Dr. Solano's seminar
Tuesday 19 February 2019 dalle ore 9:00 am alle ore 10:00 am
Room 6302 PRC Building
TITLE: A Method for the Analysis of Small Crop Fields in Sentinel-2 Dense Time Series
SPEAKER: Y.T. Solano Correa - Fondazione Bruno Kessler, Povo, TN
ABSTRACT: Satellite Image Time Series (STIS), like the Sentinel-2 (S2) ones, provide a large amount of information, due to the better compromise between temporal, spatial and spectral resolutions. The high revisit frequency and spatial resolution of S2 result in: i) increase of the probability to acquire cloud free images and ii) availability of detailed information for analyzing small objects. These characteristics become of interest in precision agriculture, where crop behaviors understanding benefits of dense SITS. In the past, information about agricultural practices have been collected over large regions and considering mixed/aggregated crops due to the poor trade-off between spatial and temporal resolutions. Products have been generated at low spatial resolution and daily basis; or at high spatial resolution and weekly/monthly basis. They are meaningful for large agricultural fields, whereas they are limiting when fields show small average size. In this context, S2 characteristics allow for both high spatial and temporal resolutions products. However, no automatic method exists able to: i) effectively separate small fields from each other in an unsupervised way; and ii) deal with irregularly sampled data in time. Thus, this seminar presents a method suitable for the analysis of small crop fields in S2 dense SITS that accounts for S2 characteristics. The method: (i) pre-processes the S2-SITS, (ii) fuses spatio-temporal information, (iii) analyzes the spatio-temporal evolution of the data; and (iv) extracts relevant spatio-temporal information. The effectiveness of the proposed method was corroborated by experiments carried out on S2-SITS acquired over an area located in Barrax, Spain.
Yady Tatiana Solano Correa received the Bachelor (B.S.) degree in Physics Engineering (honorable mention) at the University of Cauca, Cauca, Colombia, in 2011; and the Ph.D. (cum laude) in Communication and Information Technologies from the University of Trento, Dept. of Information Engineering and Computer Science (Trento, Italy), in 2018. She worked as a researcher for the research groups: Optics and Laser Group (GOL) and Environmental Studies Group (GEA) from 2009 to 2013. She is currently a postdoctoral researcher at the Remote Sensing for Digital Earth unit at FBK, Trento, Italy. She is a member of the RSLab at the University of Trento, Italy.
Her research interests include remote sensing environmental applications, change detection, both on medium resolution multispectral images (e.g. Landsat, Sentinel-2) and Very High Resolution (VHR) images, multitemporal analysis of short and long-time series, multisensor multitemporal image pre-processing and information extraction, pattern recognition and image classification. She works, and has worked, within the context of several projects with focus on analysing information for climate change and developing advanced change detection techniques for optical satellite time series data.
She is a referee for the IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING and IEEE GEOSCIENCE AND REMOTE SENSING LETTERS journals. She won the best student oral presentation award at the “MultiTemp 2017 Conference”, held in June 2017 in Bruges, Belgium.
HOST: Michele Dalponte
Download flyer here: