Posted on: March 22, 2017
Date: March 28th, 2017
Time: 12:00pm - 1:00pm
Location: Steinman T107, Grove School of Engineering, City College of New York
Remote sensing is increasingly employed in environmental monitoring studies. A wide variety of analysis methods and data are used to monitor status and changes in land, water, and atmosphere. The selection of the appropriate methods and data for each task is rarely straightforward and depends on various factors and trade‐offs. In this talk, I will present a number of use cases, where we involved different methods and data tailored to the particular characteristics of each application. First, I will discuss a machine learning‐based methodology to estimate vegetation height based only on texture features of a single multispectral image. Next, I will present our recently developed optical flow‐based motion estimation method for fine‐scale Arctic sea ice monitoring, and its combination with super‐resolution techniques. Finally, I will demonstrate preliminary results of sea ice motion prediction based on cutting‐edge deep learning techniques.
Provide your email address and we'll notify you of our news and events