Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
Not a member yet
1071 research outputs found
Sort by
Visual Impaired Assistance for Object and Distance Detection Using Convolutional Neural Networks
Get PDFVision is a very valuable gift from God; Most aspects of human needs in the body are dominated by vision. Based on data from the World Health Organization (WHO) there are around 180 million people in the world experiencing visual impairment, while the prevalence of blindness in Indonesia reaches 3 million people (1.5% of Indonesia's population), so we designed a system in the form of a prototype that could detect objects around the user and convey data in the form of sound to the user. This research discusses the application of a machine learning model using the Convolutional Neural Network method to detect objects optimally. The objects that have been collected will be trained on machine learning and produce a model to be embedded in the system's main machine, namely the Raspberry PI 4B. The training of the machine learning model was carried out several times by changing the compositions of several layers until a model with optimal accuracy was obtained; however, the size of the resulting model was quite large, so the researchers carried out SSDMobileNetV2 transfer learning to obtain the optimal model. The optimal model was obtained with a model precision of 92% and a model size of 18 MB. Object detection tests carried out under 3 test conditions resulted in an average object detection accuracy of 84.3%, and distance detection tests carried out under 10 conditions resulted in an average distance detection error of 2.1 cm. The results show that the system was accurate and effective
Digital Image Object Detection with GLCM Multi-Degrees and Ensemble Learning
Get PDFObject detection in digital images has been implemented in various fields. Object detection faces challenges, one of which is rotation problems, causing objects to become unknown. We need a method that can extract features that do not affect rotation and reliable ensemble-based classification. The proposal uses the GLCM-MD (Gray-Level Co-occurrence Matrix Multi-Degrees) extraction method with classification using K-Nearest Neighbours (K-NN) and Random Forest (RF) learning as well as Voting Ensemble (VE) from two single classifications. The main goal is to overcome the difficulty of detecting objects when the object experiences rotation which results in significant visualization variations. In this research, the GLCM method is used to produce features that are stable against rotation. Furthermore, classification methods such as K-Nearest Neighbours (KNN), Random Forest (RF), and KNN-RF fusion using the Voting ensemble method are evaluated to improve detection accuracy. The experimental results show that the use of multi-degrees and the use of ensemble voting at all degrees can increase the accuracy value, and the highest accuracy for extraction using multi-degrees is 95.95%. Based on test results which show that the use of features of various degrees and the ensemble voting method can increase accuracy for detecting objects experiencing rotation
 
Convolutional Neural Networks for Classification Motives and the Effect of Image Dimensions
Get PDFAlthough Indonesian batik patterns vary by location, they usually depict local customs and cultures. Each batik has a unique quality and, to correctly identify the batik designs, you need to understand the design patterns. However, many people struggle to identify and categorize these kinds of motivation because they don't have the requisite knowledge, understanding, or access to sufficient information. This study used photo data to classify batik patterns into 15 different groups. Batik Kawung, Megamendung, Lasem, Pole, Machete, Gills, Nutmeg, Karaswasih, Cendrawasih, Geblek Renteng, Bali, Betawi, and Dayak are all included in this category. 1,350 images were used in the research. Google supports the collection of data. To provide the highest level of precision and to evaluate how image dimensions affect the classification of batik designs, this study employs convolutional neural networks (CNNs). The results of this study show that Multi-Layer Perceptron (MLP) is a well-liked deep learning method for data classification, especially in domains where picture classification is involved. The size of the images utilized affects the accuracy of computational neural network (CNN) algorithms. The results showed that the test using training data comparisons of 60%, 30% and 10% resulted in a 01.89% loss of 1.18% and a 100% improvement in accuracy
Enhancing Weighted Averaging for CNN Model Ensemble in Plant Diseases Image Classification
Get PDFDeep learning, especially convolutional neural networks (CNN), has gained traction in the field of image classification. In the specific case of plant disease classification, improving the accuracy and reliability of image classification is paramount. This paper delves into the ensemble prediction technique using a weighted soft-voting method. Instead of assigning a generalized weight to each CNN model, our approach emphasizes giving weights to each label's prediction within every individual model. We employed three respected CNN architectures for our experiments: DenseNet201, InceptionV3, and Xception focus on classifying various diseases that affect grapes. By harnessing transfer learning coupled with end-to-end fine-tuning, we achieved a streamlined and efficient training process. In particular, the f1-score for each grape disease class was used as a parameter for weight determination and as a metric for the final evaluation. In our study, the newly proposed method was tested across various datasets and ensemble scenarios, demonstrating its effectiveness by not only outperforming the conventional soft-voting and prevalent weighted soft-voting methods, which achieved best scores of 95.68% and 95.81% respectively, but also by achieving a remarkable accuracy of 96.56%. The efficacy of this method is enhanced when the ensemble models exhibit distinct characteristics; the more varied the model characteristics, the more enhanced the ensemble results
Convolutional Neural Network and LSTM for Seat Belt Detection in Vehicles using YOLO3
No full textThe application of an electronic violation detection system has begun to be implemented in many countries using CCTV cameras installed at highway and toll road points. However, the development of a violation detection system using data in the form of images that have a high level of accuracy is still a challenge for researchers. Several types of violations detected include the use of seat belts, the use of cell phones while driving, which is influenced by the number of vehicles, vehicle speed and lighting, which can increase the difficulty in the detection process. This research developed a traffic violation detection system using a hybrid model, namely the CNN and LSTM algorithms for the application of discipline using seat belts. The dataset was obtained from RoboFlow Universe with a total of 199 front view car images consists of 82 using seatbelts and 78 not using seatbelts for the training process. The CNN algorithm plays a role in the feature extraction process from input image data, while the LSTM algorithm plays a role in the prediction process. Additionally, the performance evaluation of the CNN+LSTM algorithm will be measured using the accuracy value to measure the performance of the training process and testing process. When measuring the performance of the training process, it will be compared with several basic detection models used, such as CNN, VGG16, ResNet50, MobileNetV2, Yolo3, Yolo3+LSTM. The test results show that Yolo3+LSTM has a higher accuracy compared to the others, at 89%. Next, in the testing process, the CNN+LSTM model will be compared with the basic method, namely CNN. The test results show that the CNN+LSTM models have a higher accuracy of 89%. Meanwhile, in the basic CNN model, the resulting accuracy was 85%
Advanced Earthquake Magnitude Prediction Using Regression and Convolutional Recurrent Neural Networks
No full textEarthquake magnitude prediction is critical in seismology, with significant implications for disaster risk management and mitigation. This study presents a novel earthquake magnitude prediction model by integrating regression analysis with Convolutional Recurrent Neural Networks (CRNNs). It utilises Convolutional Neural Networks (CNNs) for spatial feature extraction from 2-dimensional seismic signal images and Long Short-Term Memory (LSTM) networks to capture temporal dependencies. The innovative model architecture incorporates residual connections and specialised regression techniques for sequential data. Validated against a comprehensive seismic dataset, the model achieves a Mean Squared Error (MSE) of 0.1909 and a Root Mean Squared Error (RMSE) of 0.4369, with a coefficient of determination of 0.79772. These metrics, alongside a correlation coefficient of 0.8980, demonstrate the model's accuracy and consistency in predicting earthquake magnitudes, establishing its potential for enhancing seismic risk assessment and informing early warning systems
Designing a Knowledge-Based Chatbot to Elevate Business Licensing Services in Indonesia
No full textThe business licensing process in Indonesia often faces several challenges, including lack of information, unstable system, complicated procedure, and slow response to complain. These issues can hinder economic growth and limit access for businesses. This research aims to design a knowledge-based chatbot to elevate business licensing services in Indonesia. The proposed chatbot will utilize natural language processing (NLP) technology and a structured knowledge base to provide accurate information, assist in form filling, and offer step-by-step guidance to users. This research employes a User-Centered Design (UCD) approach to ensure that the developed chatbot meets the needs and preferences of its users. The research stages involve user requirements analysis, UML design, system design, and iterations based on feedback obtained. Data will be collected through questionnaires, interviews, and literature studies. Leveraging the proposed architecture, we demonstrate how the resulting knowledge-based chatbot is expected to enhance business licensing services. The findings identified 8 key features expected in the chatbot, including real-time information access, problem reporting, business licensing guidance, a tracking system, personalized simulation, a feedback mechanism, multilingual support, and the ability to connect with a contact center agent. By implementing these features, the proposed chatbot is anticipated to significantly reduce processing times, streamline user interactions, and enhance user satisfaction by providing real-time assistance and reducing errors in form submissions. This will contribute to a more efficient licensing process, fostering economic growth and improving the business environment in Indonesia
Recommendation for Scrum-Based Software Development Process with Scrum at Scale: A Case Study of Software House XYZ
No full textSoftware House XYZ employs Scrum as one of its software development processes. However, the company faces several challenges in the implementation of Scrum, leading to delays in its product releases. Two specific problems are the control of a large-scale Scrum team and the management of team commitments. To address these issues, the Scrum at Scale framework has been chosen as a solution. Before implementing Scrum at Scale, an assessment of the current Scrum maturity level at Software House XYZ is deemed necessary. The Scrum Maturity Model, adapted to the Scrum Guide 2020, is selected as the method to evaluate how effectively the company is implementing Scrum. A questionnaire comprising 81 practices was distributed to development teams, with 10 valid responses collected. Based on the assessment using the Scrum Maturity Model, the current Scrum implementation maturity at Software House XYZ is rated at level 1, Initial. A total of 61 practices are proposed for improvement in the Scrum process. Scrum at Scale can be implemented once the suggested Scrum process improvements have been made. These recommendations are structured following the framework outlined in the Scrum at Scale Guide 2022. The validation of the Scrum-at-Scale recommendations was conducted by us through interviews with representatives from Software House XYZ. From the validation results, the company expresses interest in trying to implement Scrum at Scale. However, the company agrees to enhance the existing Scrum process within the organization before fully adopting Scrum at Scale
Integration of YOLOv5 Algorithm and OpenCV in Innovative Smart Parking Management Approach
Get PDFThe problem of automatic parking lot identification and vehicle detection in open areas is becoming increasingly important due to the increase in the number of vehicles in Indonesia, particularly in big cities, resulting in difficulties in finding parking spaces during peak hours. In this condition, drivers often have to compete for parking spaces. This research aims to develop a smart parking system that integrates YOLOv5 and OpenCV algorithms. This approach thoroughly combines both algorithms to identify parking spaces and detect vehicles in real time in various parking scenarios. It is carried out in an open area with reference to parking conditions at the BRIN Bandung office. This study collected data from three different parking lot conditions, namely empty, partially occupied, and full. In each condition, the system successfully detected the parking lots and vehicles accurately. The novel contribution of this research is the development of a smart parking system that uses an integrated approach, providing an effective solution to the challenges of parking lot availability and vehicle detection. Using the advantages of both algorithms, we successfully created a system that can identify parking spaces and detect vehicles accurately and efficiently under various parking circumstances. Therefore, this research makes a significant contribution to the development of smart and adaptive parking management technology.
 
Metaheuristics Approach for Hyperparameter Tuning of Convolutional Neural Network
No full textDeep learning is an artificial intelligence technique that has been used for various tasks. Deep learning performance is determined by its hyperparameter, architecture, and training (connection weight and bias). Finding the right combination of these aspects is very challenging. Convolution neural networks (CNN) is a deep learning method that is commonly used for image classification. It has many hyperparameters; therefore, tuning its hyperparameter is difficult. In this research, a metaheuristic approach is proposed to optimize the hyperparameter of convolution neural networks. Three metaheuristic methods are used in this research: ant colony optimization (ACO), genetic algorithm (GA), and Harmony Search (HS). The metaheuristics methods are used to find the best combination of 8 hyperparameters with 8 options each which creates 1.6. 107 of solution space. The solution space is too large to explore using manual tuning. The Metaheuristics method will bring benefits in terms of finding solutions in the search space more effectively and efficiently. The performance of the metaheuristic methods is evaluated using MNIST datasets. The experiment results show that the accuracy of ACO, GA and HS are 99,7%, 97.7% and 89,9% respectively. The computational times for the ACO, GA and HS algorithms are 27.9 s, 22.3 s, and 56.4 s, respectively. It shows that ACO performs the best among the three algorithms in terms of accuracy, however, its computational time is slightly longer than GA. The results of the experiment reveal that the metaheuristic approach is promising for the hyperparameter tuning of CNN. Future research can be directed toward solving larger problems or improving the metaheuristics operator to improve its performance