Skip to main content

2022 | Book

Advances on P2P, Parallel, Grid, Cloud and Internet Computing

Proceedings of the 16th International Conference on P2P, Parallel, Grid, Cloud and Internet Computing (3PGCIC-2021)


About this book

This book provide latest research findings, innovative research results, methods and development techniques from both theoretical and practical perspectives related to P2P, grid, cloud and Internet computing as well as to reveal synergies among such large-scale computing paradigms. P2P, grid, cloud and Internet computing technologies have been very fast established as breakthrough paradigms for solving complex problems by enabling aggregation and sharing of an increasing variety of distributed computational resources at large scale.
Grid computing originated as a paradigm for high performance computing, as an alternative to expensive supercomputers through different forms of large-scale distributed computing. P2P computing emerged as a new paradigm after client-server and web-based computing and has shown useful to the development of social networking, Business to Business (B2B), Business to Consumer (B2C), Business to Government (B2G), Business to Employee (B2E) and so on. Cloud computing has been defined as a “computing paradigm where the boundaries of computing are determined by economic rationale rather than technical limits.” Cloud computing has fast become the computing paradigm with applicability and adoption in all application domains and providing utility computing at large scale. Finally, Internet computing is the basis of any large-scale distributed computing paradigms; it has very fast developed into a vast area of flourishing field with enormous impact on today’s information societies serving thus as a universal platform comprising a large variety of computing forms such as grid, P2P, cloud and mobile computing.

Table of Contents

Performance Analysis of RIWM and LDVM Router Replacement Methods for WMNs by WMN-PSOSA-DGA Hybrid Simulation System Considering Stadium Distribution of Mesh Clients

Wireless Mesh Networks (WMNs) have many advantages such as: easy maintenance, low upfront cost and high robustness. The connectivity and stability affect directly the performance of WMNs. However, WMNs have some problems such as node placement problem, hidden terminal problem and so on. In our previous work, we implemented a simulation system to solve the node placement problem in WMNs considering Particle Swarm Optimization (PSO), Simulated Annealing (SA) and Distributed Genetic Algorithm (DGA), called WMN-PSOSA-DGA. In this paper, we compare the performance of Random Inertia Weight Method (RIWM) and Linearly Decreasing Vmax Method (LDVM) for WMNs by using WMN-PSOSA-DGA hybrid simulation system considering Stadium distribution of mesh clients. Simulation results show that LDVM has better performance than RIWM.

Admir Barolli, Shinji Sakamoto, Leonard Barolli, Makoto Takizawa
A Transfer Learning-Based Object Detection and Annotation System: Performance Evaluation for Vehicle Objects from Onboard Camera

It is a challenge to collect the required image data rapidly and effectively for object detection in emergency disaster situations. In this work, we focus on improving the classification accuracy and duration by collecting object images parallelly in the target disaster environment. In this paper, we propose a Transfer Learning (TL)-based object detection and annotation system. Our system has a novel image selection function to reduce the similar images for preparing a dataset that can improve the accuracy and model training duration. Our system train the model considering vehicular objects and their corresponding labels from video data of onboard camera. From the evaluation results, we observed that our proposed image selection function for training can reduce the number of sequential similar images to about 23%. Also, our TL-based object detection system can improve the detection performance compared with conventional learning method.

Yoshiki Tada, Masahiro Miwata, Shota Uchimura, Makoto Ikeda, Leonard Barolli
Fuzzy Randomized Load Balancing for Cloud Computing

In cloud computing different resources are grouped to work collectively to service users from remote locations. The cloud system’s responsibility is to route the user request to such resources, which are less utilized to achieve the least response time. This concept of distributing the workload from cloud users transparently among all the resources is called load balancing. In this paper, we discuss various algorithms for support load balancing. We propose a Fuzzy-based Randomized Load Balancing(FRLB) model that overcomes shortcomings of throttled load balancing. The simulation results are analyzed and compared to other existing load balancing algorithms proving better efficiency over them.

Saurabh Sagar, Mushtaq Ahmed, Mohammed Yaseen Husain
A LiDAR Based Mobile Area Decision Method for TLS-DQN: Improving Control for AAV Mobility

The Deep Q-Network (DQN) is one of the deep reinforcement learning algorithms, which uses deep neural network structure to estimate the Q-value in Q-learning. In the previous work, we designed and implemented a DQN-based Autonomous Aerial Vehicle (AAV) testbed and proposed a Tabu List Strategy based DQN (TLS-DQN). In this paper, we propose a LiDAR Based Mobile Area Decision Method for TLS-DQN to improve the control for AAV Mobility. The evaluation results show that the proposed method makes a good decision for the destination and mobile area based on LiDAR.

Nobuki Saito, Tetsuya Oda, Aoto Hirata, Chihiro Yukawa, Elis Kulla, Leonard Barolli
The Comparative Study of Algorithms in Building the Green Mobile Cloud Computing Environment

The usage of Mobile Cloud Computing over the year is directly proportional to the increase in energy consumption. This problem is then solved with the green approach, where many algorithms or methods were intensively developed to achieve an optimal state of Quality of Services performance. In this article, we conducted a Systematic Literature Review to find the latest algorithms and their respective performance for the Green Mobile Cloud Computing. From 25 papers, we conclude that heuristic and metaheuristic algorithms are the most widely applied for the computation offload and resource scheduling cases, respectively. Most articles we found used energy consumption rate and completion time as their Quality of Services measurement.

Nicholas Dominic, Jonathan Sebastian Prayoga, Daniel Kumala, Nico Surantha, Benfano Soewito
Wearable Internet-of-Things Device for COVID-19 Detection, Monitoring and Prevention: A Review

COVID-19, a new infectious coronavirus which initiated a pandemic outbreak lately, has challenged technology innovators to initiate more research regarding medical healthcare. Yet, detection for COVID-19 itself requires people to take tests in the hospital, which incites a higher risk of infection and the hospital itself gets overloaded. Nowadays, researches regarding the Internet of Things (IoT) in medical sectors especially COVID-19 are providing many solutions. One of the solutions is wearable IoT devices to detect symptoms using vital signs and to prevent users from being further infected. This paper will be conducted as a systematic literature review which is started by collecting references using predefined keywords and selecting references based on reliable journals. In this paper, we review the use of wearable IoT devices in COVID-19 detection using users’ vital signs and monitor social distancing as a preventive method.

Nico Surantha, Gary Nico, Michael Henry, Wiryanata Chandra, Benfano Soewito
Modern Cognitive Solutions for Advanced Information Processing

This paper will discuss modern cognitive solutions dedicated to information processing. Cognitive processes, as the most individual methods of information processing, are characteristic of every human being, and thus transferred to the ground of machine solutions, they offer the possibility of deep, meaningful data analysis. Their distinguishing feature is that they are based on a semantic interpretation based on a linguistic description. The possibility of using such solutions in information processing will allow to show their universality and usefulness.

Urszula Ogiela, Makoto Takizawa, Lidia Ogiela
An Efficient Machine Learning System for Connected Vehicles

In online machine learning systems, a computer gathers sets of training data from some data sources and trains a machine learning model every it gets a set. In the cases that the time required to gather a set is long such as the case of the data gathering from connected vehicles, the delay for reflecting the observed environmental values included in the training data to the model lengthens. In this paper, we propose a system to reduce the delay for reflecting observed environmental values to the models suppressing the increase of the validation loss. Our proposed system cyclically broadcasts the parameters of the machine learning model to the data sources and the data sources calculate the result of the loss function for their observed training data. We evaluated the proposed system assuming a machine learning system for connected vehicles. The vehicles of that training data give a larger value to the result than a given threshold send the training data to the computer for training the machine learning model. Our experimental evaluation revealed that our proposed methods can achieve lower validation loss values than a conventional method.

Tomoki Yoshihisa
Object Tracking by Google Cloud API and Data Alignment for Front/rear Car DVR Footages

Until recently, dashboard cameras have been used primarily by law enforcement agencies around the world. With their widespread use among private users, they have recently become almost ubiquitous. Driven by the potential of their application, especially in the insurance and security industries, this rise has been contributing to the advancement of research in the fields of video analytics, leading to the development of new anomaly detection techniques [1]. In this paper we exploit the potential of the Google Video Intelligence APIs in order to perform a standard object tracking operation on front/rear videos produced by car DVRs. This technology allowed us to save the intervals in which the detection occurred, used for subsequent data alignment, a very under-researched problem in this field. Our new approach for alignment is based on calculating the period of time that objects in the video are outside the field of view of both dashcams, using the.gpx file attached to each pair of front/rear videos. after this was accomplished, the individual detections were compared by constructing a fictitious continuous interval from the rcalculated blind time. the results differ according to the traffic levels within the videos. the best performances were achieved in low traffic situations where the calculation of the blind time interval is considerably more accurate, given the nearly constant speed held by the dashcam mounted car.

Walter Balzano, Leonard Barolli, Francesco Zangrillo
Efficient Federated Learning Framework Based on Multi-Key Homomorphic Encryption

Federated learning is one of the cutting-edge research area in machine learning in era of big data. In federated learning, multiple clients rely on an untrusted server for model training in a distributed environment. Instead of sending local data directly to the server, the client achieves the effect of traditional centralized learning by sharing optimized parameters that represent the local model. However, the data used to train the model by the client holds private information of individuals. A potential adversary can steal the clients’ model parameters by corrupting the server, then recover clients’ local training data or reconstruct their local models. In order to solve the aforementioned problems, we construct an efficient federated learning framework based on multi-key homomorphic encryption, which can effectively restrict the adversary from accessing the clients’ model. In this framework, using homomorphic encryption ensures that all operations, including the server-side aggregation process, are secure and do not reveal any private information about the training data. At the same time, we consider multi-key scenario, where each client does not need to share the same public key and private key, but each of them has its own public-private key pair. It is convenient for the client to join the model update or be offline at any time, which greatly increases the flexibility and scalability of the system. Security and efficiency analysis indicates that the proposed framework is secure and efficient.

Qian Zhang, Shan Jing, Chuan Zhao, Bo Zhang, Zhenxiang Chen
Blockchain-Based Pharmaceutical Supply Chain: A Literature Review

Recently, blockchain technology was introduced to the public in order to provide a secure environment that is immutable, consensus-based and transparent in the finance technology world. However, there have been many efforts to apply blockchain to other fields where trust and transparency are a requirement. The ability to reliably share pharmaceutical information between various stakeholders is essential. The use of blockchain technology adds traceability and visibility to supply chains such as pharmaceuticals to provide all the information from end to end. Currently, the data is stored and managed by large manufacturers and pharmacy retailers using their own centralized systems. Several existing approaches and methods that allow pharmaceutical information to be stored and shared between the healthcare provider and other stakeholders in a centralized manner have been discussed in the literature. Due to the lack of comprehensive literature review studies that focus on pharmaceutical supply chain using blockchain technology, thus this paper highlights and addresses this gap. This paper reviews several studies which have applied blockchain technology in pharmaceutical supply chains. This paper overviews the knowledge on blockchain technology, discusses and explores the most recent and relevant studies that adopt blockchain technology in the field of pharmaceutical supply chains, describes the challenges associated with blockchain technology, and presents some ideas for future work.

Abeer Mirdad, Farookh Khadeer Hussain
A Comparison Study of LDIWM and LDVM Router Replacement Methods for WMNs by WMN-PSODGA Hybrid Simulation System Considering Boulevard Distribution of Mesh Clients

The Wireless Mesh Networks (WMNs) have attracted attention for different applications. They are an important networking infrastructure and they have many advantages such as low cost and high-speed wireless Internet connectivity. However, they have some problems such as router placement, covering of mesh clients and load balancing. To deal with these problems, in our previous work, we implemented a hybrid simulation system based on Particle Swarm Optimization (PSO) and Distributed Genetic Algorithm (DGA) called WMN-PSODGA. Moreover, we added in the fitness function a new parameter for the load balancing of the mesh routers called NCMCpR (Number of Covered Mesh Clients per Router). In this paper, we consider Boulevard distribution of mesh clients and two router replacement methods: Linearly Decreasing Inertia Weight Method (LDIWM) and Linearly Decreasing Vmax Method (LDVM). We carry out simulations using WMN-PSODGA hybrid simulation system and compare the performance of LDIWM with LDVM. The simulation results show that LDIWM has better load balancing than LDVM.

Peng Xu, Admir Barolli, Phudit Ampririt, Shinji Sakamoto, Elis Kulla, Leonard Barolli
A Fuzzy-Based System for Deciding Driver Impatience in VANETs

In this paper, we propose and implement an intelligent system based on Fuzzy Logic (FL) for deciding driver impatience in VANETs. The proposed system, called Fuzzy-based System for Deciding Driver Impatience (FSDDI), considers parameters that have a strong impact on the driver impatience. The input parameters include the driver’s emotional condition, the time pressure and the number of route stops. Based on the driver impatience output value, the system can invoke a certain action, which aims at improving the driver’s mood by providing the appropriate driving support. We show through simulations the effect of the considered parameters on the determination of the driver impatience and demonstrate some actions that can be performed accordingly.

Kevin Bylykbashi, Ermioni Qafzezi, Phudit Ampririt, Makoto Ikeda, Keita Matsuo, Leonard Barolli
Using Photo Images with Deep Residual Network for PM2.5 Value Estimation

Fine particles (PM2.5) become an important issue in Asia. The fine particles are related for causing of severe health problems. This paper focuses on using photo images with deep residual network for PM2.5 value estimation. The proposed framework has been designed to reduce the computational complexity and improve the estimation accuracy. Regression analysis is also introduced in the proposed framework by using LSTM with the meteorological data and the features extracted from the modified ResNet model. The images with HDR and without HDR technique are applied to the image feature extraction process. Thus, the PM2.5 value estimation process can be started using the mobile phone camera.

Anupam Kamble, Paskorn Champrasert
A Scientific Model to Support Industrial Data Management Process Using Virtualized Environments

Nowadays, scientific and industrial fields produce data at rates never seen before. They usually produce an extensive amount of raw data that should be transformed and manipulated to be stored correctly considering different storage perspectives. In many cases, due to the heterogeneity of the temporal data, these storage methods should be configured to make consistent data available for client consumption. Considering raw data management of a natural industrial environment to support specific local tasks, in this paper, an industrial effort model in the context of virtual environment data access was constructed. In this solution, the data were captured, processed, and available in almost real-time for in loco staff through VR technologies. The usage of these virtualized technologies in industrial areas is challenging and promissory due to the wide variety of applications to support locale tasks. The results presented are promising to support tasks and operational demands in complex industrial contexts.

Laércio Pioli, Thiago G. Thomé, Júlia X. M. Nunes, Douglas D. J. de Macedo, Paulo César. R. de L. Junior, Mario A. R. Dantas
An Investigation of Covid-19 Papers for a Content-Based Recommendation System

The proliferation of scientific publications is a well-known phenomenon that was recently emphasized by the publications related to the Covid-19. The number of publications Covid-19 related that PubMed added in the period between January 17 and April 18, 2020 kept rising until it reached a number of 300 publications added in a single day. There are obvious issues related to this phenomenon, such as the difficulty for researchers to find papers strongly related to their applications. When searching for related papers, there could be issues with how a paper is preferred with respect to another. A paper could be recommended based on the greater number of citations, or on the connections between authors, that is as well related to the number of citations. For such reasons, the aim of this study is to build a recommendation system based exclusively on the abstracts of these publications. We provide a comparison between classical approaches—NLP-based such as TF-IDF and n-grams—and Deep Learning approaches for content-based recommendation systems, such as Transformers. We also provide an application to graphs that shows the relationships among related papers on the basis of the results obtained from the recommendation system developed.

Leonard Barolli, Francesco Di Cicco, Mattia Fonisto
A Investigation of Suitable Data Transfer Range for Web-Based Virtual World Applications

This paper proposes a step-based data transfer method on a Web browser network using WebRTC. In our study, we proposed a way to construct a Web browser network. The target application of our method is multi-player online games that has a large number of users and higher frequently data transfer to share game data among the users. Each Web browser has some game data, and they need to inform their data to the other browsers to share the same game world. However, the data flow increases if a browser requires the data from all other browsers. Therefore, our method restricts the range of the data transfer. If our method puts a short transfer range, the browser does not need the sufficient game data. Our study investigate how long transfer rage is suitable for sharing game world.

Masaki Kohana, Shinji Sakamoto, Shusuke Okamoto
Métis - An Approach Utilized as Differentiated Authenticity Tool in an IIoT Infrastructure

Security in industrial environments is a growing concern with the integration of Industrial IoT (IIoT). The communication between devices, diverse users, and the volume of digital data transferred increase the vulnerability. Aiming to tackle this challenge, we developed studies related to the application of smart contracts with blockchain support to guarantee the integrity of identity authenticity of the digital data that travels within the Industrial IoT (IIoT) environment. Therefore, in this paper, we present the Métis proposal, which represents a differentiated authenticity approach and which was tested through simulations to provide a security landscape to a real industrial 4.0 project.

Felipe S. Costa, Mario A. R. Dantas, José M. N. David, Regina M. M. B. Villela, Mattheus S. Santos
An Intelligent System for Admission Control in 5G Wireless Networks Considering Fuzzy Logic and SDNs: Effects of Service Level Agreement on Acceptance Decision

The Fifth Generation (5G) wireless network is expected to be flexible to satisfy user requirements and the Software-Defined Network (SDN) with Network Slicing will be a good approach for admission control. In 5G wireless network, the resources are limited and the number of devices is increasing much more than the system can support. So, the slice resources should be carefully managed and suited to the user’s requests. In this paper, we consider the effect of Service Level Agreement on Admission Control for 5G Wireless Networks. We present a Fuzzy-based system for admission decision. We consider four input parameters: Quality of Service (QoS), Slice Priority (SP), Slice Overloading Cost (SOC) and Service Level Agreement (SLA) as a new parameter. From simulation results, we can see that when QoS, SP, SLA parameters are increased, the AD parameter is increased. But, when the SOC parameter is increased, the AD parameter is decreased.

Phudit Ampririt, Ermioni Qafzezi, Kevin Bylykbashi, Makoto Ikeda, Keita Matsuo, Leonard Barolli
Mixed Cooperative-Competitive Communication Using Multi-agent Reinforcement Learning

By using communication between multiple agents in multi-agent environments, one can reduce the effects of partial observability by combining one agent’s observation with that of others in the same dynamic environment. While a lot of successful research has been done towards communication learning in cooperative settings, communication learning in mixed cooperative-competitive settings is also important and brings its own complexities such as the opposing team overhearing the communication. In this paper, we apply differentiable inter-agent learning (DIAL), designed for cooperative settings, to a mixed cooperative-competitive setting. We look at the difference in performance between communication that is private for a team and communication that can be overheard by the other team. Our research shows that communicating agents are able to achieve similar performance to fully observable agents after a given training period in our chosen environment. Overall, we find that sharing communication across teams results in decreased performance for the communicating team in comparison to results achieved with private communication.

Astrid Vanneste, Wesley Van Wijnsberghe, Simon Vanneste, Kevin Mets, Siegfried Mercelis, Steven Latré, Peter Hellinckx
Learning to Communicate with Reinforcement Learning for an Adaptive Traffic Control System

Recent work in multi-agent reinforcement learning has investigated inter agent communication which is learned simultaneously with the action policy in order to improve the team reward. In this paper, we investigate independent Q-learning (IQL) without communication and differentiable inter-agent learning (DIAL) with learned communication on an adaptive traffic control system (ATCS). In real world ATCS, it is impossible to present the full state of the environment to every agent so in our simulation, the individual agents will only have a limited observation of the full state of the environment. The ATCS will be simulated using the Simulation of Urban MObility (SUMO) traffic simulator in which two connected intersections are simulated. Every intersection is controlled by an agent which has the ability to change the direction of the traffic flow. Our results show that a DIAL agent outperforms an independent Q-learner on both training time and on maximum achieved reward as it is able to share relevant information with the other agents.

Simon Vanneste, Gauthier de Borrekens, Stig Bosmans, Astrid Vanneste, Kevin Mets, Siegfried Mercelis, Steven Latré, Peter Hellinckx
Lane Marking Detection Techniques for Autonomous Driving

In this paper we address challenges facing lane marking detection and tracking. Lane marking detection along with vehicle positioning between lane boundaries are fundamental tasks to achieve safe and reliable autonomous driving systems. Despite the development of perception senors and clarity of the lane markings on roadways, the lane detection remains a challenge for researchers due to environmental factors that impact performance of lane’s recognition algorithms. In this paper we compare three different lane detection strategies based on rule and learning-based approaches using perception sensors. In contrast, we perform rule-based lane detection using camera and sensor fusion combining camera images and LiDAR’s point clouds. Moreover, we use the prominent lane detection learning-based approach LaneNet to detect lanes from images. However, when using the LaneNet, we investigate the network’s performance while excluding the perspective transformation network (H-Net). Our results show that learning-based lane detection methodologies outperforms rule-based methods and can accurately predict lanes when the vehicle is cruising on roads with steep curvatures.

Ahmed N. Ahmed, Ali Anwar, Sven Eckelmann, Toralf Trautmann, Steven Latré, Peter Hellinckx
Quality-Aware Compression of Point Clouds with Google Draco

Situational awareness is getting traction in the field of autonomous inland vessels. Large amounts of data needs to be shared in order to set up this awareness. This ranges from relatively small positional updates, to consistent streams of sensory data. Point clouds, captured by LiDAR sensors, are heavily used by inland vessels as they give a detailed sense of range. This sharing is not a major complexity when vessels are in close proximity with each other; dedicated networks could handle this consistent stream of data. However, when vessels are farther away from each other, long range networks are needed, at the cost of high bandwidth capabilities. Therefore, the sensor message size should be reduced while retaining a reasonable quality. In this paper, we investigate the trade-off between lossless and lossy point cloud compression with Google Draco and its resulting quality. The results show that a considerable size reduction can be applied while the point cloud maintains acceptable quality.

Jens de Hoog, Ahmed N. Ahmed, Ali Anwar, Steven Latré, Peter Hellinckx
Transfer Learning in Autonomous Driving Using Real-World Samples

The Sim2Real gap is a topic that has been receiving a great deal of attention lately. Many Artificial Intelligence techniques, for example Reinforcement Learning, require millions of iterations to achieve satisfactory performance. This requirement often forces these techniques to solely train in simulation. If the gap between the simulated environment and the target environment is too broad, however, the trained agents will lose out on performance when deployed. Bridging this gap lowers the performance loss during deployment, in turn improving the effectiveness of these agents. This paper proposes a new technique to tackle this issue. The technique focuses on the use of demonstration samples gathered in the target environment and is based on two transfer learning fundamentals. By combining the advantages of Domain Randomization and Domain Adaptation, agents are able to transfer training performance to the target environment more successfully. Experimental results show a strong decrease in performance loss during deployment when the agent is exposed to the demonstration samples during training. The proposed technique describes a methodology that we believe can be applied in fields other than autonomous driving in order to improve transfer learning performance.

Arne Troch, Jens de Hoog, Simon Vanneste, Dieter Balemans, Steven Latré, Peter Hellinckx
Autonomous Building Control Using Offline Reinforcement Learning

Artificial Intelligence (AI) powered building control allows deriving policies that are more flexible and energy efficient than standard control. However, there are challenges: environment interaction is used to train Reinforcement Learning (RL) agents but for building control it is often not possible to use a physical environment, and creating high fidelity simulators is a difficult task. With offline RL an agent can be trained without environment interaction, it is a data-driven approach to RL. In this paper, Conservative Q-Learning (CQL), an offline RL algorithm, is used to control the temperature setpoint in a room of a university campus building. The agent is trained using only the historical data available for this room. The results show that there is potential for offline RL in the field of building control, but also that there is room for improvement and need for further research in this area.

Jorren Schepers, Reinout Eyckerman, Furkan Elmaz, Wim Casteels, Steven Latré, Peter Hellinckx
Study on Virtual Disaster Control Headquarters System

Japan is a disaster-prone country. Every year, natural disasters such as earthquakes, typhoons, and heavy rain occur. At each incident of a natural disaster, the local government establishes disaster control headquarters. Meanwhile, disaster control headquarters is also required to deal with the novel coronavirus infection. If a staff of the disaster control headquarters is infected, the disaster control headquarters may not be able to function efficiently. Therefore, a virtual disaster control headquarters system was launched to virtualize the disaster control headquarters. This system has space-sharing, video avatar generation, and information-sharing functions.

Rihito Fuchigami, Tomoyuki Ishida
Study on Interior Layout Experience System Using Mixed Reality Technology

With the development of networks and information communication devices, there has been an increase in the services that use virtual reality (VR) and augmented reality (AR) technologies. Currently, it is possible to experience real and immersive systems using various VR and AR devices. In this study, we implemented an interior layout experience system using mixed reality technology. By superimposing various three-dimensional computer graphics interiors, such as sofa, bed, and light fixtures, on the real space, the user can visualize the interior design.

Reiya Yahada, Tomoyuki Ishida
Extension of Annotation Function in Collaborative Composition System

Due to the development of social media networks, there are now increasing opportunities for multiple people to collaboratively compose music using desktop music and digital audio workstation systems. To facilitate such efforts, we have been developing a system that supports creative, collaborative compositions, have analyzed the conditions required to support such efforts, and have proposed and implemented a collaborative composition system that satisfies these conditions. This system is an asynchronous groupware application with which communications among users are performed by exchanging character-based text comments. In this paper, we describe how we extended the comment function of this previous system to support handwritten annotations.

Meguru Yamashita, Kiwamu Satoh, Akio Doi
An Approach for Composing Facial Expressions by Collecting Face Images on a Video

Recently, many video conference systems and video on demand systems have been used to support audio-visual telecommunication. However, many of these systems do not consider the immersiveness and presence states. So, by these systems can’t be seen the state of other users as in an office and a classroom. The tele-immersion system enables telecommunication with a high presence of users and a high immersiveness into a VR (Virtual Reality) space. However, by using a VR headset, the face is hidden by the VR headset when the user is displayed on the tele-immersion system. In this paper, we consider an approach for composing facial expressions collected by face images on a video, which is required for a tele-immersion system using the VR headset. We prepare the face images from a video using a commercial software and compose some face images using our developed software.

Kaoru Sugita, Kohei Soejima
A Takagi-Sugeno Fuzzy-Based Adaptive Transmission Method in Wireless Sensor Networks

Clustering, data compression and screening of sensor data, optimization methods and other intelligent approaches to increase the lifetime of sensor networks have been widely researched. In this paper, we propose an intelligent adaptive transmission control system based on Takagi-Sugeno (T-S) fuzzy inference model in Wireless Sensor Networks (WSNs). From the evaluation results, we observed that the proposed method reduces the number of transmissions by considering multiple parameters compared with the conventional method.

Daisuke Nishii, Makoto Ikeda, Leonard Barolli
Crack Detection from Weld Bend Test Images Using R-CNN

The personnel burden is an issue with the visual inspection of welding defects that occur in bend test fragments. This study aims to construct an automatic evaluation system for welding defects that occur in bend test fragments. This paper describes the automatic detection of defective areas from bend test fragments using R-CNN. First, we have described the structure of the proposed R-CNN, followed by the experiments for evaluating R-CNN and their results. Finally, we have provided a conclusion and discussed future issues.

Shigeru Kato, Takanori Hino, Shunsaku Kume, Hajime Nobuhara
A Modified Version of K-Means Algorithm

In this work is presented a modified version of the K-Means which identifies cluster stability. The stability is defined by a threshold based on a percentage of the largest displacement of centroid at first iteration. A cluster is considered stable when the largest centroid displacement in the current iteration achieves the 10% of threshold, and objects that remains in the same cluster in two consecutive iterations are removed from the classification phase in subsequent iterations. Eight different instances were used to validate the proposal, three synthetics and five reals. The modified version was compared against the standard and three related work versions. Results shows that the proposal reduced the execution time up to 92.14% regarding the standard version with only a 3.73% in the quality reduction. Despite the new version do not has the major reduction time in all cases, the algorithm reaches the best values for quality of grouping.

A. Mexicano, J. C. Carmona, S. Cervantes, J. A. Cervantes, S. López, R. Rodríguez
Neural Network with L-M Algorithm for Arrhythmia Disease Classification

This paper presents a feedforward multilayer perceptron neural network with a Levenberg-Marquardt learning algorithm for recognizing arrhythmia disease from normal electrocardiogram (ECG) patterns. To the best of our knowledge, in the field of arrhythmia disease classification, classical approaches utilize either different QRS complex detection or feature reduction methods but not both at the same time; thus, this work provides an important contribution. A total of forty-four records were obtained from the MIT-BIH arrhythmia database to test the QRS complex detection method, and the obtained results were a specificity of 96.16% and a sensitivity of 98.03%. The best classification rate obtained using the presented approach was 98.27%.

Ricardo Rodríguez-Jorge, Jiří Bíla, Jiří Škvor
Advances on P2P, Parallel, Grid, Cloud and Internet Computing
Prof. Leonard Barolli
Copyright Year
Electronic ISBN
Print ISBN