Mohammad Abdul Qayum
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Mohammad Abdul Qayum has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Nafiul Alam Siddique
New Mexico State University
3 shared publications Abu Saleh Md. Tayeen
Klipsch School of Electrical & Computer Engineering, New Mexico State University, USA
2 shared publicationsAbdel-Hameed A. Badawy
Klipsch School of Electrical and Computer Engineering, New Mexico State University, Las Cruces, USA
1 shared publication- Acramul Haque Kabir
North South University
1 shared publication - Ahmed Faizul Haque Dhrubo
North South University
1 shared publication - Asmaul Islam Tisha1 shared publication
Latest Publications
- Open Access
- 8 September 2026
- Published by Elsevier in SSRN Electronic Journal
Automated Vision-Based Corrosion Inspection for Steel Infrastructure Using Lightweight Adaptive Deep Learning
In this work, we investigate task-adaptive asymmetric loss calibration for lightweight corrosion segmentation using a CBAM-enhanced MobileNetV2 U-Net. First, we compare DiceCE with a reference Focal Tversky-based loss and observe that the latter produces a strongly precision-oriented behavior, substantially reducing corrosion recall and overall overlap. Motivated by this limitation, we propose the Industrial Defect Asymmetric Loss (IDAL), which systematically adjusts the asymmetric weighting of the Focal Tversky component to identify a task-specific precision–recall operating point. The selected configuration is further evaluated using multi-seed experiments and five-fold cross-validation. IDAL achieves comparable corrosion IoU and Dice to DiceCE (0.498 vs. 0.497 and 0.665 vs. 0.664), while improving precision (0.637 vs. 0.583), Boundary IoU (0.114 vs. 0.104), and HD95 (45.924 vs. 48.879). This benefit is accompanied by reduced recall (0.695 vs. 0.774). A limitation is that the evaluation is conducted on a single corrosion dataset.
- 8 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAutomated Vision-Based Corrosion Inspection for Steel Infrastructure Using Lightweight Adaptive Deep Learning
In this work, we investigate task-adaptive asymmetric loss calibration for lightweight corrosion segmentation using a CBAM-enhanced MobileNetV2 U-Net. First, we compare DiceCE with a reference Focal Tversky-based loss and observe that the latter produces a strongly precision-oriented behavior, substantially reducing corrosion recall and overall overlap. Motivated by this limitation, we propose the Industrial Defect Asymmetric Loss (IDAL), which systematically adjusts the asymmetric weighting of the Focal Tversky component to identify a task-specific precision–recall operating point. The selected configuration is further evaluated using multi-seed experiments and five-fold cross-validation. IDAL achieves comparable corrosion IoU and Dice to DiceCE (0.498 vs. 0.497 and 0.665 vs. 0.664), while improving precision (0.637 vs. 0.583), Boundary IoU (0.114 vs. 0.104), and HD95 (45.924 vs. 48.879). This benefit is accompanied by reduced recall (0.695 vs. 0.774). A limitation is that the evaluation is conducted on a single corrosion dataset.
- 1 January 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Interactive intelligent agents with creative minds: Experiments with mobile robots in cooperating tasks by using machine learning
In this paper, we present an intelligent system where agents can co-ordinate creative tasks through machine learning and cooperation. For machine learning, we used commonly used pattern recognition algorithm - Principal Component Analysis (PCA). Based on recognition, we plan a task that is performed by multiple intelligent agents. In our case, task is to draw a pattern or perform a creative art by agents. The task action is divided into three phases: obtaining a design, composing a mathematical model and and performing the task by agents. In case of agents co-ordination, various feedback techniques using wireless sensors and on-board sensors are used. As for proof of concept (POC), a flower pattern is detected, which is painted on a canvas by using mobile robots. Also, person's identity and mood is detected and then a creative art is performed by mobile robots to improve the mood.
- 1 January 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Interactive intelligent agents with creative minds: Experiments with mobile robots in cooperating tasks by using machine learning
In this paper, we present an intelligent system where agents can co-ordinate creative tasks through machine learning and cooperation. For machine learning, we used commonly used pattern recognition algorithm - Principal Component Analysis (PCA). Based on recognition, we plan a task that is performed by multiple intelligent agents. In our case, task is to draw a pattern or perform a creative art by agents. The task action is divided into three phases: obtaining a design, composing a mathematical model and and performing the task by agents. In case of agents co-ordination, various feedback techniques using wireless sensors and on-board sensors are used. As for proof of concept (POC), a flower pattern is detected, which is painted on a canvas by using mobile robots. Also, person's identity and mood is detected and then a creative art is performed by mobile robots to improve the mood.
- 1 January 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
StAdHyTM: A Statically Adaptive Hybrid Transactional Memory: A scalability study on large parallel graphs
In this paper, we present a Statically Adaptive Hybrid Transactional Memory (StAdHyTM) that outperforms not only existing Hardware TM (HTM) and Software TM (STMs) but also common synchronization schemes such as locks. StAdHyTM is statically tuned to adapt to the application behavior to improve the performance. We focus in particular on large parallel graph applications. Our StAdHyTM implementation outperforms coarse-grain locks by up to 8.1× and STM by up to 2.6× in total execution time for computation kernel in the SSCA-2 benchmark. It also outperforms HTM by up to 2.1× on a 28-cores and 64GB machine. We tested large graphs of up to 268 million vertices and 2.147 billion edges on a 64-core and 128 GB machine. To the best of our knowledge, this work is the first scalability study of synchronizations involving all the TM implementations- HTM, STM, HyTM, and Adaptive HyTM.
- 1 January 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
StAdHyTM: A Statically Adaptive Hybrid Transactional Memory: A scalability study on large parallel graphs
In this paper, we present a Statically Adaptive Hybrid Transactional Memory (StAdHyTM) that outperforms not only existing Hardware TM (HTM) and Software TM (STMs) but also common synchronization schemes such as locks. StAdHyTM is statically tuned to adapt to the application behavior to improve the performance. We focus in particular on large parallel graph applications. Our StAdHyTM implementation outperforms coarse-grain locks by up to 8.1× and STM by up to 2.6× in total execution time for computation kernel in the SSCA-2 benchmark. It also outperforms HTM by up to 2.1× on a 28-cores and 64GB machine. We tested large graphs of up to 268 million vertices and 2.147 billion edges on a 64-core and 128 GB machine. To the best of our knowledge, this work is the first scalability study of synchronizations involving all the TM implementations- HTM, STM, HyTM, and Adaptive HyTM.
- 1 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 377-382
Control of autonomous cars for intelligent transportation system
This work concentrates on establishing a technology to develop a unique approach for the integration of intelligent system control into transportation systems. We have implemented an algorithm [1] that controls a car to track a predefined track which integrates a human driver with computer control to increase human performance while reducing reliance on detailed driver attention. Knowledge obtained from the optical tracking system about vehicle position and orientation provides the automatic decision making intelligence needed to follow a virtual vehicle moving on track. The result shows good performance of this model independent algorithm with fairly cheap RC cars.
- 1 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 377-382
Control of autonomous cars for intelligent transportation system
This work concentrates on establishing a technology to develop a unique approach for the integration of intelligent system control into transportation systems. We have implemented an algorithm [1] that controls a car to track a predefined track which integrates a human driver with computer control to increase human performance while reducing reliance on detailed driver attention. Knowledge obtained from the optical tracking system about vehicle position and orientation provides the automatic decision making intelligence needed to follow a virtual vehicle moving on track. The result shows good performance of this model independent algorithm with fairly cheap RC cars.
- 1 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 372-376
Future of multiprocessors: Heterogeneous Chip Multiprocessors
As computer applications are becoming complex, large and versatile; the advent of Chip multiprocessor is ubiquitous. There are numerous researches going on about the core architectures within the chip. Heterogeneous Chip Multiprocessor (CMP) is leading the innovation. Heterogeneous CMP is composed of cores of varying performance, and complexity. It gives better area to performance ratio, high throughput, and higher speed up and mitigates Amdahl's bottleneck to some extent. There are three major issues to Heterogeneous CMP- scheduling applications to different cores, configuration of cores, and Amdahl's law utilization. This paper discusses recent researches of these three issues in details and finally, some recommendations are drawn from the study.
- 1 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 372-376
Future of multiprocessors: Heterogeneous Chip Multiprocessors
As computer applications are becoming complex, large and versatile; the advent of Chip multiprocessor is ubiquitous. There are numerous researches going on about the core architectures within the chip. Heterogeneous Chip Multiprocessor (CMP) is leading the innovation. Heterogeneous CMP is composed of cores of varying performance, and complexity. It gives better area to performance ratio, high throughput, and higher speed up and mitigates Amdahl's bottleneck to some extent. There are three major issues to Heterogeneous CMP- scheduling applications to different cores, configuration of cores, and Amdahl's law utilization. This paper discusses recent researches of these three issues in details and finally, some recommendations are drawn from the study.
- 1 July 2009
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 100-103
FreePDK v2.0: Transitioning VLSI education towards nanometer variation-aware designs
This paper discusses an extension to an open source, variation aware process design kit (PDK), based on scalable CMOS design rules. This PDK is designed for 45 nm feature sizes and is utilized for use in VLSI research, computer architecture, education and small businesses. This kit includes all the necessary layout design rules and extraction command decks to capture layout dependent systematic variation and perform statistical circuit analysis. The kit also includes a standard cell library, MIPSreg processor and associated GNU-compliant compiler and the necessary support files to enable full chip place and route and verification for system on chip designs. An analog and digital system test chip is also included with this PDK-extension allowing exploration of nanometer-based VLSI designs.
- 1 July 2009
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 100-103
FreePDK v2.0: Transitioning VLSI education towards nanometer variation-aware designs
This paper discusses an extension to an open source, variation aware process design kit (PDK), based on scalable CMOS design rules. This PDK is designed for 45 nm feature sizes and is utilized for use in VLSI research, computer architecture, education and small businesses. This kit includes all the necessary layout design rules and extraction command decks to capture layout dependent systematic variation and perform statistical circuit analysis. The kit also includes a standard cell library, MIPSreg processor and associated GNU-compliant compiler and the necessary support files to enable full chip place and route and verification for system on chip designs. An analog and digital system test chip is also included with this PDK-extension allowing exploration of nanometer-based VLSI designs.