can be further improved if bio-inspired concepts are integrated into deep learning models.

The purpose of the Special Section is to present and discuss novel ideas, research, applications and results related to techniques of image processing and computer vision approaches based on bio-inspired intelligence and deep learning methodologies. It aims to bring together researchers from various fields to report the latest findings and developments in bio-inspired image-based intelligence, with a focus on deep learning methodologies and applications, and to explore future research directions.

The topics of interest include, but are not limited to, image-based methodologies, applications, and techniques such as:

• Bio-inspired deep model architectures
• Theoretical understanding of bio-inspired deep architectures, models and loss functions
• Novel bio-inspired training approaches for deep learning models
• Effective and scalable bio-inspired parallel algorithms to train deep models
• Bio-inspired deep learning techniques for modeling sequential (temporal) data
• Biologically relevant adaptation techniques for deep models
• End-to-end bio-inspired deep learning solutions
• Bio-inspired model design
• Bio-inspired visualizations and explanations of deep learning
• Applications of bio-inspired deep approaches in various domains

Note that “bio-inspired” is a crucial keyword in the above list. Thus, the submissions are expected to include a discussion about the bio-inspired background of the presented method. The authors must explain how their method and its novelty correlate with what we find in nature and/or organisms, brain, psychology and similar.

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility, downloads, and citations of articles.

Associate Editor:  Peter Peer, University of Ljubljana, Slovenia

Guest Editors:

1. 1. Carlos M. Travieso-González, University of Las Palmas de Gran Canaria, Spain
   2. Vijayan K. Asari, University of Dayton Vision Lab, USA
   3. Malay K. Dutta, Dr. A.P.J. Abdul Kalam Technical University, India

Relevant IEEE Access Special Sections:

1. Deep Learning: Security and Forensics Research Advances and Challenges
2. Scalable Deep Learning for Big Data
3. Deep Learning Algorithms for Internet of Medical Things

IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

For inquiries regarding this Special Section, please contact: peter.peer@fri.uni-lj.si.

Intelligent Logistics Based on Big Data

Submission Deadline: 20 May 2020

IEEE Access invites manuscript submissions in the area of Intelligent Logistics Based on Big Data.

The advent of the era of big data and the rapid development of e-commerce have provided a new development direction for the modern logistics industry, prompting the logistics industry to think more about data. In addition, the operation mode has gradually changed from the traditional extensive mode to the intelligent logistics one, characterized by information, data, sharing and intelligence.

Intelligent logistics based on big data has significantly improved the intelligence level of warehousing, transportation and distribution, including the intelligent location of logistics outlets, the optimal configuration of transportation routes, the highest loading rate of transportation vehicles, and the optimal distribution of the last mile, which can be used to explore greater potential business value through massive logistics data analysis.

The goal of this Special Section in IEEE Access is to provide a specific opportunity to review the state-of-the-art of intelligent logistics in big data, and bring together researchers in the relevant areas to share the latest progress, novel methodologies and potential research topics.

The topics of interest include, but are not limited to:

• Design and development of intelligent logistics system
• Data collection and knowledge management for intelligent logistics based on Big Data
• Analysis of intelligent logistics mode based on Big Data
• Development of smart logistics systems using Big Data
• Emergency logistics modeling and optimization based on Big Data
• Optimal design of manufacturing/remanufacturing logistics network
• Data-driven-based intelligent logistics management methods & technologies
• Internet-of-things-based intelligent logistics design and optimization
• Environment analysis of reverse logistics based on Big Data
• Modeling of network design for intelligent logistics using Big Data

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility, downloads, and citations of articles.

Associate Editor:  Zhiwu Li, Macau University of Science and Technology, Macau

Guest Editors:

1. 1. Guangdong Tian, Shandong University, China
   2. Di Wu, Hunan University, China
   3. MengChu Zhou, New Jersey Institute of Technology, Newark, USA
   4. Feng Chu, Univeristy of Paris-Saclay and University of Evry, France

Relevant IEEE Access Special Sections:

1. Applications of Big Data in Social Sciences
2. AI-Driven Big Data Processing: Theory, Methodology, and Applications
3. Urban Computing and Intelligence

IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

For inquiries regarding this Special Section, please contact: systemscontrol@gmail.com.

Edge Computing and Networking for Ubiquitous AI

Submission Deadline: 15 May 2020

IEEE Access invites manuscript submissions in the area of Edge Computing and Networking for Ubiquitous AI.

Edge computing has become an important solution to break through the bottleneck of emerging technology development by virtue of its advantages of reducing data transmission, decreasing service latency and easing cloud computing pressure. It can also be applied to extensive application scenarios, such as smart city, manufacturing, logistics and transportation, healthcare, and smart grid. In these scenarios, transmitting massive data and requests generated by edge devices to the cloud data center is no longer the only option, and the edge computing architecture can be complementary to the cloud. Among several application scenarios, such as network optimization, intelligent manufacturing, and real-time video analytics, the combination of Deep Learning (DL) and edge computing shows its advantages.

For example, the DL model trained for face recognition can be deployed on the edge architecture to achieve real-time identity verification. In addition, from predictive maintenance to network and resource management, many researchers are paying attention to “artificial intelligence” plus “edge computing,” aiming to enhance the computing, storage and communication capabilities of edge computing networks through artificial intelligence techniques, especially Deep Reinforcement Learning (DRL). With the increment of smart devices and the diversification needs, the network environment is becoming more complex. Traditional network technologies rely on fixed mathematical models, which are not applicable in a rapidly changing network environment. The emergence of artificial intelligence can effectively solve this problem. When network devices face some complex and fuzzy network information, artificial intelligence technology relies on its powerful learning and reasoning ability to extract valuable information from massive data, and can realize intelligent management.

However, such ubiquitous intelligence potentially enabled by both edge computing and learning still faces a major challenge, i.e., the effective deployment fashion of the learning model on the collaborated “edge-cloud” architecture is still not determined. The deployment of deep learning models should concern the training and inference of them, and the edge computing architecture shall be well devised.

The topics of interest include, but are not limited to:

• Deep learning applications enabled by edge computing
• Deep learning and deep reinforcement learning for optimizing edge computing networks
• Deep learning-based traffic offloading prediction and optimization
• Distributed and collaborative AI with edge computing and networking
• Hardware platforms and software stacks for deploying deep learning on the edge
• Data processing and business intelligence on the edge
• Offloading scheme for intensive deep learning tasks
• Architecture and orchestration of deep learning services in edge computing
• Deep learning for the management of edge computing networks
• Transfer learning for the preliminary deployment of deep learning models on the edge
• Training scheme of deep learning model at the edge
• Federated learning for massive edge devices, edge nodes and the cloud data center
• Federated learning devised for deep reinforcement learning, i.e., federated reinforcement learning
• Compression of deep learning models for deploying them on edge devices or edge nodes
• Segmentation of deep learning models for collaborative intelligence between cloud and the edge
• “Early exit of inference” of deep learning models for accelerating the edge intelligence
• Incentive-based training and inference schemes for heterogeneous devices in the edge
• The fusion of training and inference in the edge computing network
• New AI-based edge computing and networking testbed and trials

We also highly recommend the submission of multimedia with each article as it significantly increases the visibility, downloads, and citations of articles.

Associate Editor:  Victor Leung, The University of British Columbia, China

Guest Editors:

1. 1. Xiaofei Wang, Tianjin University, China
   2. Abbas Jamalipour, The University of Sydney, Australia
   3. Xu Chen, Sun Yat-sen University, China
   4. Samia Bouzefrane, Conservatoire National des Arts et Métiers, France

Relevant IEEE Access Special Sections:

1. Communication and Fog/Edge Computing Towards Intelligent Connected Vehicles (ICVs)
2. 5G and Beyond Mobile Wireless Communications Enabling Intelligent Mobility
3. Artificial Intelligence and Cognitive Computing for Communications and Networks

IEEE Access Editor-in-Chief:  Prof. Derek Abbott, University of Adelaide

Article submission: Contact Associate Editor and submit manuscript to:
http://ieee.atyponrex.com/journal/ieee-access

For inquiries regarding this Special Section, please contact:  xiaofeiwang@tju.edu.cn.