Michelle Hanlon

Michelle Hanlon
Masters Student


BSc Hons in Biological Sciences (Physiology) ,UCC 2011, Postgraduate Higher Diploma in Science in Applied Computing, ITT. 2013

Project title:

Application of Deep Neural Networks to Computer Vision

Partner Company:


Project overview:

This project proposes to apply Deep Learning Networks, a recent development in Artificial Neural Network research, to the problem of image recognition in computer vision based inspection systems. Computer vision is widely used in the manufacturing industry in Ireland having applications in inspection and control systems. In the PCB manufacturing industry these systems are used to automatically inspect the PCB’s on a production line to ensure that all the components are correctly placed on a board while applications in the Pharmaceutical Industry include the inspection of blister packs, labels and even the packaging artwork for errors. These systems, while outperforming human operators in terms of accuracy are still some way from 100% error free. Artificial intelligence techniques and in particular artificial neural networks have long been used as a technique for classification of or for the recognition of the images produced by these vision systems. Artificial neural networks are a machine learning technique modelled on an understanding of how the human brain works. Terms like neurons, connections and reinforcement learning are all part of the vocabulary associated with these techniques. A recent development in the area of Neural Networks has been Deep Learning Networks which are a significant evolution and reported results suggest that these techniques significantly outperform more traditional techniques in image recognition making them an ideal candidate for this use case. This research is being supported by industry, namely Kostal (PCB manufacturer) and the project has also been discussed with Crest Solutions (packaging systems for the Pharmaceutical Industry). It is planned that images will be supplied by these companies to test the techniques developed on images produced under real world conditions. This research is strategically relevant to IT Tralee, the Intelligent Mechatronics and RFID (IMaR) technology gateway, research Priority B (Data Analytics) and Priority N (Manufacturing Competitiveness) and importantly has support from local industry. I also have an interest in mobile applications. I designed and built an Android app as part of ICT module of Higher Diploma , the purpose of which was to help type two diabetics in the management of their disease. As part of the H.dip , I also completed the work experience requirement of the same,at Dairymaster as part of their mobile application development team. I am also in the process of obtaining diplomas in Ipad and Iphone app development and Android mobile application programming, both via distance learning from Fitzwilliams Institute Dublin.