Association for Scientic Computing Electronics and Engineering (ASCEE): Open Journal Systems
Not a member yet
785 research outputs found
Sort by
Deep Learning-Based Automated Approach for Classifying Bacterial Images
Get PDFIdentifying and classifying bacterial species from microscopic images is crucial for medical applications like prevention, diagnosis, and treatment. However, because of their diversity and variability in appearance, manually classifying bacteria is difficult and time-consuming. This work suggests employing deep learning architecture to automatically categorize bacterial species in order to overcome these difficulties and raise the accuracy of bacterial species recognition. We have evaluated our suggested approach using the Digital Images of Bacteria Species (DIBaS), a publicly accessible resource of photographs of tiny bacteria. This work uses a dataset that differs in terms of bacterial morphology, staining methods, and imaging circumstances. This paper aims to enhance the accuracy and reduce the computational requirements for Convolutional Neural Networks (CNN) based classification of bacterial species using GoogLeNet and AlexNet to train the models. This paper focuses on employing transfer learning to retrain pre-trained CNN models using a dataset consisting of 2000 images encompassing 12 distinct bacteria species known to be harmful to human health. The concept of transfer learning was utilized to expedite the network's training process and enhance its categorization performance. The results are promising, with the method achieving an accuracy of 98.7% precision, recall of 99.50%, and an F1-score of 99.45%  with classifier speed. Furthermore, the proposed bacteria classification approach demonstrated strong performance, irrespective of the size of the training data used. This paper contributes by automating bacterial classification to facilitate faster and more accurate identification of bacterial species, which facilitates the treatment of infections and related diseases, in addition to monitoring public health, and promoting the wise use of antimicrobial drugs. To improve outcomes in the future, researchers can also integrate deep learning techniques with other machine learning methods
Comparative Analysis of Sensor Fusion for Angle Estimation Using Kalman and Complementary Filters
Get PDFIn engineering, especially for robots, navigation, and biomedical uses, accurate angle estimation is absolutely crucial. Using data from the IMU6050 sensor, which combines accelerometer and gyroscope readings, this work contrasts two sensor fusion methods: the Kalman filter and the complementary filter. The aim of the research is to find the most efficient filtering method for preserving accuracy and resilience throughout several motion contexts, including low-noise (standard rotation) and high-noise (external disturbances). With an eye toward improving sensor accuracy in dynamic applications, the study contribution is a thorough investigation of filter performance under different noise levels. MATLAB quantified estimate accuracy using key metrics like root mean square error (RMSE) and mean absolute error (MAE). Under controlled noise levels, our approach included methodical error analysis of both filters. Results show that, especially under low-noise conditions, the Kalman filter beats the complementary filter in terms of lower MAE and RMSE; it also shows adaptability and robustness in high-noise environments with much fewer errors than accelerometer-only and complementary filter outputs. These results show the relevance of the Kalman filter in practical settings like robotic control, motion tracking, and possible biomedical equipment, including patient positioning systems and wheelchairs with balance control. Future studies might investigate the implementation of the Kalman filter in sophisticated systems requiring accuracy, such as telemedicine robots or autonomous navigation. This work develops sensor fusion techniques and offers understanding of consistent sensor data processing in several operating environments
Comparison of Proportional Integral Derivative and Fuzzy Logic Controllers: A Literature Review on the Best Method for Controlling Direct Current Motor Speed
Get PDFControl systems, particularly for DC motors, are a continually evolving field with various methods and techniques aimed at improving control system performance. Common issues in DC motor control, such as high overshoot and inadequate response times, highlight the need for further research into more effective tuning techniques. This study compares conventional PID and FLC methods in controlling DC motor speed, while also exploring optimization potential through new approaches like hybrid methods and the use of neural networks. The contributions of this research include a comprehensive analysis of previous studies on DC motor control performance and an in-depth assessment of the effectiveness of PID and FLC methods in addressing rise time, settling time, and overshoot issues. The methodology used in this study is a literature review, which involves collecting and analyzing various studies related to the application of both methods in DC motor control. Literature selection criteria include relevance, methodology used, and contributions to scientific advancements in motor control. The analysis shows that FLC performs better in handling overshoot, with previous studies indicating its ability to completely eliminate overshoot. Although the PID method is simpler and easier to apply in systems with linear characteristics, FLC offers better flexibility and adaptability for managing uncertainty and non-linear systems. Recommendations for further research are also presented, including a deeper exploration of integrating the two methods in a hybrid control system to enhance motor control performance
Systematic Review of Unmanned Aerial Vehicles Control: Challenges, Solutions, and Meta-Heuristic Optimization
Get PDFUnmanned Aerial Vehicles (UAVs) are powerful tools with vast potential, yet they face significant challenges. One of the primary issues is flight endurance, limited by current battery technology. Researchers are exploring alternative power sources, including hybrid systems and internal combustion engines, and considering docking stations for battery exchange or recharging. Beyond endurance, UAVs must address safety, efficient path planning, payload capacity balancing, and flight autonomy. The complexity increases when considering swarming behaviour, collision avoidance, and communication protocols. Despite these challenges, research continues to unlock UAVs’ potential, with path planning optimization significantly advanced by meta-heuristic algorithms like the Cuckoo Optimization Algorithm (COA). Whereas, meta-heuristic algorithms can be defined as system-level strategies that are used to seek suboptimal solutions to optimization problems. It uses heuristic approaches together with the exploration/exploitation scheme in order to effectively employ within large solution spaces. However, dynamic environments still present difficulties. UAVs have evolved beyond recreational use, becoming essential in industries like agriculture, delivery services, surveillance, and disaster relief. By resolving issues related to autonomy, battery longevity, and security, the benefits of UAV technology can be fully optimized. This systematic review emphasizes the importance of continuous innovation in UAV research to overcome these challenges
Adaptive Fuzzy Logic Control of Quadrotor
Get PDFIntelligent controllers are created in this work to regulate the attitude of quadrotor UAVs (Unmanned Aerial Vehicles). Quadrotors offer a wide range of real-time applications, including surveillance, inspection, search and rescue, and lowering the human force safety risks. The kinematics of quadrotor are similar to those of an inverted pendulum. To maintain balance, they must continuously adjust orientation and thrust. External disturbances, like wind or sudden movements, can easily destabilize them, necessitating sophisticated control algorithms for stable flight and precise maneuverability. This instability poses a significant challenge in designing and operating quadrotors, especially in dynamic environments where real-time adjustments are crucial for maintaining control. To avoid any form of damage, a mathematical model should be constructed first, followed by the implementation of various control systems. A thorough simulation model for a Quadrotor is presented in this project. The quadrotor is a six degrees of freedom object, it has six variables to express its position in space where (x, y and z) represent the distance of quadrotor from an earth fixed inertial form to its center of mass, main movements of roll, pitch, yaw are the Euler angles representing the orientation of the quadrotor at each axis. The proposed control techniques are applied separately: PID Controller, Fuzzy Logic PID Controller and Adaptive Fuzzy Logic PID Controller. The purpose of this work is to asses these control techniques for the motions of a Quadrotor in terms of better performance, tracking error reduction, and stability. MATLAB software is used for modeling, control, and simulation. According to the obtained results, the PID controller provided the best settling time. In addition, when we applied fuzzy logic PID control to adjust the pitch angle, the system experienced overshoot; however, with Adaptive Fuzzy Logic PID controller, the system provided the best performance according to the desired criteria
Aesthetic hegemony of the new order: A critical review of cultural politics in Indonesia (1966-1998)
Get PDFThis research studies the aesthetic hegemony of the New Order in Indonesia. Thus far, the understanding has been that aesthetic matters are not related to the capital and feudal assets that construct them. To the contrary, this research analyzes how these two assets played a strategic role in establishing the aesthetic hegemony of the New Order. The research is conducted in the qualitative tradition, through a content analysis with a critical paradigm perspective. The data include documents in the form of literature, government policies, and the results of interviews. The results show that the United States of America, through the Ford Foundation, played an active role and contributed significantly to the establishment of aesthetic hegemony in the New Order. The United States was an agent that provided scholarships for art academics, young scientists, and artists to study art and philosophy in America and Europe. The style of abstract expressionism and the philosophy of absurdism were developed in Indonesia as a result of the active role of the United States of America. On the other hand, the New Order also built cultural centers on a central and regional level, including cultural parks and arts councils that were used as tools of cooptation. Through the function of “mentoring and developmentâ€, the role of cultural parks was to control the aesthetic activities of Indonesian citizens to suit the tastes of the ruler. The research results show that the aesthetic hegemony of the New Order had a double face, one with a character of modernism based on Western aesthetics, and the other with a noble adiluhung character of traditionalism based on traditional art
Simulation and Modeling with Designing for the Proportional, Integral and Derivative Control of Industrial Robotic Arm by Using MATLAB/Simulink
Get PDFThis study aims to develop a control system for a robot arm, designed to perform precise movements along a predefined path, suitable for various industrial applications. The robot arm's movements are driven by three electric motors, each responsible for controlling a joint, enabling the arm to follow the required path accurately. To manage the complexity of multiple motors and dynamic movement requirements, an automated control system has been developed, tailored to meet the specific demands of the proposed task. A highly efficient, reliable, and safe control system design is being developed and simulated to evaluate its effectiveness in executing the required path. A simulation model is being constructed to assess the system's ability to follow the prescribed path, its responsiveness to disturbances and transient conditions, and the overall accuracy of the arm's movements. Simulation results will be analyzed to determine the system's performance across various scenarios, evaluating its adaptability to the work environment and its ability to achieve tasks with high accuracy, thereby enhancing system effectiveness
Enhancing the performance of heart arrhythmia prediction model using Convolutional Neural Network based architectures
Get PDFHeart disease is one of the diseases that exposes high mortality worldwide. This conventional way of predicting heart disease is usually expensive, time-consuming, and prone to human error. Early detection of heart disease is important as it helps to prevent deaths caused by this disease. Machine learning utilization as the non-invasive means for predicting heart disease is considered as a fast and affordable method to prevent the fatality of heart disease. This work aims at utilizing  Convolutional neural network (CNN)  to enhance the performance of an Arrhythmia prediction model. We have built an Arrythmia prediction model using neural networks comprising multiple convolutional layers and maxpooling layers. Our proposed model is trained using the MIT-BIH Arrhythmia dataset. The model performance has been evaluated and the model achieves 98.43% of performance accurac
Digitalization and down streaming of Sumatra batik motifs as symbols of cultural heritage: turtle graphics as a tool for sustainable development
Get PDFThe importance of preserving Sumatra batik motifs as a cultural heritage is facing major challenges in the digital era. The problem of this research lies in the difficulties in digitalizing and commercializing batik motifs effectively and maintaining authenticity and cultural value. This article aims to explore the use of the Turtle Graphic as a tool for digitalizing and down-streaming Sumatra batik motifs. Turtle graphics were initially crafted as an instructive instrument for educational settings, paving the way for teachers to demonstrate drawing concepts. On a different note, it also serves convenience to coders needing graphic output - offering them a solution without necessitating more complicated or external libraries in their projects. The research method involves applying graphical programming (using Turtle graphics) to produce digital representations of batik motifs. The technical aspects of digitization implemented in the Turtle Graphic are (1) data identification, which is done by creating a table to select motifs that are currently rare; (2) motifs that have been identified are revitalized through making simple sketches; (3) sketches are made in digital Turtle Graphics to make it easier for MSMEs to document and implement into prototype batik models. One example of the batik motif used in this research is the Daun Sirih Melayu batik motif. This batik motif often uses images of betel leaves that are intertwined and sticking out. The Daun Sirih Melayu batik motif was chosen as a representative example because the philosophical meaning of the betel leaf batik motif symbolizes a form of respect. Betel leaves are also important in tradition, especially at proposals, weddings, traditional title awards, and other events. The research results show that Turtle Graphics can simplify digitalization. These contribute to the preservation and socialization of Sumatra batik and support the sustainability and popularization of batik in an increasingly digital global context. Academics can also use this research as a basis for further research, while for MSMEs, it is a good opportunity to implement it in batik centers in the archipelago, especially on Sumatra island
Development of a Testbed for Autonomous Navigation of an Off-Shelf Quadrotor Based on Ultra-Wide-Band Real-Time Localization
Get PDFRecent advances in autonomous aerial vehicle research, from theoretical simulations to experimental validations, has triggered demand for reliable proof-of-concept test-beds. Although such test-beds have been developed in some advanced drone research laboratories, their cost, expertise and complexity place them out of reach for upcoming research teams. This raises the need for development of less complex and affordable testbeds for quadrotor research. The contribution of this research is provision of low-cost autonomous quadrotor test-bed for proof-of-concept. The development of the proposed testbed entails configuration of Ultra-Wide-Band (UWB) based Real-Time Localization System (RTLS) to transmit position data of multiple agents to LabVIEW software for analysis and decision making. The autonomous navigation commands for the quadrotor are generated from the LabVIEW software and relayed through customized USB interface to the flight control module. The commands alter the digital state of Arduino board pins which are connected to the flight controller hence manipulating navigation pitch and roll parameters. The validation tests performed in the test-bed involved quadrotor hover and navigation in pursuit of the ground agent. The results demonstrate that UWB based RTLS achieves high precision of 99% when the modules are stationary but the precision reduced to 90% when the modules were in motion, which may be attributed actuating signal transmission delays. The results also showed that the Arduino based electronic flight controller is capable of generating flight paths to follow the ground robot in real-time with precision deviations of under 10% which is at par with other research test beds. This novel testbed provides a costeffective and accurate solution for autonomous flight testing, with precision comparable to visual-based testbeds, but at a much lower cost. Further research is encouraged to explore how the system performs with more than two agents and on a wider test arena