Article Per Year (5 Year)
p-Index From 2021 - 2026
0.835
P-Index
This Author published in this journals
All Journal Scientific Journal of Engineering Research Methods in Science and Technology Studies
Muhammad Raza
Nanjing University of Information Science and Technology
Engineering

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
A Study of Loss Weight Balance in Lightweight Self-Distilled Crowd Counting Muhammad Raza; Atta Ur Rahman; Pandula Pallewatta; Inayat Ur Rahman; Sahib Bahadar
Scientific Journal of Engineering Research Vol. 2 No. 3 (2026): September
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/sjer.v2i3.2026.493

Abstract

Lightweight crowd counting is important for real-time surveillance and resource-constrained deployment, where both computational efficiency and effective supervision are required. Although teacher-free self-distillation can improve lightweight density-regression models by guiding intermediate representations without an external teacher, the influence of composite loss weights in such frameworks has not been sufficiently analyzed. This paper presents a focused coefficient-wise loss-weight analysis within the Lightweight Self-Knowledge Distillation framework for single-image crowd counting. Instead of proposing a new architecture, the study investigates how the coefficients α, β, γ, and λ₂ affect optimization behavior and counting accuracy under a fixed experimental setup on ShanghaiTech Part B. Specifically, α controls intermediate feature alignment, β controls consistency supervision, γ controls direct density-regression supervision, and λ₂ controls the structural similarity term in the regression loss. The results show that moderate values of α and β improve performance by providing useful internal regularization, while excessive auxiliary weighting can slightly degrade accuracy. The analysis also indicates that γ should remain dominant because direct density-map regression is the primary learning signal. The best observed configuration is α = 6.0, β = 2.0, γ = 13.0, and λ₂ = 0.2, achieving 8.94 MAE and 11.51 RMSE on ShanghaiTech Part B. These findings highlight the importance of balanced supervision design within the evaluated LSKD framework on ShanghaiTech Part B.
NRCC-LC: Noise-Robust Crowd Counting with Dynamic Label Correction under Noisy Supervision Abubakar Abdinur Hersi; Miaogen Ling; Muhammad Raza; Abdirahman Mohamed Hassan; Idris Aweis Hussien
Scientific Journal of Engineering Research Vol. 2 No. 3 (2026): September
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/sjer.v2i3.2026.494

Abstract

Crowd counting remains a challenge within computer vision due to many factors that affect the performance of available methods such as occlusion, scale variability, and perspective distortion. Additionally, many labels associated with crowd counting systems have high levels of noise caused by various real-world conditions. Although crowd counting methodologies have improved accuracy over recent years, the majority of crowd counting models still rely on clean real-time supervision and lack systems that can correct for dynamically corrupted labels, resulting in low robustness for crowd counting models when deployed in real-world applications. In this work we present a Noise-Robust Crowd Counting with Label Correction (NRCC-LC) framework to obtain reliable density estimates from noisy supervision. To accomplish this, our approach uses a combined CNN-Transformer architecture to capture both locally- and globally-relevant visual information (i.e., image content and context), along with a Noise-Robust Module (NRM) and a Dynamic Label Correction (DLC) mechanism. Our principle experimental results evaluated across four benchmark datasets: ShanghaiTech Part A, ShanghaiTech Part B, NWPU-Crowd, and JHU-Crowd++, indicate that the NRCC-LC exhibits competitive performance with respect to existing state-of-the-art crowd-counting methods; most notably, producing per-image MAEs of 97.8 and 392.3 on NWPU-Crowd. These experimental results additionally have real-world implications for improving public safety and urban planning; thus, through our novel method of noise-aware feature learning combined with iterative label correction, we can establish the potential of automated monitoring systems in complex, real-world environments to be significantly more reliable.
Trust–Privacy-Based User Acceptance Model for Location-Based Mobile Navigation Services: Conceptual Development and Research Propositions Atta Ur Rahman; Fuyang Ke; Muhammad Raza
Methods in Science and Technology Studies Vol. 2 No. 2 (2026): December Article in Process
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/msts.v2i2.2026.522

Abstract

The two most important factors for location-based mobile navigation services are navigation performance and trusted data governance, especially when location data is used repeatedly. Previous studies have tended to focus on privacy concerns, perceived risk, trust, transparency, control, and service quality separately or in settings other than consumer navigation, as is the case in this study. As a result, the interaction between privacy risk perceptions and provider assurances in influencing mobile navigation acceptance remains poorly defined. This conceptual study develops a trust- and privacy-based user acceptance framework by synthesizing the literature through a structured procedure and incorporating the Technology Acceptance Model, Privacy Calculus Theory, and Trust Theory. The framework posits that privacy concern increases perceived risk, while privacy concern and perceived risk reduce trust. In contrast, transparency, perceived control over location data, and service reliability can enhance trust. The roles of perceived usefulness and user acceptance intention in the pathway to user acceptance remain unchanged, suggesting that trust is an antecedent to perceived usefulness and user acceptance intention. The framework fills the theoretical gap between privacy-risk assessment and trust-building service attributes, and between trust in technology acceptance and repeated disclosure of the location setting. It also recognizes the need for transparent data practices, meaningful location control, and reliable service performance as provisional design priorities, which require future empirical validation. The propositions are to be used as a starting point for further testing on platforms, providers, user groups, cultures, and regulatory frameworks.
LSKD: Lightweight Self-Knowledge Distillation Framework for Fast and Robust Crowd Counting Muhammad Raza; Miaogen Ling; Atta Ur Rahman; Pandula Pallewatta; Aboubakar Abdinur Hersi; Shehan Maxwell Beruwalage; Deshan Sachintha Kannangara
Scientific Journal of Engineering Research Vol. 2 No. 2 (2026): June
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/sjer.v2i2.2026.436

Abstract

Crowd counting plays an important role in the surveillance of the safety of the people, traffic, and intelligent surveillance systems. However, the exact density estimations remain hard to achieve in highly congested scenes due to the tough occlusion, large-scale variance, and complicated background. Although the recent deep-learning methods have high performance, several of them do not need computationally efficient underlying backbone networks, and rather, they employ an external teacher-student distillation architecture, which can limit their use in resource-constrained applications. To avoid this problem, we introduce LSKD, a lightweight self-knowledge distillation network that is density map regression-specific. Unlike other conventional teacher-dependent processes, LSKD can also independently carry out internal multi-level feature alignment within a single small network that is not in need of an external teacher model. The structure integrates a Feature Matching Block (FMB) and a Context Fusion (CoFuse) block to enhance the hierarchical match of features and global awareness of context. The large experiments demonstrate that LSKD obtain competitive performance using the number of parameters as 2.65 million and GFLOPs as 10.23. Particularly, it has 63.17 MAE on ShanghaiTech Part A, 8.94 on ShanghaiTech Part B, 143.7 on UCF-QNRF, and 223.88 on UCF-CC-50, which is a good ratio between the accuracy and the efficiency of the calculations. Such results indicate that LSKD has an implementable and efficient solution to the real-time counting of crowds at the edge devices.
Parameter-Efficient Fine-Tuning for Sonar Shipwreck Segmentation: A Seed Averaged Study with SegFormer and LoRA Shehan Maxwell Beruwalage; Chunyong Yin; Muhammad Raza; Deshan Sachintha Kannangara; Sachini Amani Hendavitharana
Scientific Journal of Engineering Research Vol. 2 No. 2 (2026): June
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/sjer.v2i2.2026.454

Abstract

Accurate segmentation of shipwreck targets in sonar imagery is important for underwater archaeology, marine monitoring, and search operations, but the task remains difficult because labeled sonar masks are scarce and full adaptation of transformer models can be computationally expensive. This study evaluates whether parameter-efficient fine-tuning can provide a practical alternative for binary sonar shipwreck segmentation. Using SegFormer-B0 initialized from a pretrained checkpoint, three adaptation strategies were compared under a consistent protocol: full fine-tuning of all model parameters (FullFT), training only the segmentation head (Head-only), and LoRA-based adaptation of selected linear layers together with head training (LoRA-A+Head). Models were selected by the best validation epoch and evaluated on a held-out test set. Across three random seeds, FullFT achieved the best performance, with a Dice score of 0.614 ± 0.008 and IoU of 0.487 ± 0.007. LoRA-A+Head achieved a Dice score of 0.546 ± 0.010 and IoU of 0.401 ± 0.008 while updating only 1.57% of the parameters, whereas Head-only reached 0.494 ± 0.010 Dice and 0.354 ± 0.008 IoU. These results show a clear accuracy efficiency trade off, full fine-tuning gives the highest accuracy, whereas LoRA-A+Head offers a practical option when reducing the number of updated parameters is important. The findings support the use of parameter-efficient adaptation for sonar segmentation in compute-limited settings.
Co-Authors Abdirahman Mohamed Hassan Aboubakar Abdinur Hersi Abubakar Abdinur Hersi Atta Ur Rahman Chunyong Yin Deshan Sachintha Kannangara Fuyang Ke Idris Aweis Hussien Inayat Ur Rahman Miaogen Ling Pandula Pallewatta Sachini Amani Hendavitharana Sahib Bahadar Shehan Maxwell Beruwalage