Article Per Year (5 Year)
p-Index From 2021 - 2026
0.562
P-Index
This Author published in this journals
All Journal JTH: Journal of Technology and Health
Supatman
Program Studi Informatika, Universitas Mercu Buana Yogyakarta
Health Professions Medicine & Pharmacology Nursing

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
OPTIMIZATION OF FACIAL IMAGE RESOLUTION ON CONVOLUTIONAL NEURAL NETWORK FOR PRESENCE BIOMETRIC SYSTEMS Andhika Fajri Raihan Supadi; Supatman
JTH: Journal of Technology and Health Vol. 4 No. 1 (2026): July: JTH: Journal of Technology and Health
Publisher : CV. Fahr Publishing

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61677/jth.v4i1.895

Abstract

Facial recognition-based biometric systems are widely used in attendance and security applications because they do not require physical contact and are easy to implement on various devices. However, image resolution can affect the accuracy and computation time of facial recognition systems. This study developed a system with varying architectures and input resolutions, namely 512×512, 256×256, 128×128, 64×64, and 32×32 pixels. The dataset consisted of 480 images, including 240 face images and 240 non-face images. Evaluation was conducted using accuracy and training and testing computation times. The results showed that the 512×512 pixel resolution yielded the highest training accuracy of 55.00%, and the 512×512-pixel resolution required the longest training time of 504.23 seconds. In testing using new data, the 256×256 pixel resolution demonstrated optimal performance with an accuracy of 72.50% and a computation time of approximately 0.1602 seconds. Based on these results, the 256×256 pixel resolution can be recommended as the preferred choice for implementing a CNN based facial recognition system with limited computational resources.
CLASSIFICATION OF RICE TYPE IMAGE USING RESNET18 Khelvin Osvaldo Welldhar; Supatman
JTH: Journal of Technology and Health Vol. 4 No. 1 (2026): July: JTH: Journal of Technology and Health
Publisher : CV. Fahr Publishing

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61677/jth.v4i1.896

Abstract

Rice is the primary staple food and a source of carbohydrates for most people in Indonesia. Various types of rice available on the market have differences in shape, size, color, and texture. However, some rice varieties have similar characteristics, making the identification process through visual observation still quite difficult. This study was aimed at developing a rice classification program using the Convolutional Neural Network (CNN) method with the ResNet18 architecture based on digital images. The dataset used consisted of 3,120 images divided into six classes, namely Gentong, Naga Mas, Raja Merah, Setra Ramos Alfamart, SPHP, and Wong Tani rice. Before the training process, the images underwent preprocessing, including resizing to 256 × 256 pixels and data normalization. The model was trained using a transfer learning approach with a data split of 80% and 20% for testing. The results showed that the ResNet18 model achieved a training accuracy of 95.81% and a testing accuracy of 84.55%. Evaluation using a confusion matrix showed that the model was able to classify most rice varieties successfully, although several misclassifications were still found among rice varieties with similar shapes and textures. Based on these results, the Convolutional Neural Network (CNN) method with the ResNet18 architecture can be used to support the automatic recognition and classification of rice varieties.
YOLOV8-BASED VISITOR DENSITY DETECTION AND CLASSIFICATION SYSTEM IN ACCESSORY SHOPPING CENTERS (CASE STUDY: SUPER DAZZLE GEJAYAN) Lorensius Forbin Dapa; Supatman
JTH: Journal of Technology and Health Vol. 4 No. 1 (2026): July: JTH: Journal of Technology and Health
Publisher : CV. Fahr Publishing

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61677/jth.v4i1.897

Abstract

Visitor density monitoring in commercial areas directly affects safety and facility comfort management. This study develops a visitor density detection and classification system based on a pretrained YOLOv8s model, with a case study at Super Dazzle Gejayan, Yogyakarta. The YOLOv8s model was used without retraining, leveraging weights pretrained on the COCO dataset to detect human objects in real time. Detection results were then classified into three density categories: Quiet (≤5 people), Moderate (6–10 people), and Crowded (>10 people), using a threshold-based rule approach. Testing was conducted on 707 static images collected directly from the store environment under various visitor conditions. The system achieved an overall accuracy of 89.39%. Based on the classification report, the Crowded class achieved the best performance with an F1-score of 0.94, followed by the Quiet class (F1-score 0.91), while the Moderate class obtained the lowest F1-score of 0.82 due to its position between the two other classes. The main limitations of the system include occlusion conditions, lighting variation, and threshold sensitivity. The results indicate that the pretrained approach without retraining saves time and computational resources for the training process, while still producing competitive accuracy for visitor density classification applications in real-world environments.
Co-Authors Andhika Fajri Raihan Supadi Khelvin Osvaldo Welldhar Lorensius Forbin Dapa