Indonesian Journal of Electrical Engineering and Informatics (IJEEI)
Not a member yet
776 research outputs found
Sort by
Hand Gestures Replicating Robot Arm based on MediaPipe
Get PDFA robotic arm is any variety of programmable mechanical devices designed to operate items like a human arm and is one of the most beneficial innovations of the 20th century, quickly becoming a cornerstone of many industries. It can perform a variety of tasks and duties that may be time-consuming, difficult, or dangerous to humans. The gesture-based control interface offers many opportunities for more natural, configurable, and easy human-machine interaction. It can expand the capabilities of the GUI and command line interfaces that we use today with the mouse and keyboard. This work proposed changing the concept of remote controls for operating a hand-operated robotic arm to get rid of buttons and joysticks by replacing them with a more intuitive approach to controlling a robotic arm via the hand gestures of the user. The proposed system performs vision-based hand gesture recognition and a robot arm that can replicate the user's hand gestures using image processing. The system detects and recognizes hand gestures using Python and sends a command to the microcontroller which is the Arduino board connected to the robot arm to replicate the recognized gesture. Five servo motors are connected to the Arduino Nano to control the fingers of the robot arm; These servos are related to the robot arm prototype. It is worth noting that this system was able to repeat the user's hand gestures with an accuracy of up to 96%
Forecasting Carbon Dioxide Emission in Thailand Using Machine Learning Techniques
Get PDFMachine Learning (ML) models and the massive quantity of data accessible provide useful tools for analyzing the advancement of climate change trends and identifying major contributors. Random Forest (RF), Gradient Boosting Regression (GBR), XGBoost (XGB), Support Vector Machines (SVC), Decision Trees (DT), K-Nearest Neighbors (KNN), Principal Component Analysis (PCA), ensemble methods, and Genetic Algorithms (GA) are used in this study to predict CO2 emissions in Thailand. A variety of evaluation criteria are used to determine how well these models work, including R-squared (R2), mean absolute error (MAE), root mean squared error (RMSE), mean absolute percentage error (MAPE), and correctness. The results show that the RF and XGB algorithms function exceptionally well, with high R-squared values and low error rates. KNN, PCA, ensemble methods, and GA, on the other hand, outperform the top-performing models. Their lower R-squared values and higher error scores indicate that they are unable to accurately anticipate CO2 emissions. This paper contributes to the field of environmental modeling by comparing the effectiveness of various machine learning approaches in forecasting CO2 emissions. The findings can assist Thailand in promoting sustainable development and developing policies that are consistent with worldwide efforts to combat climate change
Protecting Attacks on Unmanned Aerial Vehicles using Homomorphic Encryption
Get PDFWith the exponential growth in the usage of unmanned aerial vehicles (UAV), often known as drones, for military, civilian, and recreational purposes. Security of internal communication modules and communication to the ground control station is considered the foremost challenge. Hacking into the system and attacking the internal communication devices with malicious code can disaster the vehicle's system. The need for having a secure communication channel between the internal modules of the vehicle and transmission of data to the ground control station is of utmost crucial. Existing mechanisms based on conventional encryption methods are highly suspectable to attacks as their keys can be broken by employing high computing power. Another challenge with these approaches is undesired high-level data communication latency affecting real-time communication. This study implements a homographic encryption-based technique for secure communication. In addition, we also propose a key regeneration algorithm based on pallier homomorphic encryption. Simulations were conducted using OMNET++ and Aerial Vehicle Network Simulator (AVENS). In this study 54 encryption attacks were collected from different sources. Compared to Digital Encryption Standard (DES) and Advanced Digital Encryption (AES), the proposed approach defended all the communication attacks between the UAV and the ground control station
Soft Robots: Implementation, Modeling, and Methods of control
Get PDFSoft robotics is a branch of robotics that focuses on technologies with physical features that are like those of live biological creatures. Additionally, they have many details that are hard, if not impossible, to realize with traditional robots which are composed of solid materials. This study concentrates on the current expansion of soft pneumatic actuators for modern soft robotics in recent years, with an emphasis on three areas: Implementation of soft robots, Modeling, and Methods of control systems. Therefore, numerous soft robotic designs and ways to make them suitable for medical, manufacturing, and agricultural applications have been presented. Moreover, a set of functional and technological aspects have been given to review models similar to human hand functionality and motions. To realize the advanced soft robotic hand manipulation function, robotic hands must be equipped with tactile sensing, that sensing is required to provide continuous data on the volume and direction of forces at all contact locations. The research examines achievements in material science, actuation, sensing techniques, and manufacturing technologies, as well as how to model and control a soft robot's motion, all of which are scientifically challenging and, more importantly, practical
Pectoral Muscle Removal in Digital Mammograms Using Region Based Standard Otsu Technique
Get PDFMammography is usually the first preference of imaging diagnostic modalities used for detection of breast cancer in the early stage. Two projections Cranio Caudal (CC) and Medio-Lateral Oblique (MLO) which depict different degrees for visualizing the breast are used during digital mammogram acquisition and the MLO view shows more breast tissue and Pectoral Muscle (PM) area when compared to CC view. Although, the PM is a criterion used to show proper positioning, it can result in biased results of mammographic analysis like: cancer detection and breast tissue density estimation, because the PM area has similar or even higher intensity than breast tissue and breast lesions if present. This paper proposed a Region Based Standard Otsu thresholding method for the elimination of PM area present in MLO mammograms. The proposed algorithm was implemented using 322 digital mammograms from the Mammographic Image Analysis Society (MIAS) database, and the difference between the PM detected and the manually drawn PM region by an expert was evaluated. The results showed an average: Jaccard Similarity Index, False Positive Rate (FPR) and False Negative Rate (FNR) of 93.2%, 3.54% and 5.68% respectively and also an acceptable rate of 95.65
ACRMiner: An Incremental Approach for Finding Dense and Sparse Rectangular Regions from a 2D Interval Dataset
Get PDFIn many applications, transactions are associated with intervals related to time, temperature, humidity or other similar measures. The term "2D interval data" or "rectangle data" is used when there are two connected intervals with each transaction. Two connected intervals give rise to a rectangle. The rectangles may overlap producing regions with different density values. The density value or support of a region is the number of rectangles that contain it. A region is closed if its density is strictly bigger than any region properly containing it. For rectangle dataset, these regions are rectangular in shape.In this paper an algorithm named ACRMiner has been proposed that takes as input a sequence of rectangles and computes all closed overlapping rectangles and their density values. The algorithm is incremental and thus is suitable for dynamic environment. Depending on an input threshold the regions can be classified as dense and sparse.Here a tree-based data structure named as ACR-Tree is used. The method has been implemented and tested on synthetic and real-life datasets and results have been reported. Few applications of this algorithm have been discussed. The worst-case time complexity the algorithmis O(n5) where n is the number of input rectangles
Energy Management Analysis of Residential Building Using ANN Techniques
Get PDF The process of limiting the amount of energy that is utilized is known as energy conservation. This can be accomplished by making more effective use of the energy that is available. As a result, there is a requirement for more effective management of the consumption of energy in buildings. It is essential to have an accurate load calculation for a residential building because the loads for heating and cooling add up a significant portion of the total building loads. In this study, the load analysis of the HVAC (Heating, Ventilation, and Air Conditioning) system in a residential building was carried out by taking into consideration three different neural networks. These networks are known as the feed forward network, the cascaded forward back propagation network, and the Elman back propagation network. During the process of conducting a load study of the heating and cooling loads on an HVAC system, performance measurements like MAE (mean absolute error), MSE (mean square error), MRE (mean relative error), and MAPE (mean absolute percentage error) are taken into consideration. It has been discovered that the cascaded forward back propagation method is the most effective method, with MAE, MSE, MRE, and MAPE values of 0.08, 0.0336, 0.0051, and 0.51% respectively for heating load and MAE, MSE, MRE, and MAPE values of 0.0975, 0.0406, 0.0053, and 0.53% respectively for cooling load
Optimizing U-Net Architecture with Feed-Forward Neural Networks for Precise Cobb Angle Prediction in Scoliosis Diagnosis
Get PDFIn the burgeoning field of Artificial Intelligence (AI) and its notable subsets, such as Deep Learning (DL), there is evidence of its transformative impact in assisting clinicians, particularly in diagnosing scoliosis. AI is unrivaled for its speed and precision in analyzing medical images, including X-rays and computed tomography (CT) scans. However, the path does not lack obstacles. Biases, unanticipated outcomes, and false positive and negative predictions present significant challenges. Our research employed three complex experimental sets, each focusing on adapting the U-Net architecture. Through a nuanced combination of feed-forward neural network (FFNN) configurations and hyperparameters, we endeavored to determine the most effective nonlinear regression model configuration for predicting the Cobb angle. This was done with the dual purpose of reducing AI training time without sacrificing predictive accuracy. Utilizing the capabilities of the PyTorch framework, we meticulously crafted and refined the deep learning models for each of the three experiments, focusing on an FFFN dropout rate of p=0.45. The Root Mean Square Error (RMSE), the number of epochs, and the number of nodes spanning three hidden layers in each FFFN were utilized as crucial performance metrics while a base learning rate of 0.001 was maintained. Notably, during the optimization phase, one of the experiments incorporated a learning rate scheduler to protect against potential pitfalls such as local minima and saddle points. A judiciously incorporated Early Stopping technique, triggered between the patience range of 5-10 epochs, ensured model stability as the Mean Squared Error (MSE) plateau loss approached approximately 1. Consequently, the model converged between 50 and 82 epochs. We hypothesize that our proposed architecture holds promise for future refinements, conditioned on assiduous experimentation with an array of medical deep learning paradigms
Compact Dual-Frequency Slot Antenna for C-Band Applications Based on Substrate Integrate Waveguide
Get PDFThis paper provides a compact substrate-integrated waveguide-based dual-frequency H-fractal slot antenna for the C-band (SIW). From the view of fractal slot antenna, two H-fractal slot shaped elements with 3rd iteration are used to cut currents in two TE modes, which leads to dual-frequency performance and reduces the size. The antenna operating at dual frequency of 4 GHz and 5.7 GHz with gain greater than 5 dB is designed and fabricated. Measured and simulated response of the antenna are introduced as well. The responses showed that the proposed antenna achieved stable dual-frequency performance with total size of 23 × 11 mm², which may be applied for C-band communication systems. The proposed antenna was simulated, analysed, and optimized using computer simulation technology (CST) software
Autoencoder-Based Representational Learning for the Determination of Corrosion Severity
Get PDFAutomatic determination of corrosion severity is an important task that has not received adequate study due to the non-availability of datasets. This study explores corrosion severity detection by leveraging representation learning for the development of lightweight, shallow machine learning models. A variational autoencoder was used for feature extraction. Four classifiers, specifically RF, SVC, k-NN, and XGBoost, were trained using the encoded representation without any additional processing. Voting classifiers were also constructed using the trained models. Except for the slight drop in precision (6.06%) of the soft voting classifier, augmentation produces a general positive influence on the precision, recall, and F1-score of the models understudied; it improved the precision, recall, and F1-metrics of K-NN and hard-voting classifiers most remarkably. For K-NN, the improvement reached 29.17%, 20.93%, and 30.77% for precision, recall, and F1-metrics, respectively, and 30.36%, 26.42%, and 35.29%, respectively, for the hard voting classifier. Whereas our experiments could not produce state of-the art results, it provides adequate motivation to further study VAE as a data preprocessing unit for the development of simple, efficient, lightweight models that can be deployed on resource constrained devices, which will in turn advance the development and deployment of corrosion monitoring systems on low-cost devices