Publication Researches: 

Conference Publications:

1. "Enhancing Bangla Local Speech-to-Text Conversion Using Fine-Tuning Wav2vec 2.0 with OpenSLR and Self-Compiled Datasets Through Transfer Learning.". 7th IEOM Bangladesh International Conference on Industrial Engineering and Operations Management. Link

2. "A Comprehensive Approach to Gold Price Prediction Using Machine Learning and Time Series Models" – Accepted at the 3rd IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IEEE IATMSI-2025). 

Journal Publications: 

1. "Enhanced Brain Tumor Detection Using YOLOv11 with Attention Mechanisms and Lightweight Design for Real-Time Applications" ( Submitted in Scientific Research )

Book Chapters:

1.  "Innovative Approaches to Tomato Leaf Disease Detection Bridging Tradition and Technology" (Submitted in IGI Global )

Accepted Paper PDF

"Saving Lives at Sea with Drones: Adaptive Swimmer Detection for Safer Rescue" (Accepted in ICCIT-2025)

"An Explainable Ensemble Learning Framework for Crop Classification with Optimized Feature Pyramids and Deep Networks" (Accepted in ICCIT-2025)"

Ongoing Researches: 

Accepted Manuscripts 

• Advancements in EEG Signal Decoding for Cognitive and Motor Task Classification: This study explores advancements in EEG signal decoding for cognitive and motor task classification, focusing on signal preprocessing, feature extraction, and classification techniques. A robust framework employing both traditional machine learning and deep learning models was implemented, demonstrating the efficacy of hybrid architectures like CNN-LSTM in capturing spatial and temporal features. Notable results include high classification accuracies across various tasks, achieved through optimized model training and preprocessing strategies. The study underscores the significance of incorporating techniques such as Principal Component Analysis and Discrete Wavelet Transform for feature reduction and time-frequency analysis, respectively. These findings highlight the potential of EEG-based BCIs to improve communication and control for individuals with neurological impairments, while addressing challenges in signal interpretation and system adaptability. Future work aims to enhance the scalability and robustness of BCI systems for real-world applications. Advancements in EEG Signal Decoding for Cognitive and Motor Task Classification

• Robust Object Detection Using Underwater Imaging to Investigate Various Activities : Object Detection in underwater environment still an unsolved research problem while it is fully integrated into the mapping payload of the Autonomous Underwater Vehicle (AUV). Noise issues due to various intensity scattering and absorption present huge challenge for underwater imaging. Aim of this research is to propose enhanced method to improve underwater imagery for various object detection. Expected outputs of this research intends to facilitate investigation of human activities underwater for various life search for later classification purpose. 

Ongoing Researches 

• An Enhanced Model for Bengali Dialect Speech-to-Text System

Description: Developing a robust model for dialect-specific speech recognition in Bengali.
👥 Collaborators: Dr. Muhammad Firoz Mridha, Dipta Gomez, Md. Khairul Alam Mazumder, Rahul Biswas. 

• Performance Analysis Model for Brain Tumor Detection

Description: Analyzing and improving model performance for brain tumor detection using advanced techniques.
👥 Collaborators: Dr. Muhammad Firoz Mridha, Dipta Gomez, Md. Khairul Alam Mazumder

• Movie Plot Building and Analysis

 Description: Designing systems for movie plot generation and analysis using NLP and Machine Learning.
👥 Collaborator: Dr. Debajyoti Karmaker

• Optimizing Image Processing with a Proposed Model and Enhanced Loss Function

Description: Proposing a novel image processing model optimized with advanced loss functions.
👥 Collaborator: Dr. Debajyoti Karmaker

• A Comprehensive Review of Vision-Language Models: Advances, Challenges, and Future Directions

 Description: This paper provides an in-depth review of Vision-Language Models (VLMs), analyzing recent advancements, key challenges, and emerging trends. It explores state-of-the-art architectures, benchmark datasets, multimodal learning techniques, and real-world applications, offering insights into the future development of VLMs.
👥 Collaborator:  Kishor Morol

• Advancing Guava Disease Detection Using AI-Driven Image Analysis

Description: This research focuses on developing an AI-driven approach for detecting and classifying guava diseases using advanced image processing techniques and machine learning models. By leveraging deep learning and explainable AI, the study aims to enhance precision in disease identification, aiding farmers in early diagnosis and crop management.
👥 Collaborators: Dr. Muhammad Firoz Mridha, Dipta Gomez, Md. Khairul Alam Mazumder