APPENDIX A
Robust Statistics
This chapter introduces the subject of robust statistics, which is the branch of statistics that aims to accurately discriminate between inliers and outliers, thereby permitting soundly based models to be made of the incoming data. First, the chapter clarifies the notion of robustness and arrives at definitions of breakdown point and relative efficiency. Then it differentiates between the main types of estimator and analyzes the various types of influence function arising under the M-estimator heading. It points out that this category includes functions for determining the mean and median of a distribution, and shows why the median filter is far more robust than the mean filter in eliminating impulse noise. It goes ...
Get Computer and Machine Vision, 4th Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
