Lecturer:Yuan Xiaohui

School:University of North Texas

Organizer:School of Software,Hefei University of Technology

Report topic:Classification of urban functions based on social media data

Report introduction:

The classification of urban land function is generally realized through field investigation or remote sensing technology. This classification method is time-consuming and labor-intensive. With the widespread development of social networks such as Weibo, images of cities are taken by residents or tourists and uploaded to the Internet. These images provide a new source of data for urban land use classification. In this paper, we propose an automatic method to recognize the public geographic indication images in the network platform, so as to realize the classification of urban land use. Using the classification of streets, we divide the city into different levels of districts. Thus, the recognition of unevenly distributed street view images is extrapolated to different levels of blocks. The experiment was conducted using geotagged photos of downtown areas and OpenStreetMap. The proposed method achieves an overall recognition accuracy of 76.5% for five types of land use. More importantly, our approach deals with a mix of land use types.

Introduction to the Reporter:

Yuan Xiaohui has been appointed tenured Professor at the University of North Texas, Director of the Computer Vision and Intelligent Systems Laboratory, and a Senior Member of IEEE. He received his Bachelor's degree in Electrical Engineering from Hefei University of Technology in 1996 and his doctorate from Tulane University in 2004. In the same year, he joined the National Institutes of Health in medical imaging and image analysis. He was hired as an assistant professor at the University of North Texas in 2006 and tenured as an associate professor in 2012. He founded and directs the Computer Vision and Intelligent Systems Lab at the University of North Texas. His research focuses on computer vision, data mining, machine learning, and artificial intelligence. To date, he has led more than 20 research projects supported by the U.S. Air Force Laboratory, the U.S. Science Foundation, the Texas Advanced Research Project, the Oak Ridge University Consortium, the University of North Texas, and technology companies. His findings have been published in 200 international conferences and journals. Among them, more than 80 papers were retrieved by SCI and more than 70 by EI. In Google Scholar, h-index: 35, i10-index: 77. Professor Yuan received the Ralph E. Powe Professor Award in 2008 and the U.S. Air Force Visiting Professor Award in 2011, 2012, and 2013. He has also received two research awards and a teaching award from the University of North Texas in 2007, 2008, and 2012, and was inducted into Who's Who in America in 2010. He has served as associate editor, editorial board member, and guest editorial board member in several International journals, including IAES International Journal of Artificial Intelligence, Journal of Biomedical Graphics and Computing, American Journal of Science and Engineering, Journal of Supercomputing. He has also served as an organizing member and conference chair at several international conferences, Including International Conf. on Urban Intelligence and Applications, IEEE BigDataService, ICCCNT, IEEE GRC, BIBM, etc. He has been invited to serve as a project judge for scientific research projects at NASA, NSF, NIH and the Louisiana State Board of Education. He has supervised 5 PHDS, graduated 6 PHDS, 4 masters (thesis), more than 40 course masters, and 24 visiting scholars from China, Egypt, India and other countries.