International Journal of Information Technology and Computer Science @ijitcs
Journal articles - International Journal of Information Technology and Computer Science
All articles: 1317
Real Time Implementation of Audio Source Localization on Blackfin ADSP-BF527 EZ-KIT
Scientific article
The present paper discusses the implementation of direction of arrival estimation using Incoherent Wideband Music algorithm. The direction of arrival of an audio source is estimated using two microphones plugged in “Line-in” input of a DSP development board. A solution to the problem of fluctuating estimation of the angle of arrival has been proposed. The solution consists on adding an audio activity detector before going on processing. Only voiced sound frames are considered as they fulfill the theoretical constraint of the used estimation technique. Furthermore, the latter operation is followed by integration over few frames. Only two sensors are used. For such a reduced number of sensors, the obtained results are promising.
Free
Real Time Object Tracking using FPGA Development Kit
Scientific article
The main idea of this work is object tracking using real time video processing. For this purpose we designed an embedded system that performs the object tracking algorithm for accurate tracking of defined object. The theme may be implemented for the security companies, sports and the armed forces to make them more equipped and advanced. The heart of the system is a Field Programmable Gate Arrays development kit. It controls the whole system by receiving the video signal from camera, processes it and sends the video signal to the Liquid Crystal Display or monitor. After receiving video of intended object, target selection is performed to select the target to track and then the tracking algorithm is implemented using image processing algorithms implemented using Field Programmable Gate Arrays development kit. We also interfaced the DC gear motor to control the movement of the camera in order to track the selected object. In order to design the standalone application we transformed our algorithm in Field Programmable Gate Array kit.
Free
Scientific article
In recent years, with an increasing number of requests, energy, power and temperature have been important keys in embedded systems, which decrease the lifetime of both CPUs and hard disks. The energy consumption is an important issue in computer systems, particularly real-time embedded systems. The frequency and the Revolutions Per Minute are major factors in the reduction of energy consumption in both processors and hard disk drives. Therefore, the main goal of this paper is to present a scheduling mechanism for a real time periodic task that can save more energy. This mechanism is based on increasing, as much as possible, the execution time of the CPU and/or the Read/Write time of the hard disk without passing the task deadline. This will be done by dynamically changing the CPU frequency and/or the RPM of hard disk. Our experimental results demonstrate that the proposed algorithm manages to lower energy consumption by an average of 25% and to reduce the number of missed tasks by 80%.
Free
Real-Time Air Pollutants Rendering based on Image Processing
Scientific article
This paper presents a new method for realistic real-time rendering of air pollutants based on image processing. The air pollutants’ variable density can create many shapes of mist what can add a realistic environment to virtual scene. In order to achieve a realistic effect, we further enhance thus obtained air pollution data getting from monitor in spatial domain. In the proposed method we map the densities of air pollutants to different gray levels, and visualize them by blending those gray levels with background images. The proposed method can also visualize large-scale air pollution data from different viewpoints in real-time and provide the resulting image with any resolution theoretically, which is very important and favorable for the Internet transmission.
Free
Real-Time Group Face-Detection for an Intelligent Class-Attendance System
Scientific article
The traditional manual attendance system wastes time over students’ responses, but it has worked well for small numbers of students. This research presents a real-time group face-detection system. This system will be used later for student class attendance through automatic student identification. The system architecture and its algorithm will be described in details. The algorithm for the system was based on analyzing facial properties and features in order to perform face detection for checking students’ attendance in real time. The classroom’s camera captures the students’ photo. Then, face detection will be implemented automatically to generate a list of detected student faces. Many experiments were adopted based on real time video captured using digital cameras. The experimental results showed that our approach of face detection offers real-time processing speed with good acceptable detection ratio equal to 94.73%.
Free
Real-Time Tree Counting Android Application and Central Monitoring System
Scientific article
In this study, a cloud-based android application and centralized tracking software were developed to perform an accurate and uninterrupted tree count across open lands. The application is used to count the desired number of trees and species at the same time. User-logged data and location information are saved in real-time to the application's cloud database. The application can work online and with offline mod. In cases where there is no internet connection, it inserts the data to the local SQLite database. After the connection is established, the pairing continues. It's used Google Firebase on the cloud server for data storage. The processing of target locations and GPS coordinates was developed with the Google Map Library. The tree counting application automatically picks up the user's current location when it is first opened. The counting starts after the tree and tree species that the user has selected from the menu. The software developed shows that tree counting is done simultaneously at the desired point. It also solves confusion caused by different tree species during the counting. We've received feedback from 100 people using the application. The users answered five questions. As a result, it is aimed to provide a comfortable transition between tree species and its users with its simple use to eliminate the complexity of counting and save time.
Free
Real-time Flame Rendering with GPU and CUDA
Scientific article
This paper proposes a method of flame simulation based on Lagrange process and chemical composition, which was non-grid and the problems associated with there grids were overcome. The turbulence movement of flame was described by Lagrange process and chemical composition was added into flame simulation which increased the authenticity of flame. For real-time applications, this paper simplified the EMST model. GPU-based particle system combined with OpenGL VBO and PBO unique technology was used to accelerate finally, the speed of vertex and pixel data interaction between CPU and GPU increased two orders of magnitude, frame rate of rendering increased by 30%, which achieved fast dynamic flame real-time simulation. For further real-time applications, this paper presented a strategy to implement flame simulation with CUDA on GPU, which achieved a speed up to 2.5 times the previous implementation.
Free
Real-time realistic telepresence using a 360° camera and a virtual reality box
Scientific article
This study focuses on the transmission of live 360 video using Ricoh Theta S which is transmitted to Android phone mounted in BoboVR. The proposed solution is to use a 360 camera that captures the scene in all direction and create an application which streams the captured 360 live video from Laptop (Server) into the Android Phone (Client). The Android phone’s IMU sensor is responsible for the corresponding viewport selected from the 360 environment. The viewport is subject for Stereoscopic SBS to show the 3D effect in accordance with the user’s perception The viewport is subject for Stereoscopic SBS to show the 3D effect in accordance with the user’s perception. A 3D video can be produced by applying the Stereogram SBS. Also, depth can be perceived due to the varying distance between the focal baseline and focal length.
Free
Recognition of Marrow Cell Images Based on Fuzzy Clustering
Scientific article
In order to explore the leukocyte distribution of human being to predict the recurrent leukemia, the mouse marrow cells are investigated to get the possible indication of the recurrence. This paper uses the C-mean fuzzy clustering recognition method to identify cells from sliced mouse marrow image. In our image processing, red cells, leukocytes, megakaryocyte, and cytoplasm can not be separated by their staining color, RGB combinations are used to classify the image into 8 sectors so that the searching area can be matched with these sectors. The gray value distribution and the texture patterns are used to construct membership function. Previous work on this project involves the recognition using pixel distribution and probability lays the background of data processing and preprocessing. Constraints based on size, pixel distribution, and grayscale pattern are used for the successful counting of individual cells. Tests show that this shape, pattern and color based method can reach satisfied counting under similar illumination condition.
Free
Recommendation of Move Method Refactoring to Optimize Modularization Using Conceptual Similarity
Scientific article
Placement of methods within classes is one of the most important design activities for any object oriented application to optimize software modularization. To enhance interactions among modularized components, recommendation of move method refactorings plays a significant role through grouping similar behaviors of methods. It is also used as a refactoring technique of feature envy code smell by placing methods into correct classes from incorrect ones. Due to this code smell and inefficient modularization, an application will be tightly coupled and loosely cohesive which reflect poor design. Hence development and maintenance effort, time and cost will be increased. Existing techniques deals with only non-static methods for refactoring the code smell and so are not generalized for all types of methods (static and non-static). This paper proposes an approach which recommends 'move method' refactoring to remove the code smell as well as enrich modularization. The approach is based on conceptual similarity (which can be referred as similar behavior of methods) between a source method and methods of target classes of an application. The conceptual similarity relies on both static and non-static entities (method calls and used attributes) which differ the paper from others. In addition, it compares the similarity of used entities by the source method with used entities by methods in probable target classes. The results of a preliminary empirical evaluation indicate that the proposed approach provides better results with average precision of 65% and recall of 63% after running it on five well-known open projects than JDeodorant tool (a popular eclipse plugin for refactorings).
Free
Regression Test Case Selection for Multi-Objective Optimization Using Metaheuristics
Scientific article
A new heuristic algorithm is proposed by this paper, on multi-objective optimization using metaheuristics and TSP (travelling salesman problems). Basic thinking behind this algorithm is minimizing the TSP path or tour by dividing the entire tour into blocks that are overlapped to each other and then improve each individual block separately. Although it is unproven that a good solution have small improvement chances if a node moved far way to a position compared to its original solution. By intensively searching each block, further improvement is possible in TSP path or tour that never be supported in various search methods and genetic algorithm. Proposed algorithm and computational experiment performance was tested, and these tests are carried out with support of already present instances of problem. According to the results represented by paper, the computation verifies that proposed algorithm can solve TSPs efficiently. Proposed algorithm is then used for selecting optimal test cases, thousands of those test cases which are selected after confirming that they identify bugs and they itself selected from a repository of test cases; these thousand test cases are those test cases which are selected from several thousand test cases because they detect bugs. Few test cases from repository act as milestones (nodes) and having certain weight associated with each, proposed algorithm based on TSP implemented over selected result and select the optimal result or path or solution. These selected optimal test cases or selected path are further used to perform the regression testing, by applying those test cases selected by proposed algorithm in order to remove most of the faults or bugs effectively, i.e. take less time and identify almost all the bugs with few test cases. Hence this proposed algorithm assures most effective solution for regression testing test case selection.
Free
Scientific article
Reinforcement Learning (RL) is a successful and established Artificial Intelligence (AI) method, particularly with recent groundbreaking progress in Deep Reinforcement Learning (DRL). Reinforcement learning is very well suited for sequential decision-making tasks, wherein a learned agent learns an optimal policy after many interactions with an environment. The present paper examines the application of reinforcement learning for automating screening of literature in academic research, particularly in the fields of computer science and e-learning. Keyword filtering techniques, while predominantly applied, are found to be inflexible as well as unable to capture the dynamic nature of research themes. To overcome such constraints, we present a Deep Q-Network (DQN)-based reinforcement learning model that combines reinforcement learning with the Semantic Scholar API to enhance research paper classification based on dynamically acquired decision rules. The proposed reinforcement learning model was trained and tested with a dataset of 8,934 research papers, accessed by systematic searching. The agent filtered and picked 11 effective papers depending on improved selection criteria like publication date, keyword relevance, and scholarly topic provided. The model iteratively optimizes the decision-making process through reward-based learning and therefore maximizes categorization accuracy over time. Test experiments demonstrate utilization of RL-based suggested framework yields classification accuracy at 91.5%, recall at 86.3%, and precision at 89.7%. A comparison test demonstrates that the approach performs 12.5% better on recall and 9.8% better on accuracy compared to traditional keyword-filtering methods. The finding confirms the power of the model in minimizing false positives and false negatives for screening literature, hence proving the scalability and adaptability of reinforcement learning in managing high academic data. This work offers a scalable, cognitive approach to conducting systematic reviews of literature through the application of reinforcement learning to programmatically execute work in academic research. The work shows the promise of reinforcement learning to further enhance research methodology, make literature reviews more effective, and facilitate more knowledgeable decision-making in fast-changing scientific disciplines. Further research will be focused on incorporating hybrid AI models with multi-agent systems of reinforcement learning for responsiveness and classification enhancements.
Free
Relative Performance of Certain Meta Heuristics on Vehicle Routing Problem with Time Windows
Scientific article
Solving Vehicle Routing Problem (VRP) and its variants arise in many real life distribution systems. Classical VRP can be described as the problem of finding minimum cost routes with identical vehicles having fixed capacity which starts from a depot and reaches a number of customers with known demands with the proviso that each route starts and ends at the depot and the demand of each customer does not exceed the vehicle capacity is met. One of the generalizations of standard VRP is Vehicle Routing Problem with Time Windows (VRPTW) with added complexity of serving every customer within a specified time window. Since VRPTW is a NP hard meta heuristics have often been designed for solving it. In this paper we compare the performance of Simulated Annealing (SA), genetic Algorithm (GA) and Ant Colony Optimization (ACO) for solving VRPTW based on their performance using different parameters taking total travel distance as the objective to be minimized. The results indicate that ACO is in general slightly more efficient then SA and GA.
Free
Relevant XML Documents - Approach Based on Vectors and Weight Calculation of Terms
Scientific article
Three classes of documents, based on their data, circulate in the web: Unstructured documents (.Doc, .html, .pdf ...), semi-structured documents (.xml, .Owl ...) and structured documents (Tables database for example). A semi-structured document is organized around predefined tags or defined by its author. However, many studies use a document classification by taking into account their textual content and underestimate their structure. We attempt in this paper to propose a representation of these semi-structured web documents based on weighted vectors allowing exploiting their content for a possible treatment. The weight of terms is calculated using: The normal frequency for a document, TF-IDF (Term Frequency - Inverse Document Frequency) and logic (Boolean) frequency for a set of documents. To assess and demonstrate the relevance of our proposed approach, we will realize several experiments on different corpus.
Free
Reliability Assessment for Open-Source Software Using Deterministic and Probabilistic Models
Scientific article
Nowadays, computer software plays a significant role in all fields of our life. Essentially open-source software provides economic benefits for software companies such that it allows building new software without the need to create it from scratch. Therefore, it is extremely used, and accordingly, open-source software’s quality is a critical issue and one of the top research directions in the literature. In the development cycles of the software, checking the software reliability is an important indicator to release software or not. The deterministic and probabilistic models are the two main categories of models used to assess software reliability. In this paper, we perform a comparative study between eight different software reliability models: two deterministic models, and six probabilistic models based on three different methodologies: perfect debugging, imperfect debugging, and Gompertz distribution. We evaluate the employed models using three versions of a standard open-source dataset which is GNU’s Not Unix Network Object Model Environment projects. We evaluate the employed models using four evaluation criteria: sum of square error, mean square error, R-square, and reliability. The experimental results showed that for the first version of the open-source dataset SRGM-4 based on imperfect debugging methodology achieved the best reliability result, and for the last two versions of the open-source dataset SRGM-6 based on Gompertz distribution methodology achieved the best reliability result in terms of sum of square error, mean square error, and R-square.
Free
Reliability Evaluation Metrics for Internet of Things, Car Tracking System: A Review
Scientific article
As technology continues to advance, the need to create benchmark or standards for systems becomes a necessity so as to ensure that these new advanced systems functions at its maximum capacity over a long period of time without any failure, fault or errors occurring. The internet of things technology promises a broad range of exciting products and services, with car tracking technology as part of the broad range of technological concept under the internet of things paradigm. The car tracking technology involves deploying some basic internet of things components into the tracking of important transportation component; the basic principle behind any technological concept involves delivery of high quality product that conforms to specifications. In this paper, the concept and technological description about the internet of things is discussed with emphasis on the principal functional component, this is to enable a broaden conceptualization about car tracking technology because it needs to function correctly, at all time. The concept of reliability engineering is also discussed in respect to an important quality factor, which entails that systems must function correctly without fault, failure or errors, it provides benchmark, principles, or standards in which the internet of things system must possess for an increased quality assurance.
Free
Removing Noise from Speech Signals Using Different Approaches of Artificial Neural Networks
Scientific article
In this research, four ANN models: Function Fitting (FitNet), Nonlinear AutoRegressive (NARX), Recurrent (RNNs), and Cascaded-ForwardNet were constructed and trained separately to become a filter to remove noise from any speech signal. Each model consists of input, hidden and output layers. Two neurons in the input layer that represent speech signal and its associated noise. The output layer includes one neuron that represent the enhanced signal after removing noise. The four models were trained separately on stereo (noisy and clean) audio signals to produce the clean signal. Experiments were conducted for each model separately with different: architecture; optimization training algorithms; and learning parameters to identify model with best results of removing noise from speech signal. From experiments, best results were obtained from FitNet and NARAX models respectively. TrainLM is the best training algorithm in this case. Finally, the results showed that the suggested architecture of the four models have filtering ability to remove noise form both trained and not trained speech signals samples.
Free
Representation of Dynamic Basic Block in Software Evolution Using Incidence Matrix
Scientific article
Software evolution is a continuous process that transforms changing user requirements into improved software systems. Establishing a clear and well-structured development process is widely recognized as an effective means to enhance software maintainability, quality, and productivity. Tailoring software processes from existing process patterns and standards is essential for improving process performance, ensuring product quality, reducing development risks, and minimizing rework. Despite its importance, current research lacks a systematic and formally grounded method for tailoring software evolution processes. In this paper, we propose a structured approach based on Petri Net (PN) theory to address this limitation. There are four fundamental process constructs: sequence, concurrency, selection, and iteration are identified as basic building blocks for modeling software evolution processes. Using these constructs, four tailoring operations, namely adding, deleting, splitting, and merging, are formally defined. We study on the scalable process composition, matrix-based representations of Petri Nets (PNs) are employed. Incidence and related matrices provide a concise and mathematically tractable representation of both place/transition nets and restricted PNs, enabling the identification of essential structural properties of software processes. Also, we prove the reachability analysis and firing rules are utilized to derive a mathematical behavioral notation that captures binary relationships between input and output variables. This notation facilitates precise analysis of dynamic behavior for systematic software process tailoring.
Free
Research and Compare Standards of E-Learning Management System: A Survey
Scientific article
Nowadays, using of e-learning has a special place in organizations and universities. Understanding the efficiency and effectiveness of this type of education, scientific and professional assemblies try to provide effective tools and strategies to operate this kind of training better. E-Learning management system as one of the basic requirements of the system plays a special role in this field. Therefore all companies are looking for a system that meets their needs in the field of e-learning. Standards of content and structure of e-learning must be set so that access to possibilities such as content reuse or gathering or discriminating subject from various sources at different times is possible. This paper reviews and compares some of the most important standards in the field of e-learning.
Free
Scientific article
Work-based learning is what equips students with practical skills. All higher learning institutions (HLIs) have a specified period of time for students to carry out field based practices in companies which are relevant to their fields of study. As the number of students in Tanzanian HLIs become larger, coordination and allocation of students to relevant companies is becoming tougher. This study therefore intended to examine a better method to facilitate coordination and allocation of students to relevant companies through development of an online computer system. The research study to determine systems’ requirements was conducted in Arusha and Kilimanjaro regions by involving 62 HLI students, 3 HLIs and 5 companies. Data were collected using key informant interviews, observation and workshop. Both informative and descriptive information regarding current practices and desired features were collected and analyzed. It was found that, a platform for registering students’ profiles and companies’ information has advantages to all three main stakeholders who are HLIs, students and companies. Prior to actual implementation, collaborative prototype was designed using pencil software and shared to 5 users from each group of stakeholders to evaluate the tasks. Responses from users were used to refine the requirements and design the final prototype. The final prototype design was used to develop a Field Attachment Management System (FAMS). FAMS indicated to have improved access of students to relevant companies, reports generation, students’ assessment and follow-up conducted by HLIs to their students.
Free
