Authors: Yuhong He*, University of Toronto Mississauga, Bing Lu, University of Toronto Mississauga, Jian Yang, University of Toronto Mississauga, Cameron Proctor, University of Toronto Mississauga
Topics: Remote Sensing, Natural Resources
Keywords: High spatial optical remote sensing, vegetation
Session Type: Paper
Start / End Time: 5:20 PM / 7:00 PM
Room: Lafayette, Marriott, River Tower Elevators, 41st Floor
Presentation File: No File Uploaded
The increasing availability of high spatial resolution optical sensors has provided a novel data sources to answer environmental questions with an unprecedented level of detail. However, increased spatial resolution exasperates inter-class and intra-class variability. Mixed pixels are increasingly common, as the number of detectable entities or classes increases with spatial resolution. Traditional information extraction approaches are ill-suited due to the high data volume, spurring the need for developing innovative image processing techniques. To effectively utilize the data contained in high spatial resolution imagery, some key questions must be addressed, including: (1) how can useful information be extracted from high spatial resolution images for a specific application? (2) will multi-temporal high spatial resolution images improve the quantification and characterization of vegetative processes? and (3) what are the suitable methods to evaluate accuracy of the fine-level information extracted from high spatial resolution images? My talk will discuss multi-temporal high spatial resolution surveys of grasslands and forests, with a focus on the opportunities and challenges provided by multispectral and hyperspectral equipped drones, manned helicopters, and satellites.