It is estimated that there are 110 million active land mines in over 70 countries, roughly one land mine for every 52 people. Another 110 million land mines are stored, ready to be used. Clearing mines is very dangerous work; for every 5,000 mines that are removed, one person is killed, and two more are are injured.
Dirk Gorissen demonstrates how to use machine learning to detect land mines from a drone-mounted ground-penetrating radar sensor. While this involves many challenges on the hardware front, Dirk focuses specifically on the machine-learning problem of identifying land mines from background clutter in GPR data—an extremely challenging but rewarding problem.
Dirk Gorissen is the head of R&D at Skycap and a consultant for the World Bank. Dirk has worked in research labs in the US, Canada, Europe, Africa and worked with a large range of industrial partners, including Rolls-Royce, BMW, ArcelorMittal, NXP, and Airbus. His interests span machine learning, data science, and computational engineering, particularly in the unmanned systems domains. Dirk holds master’s degrees in computer science and artificial intelligence and a PhD in computational engineering. After eight years in academia, he joined BAE Systems Research, where he worked on big data analysis, deep learning, integrated vehicle health management, and autonomous systems-related topics, before going into business for himself, dividing his time between the World Bank (where he advises the Tanzanian government), UAV startup Skycap (where he leads the R&D activities), and a number of other startups. Dirk is an active STEM Ambassador and organizer of the London Big-O Algorithms & Machine Learning meetups and is active in the Tech4Good/ICT4D space.
©2016, O’Reilly UK Ltd • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org
Apache Hadoop, Hadoop, Apache Spark, Spark, and Apache are either registered trademarks or trademarks of the Apache Software Foundation in the United States and/or other countries, and are used with permission. The Apache Software Foundation has no affiliation with and does not endorse, or review the materials provided at this event, which is managed by O'Reilly Media and/or Cloudera.