Jurnal Elektro

Klasifikasi Gender Berdasarkan Gambar Menggunakan Metode Deep Learning Pada MATLAB

Haenuki Sachi, Linda Wijayanti, Sandra Octaviani — Mon, 15 Jan 2024 00:00:00 +0100

In the present era, machine intelligence, also known as Artificial Intelligence (AI), is demanded not only to execute specific commands but also to recognize, analyze, or even make decisions, thereby providing desired outputs. By harnessing the power of AI, it is anticipated that desired outcomes will be more accurate and goal achievement will be optimized, minimizing losses. With the capabilities of AI in mind, a research study has been conducted on AI's ability to analyze and make decisions based on specific data. In this study, data in the form of images of men and women were utilized. The objective of this research is to analyze the ability of AI, particularly in gender classification. The method employed in designing this system is Deep Learning, with GoogLeNet as the Convolutional Neural Network utilized. In testing, the data accuracy ranged from 61.8% to 100% for the system without training algorithm options and from 97.5% to 100% for the system with training algorithm options. Testing was also carried out on a smaller set of training data and grayscale images, yielding lower accuracy ranges. From this research, it can be concluded that the quantity of training data, image preprocessing, and training algorithm options are crucial indicators for enhancing prediction accuracy.

Sistem Pemberi Makan dan Minum Pintar Pada Hewan Peliharaan Berbasis Website dan Aplikasi Android

Kevin Setiawan, Lukas Lukas, MM Lanny Panjaitan — Mon, 15 Jan 2024 00:00:00 +0100

Based on a survey by “Mars”, about 59% of dogs and 52% of cats weighed more than they should. This is the motivation of this research in designing a feeding and drinking automation for pets so that it is easier for owners to take care their pet's health diet, and so that owners don't have to worry about forgetting to feed their pets. To connect to the internet, ESP32 is used. The supporting components consist of load cell (weight sensor), three ultrasonic sensors (proximity sensor), servo motor and pump motor. These components are combined to carry out the function of feeding and drinking automatically. The device will run after being given an order via website or application. For the feeding system, the animal owner can provide input on the schedule, frequency and portion of feeding. While the drinking system, the owner can press the calibration button and the pause button. Tests are carried out on each sensor, to measure its accuracy and testing of the entire system is carried out to ensure all components work together to carry out their functions. The test results show that every hardware and software component has worked in accordance with the designed objectives.

Simulasi Pengendali Suhu Menggunakan Algoritma Proportional, Integral, Derivative Berbasis Programmable Logic Controller Pada Modul Plant Honeywell

Odillia Kanaya, Melisa Mulyadi, Kumala Indriati — Mon, 15 Jan 2024 00:00:00 +0100

The process of heating in the food industry is often can be found, start from inserting water and the heating process begin to reach a certain temperature, so the process of controlling temperature is one of the important things in the food industry. Temperature controller is an important thing, to maintain certain temperature, so that it did not exceed or lower than needed. This system uses Honeywell programmable logic controller (PLC) as the automation based system. In this research, Honeywell PLC controller will be using proportional, integral, derivative (PID) algorithm to maintain the required temperature. The research will be using the Honeywell module in Automation Laboratorium, Engineering Faculty, Atma Jaya Catholic University of Indonesia. The heating process will begin if the water level in the tank is filled to the required level. System will turn on the heater based on the PID output in the PLC program, so the temperature is maintained to a required setpoint 35^oC. The simulation system can be controlled through a human machine interface (HMI) HCix08 that was made with XDesigner Plus software. The maintained temperature has ±5% tolerance or ±1,75˚C. The result of the research shows that the best parameter for setpoint 35˚C is using a PD controller with the values for Kp is 8 and Kd is 2.

Implementasi Convolutional Neural Network Pada Robot Tikus Micromouse Untuk Navigasi Labirin Dengan Pendeteksi Garis Menggunakan Raspberry Pi

Devin Pangestu, Ferry Rippun Gideon Manalu — Mon, 15 Jan 2024 00:00:00 +0100

Micromouse is a science and technology project competition that began in the late 1970s. Simply put, the micromouse competition is an event where robotic mice can navigate a 16x16 maze without human assistance. Convolutional Neural Network, also known as CNN, is an extension of the Artificial Neural Network (ANN) designed to process two-dimensional data and is often used for classification, recognition, and prediction tasks. In the conducted research, a robotic mouse was designed using Raspberry Pi to control servo motor movements and utilize a camera to detect maze obstacles, such as intersections, through image scanning. The robotic mouse was tested using a CNN architecture model on TensorFlow, assisted by a line detection algorithm. The testing results indicated that the CNN model architecture showed signs of overfitting (the model learned features too well from the training data) with an accuracy of 96.53%, and the model accuracy evaluation result was 97.97% on the designed dataset. As part of the testing, a maze size of 5x5 was used. The testing of the Convolutional Neural Network model and line detection algorithm, when applied to the robotic mouse, demonstrated that the mouse could navigate according to the path. However, the recorded data was influenced by external light reflecting off the surface of the path, causing the line detection algorithm to perceive it as an obstacle, leading the robot to turn back..

Designing a Website-Based Cooperative Application

Jean Paul Ferdinand, Henoch Juli Christanto, Catherine Olivia Sereati — Mon, 15 Jan 2024 00:00:00 +0100

The development of the times in the field of technology allows information to be obtained quickly and easily. We often encounter exchange of information in the world of work, one of the business entities that is very dependent on the exchange of information is cooperatives. In traditional cooperatives, the use of services requires users to come directly to the physical location of the cooperative. The process of recording user data and transaction history is also still done manually, namely by writing it on a book, so it takes a long time. The cooperative system created is a web-based application with the Laravel framework and cooperative data managed using MySQL. Designing a web-based application will provide easy access to cooperative services and decrease the time required for transaction processing for users and increase the efficiency of cooperative management for cooperative managers.