Skill Morph Research Lab.
Skill Morph Research Lab.
Embedded Files
Skill Morph Research Lab.
Skill Morph Research Lab. is a visionary research lab in Bangladesh, dedicated to transforming intelligent ideas into high-impact, globally recognized research. Founded with a passion for scientific excellence and academic innovation, the lab stands as a beacon for aspiring researchers, students, and professionals who seek to make a meaningful contribution to the world of knowledge.
At Skill Morph Research Lab., we foster an ecosystem where deep curiosity meets rigorous experimentation, driving groundbreaking discoveries in Artificial Intelligence, Machine Learning, Deep Learning, Federated Learning, and Data Science. We believe that great research is not born in isolation, it thrives in a culture of collaboration, mentorship, and shared purpose.
One of our core strengths is our ability to bridge Data Science with virtually any academic domain or subject. Whether your background is in Medicine, Public Health, Agriculture, Business, Economics, Social Science, Engineering, or any other field, we help you harness the power of Data Science to conduct rigorous, meaningful, and publishable research in your own domain. We believe that every field has data, and every piece of data has a story worth telling through research.
Our Mission
The mission of Skill Morph Research Lab. is to produce high-impact, globally recognized research that advances science, technology, and human knowledge, and to make the journey of research accessible to every curious mind, regardless of their academic background or prior experience.
We are committed to empowering students, early-career researchers, and professionals from all disciplines by equipping them with the tools, knowledge, and mentorship needed to conduct rigorous, meaningful, and publishable research. Through the power of Data Science, Machine Learning, Deep Learning, and Federated Learning, we help researchers from any domain, be it Medicine, Engineering, Agriculture, Business, Social Science, or beyond, transform their ideas into high-impact publications in the world's leading journals and conferences.
At the heart of our mission are four core commitments:
Research Excellence — We uphold the highest standards of scientific rigor, reproducibility, and academic integrity in every research project we undertake.
Inclusive Research Culture — We believe research belongs to everyone. We open the doors of high-impact research to all, irrespective of discipline, institution, or experience level.
End-to-End Guidance — We walk alongside every researcher from problem formulation to publication — providing complete, step-by-step supervision at every stage of the research journey.
Global Impact — We strive to publish in the world's most reputable venues, Scopus, SCI, Web of Science, IEEE, Elsevier, Springer, and Nature, ensuring our research reaches and influences the global scientific community.
Together, we work to build a generation of confident, capable, and published researchers who are ready to shape the future, one discovery at a time.
Our Vision
To be a leading research lab in Bangladesh and across the world, recognized globally for producing high-impact, data-driven research across diverse domains, and for nurturing a new generation of skilled, confident, and published researchers who shape the future of science and technology.
Message from Founder and Director
Message from Founder and Director
Welcome to Skill Morph Research Lab., a place where ideas are transformed into high-impact, globally recognized research.
We founded this lab with one dream: to make world-class research accessible to every student, researcher, and curious mind in Bangladesh and beyond.
Whether you are a student taking your first step into research, an experienced researcher seeking a collaborative and intellectually rich environment, or a university, research organization, or industry partner looking for meaningful joint initiatives, Skill Morph Research Lab. welcomes you. We guide, collaborate, and grow together to create real-world impact through high-quality, published research.
Bangladesh has the talent and potential to stand tall on the global research stage. Skill Morph Research Lab. is here to unlock that potential.
I personally invite you to come, research, publish, and grow with us.
Welcome to Skill Morph Research Lab., where passion for knowledge meets the pursuit of scientific excellence.
As a co-founder of this lab, my journey in academia and research has taught me one fundamental truth, that great research is not reserved for a privileged few. It belongs to anyone who is willing to learn, work hard, and stay committed. That belief is the foundation upon which Skill Morph Research Lab. was built.
Whether you are a student publishing your very first paper, a researcher seeking collaboration, or an institution looking for a meaningful research partnership, we are here for you. We are committed to guiding every individual through the complete research journey, with honesty, dedication, and academic integrity at every step.
I warmly invite you to be a part of our growing research community. Let us learn together, discover together, and contribute together to the advancement of science and technology — for Bangladesh and for the world.
Research Projects
Research Projects
(Advancing the Field of AI and Data Science)
Federated Learning Based UAV Intrusion Detection
Federated Learning Based UAV Intrusion Detection
As UAVs are increasingly deployed across critical domains, including military surveillance, disaster management, agriculture, and smart logistics, securing their communication networks has become a pressing challenge.
This project proposes a privacy-preserving, decentralized intrusion detection framework where UAV nodes collaboratively train a shared detection model without exchanging raw data, ensuring both data privacy and network security. Advanced Machine Learning and Deep Learning architectures are leveraged within a Federated Learning paradigm to detect a wide range of network intrusions and cyber threats in real time, even under non-IID data distributions across UAV nodes.
The outcomes of this research are expected to produce high-impact publications in leading Scopus-indexed journals and IEEE conferences.
Computer Vision Framework for Real-Time Object Detection and Classification in UAV Aerial Imagery
Computer Vision Framework for Real-Time Object Detection and Classification in UAV Aerial Imagery
As UAVs are increasingly used for surveillance, disaster response, precision agriculture, and infrastructure inspection, the ability to accurately detect and classify objects from aerial imagery in real time has become a critical research challenge.
This project proposes a robust Computer Vision framework powered by advanced Deep Learning architectures, including Convolutional Neural Networks (CNN), YOLO-based detectors, and attention mechanisms, to perform real-time object detection, segmentation, and classification from UAV-captured aerial images. The framework is designed to handle challenges unique to aerial imagery, including varying object scales, occlusion, complex backgrounds, and high-altitude distortion.
A Zero-Shot and Few-Shot Learning Framework for Generalized Image Classification Across Unseen Categories
A Zero-Shot and Few-Shot Learning Framework for Generalized Image Classification Across Unseen Categories
Traditional Deep Learning models require large amounts of labeled training data to perform well, a constraint that is often impractical in real-world scenarios where data is scarce, expensive to annotate, or entirely unavailable for certain categories.
This project proposes a generalized Zero-Shot and Few-Shot Learning framework that enables models to recognize and classify images from unseen or rarely seen categories by leveraging semantic embeddings, meta-learning strategies, and knowledge transfer techniques. By learning transferable visual representations from seen classes, the model generalizes effectively to new categories with little to no labeled examples, mimicking the way humans learn from minimal information.
Publications
Publications
(Published Research Articles)
A novel unified lightweight temporal-spatial transformer approach for intrusion detection in drone networks (Scientific Reports, Q1)
A novel unified lightweight temporal-spatial transformer approach for intrusion detection in drone networks (Scientific Reports, Q1)
Tarun Kumar Biswas, Ashrafun Zannat, Waqas Ishtiaq, Md. Alamgir Hossain
An Ensemble Learning Approach for Multi-Class Intrusion Detection with Comparative Performance and Explainability Analysis (International IEEE Conference)
An Ensemble Learning Approach for Multi-Class Intrusion Detection with Comparative Performance and Explainability Analysis (International IEEE Conference)
Md. Shafiqul Islam, Md. Alamgir Hossain, Md. Samiul Islam
Advanced explainable ensemble models for multi-class intrusion detection in heterogeneous drone and industrial networks (Journal on Information Security, Q2)
Advanced explainable ensemble models for multi-class intrusion detection in heterogeneous drone and industrial networks (Journal on Information Security, Q2)
Md. Shafiqul Islam, Faraz Ahmed, Waqas Ishtiaq, Md. Alamgir Hossain, Md. Samiul Islam, Muhammad Masud Tarek
Join Our Research Community
We are always looking for passionate, curious, and dedicated individuals to collaborate with us and advance the frontiers of Artificial Intelligence, Machine Learning, Data Science, and beyond.
Page updated
Google Sites
Report abuse