GEOG Seminar 10/9: Sergii Skakun, “The Impact of Map Accuracy on Area Estimation ...”

Join us at 3:30 p.m. on Thursday, Oct. 9 for this seminar at River Road and via Zoom with Associate Professor Sergii Skakun.

Associate Professor Sergii Skakun will discuss "the impact of map accuracy on area estimation with remotely sensed data within the stratified random sampling design."

Dr. Skakun is an Associate Professor with a joint appointment at the Department of Geographical Sciences and the College of Information (INFO). His current research focus is to advance methods, models, and emerging technologies in the area of data science for heterogeneous remote sensing data fusion, processing, and analysis, as well as their applications to Earth System Science and areas of societal benefit. Within the Committee on Earth Observation Satellites (CEOS), he is co-leading a Cloud Masking Inter-comparison Exercise. He is currently an Associate Editor for the journal Remote Sensing of Environment.

Satellite-based classification maps are essential for area estimation but inevitably contain errors stemming from imperfect inputs, incomplete coverage, and class confusion, making pixel-counting a biased estimator. Sample-based approaches aim to provide a means for statistical inference from the maps. One such approach is a stratified random sampling design, in which classification maps could be used for stratification in the sampling design, and areas are estimated from the sample data. In this seminar, Dr. Skakun will show how map quality impacts the efficiency of stratification. A more accurate map will require a smaller sample size to reach the target variance of the estimate, or it will yield improved precision if the sample size is fixed. The findings provide a quantitative basis for benchmarking classification algorithms when area estimation is the primary objective.

Location: For Zoom details, please visit the GEOG Department Calendar.