Skip to content
012 projects

Selected work

Work carried past the training run: to evaluation figures and conference material, or to a deployed application on hardware. Open an entry for the full case study.

  1. Discipline
    Computer Vision / Research
    Context
    Academic research and conference publication material

    ThoraxVision

    A computer vision research project for early tuberculosis detection from chest X-ray images, built around a DenseNet121 classifier and benchmarked against two other convolutional backbones.

    4,784 images · Local chest X-ray dataset

    • PyTorch
    • DenseNet121
    • ResNet50
    • VGG19
    • Image preprocessing
    • Augmentation
    • Hyperparameter experimentation
    • ROC / AUC
    • Confusion matrix
    • Grad-CAM
  2. Discipline
    Edge AI / Deployment
    Context
    Client project
    Timeline
    20 – 30 June

    MelonVision AI

    An AI-based melon plant detection system that runs inference on an ESP32-CAM and reports into a FastAPI service, taking a trained model through to a deployed, working application.

    1,250 images · Melon plant image dataset

    • MobileNetV2 FOMO
    • INT8 quantised
    • TensorFlow Lite
    • FastAPI
    • PostgreSQL
    • Web integration
    • ESP32-CAM
    • Docker
    • Linux VPS