Alzayat Saleh

Computer Scientist · Computer Vision & Deep Learning

Alzayat Saleh

Research Fellow, James Cook University, Australia · PhD in Computer Science

I’m a computer scientist working in computer vision and deep learning. I build systems that work outside the lab, and apply them in different domains: murky water, bright paddocks and clinical scans. My core research asks how to train reliable models when labelled data is scarce, using self-supervised, semi-supervised and weakly supervised learning, and how to make those models small and fast enough to deploy on real equipment.

Open-source tools and side projects

Favabeansfavabeans.app Calm, free apps and games for everyday life. No accounts, no ads, and your data stays on your device. Ask my paperspapers.alzayat.com An AI assistant that answers questions about my research, with links to the exact page in each paper. UNIQATOpen-source Python toolkit · with AIMS Scores the quality of underwater survey images automatically, at 7.9 4K images per second. Published in Methods in Ecology and Evolution. DeepFishOpen dataset and benchmark code 40,000 underwater images with labels for counting, locating and segmenting fish. UDnetOpen-source PyTorch code · MIT licence Makes murky, blue-green underwater photos clear, and learns without clean reference images. Tested on eight public datasets. Published in Expert Systems with Applications.

Research

Across all of it Learning with few labels (self-, semi- and weakly supervised), realistic benchmark datasets, and lightweight networks for deployment in the field.

Publications

26 papers, including 20 journal articles. Full record on Google Scholar.

2026

  1. UNIQAT: An open-source toolkit for reproducible image quality assessment in marine surveys

    A. Saleh, A. Chennu

    Methods in Ecology and Evolution

    DOI Scholar

  2. Lessons from Building a Semantic 3D Gaussian Splatting Pipeline for Multi-Year Coral Reef Monitoring

    A. Saleh, K. Han, R. Ferrari

    RSS 2026 Workshop: Mapping the Reef: Underwater 3D Reconstruction for Coral EcosystemsConference

    Best Application Award

    Paper Scholar

  3. Weed Detection in Challenging Field Conditions: A Semi-Supervised Framework for Overcoming Shadow Bias and Data Scarcity

    A. Saleh, S. Hatano, M. Rahimi Azghadi

    Computers, vol. 15, pp. 171

    Scholar

  4. Depth-Resolved Coral Reef Thermal Fields from Satellite SST and Sparse In-Situ Loggers Using Physics-Informed Neural Networks

    A. Saleh, M. Rahimi Azghadi

    PreprintPreprint

    Paper Scholar

  5. Linking fish activity and turbidity through visual and sensor data fusion and deep learning

    M. Jahanbakht, A. Tiernan, A. Saleh, N. Stokes, O. Langham, M. R. Azghadi, N. J. Waltham

    Marine Pollution Bulletin, vol. 223, pp. 119070

    DOI Scholar

  6. Multimodal language models in agriculture: A tutorial and survey

    M. Haghighat, A. Saleh, M. R. Azghadi

    Information Fusion, vol. 129, pp. 104042

    DOI Scholar

  7. Multi-label Instance-level Generalised Visual Grounding in Agriculture

    M. Haghighat, A. Saleh, M. Rahimi Azghadi

    European Conference on Computer Vision, pp. 448–467Conference

    Scholar

  8. Artificial intelligence-enabled aquacultural engineering systems: A design-oriented review of sensing, modelling, and deployment architectures

    V. Sankar, A. Saleh, A. Ehrampoosh, P. Arbon, D. R. Jerry, M. R. Azghadi

    Computers and Electronics in Agriculture, vol. 256, pp. 112353

    Scholar
2025

  1. Semi-supervised weed detection for rapid deployment and enhanced efficiency

    A. Saleh, A. Olsen, J. Wood, B. Philippa, M. R. Azghadi

    Computers and Electronics in Agriculture, vol. 236, pp. 110410

    DOI Scholar

  2. Adaptive deep learning framework for robust unsupervised underwater image enhancement

    A. Saleh, M. Sheaves, D. Jerry, M. R. Azghadi

    Expert Systems with Applications, vol. 268, pp. 126314

    DOI Scholar

  3. Overcoming annotation bottlenecks in underwater fish segmentation: a robust self-supervised learning approach

    A. Saleh, M. Sheaves, D. Jerry, M. R. Azghadi

    Signal, Image and Video Processing, vol. 19, pp. 1–10

    DOI Scholar

  4. FieldNet: Efficient real-time shadow removal for enhanced vision in field robotics

    A. Saleh, A. Olsen, J. Wood, B. Philippa, M. R. Azghadi

    Expert Systems with Applications, vol. 279, pp. 127442

    DOI Scholar

  5. Precision robotic spot-spraying: Reducing herbicide use and enhancing environmental outcomes in sugarcane

    M. R. Azghadi, A. Olsen, J. Wood, A. Saleh, B. Calvert, T. Granshaw, E. Fillols, B. Philippa

    Computers and Electronics in Agriculture, vol. 235, pp. 110365

    DOI Scholar
2024

  1. Applications of deep learning in fish habitat monitoring: A tutorial and survey

    A. Saleh, M. Sheaves, D. Jerry, M. R. Azghadi

    Expert Systems with Applications, vol. 238, pp. 121841

    DOI Scholar

  2. Prawn morphometrics and weight estimation from images using deep learning for landmark localization

    A. Saleh, M. M. Hasan, H. W. Raadsma, M. S. Khatkar, D. R. Jerry, M. R. Azghadi

    Aquacultural Engineering, vol. 106, pp. 102391

    DOI Scholar

  3. WeedCLR: Weed contrastive learning through visual representations with class-optimized loss in long-tailed datasets

    A. Saleh, A. Olsen, J. Wood, B. Philippa, M. R. Azghadi

    Computers and Electronics in Agriculture, vol. 227, pp. 109526

    DOI Scholar

  4. How to track and segment fish without human annotations: a self-supervised deep learning approach

    A. Saleh, M. Sheaves, D. Jerry, M. Rahimi Azghadi

    Pattern Analysis and Applications, vol. 27, pp. 4

    DOI Scholar

  5. Learning from the Giants: A Practical Approach to Underwater Depth and Surface Normals Estimation

    A. Saleh, M. Olsen, B. Senadji, M. R. Azghadi

    arXiv preprint arXiv:2410.02072Preprint

    Paper Scholar
2023

  1. MFLD-net: a lightweight deep learning network for fish morphometry using landmark detection

    A. Saleh, D. Jones, D. Jerry, M. R. Azghadi

    Aquatic Ecology, vol. 57, pp. 913–931

    DOI Scholar
2022

  1. Computer vision and deep learning for fish classification in underwater habitats: A survey

    A. Saleh, M. Sheaves, M. Rahimi Azghadi

    Fish and Fisheries, vol. 23, pp. 977–999

    DOI Scholar
2021

  1. Weakly supervised underwater fish segmentation using affinity LCFCN

    I. H. Laradji, A. Saleh, P. Rodriguez, D. Nowrouzezahrai, M. R. Azghadi, D. Vazquez

    Scientific reports, vol. 11, pp. 17379

    DOI Scholar

  2. A deep learning localization method for measuring abdominal muscle dimensions in ultrasound images

    A. Saleh, I. H. Laradji, C. Lammie, D. Vazquez, C. A. Flavell, M. R. Azghadi

    IEEE Journal of Biomedical and Health Informatics, vol. 25, pp. 3865–3873

    DOI Scholar
2020

  1. A realistic fish-habitat dataset to evaluate algorithms for underwater visual analysis

    A. Saleh, I. H. Laradji, D. A. Konovalov, M. Bradley, D. Vazquez, M. Sheaves

    Scientific reports, vol. 10, pp. 14671

    DOI Scholar
2019

  1. Underwater fish detection with weak multi-domain supervision

    D. A. Konovalov, A. Saleh, M. Bradley, M. Sankupellay, S. Marini, M. Sheaves

    2019 international joint conference on neural networks (ijcnn), pp. 1–8Conference

    DOI Scholar

  2. Automatic weight estimation of harvested fish from images

    D. A. Konovalov, A. Saleh, D. B. Efremova, J. A. Domingos, D. R. Jerry

    2019 Digital image computing: Techniques and applications (DICTA), pp. 1–7Conference

    DOI Scholar
2018

  1. Estimating mass of harvested Asian seabass Lates calcarifer from images

    D. A. Konovalov, A. Saleh, D. B. Efremova, J. A. Domingos, D. R. Jerry

    World Journal of Engineering and Technology, vol. 6, pp. 15

    DOI Scholar

Contact

I am open to collaborations in computer vision for ecology, aquaculture, agriculture and robotics.