About me
Ferdous Sohel is a Professor in Information Technology at Murdoch University, Australia. He was a Research Assistant Professor/Research Fellow at The University of Western Australia. He was the inaugural lead of the Agricultural Technologies program of the Western Australian Agricultural Research Collaboration. His research interests include artificial intelligence, image processing, machine learning, computer vision, digital agriculture, and digital medicine. Prof Sohel is a recipient of many national, state, international, and industry research projects/grants. He is a recipient of the VC’s Early Career Investigators award (UWA) and a best PhD thesis medal from Monash University. He has won several best paper awards. He is an Associate Editor of IEEE Transactions on Multimedia, IEEE Signal Processing Letters, and Computers and Electronics in Agriculture journals. He is a Fellow of the Higher Education Academy and a senior member of the Australian Computer Society and IEEE.
Research: Ferdous has strong experience in research and leadership with engagements with academia and industry. His research focuses on empowering digital technologies with advanced artificial intelligence and multimodal sensing in the areas of computer vision and machine learning and transdisciplinary areas of agriculture, health, environment, and remote sensing.
Key areas: Computer vision, machine learning, artificial intelligence, digital agriculture and biosecurity, marine image analysis, and digital health & medicine.
Major contributions:
Artificial Intelligence and Machine Learning: AI and deep learning models for a raft of applications, addressing data scarcity (few-shot, zero-shot learning), class imbalance, missing and uncertain data, hierarchical classification, level of supervision, and explainability.
Computer Vision: 2D and 3D scene understanding, biometric recognition, action recognition, human-computer interactions, and robot vision.
Audiovisual Analysis: Audiovisual speech and speaker recognition, scene understanding, polyphonic event detection and recognition.
Digital Agriculture: Plant phenotyping, weed, pest, and disease detection, stress (frost, heat, salt, metal) analysis, 3D sensing, animal welfare monitoring, and animal feeding and growth analytics.
Spectral and Remote Sensing: Multispectral analysis for agricultural and mining solutions.
AI for Health and Medical: Eye health, phenotyping, and health informatics for cardiovascular disease, microscopic high-resolution imaging for cancer detection, and fine-grained anomaly detection.
AI for Environment and Biosecurity: Remote (drone and satellite) sensing for monitoring marine and land species; invasive and endangered species; and underwater imaging.