Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal PIKSEL : Penelitian Ilmu Komputer Sistem Embedded and Logic Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Jurnal Riset Informatika
Akhiyar Waladi
Universitas Jambi
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Energy Engineering

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

Found 3 Documents
Search

 1 
Parameter-Efficient Models for Malaria Detection and Classification Using Small-Scale Imbalanced Blood Smear Images Akhiyar Waladi; Hasanatul Iftitah; Nindy Raisa Hanum; Yogi Perdana; Fitra Wahyuni; Rahmad Ashar
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 11, No. 2, May 2026
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v11i2.2558

Abstract

Malaria diagnostic automation faces critical challenges, including severe class imbalance with ratios of up to 54:1, limited datasets containing 200 to 500 images, and computational inefficiency resulting from the need to train separate models for each detection-classification combination. This study developed a multi-model framework with a shared classification architecture that trains classification models once on ground-truth crops and reuses them across all detectors. The framework systematically evaluated three YOLO Medium architectures for parasite detection and six CNN architectures for lifecycle and species classification across four complementary malaria datasets totaling 1,544 microscopy images. Detection achieved mAP@50 scores ranging from 70.84% to 96.27%, with high recall values of 71.05% to 93.12% minimizing missed parasite detections. Classification results demonstrated the importance of dataset-dependent model selection, with parameter-efficient EfficientNet models containing 5.3M to 9.2M parameters consistently outperforming ResNet variants with up to 44.5M parameters. EfficientNet-B1 achieved accuracies of 91.51% on the IML Lifecycle dataset and 98.28% on the MP-IDB Species dataset, while EfficientNet-B0 achieved 86.45% on the multi-patient MD-2019 dataset. ResNet50 achieved 96.13% accuracy on severely imbalanced MP-IDB Stages dataset. Focal Loss optimization with alpha = 1.0 and gamma = 1.5 enabled robust minority-class performance, achieving F1-scores between 0.44 and 1.00 on ultra-minority classes and demonstrating effective handling of class imbalance. The compact models, with sizes ranging from 46 MB to 89 MB, enable practical deployment on resource-constrained hardware.
Comparative Analysis of Triangulation Methods for Optimal Solutions to the Art Gallery Problem Jefri Marzal; Niken Rarasati; Akhiyar Waladi; Yogi Perdana
PIKSEL : Penelitian Ilmu Komputer Sistem Embedded and Logic Vol. 13 No. 1 (2025): Maret 2025
Publisher : LPPM Universitas Islam 45 Bekasi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33558/piksel.v13i1.10749

Abstract

Triangulation is the process of breaking down an n-sided polygon into triangles and it is necessary in deciding the optimal count and the position of guards in the Art Gallery Problem (AGP) There is a theoretical limit that has been established which states that the number of required guards needed to keep an eye on such a polygon is ⌊n/3⌋ and this research considers this as the limit. Among various triangulation methods, Ear Clipping and Minimum Weight are two primary approaches frequently used to achieve optimal solutions. Nonetheless, its comparison with other methods, more particularly the amount of guards required for the maximum theoretical figure, is still a gap in literature. The aim of this research is to create an AGP simulation program and test it against the theoretical upper bound, determining the number of guards required. 228 simple polygons with vertices varying between 10 and 110 were utilized in this research. The polygons were classified into three groups based on the ratio of convex to concave vertices: less concave vertices, equal amount of concave and convex vertices and vice versa. Result study shows that the Ear Clipping method is significantly superior to Minimum Weight in reducing guard requirements. Practically speaking, these advancements are important for the design of engineering systems such as surveillance systems and the surveillance of public spaces. In the context of building security system design and monitoring of large areas, these conclusions are of utmost importance.
TRANSFER LEARNING ARCHITECTURE SELECTION FOR REMOTE SENSING SCENE CLASSIFICATION Akhiyar Waladi; Hasanatul Iftitah
Jurnal Riset Informatika Vol. 8 No. 3 (2026): Juni 2026
Publisher : Kresnamedia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34288/jri.v8i3.515

Abstract

Selecting a deep learning architecture for classifying remote sensing scenes usually involves comparing published accuracy across papers that each use different training protocols, making it unclear whether accuracy gaps reflect architecture or training differences. We isolate the architecture variable by evaluating eight models from three design families, five classical CNNs (ResNet-50, ResNet-101, DenseNet-121, EfficientNet-B0, EfficientNet-B3), two vision transformers (ViT-B/16, Swin Transformer), and one modernized CNN (ConvNeXt-Tiny), under identical training conditions on EuroSAT (10 classes, 27,000 Sentinel-2 patches) and UC Merced (21 classes, 2,100 aerial photographs). Every model shares the same ImageNet-1K initialization, AdamW optimizer, augmentation pipeline, and early stopping rule. ConvNeXt-Tiny reached the highest accuracy on EuroSAT (99.11%) and Swin-T on UC Merced (99.76%), but the accuracy range on EuroSAT was only 0.41 percentage points (1.66 on UC Merced). McNemar's test confirmed that most pairwise differences were not significant. EfficientNet-B0, the smallest model at 4.0M parameters, reached 98.76% and 99.52% while using 21x fewer parameters than ViT-B/16. On these two well-studied benchmarks, a single uniform training configuration was sufficient to bring all architectures to near-identical performance. This convergence, observed under one fixed protocol and a single data partition, suggests that on saturated classification tasks the choice of architecture may be secondary to the choice of training procedure. Whether this convergence holds on harder benchmarks, under architecture-specific optimal configurations, or with domain-specific pretraining remains to be tested
Co-Authors Fitra Wahyuni Hasanatul Iftitah Jefri Marzal Niken Rarasati Nindy Raisa Hanum Rahmad Ashar Yogi Perdana