cover
Contact Name
Ramdan Satra
Contact Email
Ramdan Satra
Phone
-
Journal Mail Official
ramdan@umi.ac.id
Editorial Address
-
Location
Kota makassar,
Sulawesi selatan
INDONESIA
ILKOM Jurnal Ilmiah
ISSN : 20871716     EISSN : 25487779     DOI : -
Core Subject : Science,
ILKOM Jurnal Ilmiah is an Indonesian scientific journal published by the Department of Information Technology, Faculty of Computer Science, Universitas Muslim Indonesia. ILKOM Jurnal Ilmiah covers all aspects of the latest outstanding research and developments in the field of Computer science, including Artificial intelligence, Computer architecture and engineering, Computer performance analysis, Computer graphics and visualization, Computer security and cryptography, Computational science, Computer networks, Concurrent, parallel and distributed systems, Databases, Human-computer interaction, Embedded system, and Software engineering.
Arjuna Subject : -
Articles 633 Documents
Improved Coffee Beans Detection Using Contrast Limited Adaptive Histogram Equalization and Unsharp Masking Srivan Palelleng; Nugra Tasik Allo; Juprianus Rusman
ILKOM Jurnal Ilmiah Vol 18, No 2 (2026)
Publisher : Prodi Teknik Informatika FIK Universitas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/ilkom.v18i2.3120.351-362

Abstract

Contrast Limited Adaptive Histogram Equalization (CLAHE) is widely used to enhance image contrast, but it has limitations in improving edge sharpness and may amplify noise. Meanwhile, Unsharp Masking (USM) is effective for edge enhancement, yet it is less optimal when applied to low-contrast images. To address these weaknesses, this study proposes a hybrid image enhancement method combining CLAHE and USM for coffee beans quality analysis. The dataset consists of four classes: Biji-Kopi Normal (normal), Biji-Kopi Pecah (broken), Biji-Kopi Hitam (black), and Biji-Kopi Berlubang (hollow). All images were trained at a resolution of 512×512 pixels for 2000 iterations without transfer learning to ensure fair and unbiased evaluation. Histogram-based analysis demonstrates significant improvements after enhancement, including a 50.19% increase in mean intensity, a 217.9% increase in variance, a 78.38% rise in standard deviation, and a 22% increase in skewness, along with a 37.42% reduction in kurtosis. Performance evaluation using YOLOv4-Tiny shows that the proposed method improves AP50 from 96.51 to 97.7 and AP75 from 59.91 to 66.24, while AP95 remains unchanged at 0.01. The most notable improvements are observed at AP70 to AP90, indicating that the hybrid approach not only enhances classification performance but also strengthens object localization accuracy, making it effective for coffee bean defect detection.
Forensic Analysis for Detecting Deep-Fake Images Using A Convolutional Neural Network (CNN) and The National Institute of Standards and Technology (NIST) Methods Muhammad Na'im Al Jum'ah; Hamid Wijaya; Muh. Hajar Akbar; Suwito Pomalingo
ILKOM Jurnal Ilmiah Vol 18, No 2 (2026)
Publisher : Prodi Teknik Informatika FIK Universitas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/ilkom.v18i2.3178.281-291

Abstract

The development of Artificial Intelligence (AI) has significantly influenced audio, video, and image manipulation techniques, commonly known as deepfakes. Image forensics faces an urgent challenge in identifying and mitigating the impact of deepfake content to maintain the integrity and credibility of digital information. This research aims to perform forensic analysis in accordance with NIST standards and to implement Convolutional Neural Network (CNN) methods to detect deepfake images. Based on the test results, the Convolutional Neural Network (CNN) method can be effectively applied to deepfake image detection. The CNN architecture used can identify the distinct visual characteristics of deepfake images with high performance. The model demonstrates the ability to learn and minimize prediction errors on training data. Accuracy graphs indicate that the model has successfully learned data patterns, as evidenced by consistent improvements in both training and validation data as the number of epochs increases. Furthermore, the loss graph shows a downward trend, signifying a continuous reduction in model error. The precision graph demonstrates the model's effectiveness in reducing false positives, thereby minimizing errors in detecting the original data. The recall graph also indicates improved detection performance on the training data. The ROC curve suggests that the model possesses superior classification capabilities compared to random guessing. Additionally, the Area Under the Curve (AUC) of 0.6544 serves as a quantitative indicator of performance, indicating that the model has moderate capability for class differentiation. Detection results from the CNN model on a dataset of real and deepfake images show that the Confidence and Raw Score values can distinguish between the two; however, the confidence levels still fluctuate around the classification threshold. Low confidence values in certain images suggest that the extracted features are not yet optimal at distinguishing between real faces and manipulated images. Moreover, the application of the National Institute of Standards and Technology (NIST) standards (Collection, Examination, Analysis, and Reporting) for forensic analysis ensures that the evidence gathered is legally accountable in court. Thus, these standards can serve as a scientific reference to ensure a more structured and standardized investigation process for deepfake images.
An Optimized YOLOv7-Based Object Detection Framework Leveraging Low-Light Image Enhancement to Improve Accuracy Rasim Rasim; Farhan Nurzaman; Yaya Wihardi; Herbert Siregar; Samialloi Nusratullo
ILKOM Jurnal Ilmiah Vol 18, No 2 (2026)
Publisher : Prodi Teknik Informatika FIK Universitas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/ilkom.v18i2.3102.207-220

Abstract

Low-light environments remain a persistent challenge in computer vision, often leading to notable degradation in object detection performance. This is primarily caused by reduced contrast, increased noise, and the loss of critical visual details, all of which hinder reliable feature extraction. To address these limitations, this study proposes an integrated framework that combines Zero-Reference Deep Curve Estimation (Zero-DCE) for adaptive image enhancement with the YOLOv7 architecture for efficient and accurate object detection. The study was conducted through a structured pipeline consisting of several key stages: (1) preparation of the ExDark, NOD, and LOD datasets; (2) preprocessing, including annotation and labelling; (3) low-light image enhancement using Zero-DCE; (4) dataset selection, partitioning, and utilization; (5) image resizing to ensure model compatibility; (6) model development, comprising both a baseline YOLOv7 model and an enhanced Zero-DCE + YOLOv7 configuration; and (7) performance analysis and evaluation using mean Average Precision (mAP) as the primary metric. Experimental results demonstrate that the integration of Zero-DCE with YOLOv7 improves detection performance, with mAP@0.5 increasing from 0.785 in the baseline model to 0.794. Although the improvement is modest, it is consistent and indicates the effectiveness of incorporating illumination enhancement into the preprocessing stage. In addition, this study’s contribution lies in demonstrating that the proposed framework enhances the robustness of object detection systems under challenging lighting conditions without incurring significant computational overhead.