Ultima Computing : Jurnal Sistem Komputer

Liquid Petroleum Gas (LPG) Cylinder Leak Detection Tool Using MQ-2 Sensor Based on Internet of Things (IoT)

2024-12-31T16:38:19+00:00

The widespread use of LPG gas cylinders brings the risk of gas leaks that can cause serious hazards, including fires and explosions. Therefore, an effective system is needed to detect gas leaks and provide early warnings to users. This study aims to develop an LPG gas cylinder leak detection device using an MQ-2 sensor based on the Internet of Things (IoT). The system consists of an MQ-2 sensor capable of detecting LPG gas, a microcontroller module for data processing, and an IoT communication module to send alerts to user devices via the internet. When the MQ-2 sensor detects a gas concentration that exceeds the predetermined threshold, the system sends an alert in the form of a notification to the user's mobile application. Additionally, the system is equipped with an audible alarm for direct on-site warnings. Test results indicate that this system can detect gas leaks with high accuracy and send alerts promptly. The implementation of IoT technology allows for real-time monitoring and handling of gas leaks, thereby enhancing the safety of LPG gas cylinder users. Thus, this leak detection device is expected to reduce the risk of accidents due to gas leaks and provide a sense of security for users.

Analysis of Noise Removal Performance in Speech Signals through Comparison of Median Filter, Low FIR Filter, and Butterworth Filter: Simulation and Evaluation

2024-12-31T16:38:19+00:00

This research aims to analyze the performance of three types of filters, namely median filters, low FIR filters, and Butterworth filters, in eliminating noise in sound signals. Evaluation is carried out through simulation and evaluation using the Mean Squared Error (MSE) and Signal-to-Noise Ratio (SNR) parameters. The simulation results show that the three filters are able to produce signal estimates that are close to the original signal with low MSE values. The median filter shows the best performance with an MSE of 0.015833 and the highest SNR of 51.6334 dB, indicating its ability to reduce noise without sacrificing signal clarity. FIR and Butterworth filters also provide good results, although with slightly lower levels of accuracy. In conclusion, median filters are the optimal choice for noise removal in speech signals, while FIR and Butterworth filters remain good alternatives depending on application requirements. Further research and practical testing are needed for validation in real-world situations

Development of Cavendish Banana Maturity Detection and Sorting System Using Open Source Computer Vision and Loadcell Sensor

2024-12-31T16:38:20+00:00

This research aims to develop a system of detecting the maturity and sorting of cavendish bananas using Open Source Computer Vision (OpenCV) and also assisted by a loadcell sensor. The problem experienced at this time is that fruit sorting is still manual which is less efficient and inaccurate in distinguishing banana maturity based on the color of the skin. This is because the human eye is sensitive to changes in lighting and fatigue. This designed system will use webcam for image processing and loadcell for fruit weight measurement, controlled by Arduino Uno microcontroller. While the algorithm used to determine the color of the ripeness of the banana fruit itself is HSV. The test results show an average weight error of 0.08% for ripe bananas, 0.71& for unripe bananas, while the color detection produces an accuracy of 47.34% on average in bright lighting conditions. In conclusion, this system is successful in improving sorting efficiency with adequate accuracy results, but further development is needed so that the accuracy level increases.

Air Filtration System Utilizing Biomimetic Technology and IoT for Air Quality Improvement

2024-12-31T16:38:20+00:00

The "Hepix" smart air filtration system, developed with biomimetic and Internet of Things (IoT) technology, aims to address the urgent issue of poor indoor air quality, particularly in high-mobility urban areas. This system integrates advanced sensors (MQ135 and BME680) and biomimetic filtration inspired by leaf stomata to monitor and filter air pollutants. Tested across three locations—Cilame, Jatinangor, and Cibiru—the system achieved an approximate 24.4% reduction in pollutant levels, as well as stable control of humidity and air pressure. Real-time data is continuously monitored through a mobile and web interface, supported by Google Assistant integration for voice commands. The results demonstrate that "Hepix" effectively improves air quality, offering a practical solution for healthier indoor environments in urban areas.

Microscopic Sand Image Classification Using Convolutional Neural Networks

2024-12-31T16:38:20+00:00

Abstract— This research paper reviews the use of Convolutional Neural Networks (CNNs) to categorize diverse sand type using microscopic images, with an objective of improving quality control in construction materials. The paper compares three CNN architectures—LeNet-5, Inception v3, and ResNet50—for discriminating between specific sand categories, such as two river sands (Cipongkor and Citarum) and three types of silica sand (brown, cream, and white). Each model was trained and tested on different dataset splits, with images pre-processed to highlight specific microscopic properties. To achieve a thorough comparison, each model's performance was measured using a variety of measures such as F1-score, accuracy, recall, and precision. These measurements enabled a comprehensive evaluation of how accurately and reliably each CNN model categorized the various sand types. ResNet50 consistently delivered the highest accuracy, achieving perfect classification in some instances, showcasing its effectiveness in capturing fine details in sand textures. These results highlight the potential of CNN-based approaches for precise and automated sand classification, which supports increased quality assurance in construction and related areas.

Index Terms— Convolutional Neural Network (CNN); sand classification; LeNet-5; Inception v3; ResNet50