Research Interests

My research spans machine learning and deep learning, with a focus on computer vision and image processing. This includes object detection, segmentation, tracking, pattern recognition, pose and depth estimation, and motion capture. I work with neural networks and attention mechanisms like CNNs, transformers (ViTs), convolutional attention, and models using cross-feature dependencies. I’m also interested in semi- and weakly-supervised learning approaches. I explore vision-language models and generative AI (VLMs, LLMs), and work with 3D and multimodal data—such as LiDAR, point clouds, volumetric imaging, and temporal data. Much of my work is applied in scientific and real-world contexts, including medical imaging, microscopy, diagnostic systems, anomaly detection, surveillance, and OCR.

Publications

• M. Almahasneh, A. Paiement, X. Xie, J. Aboudarham, J. Deng (2019). Solar Active Regions Localization over Multi-Spectral Observations. Machine Learning in Heliophysics.
• M. Almahasneh, A. Paiement, X. Xie, J. Aboudarham (2021). MLMT-CNN for Object Detection and Segmentation in Multi-Layer and Multi-Spectral Images. Journal of Machine Vision and Applications.
• M. Almahasneh, A. Paiement, X. Xie, J. Aboudarham (2021). Active Region Detection in Multi-Spectral Solar Images. International Conference on Pattern Recognition Applications and Methods (ICPRAM).
• M. Almahasneh, A. Paiement, X. Xie, J. Aboudarham (2022). MSMT-CNN for Solar Active Region Detection with Multi-Spectral Analysis. Springer Nature Computer Science.
• M. Almahasneh (2022). Localization in 3D Images Using Cross-Feature Correlation Learning. Doctoral Thesis, Swansea University. [Available at: confra.swansea.ac.uk]
• M. Almahasneh, B. Li, H. Cai, N. Rajabi, L. Davies, Q. Meng (2025). Herbicide Efficacy Prediction Based on Object Segmentation of Glasshouse Imagery. International Conference on Computer Vision Theory and Applications (VISAPP).
• M. Almahasneh, X. Xie, A. Paiement (2025). AttentNet for Pulmonary Lung Nodule Detection Using 3D Convolutional Attention and 3D Vision Transformer Networks. Springer Nature Computer Science.
• M. Almahasneh, B. Li, H. Cai, N. Rajabi, L. Davies, Q. Meng (2025). Deep Learning-Based Herbicide Activity Prediction Using Object Localization and Contrastive Analysis. Springer Computer Science (submitted).