The presence of cloud shadows in satellite imageries decreases the reflectance of the objects under the shades to relatively low intensities, leads to identification errors. Thus, cloud shadows detection is crucial in image processing steps. We integrated solar position modelling, projective geometry modelling, and spectral thresholding to detect cloud shadows in Landsat 8 imageries. We evaluated the algorithm using the window area of Mount Halimun-Salak, Bogor, West Java, Indonesia. The best rate accuracies of cloud shadow detection using the algorithm was obtained at producer's accuracy, user's accuracy and κ of 63.79%, 70.58%, and 0.66, respectively. Possibility of improving the algorithm for correcting the reflectance of the objects under the shades instead of removing is discussed.
Cloud computing has a fundamental impact on information technology. It is a delivery model in which computing resources (such as computer networks, servers, storage, applications, and services) are made available as virtual resources in on-demanf manner. A major challenge for cloud computing suppliers is to provide adequate security levels to protect users' data. A lot of research has been carried out with this target. In this paper, we evaluated public cloud performance considering different security schemes of firewalls and VPNs c under four scenarios: no firewall no VPN, firewall-only, VPN-only and a combination of firewall and VPN. A simulated cloud model was created to evaluate the public cloud performance under the above scenarios. The results suggest that the cloud server throughput and point-to-point utilization are agnostic to the implemented security sheme. On the other hand, the queing delay is slightly affected when firewalls are used.
The success of implementing Total Quality Management (TQM) depends heavily on its degree of integration with the local environment. To improve the efficiency in deploying TQM in China, it is required that Chinese enterprises find proper focuses in the management system that can remarkably raise quality level and improve enterprise competitiveness. This study employs the Grounded Theory methodology to analyze in depth the TQM practice of Chinese manufacturing enterprises that have won the China Quality Award, and summarizes from those cases the critical factors that Chinese manufacturing enterprises should lay emphasis on when deploying TQM. This paper then compares the factors of TQM practice between Chinese and Western enterprises from both social and cultural perspectives, and explains the changes and new features of QM development in Chinese enterprises.
Global warming, the decline of oil resources, as well as the introduction of emission regulations, has led to a research focus in new drive technologies. Whereas hybrid electric vehicles (HEV) can only be considered as a transition technology, electric vehicles (EV) and fuel cell vehicles (FCV) are zero emission technologies; HEV are only considered a bridge technology. FCV enable longer ranges and faster refills at the cost of a moderate increase in fuel cost. Challenges for an overall acceptance of fuel cell vehicles are the degradation mechanisms of the fuel cell that lead to a very limited lifetime. Current research focuses on energy management strategies to reduce the overall energy demand by predicting the vehicles coarse power consumption on the overall trip. This prediction is used to optimize the control of the drivetrain components. This paper reviews energy management strategies and points out the lack of research in the prediction of short-term speed changes. The knowledge of short-term speed changes can drastically reduce degradation mechanisms by avoiding power fluctuations in the fuel cell as well as in the battery. The proposed hierarchical model predictive control strategy is able to incorporate the knowledge of long-term energy consumption to minimize the energy demand as well as the short-term speed predictions to avoid degradation mechanisms. The suggested system will lead to longer lasting vehicles and to a better acceptance of fuel cell vehicles. To incorporate a drive data pool, this paper describes the development of onboard micro trips by evaluating the driving information and splitting them into sub-trips.
In a group of individuals aiming for a collective decision, that decision is influenced by individual competences, but these are in turn subject to a reciprocal influence. Such an influence may lead to less competent agents exerting an unduly influence, a phenomenon known as equality bias. In this paper an agent-based model is proposed to investigate the evolution of competences under such a reciprocal influence. Through MonteCarlo simulation it is shown that: a) the average competence at steady state diminishes as the degree of interaction among the agents and/or the number of agents grow, both in the case of just positive influence (more competent agents increasing the competence of less competent one) and when both positive and negative influences are present; b) the convergence towards a steady state value is slow and characterized by oscillations.
Automated classification algorithms for satellite imageries require spectral correction from terrain effects due to shading. Such terrain effects can produce reflectance bias of pixels in the same category. This study was aimed at exploring robust algorithms for correcting satellite imageries from terrain effects, applicable for either Landsat 8 or Sentinel-2A imageries. Mount Halimun-Salak and Mount Gede-Pangrango, Bogor, West Java, Indonesia, were selected as the window areas to evaluate the algorithm. We developed algorithm, which combined solar position modelling, illumination modelling, and simple statistical model to remove the terrain effects. The algorithm was proven to be able to solve over correction problems and result relatively consistent sensitivities in SWIR, NIR and blue bands of either Landsat 8 or Sentinel-2A imageries in both window areas from different acquisition dates and times.
The multiplier undoubtedly is one of the most critical digital logic components in computer architecture. To achieve a faster response from a system, digital logic components need to respond faster with negligible error rates. Important factors in the consideration for implementing high performance/ speed multipliers are reduction in time delay, power at maximum speed and power delay product. This paper implements and compares high performance multipliers using various algorithms and techniques. It also analyzes the performance of 2, 4 & 8-bit multipliers based on Vedic Algorithm and 4 & 8-bit multipliers based on the Modified Booth Algorithm, and Wallace Tree. The conclusion on the most preferred choice of algorithm across the product dimension is made based on the maximum delay. The multiplier algorithms minimize the delay by reducing the number of partial products. Implementation and analysis of the results have been carried out using Verilog on Xilinx Vivado IDE.
In this work we present a software named Neuron Analyzer and Simulator (NAS). There are some other programs in the field of computational neuroscience which are widely used and had and still have a huge scientific importance. Nevertheless, we feel that many newcomers to the field would benefit from a simpler visual user-friendly software for modeling and simulating realistic reconstructed neurons. Therefore, this work represents an effort to bring neuronal computer simulation to prospective researchers in the field.
In the article there are substantiation of architectural and technical solutions, with the basis of the universal CASE-tool for describing ("programming") the behavior of mobile robots. The development tool intended for carrying out experiments in the field of artificial intelligence and it is based on multi-agent technology. In addition, the toolkit will be the maximum possible reuse of elements (tasks, processes, etc.). The basis for the development is the idea of combining, within the framework of one tool, both the real execution of the algorithm by the robot, and its simulation. It allows talking about testing partially implemented hardware (sensors and actuators). Development is carried out based on open source technology; all texts of programs are available at web-source: https://github.com/unclesal/tenguai.
This paper is about the assessment of voice quality as required routinely in hospital voice clinics. It describes a computer application capable of analysing recordings of a patient's voice and producing quantitative assessments of its quality, simulating those traditionally made by trained speech and language therapists (SLTs). Adopting a machine learning approach based on a database of recordings and assessments by a team of SLTs required measurements of consistency to be taken into account. The means of doing this, details of the machine learning approaches and the performance of the resulting algorithms are presented.
This paper presents the empirical comparison of boosting implementation by reweighting and resampling methods. The goal of this paper is to determine which of the two methods performs better. In the study, we used four algorithms namely: Decision Stump, Neural Network, Random Forest and Support Vector Machine as base classifiers and AdaBoost as a technique to develop various ensemble models. We applied 10-fold cross validation method in measuring and evaluating the performance metrics of the models. The results show that in both methods the average of the correctly classified and incorrectly classified are relatively the same. However, average values of the RMSE in both methods are insignificantly different. The results further show that the two methods are independent of the datasets and the base classier used. Additionally, we found that the complexity of the chosen ensemble technique and boosting method does not necessarily lead to better performance.
The aim of this research is to investigate the effectiveness between Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) along with neural network (NN) in classifying the gait of autistic children as compared to control group. Twelve autistic children and thirty two normal children participated in this study. Firstly the walking gait of these two groups are acquired using VICON Motion Analysis System to extract the three dimensional (3D) gait features that comprised of 21 gait features namely five features from basic temporal spatial, five features represented the kinetic parameters and twelve features from kinematic. Further, PCA and LDA are utilized as feature extraction in determining the significant features among these gait features.With NN as classifier, results showed that LDA as feature extraction outperform PCA for classification of autism versus normal children namely kinematic gait patterns attained 98.44% accuracy followed by basic temporal spatial gait features with accuracy of 87.5%.
This paper considers the following methods of the work scheduling: network planning techniques (critical path method, program evaluation and review technique, and graphical evaluation and review technique), method of agents cooperation in the needs-and-means networks proposed by Skobelev P.O., method of simulation and genetic algorithms integration proposed by Kureichik V.V., and method of multiagent genetic optimisation developed by the authors based on the Kureichik method. As a result of the comparative analysis, the advantages of the method of multiagent genetic optimisation in terms of solving the problem of subcontracting scheduling have been revealed. The multiagent genetic optimisation method takes into account the non-renewable resources, allows implementing different resource allocation strategies using simulation and multiagent modeling, and allows optimising subcontract resources via analysis of alternative work schedules using genetic algorithms and simulation.
This paper presents preliminary work from a current study on large refrigeration pack network. In particular, the simulation model of a typical refrigeration system with a single pack of 6 compressor units operating as fixed volume displacement machines is presented, and the potential of delivering static FFR with a large population of such packs is studied. Tuning of the model is performed using experimental data collected at the Refrigeration Research Centre in Riseholme, Lincoln. The purpose of modelling is to monitor the essential dynamics of what resembles a typical supermarket convenience-type store and to measure the capacity of a massive refrigeration network to hold off a considerable amount of load in response to FFR DSR event. This study focuses on investigation of the aggregated response of 150 packs (approx. 1 MW capacity) with refrigeration cases on hysteresis and modulation control. The presented model captures interconnected dynamics (refrigerant flow in the system linked to temperature control and the system's refrigerant demand and to compressors' power consumption). Type of refrigerant used for simulation is R407F. Refrigerant properties such as specific enthalpy, pressure and temperature at different state points are computed on each time step of simulation with REFPROP.
Expert system DNA analysis is not a new application for the medical world, this expert system has long been used as a tool for analyzing the structure of DNA. The function of this expert system is to assist medical personnel of analyzing patterns of DNA sequence trends, isolate grouping, tree molecular clock isolate, and detection of the disease from the isolate. So far, the existing expert system is able to process the isolate data onto the FASTA and create a closeness tree from among isolating. Based on the scheme tree an isolated from the primary can be known ancestors of a virus or disease. So it can be known pattern of DNA virus mutations or the disease from year after year. DNA analysis applications can be web-based or only accessible online, some are open source, as well as local software whose ownership is paid. Level maturity of various expert system is also diverse, there are able to generate FASTA independently, there is also a need to collaborate with other expert systems. But whether the current expert system is able to accommodate the needs of today's medical world? How is the expert system model appropriate for future needs? This study examines how current expert DNA analysis systems work as well as the extent to which they are processed. The results of the review can formulate a recommendation of the system and service features as the basis of software improvement for the future of DNA expert system in Indonesia.
Spatial and temporal isolation is a key issue in embedded systems running multiple tasks with different criticality levels (i.e. mixed-criticality embedded systems). This is particularly relevant for embedded systems based on multi core/processor (i.e. parallel) architectures. In such a context, this work focuses on Network on Chip (NoC) architectures and proposes a hardware mechanism designed to be introduced directly into network interfaces. Such a mechanism supports the isolation of multiple tasks with different criticality levels by controlling the exchange of messages while introducing very limited overhead on the monitored NoC. Then, the proposed mechanism is analyzed by means of two different simulations approaches for behavior validation, feasibility check, and scalability evaluation. The overall results show that the proposed mechanism is suitable to support mixed-criticality parallel embedded systems based on NoC architectures.
Global warming, the decline of oil resources as well as the introduction of emission regulations have led to a research focus in new drive technologies. Within this group of alternative drive technologies, fuel cell hybrid electric vehicles (FHEV) are considered to be especially effective. Nevertheless, in order to achieve an efficient operation, an energy management system (EMS) is required. Since system efficiency as well as the operation characteristics is determined by the chosen EMS scheme, current research focuses on new EMS approaches. This paper reviews and evaluate three widely accepted state of the art EMS schemes: classical proportional-integral control, state machine approach and online optimization based equivalent consumption minimization strategy (ECMS). The evaluation is done based on an a physical model of a FHEV. Since, use cases of the vehicle also have significant influence, real word driving was used to generate test cases. Thus, a method to cluster recorded driving data with regards to drive scenarios is proposed. Finally, the extracted reference cycles in combination with the physical model form a virtual test bench, used to evaluate the three EMS approaches under test.
Data dissemination is the core of the Vehicular Ad-hoc Networks(VANETs). Current media through which information is disseminated in the vehicular environment pose number of challenges. Some of these challenges include bandwidth and data latency. In the light of these challenges, this paper presents a review on current data dissemination technologies and their associated problems. The paper goes further to propose the adoption of Light Fidelity(LiFi), which is an emerging data dissemination technology, as an alternative medium through which data can be disseminated in the vehicular environment.
The paper investigates the use of a space deformation technique for the 3D manipulation of textile antenna deformations. The results obtained using Green Coordinates for space deformations are compared against published results for antennas bent over a cylindrical surface. The method is also applied to model crumpling and twisting of textile antennas.
The paper focuses on the problem of the logistical department of the hardware stores chain that is related to the delivery route planning, effective load of transportation vehicles, and decreasing of the idle time while loading at the warehouse. Thus, an actual problem is identification and application of the new principles for building and analysis of the multi-agent models of the resource conversion processes. The method for multi-agent delivery planning and the vehicle loading is based on the dynamic programming.
