Title :
Applying data mining to false alarm reduction in an aviation explosives detection system
Author :
Merzbacher, Matthew ; Gable, Todd
Author_Institution :
Morpho Detection, Inc., Newark, CA, USA
Abstract :
Data mining techniques were applied for the reduction of false positives in aviation explosives detection CT (computed tomography) imaging systems. An inductive post-detection classifier (PDC) was trained, implemented, and fielded. The PDC can only eliminate alarms generated from the existing detection system - it does not detect new alarms.
Keywords :
aerospace industry; computerised tomography; data mining; explosives; learning (artificial intelligence); object detection; pattern classification; aviation explosives detection; classifier voting; computed tomography; data mining; data set training; false alarm reduction; false positive; post detection classifier; Classification algorithms; Computed tomography; Correlation; Data mining; Detection algorithms; Explosives; Robustness; Data mining; aviation security; classification; image processing; robustness; voting;
Conference_Titel :
Security Technology (ICCST), 2010 IEEE International Carnahan Conference on
Conference_Location :
San Jose, CA
Print_ISBN :
978-1-4244-7403-5
DOI :
10.1109/CCST.2010.5678738
