Article Per Year (5 Year)
p-Index From 2021 - 2026
2.217
P-Index
This Author published in this journals
All Journal Jurnal Edukasi dan Penelitian Informatika (JEPIN) Jurnal Teknologi Informasi dan Multimedia Cyber Security dan Forensik Digital (CSFD) Masyarakat Berdaya dan Inovasi Jurnal Computer Science and Information Technology (CoSciTech) Humanism : Jurnal Pengabdian Masyarakat Prosiding Seminar Nasional Pengabdian Kepada Masyarakat KOMMAS: Jurnal Pengabdian Kepada Masyarakat Jurnal Informatika Polinema (JIP) Scientific Journal of Informatics INOVTEK Polbeng - Seri Informatika
Murinto
Unknown Affiliation
Religion Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Energy Engineering Environmental Science Law, Crime, Criminology & Criminal Justice Library & Information Science Nursing Public Health Social Sciences Other

Published : 14 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 14 Documents
Search

 1   2 
Siamese Neural Networks with Chi-square Distance for Trademark Image Similarity Detection Suyahman; Sunardi; Murinto; Arfiani Nur Khusna
Scientific Journal of Informatics Vol. 11 No. 2: May 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i2.4654

Abstract

Purpose: The objective of this study is to address the limitations of existing trademark image similarity analysis methods by integrating a Chi-square distance metric within a Siamese neural network framework. Traditional approaches using Euclidean distance often fail to accurately capture the complex visual features of trademarks, leading to suboptimal performance in distinguishing similar trademarks. This research aims to improve the precision and robustness of trademark comparison by leveraging the Chi-square distance, which is more sensitive to image variations. Methods: The approach involves modifying a Siamese neural network traditionally employing Euclidean distance with the use the Chi-square distance metric instead. This alteration allows the network to better capture and analyze critical visual features such as color and texture. The modified network is trained and tested on a comprehensive dataset of trademark images, enabling the network to learn and distinguish between similar and dissimilar trademarks based on subtle visual cues. Result: The findings from this study show a significant increase in accuracy, with the modified network achieving an accuracy rate of 98%. This marks a notable improvement over baseline models that utilize Euclidean distance, demonstrating the effectiveness of the Chi-square distance metric in enhancing the model's ability to discriminate between trademarks. Novelty: The novelty of this research lies in its application of the Chi-square distance in a deep learning framework specifically for trademark image similarity detection, presenting a novel approach that yields higher precision in image-based comparisons.
Comparative Analysis of CNN Architectures in Siamese Networks with Test-Time Augmentation for Trademark Image Similarity Detection Suyahman; Sunardi; Murinto
Scientific Journal of Informatics Vol. 11 No. 4: November 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i4.13811

Abstract

Purpose: This study aims to enhance the detection of trademark image similarity by conducting a comparative analysis of various Convolutional Neural Network (CNN) architectures within Siamese networks, integrated with test-time augmentation techniques. Existing methods often face challenges in accurately capturing subtle visual similarities between trademarks due to limitations in feature extraction and generalization capabilities. The research seeks to identify the most effective CNN architecture for this task and to assess the impact of test-time augmentation on model performance. Methods: The study implements Siamese networks utilizing three distinct CNN architectures: VGG16, VGG19, and ResNet50. Each network is trained on a dataset of trademark images to learn deep feature representations that can discriminate between similar and dissimilar trademarks. During the evaluation phase, test-time augmentation (TTA) is applied to enhance model robustness by averaging predictions over multiple augmented versions of the input images. TTA includes transformations such as random rotations (up to 40%), width and height shifts (up to 20%), random shear transformations, zooming (up to 20%), horizontal and vertical flips, and random brightness adjustments. Result: Experimental findings reveal that the Siamese network based on VGG19 achieves the highest accuracy at 98.82%, outperforming the VGG16-based network with an accuracy of 97.07% and the ResNet50-based network with 50.00% accuracy. The application of TTA has improved performance across all models, with the VGG19 model receiving the highest improvement. The extremely low accuracy of ResNet50 can be attributed to its misinterpretation of original trademark images as close-forged ones, probably due to overfitting or lack of an efficient ability in generalizing very fine visual features. Novelty: The study conducted a comparative analysis of CNN architectures, namely VGG16, VGG19, and ResNet50 in Siamese networks for trademark image similarity detection.
Peningkatan Kualitas Citra Hilal Berdasarkan Kontras Menggunakan Metode Histogram Equalization, AHE, dan CLAHE Suprayitno, Ady; Murinto; Kartika Firdausy
Computer Science and Information Technology Vol 6 No 3 (2025): Jurnal Computer Science and Information Technology (CoSciTech)
Publisher : Universitas Muhammadiyah Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37859/coscitech.v6i3.10376

Abstract

The determination of the beginning of the Hijri month is often aided by digital imaging technology, but the quality of the crescent images produced often faces the challenge of very low contrast. The faint light of the crescent is difficult to distinguish from the still bright background of the evening sky, exacerbated by atmospheric conditions and camera sensor noise that reduce visual quality. To improve the image, many still perform manual contrast enhancement. On the other hand, the selection of contrast enhancement methods is often without a measurable basis. This study aims to conduct a comparative performance evaluation between three contrast enhancement methods: Histogram Equalization (HE), Adaptive Histogram Equalization (AHE), and Contrast Limited Adaptive Histogram Equalization (CLAHE). The goal is to identify the most suitable technique for improving the quality of crescent images, the specific application of which has not been widely explored. A total of 30 crescent images were tested through a quantitative evaluation approach using the Mean Squared Error (MSE) and Peak Signal-to-Noise Ratio (PSNR) metrics. The results show that CLAHE provides the best performance with the lowest average MSE (89.97) and the highest PSNR (30.92 dB), demonstrating the best ability to balance contrast enhancement and distortion reduction. In contrast, the HE and AHE methods produce high MSE and low PSNR values, indicating significant visual distortion. Thus, CLAHE is recommended as the most reliable method for improving the quality of crescent images based on contrast in digital technology-based observation systems. For further research, it is recommended to explore the automatic determination of CLAHE parameters and the use of additional evaluation metrics such as SSIM (Structural Similarity Index Measure).
Deteksi Objek Sampah Organik dan Non-Organik Menggunakan YOLO26n Silviana, Yorissa; Murinto; Firdausy, Kartika
JEPIN (Jurnal Edukasi dan Penelitian Informatika) Vol. 12 No. 2 (2026): Volume 12 No 2
Publisher : Program Studi Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Pemilahan sampah secara manual membutuhkan waktu yang lama dan tingkat konsistensi pengamatan yang tinggi, terutama ketika objek memiliki bentuk, warna, dan tekstur yang beragam. Penelitian ini bertujuan untuk mengevaluasi kinerja YOLO26n dalam mendeteksi dan melokalisasi sampah organik dan non-organik. Dataset yang digunakan terdiri atas 2.000 citra dengan distribusi seimbang, yaitu masing-masing 1.000 citra organik dan non-organik yang diperoleh dari pengumpulan mandiri dan dataset publik. Data dianotasi menggunakan bounding box dan dibagi menjadi data pelatihan, validasi, dan pengujian dengan rasio 80:10:10. Model dilatih selama 100 epoch menggunakan citra berukuran 640 × 640 piksel dengan batch size 32. Hasil evaluasi menunjukkan nilai precision sebesar 0,9292, recall sebesar 0,8488, F1-score sebesar 0,8870, mAP@0.5 sebesar 0,8967, dan mAP@0.5:0.95 sebesar 0,8034. Kelas non-organik memperoleh performa lebih tinggi dengan mAP@0.5 sebesar 0,928 dan mAP@0.5:0.95 sebesar 0,894, dibandingkan kelas organik yang masing-masing sebesar 0,865 dan 0,713. Perbedaan tersebut menunjukkan bahwa variasi bentuk, tekstur, dan batas visual objek organik masih menjadi tantangan dalam proses deteksi dan lokalisasi. Model juga mencatat waktu inferensi sebesar 0,7 ms per citra pada GPU NVIDIA RTX PRO 6000. Secara keseluruhan, YOLO26n mampu mendeteksi dan melokalisasi sampah organik dan non-organik dengan kinerja yang baik, namun masih memerlukan pengujian lebih lanjut pada data independen dan perangkat edge sebelum diterapkan pada sistem pemilahan sampah secara real-time.
Co-Authors Abdul Fadlil Abdul Fadlil Agung Tri Lestari, agungyappi Agus Akhtawan Andhi Yanto Wibowo Arfiani Nur Khusna Asno Azzawagama Firdaus Dedi Wijayanti Deni Kuswandani Dianda Rifaldi Herman Imam Riadi Kartika Firdausy Kencana, Nagala Wangsa Nadia Wati Aprianti Nilam Tri Astuti Prambudi, Rizal Prambudi Rinday Zildjiani Salji Rusydi Umar Salsabilla Azahra Putri Silviana, Yorissa Sunardi Suprayitno, Ady Suyahman Suyahman