Man vs. machine: New Leicester research to compare human eye-tracking with AI in first study of its kind

The human brain’s ability to track and process hazards will be pitted against an innovative artificial intelligence (AI) tool in a new interdisciplinary research study at the University of Leicester.

Researchers in the School of Computing and Mathematical Sciences have combined efforts with visual perception experts from Leicester’s Department of Neuroscience, Psychology and Behaviour to shape the study, which will track participants’ eye movements when faced with multiple fast-moving hazards simultaneously.

The research will be used to inform the next generation of a rail safety device being developed as part of an Innovate UK Knowledge Transfer Partnership (KTP) between Leicester experts and systems engineering specialists Synoptix.

The OPTIMUS prototype – installed at a Network Rail-operated level crossing near Cheltenham earlier this year – uses machine learning and an AI-based object detection system, hosted locally on the small edge-based device, to identify and quantify different types of traffic.

Now, this new interdisciplinary aspect of the project will allow researchers to compare both the accuracy and speed of its detection capability to a human completing the same task. It is believed that the study is the first of its kind in comparing humans and AI for the task of ‘visual census’ in this way.

George Leete is KTP Research Associate within the Artificial Intelligence, Data Analytics, and Modelling (AIDAM) Centre at the University of Leicester, and leads development of the machine learning aspect of the project under the supervision of Professor Ivan Tyukin. He said: “The question was asked on the project ‘How good is good?’, in relation to the performance of the OPTIMUS system at the crossing. I’m confident that the results of this study will lend us some valuable insight into this question, firstly how good humans are at the current task, and secondly whether our current system holds up to human standards. We believe this is the first time an AI-based system will be validated against a standard in this way, opening the door for other systems to be validated via a similar method.”

Dr David Souto, Dr Doug Barrett and Dr Claire Hutchinson make up the team of Leicester psychologists lending their expertise to the study.

Dr Claire Hutchinson, Associate Professor in Experimental Psychology at the University of Leicester said: “We are really excited to be involved in this interdisciplinary work. Comparing AI and human performance in this way will help us understand how the human brain identifies and weights visual information so that we can safely navigate the world around us.”

Stephen Measures, Project Lead at Synoptix, added: “Synoptix are delighted to have the opportunity to work with both Leicester’s School of Computing and Mathematical Sciences and the Department of Neuroscience, Psychology and Behaviour on this breakthrough study. The results of this will be crucial to help our project team validate the performance and effectiveness of the Optimus prototype system and its AI model, when compared to a human based approach to object census.”

Justin Mountjoy, Programmes Director at Synoptix, added: “We feel very privileged to be part of this Collegiate working group as we seek to push the boundaries of this project which has the potential to be a market leader in AI and Data Analytics; not only will it have a huge impact on Artificial Intelligence technologies, but it has real significance in being a fully interactive element to increase safety across Network Rail’s infrastructure.”

According to Network Rail, which is responsible for the country’s rail infrastructure, there are around 6,000 level crossings in the UK. Figures for 2019/20 show that there were 316 near misses with pedestrians on UK level crossings, and two pedestrian fatalities.

The OPTIMUS prototype was installed at a site on the Cross Country Route in January 2022 and has already identified hundreds of thousands of movements on the crossing, including pedestrians, cyclists and other road traffic. As the identification and categorisation of users occurs locally on the device – and the only data transmitted is of traffic numbers and types – the privacy of crossing users is protected in line with data protection guidelines.

Synoptix provides multi-disciplinary systems engineering support across various technical industries. The organisation is certified by the Railway Industry Supplier Qualification Scheme (RISQS) for the provision of Systems and Safety Engineering Consultancy to the Rail Industry. The KTP is funded by Innovate UK.