9 February 2006 Statistical learning with imbalanced training set in a machine vision application: improve the false alarm rate and sensitivity simultaneously
Author Affiliations +
Abstract
We studied the statistical learning methods with imbalanced training data sets. Imbalanced training sets are very common in industrial machine vision applications. The minority class contains the defects or anomalies we try to catch. The majority class contains the "regular" objects. We need a method that performs well at both false positive and false negative error rates. Traditional methods such as classification tree yield unsatisfactory results. We propose a two-stage classification scheme. We first use a subset selection method to remove redundant examples from the majority class. As a result, the training sample becomes more balanced without losing critical boundary information. The computation-intensive methods such as boosted classification trees are then applied to further improve both error rates.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jonathan Qiang Li, Jonathan Qiang Li, } "Statistical learning with imbalanced training set in a machine vision application: improve the false alarm rate and sensitivity simultaneously", Proc. SPIE 6070, Machine Vision Applications in Industrial Inspection XIV, 607002 (9 February 2006); doi: 10.1117/12.643444; https://doi.org/10.1117/12.643444
PROCEEDINGS
5 PAGES


SHARE
Back to Top