Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal International Journal of Advances in Applied Sciences JOIV : International Journal on Informatics Visualization
Rahman, Shakila
Unknown Affiliation
Computer Science & IT Earth & Planetary Sciences Environmental Science Materials Science & Nanotechnology Mathematics Physics

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Comparative Study of two Region-based Detection Models, Faster R-CNN and R-FCN in detecting Smoke Region from Several Environmental Conditions Benta Hasan, Sumayea; Rahman, Shakila; Khaliluzzaman, Md; Binti Abdul Aziz, Nor Hidayati; Jakir Hossen, Md
JOIV : International Journal on Informatics Visualization Vol 9, No 6 (2025)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.9.6.3764

Abstract

Detecting smoke covering different environmental situations is an explanatory task while mitigating the loss of lives or reducing the fire-based disaster. Various approaches were experimented with to tackle this task. A few of them are convenient, whereas many fail in outcomes. To that end, this research presents a comparative analysis of two region-based detection models based on deep neural networks. Faster R-CNN denotes the Region-based Convolutional Neural Network, and R-FCN, which constitutes the Region-based Fully Convolutional Network, is employed in this research to evaluate detection performance for the smoke detection task. The analysis demonstrates these models with respect to detection accuracy, precision, recall, Intersection over Union, detection speed, and resilience to challenging conditions (e.g., variations in lighting, weather, and complex backgrounds). Research results highlight Faster R-CNN's accuracy, which supports applications in fire-smoke prevention, whereas R-FCN focuses on detection speed, which is relevant to the smoke-monitoring sector. The key issue to consider is the trade-off between computational efficiency and detection accuracy. Performance in extreme environmental conditions can be enhanced through further advances with regard to data variability and typical challenging scenarios. Moreover, Faster R-CNN and R-FCN achieved Precision values of 96.72% and 95.46%, Recall values of 97.66% and 95.73%, and F1-Score values of 97.18% and 95.59%, respectively. This study assumes further assessments to ensure the safety of the living.
Comparative analysis of YOLO variants and EfficientNet for detecting bone fractures in X-ray images Sarker, Shatabdi; Roy, Avizit; Sharmin, Shaila; Rahman, Shakila; Uddin, Jia
International Journal of Advances in Applied Sciences Vol 15, No 1: March 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijaas.v15.i1.pp155-167

Abstract

A bone fracture is a serious medical problem, and accurate and prompt diagnosis is crucial for optimal treatment. This study highlights the progress of automatic bone fracture detection using deep learning (DL) models. A dataset containing 17 different fracture classes was used to train and evaluate the models. The dataset had class imbalance and minor fracture detection challenges. Extensive preprocessing, including data augmentation and resizing, has been applied to solve these problems, which has helped to increase the robustness of the model. Seven state-of-the-art models—you only look once (YOLO)v8, YOLOv9, YOLOv10, YOLOv11, EfficientNetB0, DenseNet169 and ResNet50—are trained and evaluated. Precision, recall, F1-score, and mean average precision (mAP) were used to evaluate the performance of the models. Among all models, YOLOv11 leads the others by achieving the highest precision, mAP, and precision-recall balance. YOLOv11 adds architectural improvements such as a deep backbone network and hybrid feature fusion, which make the model more reliable in different types of fracture detection. It is capable of reducing false detections and maintaining stable memory usage consistency even under different imaging conditions. Overall, YOLOv11 showed promising results and highlighted the potential of AI-powered diagnostic tools to improve clinical processes and patient care. As future work, the application field of the model can be extended to larger medical imaging tasks, and it can be further refined for effective use in resource-limited environments.
Co-Authors Benta Hasan, Sumayea Binti Abdul Aziz, Nor Hidayati Jakir Hossen, Md Khaliluzzaman, Md Roy, Avizit Sarker, Shatabdi Sharmin, Shaila Uddin, Jia