Remote Monitoring of our Environment: Its Really a Data Fusion Problem

Remote monitoring of our environment can be done, even in the harshest conditions, using a combination of sensors. The health and wellness of a river can be monitored by measuring water depth, turbidity, pH, conductivity, and nutrient content (nitrogen and phosphate), some of these being cheap and off-the-shelf, others being more expensive, and all suffering from problems of calibration and biofouling. Similarly for a coastal area like a bay, we can measure sea surface temperature, wave height, chlorophyll content and water composition with the same issues of calibration and biofouling.  Each of the sensors we use to monitor environments, especially water-based environments, produce streams of data values which need to be woven together in order to get an holistic overview of the health of the area being monitored. In this presentation I will describe how environmental monitoring using multiple sensor streams is really a problem of data fusion where each of the incoming data streams has accuracy and reliability issues. I will also describe how we have developed, and deployed, a trust and reputation framework to address these issues and how we have put this into effect in remote monitoring of two coastal regions.

by Prof. Alan F. Smeaton, Director, Insight Centre for Data Analytics, DCU

Since 1997, Alan Smeaton has been a Professor of Computing at Dublin City University where he has previously been Head of School and Dean of Faculty. His early research covered the application of natural language processing techniques to information retrieval and this broadened to cover content-based retrieval of information in all media, text, image, audio and digital video. From 2013 to 2019 he is Director of the INSIGHT Data Analytics Research Centre at DCU, a €70m+ centre funded by SFI and by industry based mostly at 4 Irish Universities and with 30+ industry partners, representing the largest single investment in research made by the government of Ireland.  Alan has published over 550 refereed papers in journals, conferences or workshops, has more than 9,300 citations and a h-index of 45 (Google Scholar, February 2014).  Alan is co-founder and coordinator of TRECVid which, since 2001, has grown to be the largest multimedia benchmarking activity worldwide with about 80 participating research groups each year and over 1,200 individual researchers over the last decade. Alan is a member of the ACM, a Senior Member of the IEEE and a Fellow of the Irish Computer Society.