Successful detection of antipersonnel landmines often requires decision making on the basis of more than one decisive feature. The fusion of the available features should be done in a statistically optimal way. Theoretically this might be achieved by Bayesian-based criteria. However application of Bayes' criteria in general form is computionally difficult and may lead to overtraining of classification algorithms. To avoid this situation linear and quadratic fusion rules are often used in practice. However these rules produce optimal results only when the decisive feautures are normally distributed, which is not always guaranteed in practice. To improve the performance of the linear and quadratic detectors we suggest a normalization of the decisive features by means of Johnson's transform prior to the fusion. We describe the normalization algorithm and demonstrate that it improves performance of the linear and quadratic classifiers. The performance is judged in terms of the ROC curves.
|Title of host publication||Proceedings of the 3rd European Radar Conference and Proceedings of the European Microwave Week 2006|
|Place of Publication||Manchester|
|Number of pages||4|
|Publication status||Published - 2006|
|Event||3rd European Radar Conference and Proceedings of the European Microwave Week 2006 - Manchester, UK, Manchester, United Kingdom|
Duration: 13 May 2006 → 15 Sep 2006
|Conference||3rd European Radar Conference and Proceedings of the European Microwave Week 2006|
|Period||13/05/06 → 15/09/06|
- conference contrib. refereed
- Conf.proc. > 3 pag
Kovalenko, V., Yarovoy, A., & Ligthart, LP. (2006). Normalization of feature distributions for linear-Quadratic fusion in landmine detection using gpr. In s.n. (Ed.), Proceedings of the 3rd European Radar Conference and Proceedings of the European Microwave Week 2006 (pp. 257-260). Horizon House.