Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Scientific Journal of Engineering Research
Ijaz, Ahmad
Unknown Affiliation
Engineering

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

Found 2 Documents
Search
Journal : Scientific Journal of Engineering Research

 1 
AVNPR-Net: A Real-Time Deep Learning Framework for Robust Vehicle Number Plate Detection and Recognition Ijaz, Ahmad; Sarfraz, Tayyba; Bibi, Tanzeela; Usman, Muhammad
Scientific Journal of Engineering Research Vol. 2 No. 3 (2026): September (in Process)
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/sjer.v2i3.2026.495

Abstract

AVNPR systems are critical in intelligent transportation, monitoring, and law enforcement systems. Nevertheless, the current systems are usually challenged by the issues of dissimilar illumination, obstruction, and the diversity of plate formats, which restrict their practical applicability. To solve these problems, this paper suggests a real-time deep learning-driven AVNPR framework that incorporates effective detection and recognition systems.  The proposed system employs the YOLOv8 object detector model to localize number plates with high accuracy and speed, as well as a lightweight recognition module to identify alphanumeric characters. A custom dataset with different types of vehicles in different environmental conditions was created and improved with the help of preprocessing and data augmentation methods to make the model more robust. In the experiments, the proposed system demonstrated an overall system accuracy of 98.7%, representing the combined number plate detection and character recognition results. The mAP@0.5 is 97%, and mAP 0.5-0.95 is 91%, as well as high precision, recall, and F1-score, which suggests that it shows potential applicability across varying conditions in the assessed dataset and suggests that it may be suitable for real-world applications. The system is also implemented with a Flask-based web application, and it supports image based and real-time webcam detection. The results indicate that the proposed framework provides a viable, efficient, and deployable solution to AVNPR applications. The work will lead to the creation of scalable and real-time intelligent transportation systems and give a basis for future advancement in the improvement of robust vehicle recognition in challenging conditions.
BReMS-Net: Prediction-Guided Coarse-to-Fine Refinement with Boundary-Aware Multi-Scale Dilated Fusion for Robust Breast Mass Segmentation Sarfraz, Tayyba; Ling, Tan; Ijaz, Ahmad
Scientific Journal of Engineering Research Vol. 2 No. 3 (2026): September (in Process)
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/sjer.v2i3.2026.489

Abstract

Breast masses in mammograms are important to segment for computer-aided diagnosis (CAD) to enhance early detection and treatment decisions. Current approaches face challenges in segmenting lesions with low lesion-to-tissue contrast and diverse textures, resulting in misclassification or poor segmentation accuracy. To overcome this challenge, this paper introduces BReMS-Net, a multi-stage segmentation network to improve contextual learning and refined boundaries. We used an MBA-Net backbone with two major components: a Multi-scale Hybrid Dilated Convolution (MHD) module to extract multi-scale contextual features, and a Boundary Feature Auxiliary (BFA) module to strengthen boundary representations via coarse-to-fine feature fusion. Furthermore, a lightweight Prediction-Guided Refinement Module (PRM) uses initial predictions to produce attention maps, remove background clutter, and progressively refine boundary areas. The model has been evaluated on a cross-dataset basis, trained on the CBIS-DDSM dataset and tested on the INbreast dataset, and the results show that the BReMS-Net produces a Dice coefficient of 93.12% and an HD95 of 0.9826, which demonstrate competitive performance compared to several state-of-the-art deep learning methods. These results underline its generalization and robustness. Overall, the framework provides a robust and efficient approach to breast mass segmentation and has important implications for the performance and clinical relevance of automatic breast cancer diagnosis systems.
Co-Authors Bibi, Tanzeela Ling, Tan Muhammad Usman, Muhammad Sarfraz, Tayyba