Statistics for Imaging and Sensor Data

Level: Introductory Length: 7 hours Format: In-Person Lecture Intended Audience: This course is intended for participants who need to incorporate fundamental statistical methods in their work with imaging data. Participants are expected to have some experience with analyzing data. Description: The purpose of this course is to survey fundamental statistical methods in the context of imaging and sensing applications. You will learn the tools and how to apply them correctly in a given context. The instructor will clarify many misconceptions associated with using statistical methods. The course is full of practical and useful examples of analyses of imaging data. Intuitive and geometric understanding of the introduced concepts will be emphasized. The topics covered include hypothesis testing, confidence intervals, regression methods, and statistical signal processing (and its relationship to linear models). We will also discuss outlier detection, the method of Monte Carlo simulations, and bootstrap. Learning Outcomes: - explain the basics of statistical signal processing and its relationship to linear regression models - implement the methodology of Monte Carlo simulations - construct confidence intervals for a variety of imaging applications - fit predictive equations to your imaging data - apply the statistical methods suitable for a given context - construct confidence and prediction intervals for a response variable as a function of predictors - demonstrate the statistical significance of your results based on hypothesis testing - perform correct analysis of outliers in data Instructor(s): Peter Bajorski is Professor of Statistics at the Rochester Institute of Technology. He teaches graduate courses in statistics including a course on Multivariate Statistics for Imaging Science. He also designs and teaches short courses in industry, with longer-term follow-up and consulting. He performs research in statistics and in hyperspectral imaging. Dr. Bajorski wrote a book on Statistics for Imaging, Optics, and Photonics published in the prestigious Wiley Series in Probability and Statistics. He is a senior member of SPIE and IEEE. Event: SPIE Defense + Commercial Sensing 2019 Course Held: 14 April 2019