Improving UXO Detection and Discrimination in Magnetic Environments
Abstract
This final report for the SERDP project MM-1414 covers research results accomplished since the projects inception in 2005. The basic premise of the project is the theoretical understanding achieved through SERDP SEED project UX-1285, which predicted that spatially variable frequency-dependent susceptibility in soils hosting UXO can produce strong background response in both electromagnetic and magnetic data. Such responses can mask UXO signatures or produce false alarms. Improving UXO detection and discrimination in such strongly magnetic environments requires the characterization and removal of these responses. The overall objective has therefore been to understand and characterize the spatial variation and frequency dependence of soil magnetic susceptibility so effective means can be developed for characterizing and removing soil responses. We have modified and implemented two filtering approaches for UXO magnetic data in order to separate the response of strong magnetic geology from UXO anomalies. The two methods are an iterative Wiener filter approach, and a wavelet filter approach. Finally, we have modified an approach, originally developed under a separate project, for automatic detection of UXO anomalies in total-field magnetic data. The method is based on the concept of structural index (SI). We further developed the method by implementing it in combination with pre-processed data using the iterative Wiener and wavelet filters for removing strong geologic responses in the magnetic data. The goal of this component has been to better understand and develop practical procedures, appropriate to a geologic setting, for integrating new filtering technologies with recently developed methodologies for detecting UXO in magnetic environments. In parallel to development and implementation of filters for UXO magnetic data sets, MM-1414 utilized data inversion simulations to quantify the effectiveness of different processing techniques.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 2010
- Accession Number
- ADA572266
Entities
People
- Bruce Harrison
- Doug Oldenburg
- Len Pasion
- Remke Van Dam
- Richard Krahenbuhl
- Steve Billings
- Todd Meglich
- Yaoguo Li
Organizations
- Colorado School of Mines