Image recognition is the ability of computer software to recognize objects, such as people and animals, by analyzing the patterns in optical images. Its applications have gained a large amount of popularity in the recent decade, such as face recognition for photographing, internet content filtering, advanced driving assistance system, and self-driving vehicles. All these applications will not be feasible without the recent development of neural networks and the computational capable hardware. In this case study, we will dive into the fundamental of a popular image classification structure – the Convolutional Neural Networks (CNN) – and implement it in Matlab to recognize hand-written digits.

© Copyright. McMaster University. May be used for educational purposes without written permission but with a citation to this source.

Download Case Study