Deeper Networks for Image Classification
This is an analysis of Convolutional Neural Networks (CNNs) on the MNIST and CIFAR10 datasets. The project evaluates the performance of four models: InceptionV3, VGG16, ResNet50, and MobileNetV2
Selected work
Projects
Vision Toolkit
Vision Toolkit is a project showcasing various basic computer vision tools. The toolkit includes: 1. Image Rotation, 2. Convolution and Filters, 3. Color Histogram, 4. Similarity Descriptor, 5. Background Extractor
AI Route Finder
This project is an AI route finder for the London Tube using an agenda-based search mechanism. It implements and compares four algorithms: 1. Breadth-First Search (BFS), 2. Depth-First Search (DFS), 3. Uniform Cost Search (UCS), 4. Best-First Search (Heuristic Search)
Apes Anticipation Simulation
ROS package simulates an experiment to test if great apes can anticipate human false beliefs using a Bayesian network.
Fruit Fly Application
A Flutter mobile application that allows identification of fruit flies using Deep Learning.
Infinity Robot
An automatic seeding machine that makes use of a rover, a robot arm and sensors. This project won the first prize in UoM Hackathon 2020 and was presented for UoM Robotics Lab launching ceremony.