Continued Discrimination Demonstration Using Advanced EMI Models at Live UXO Sites: Data Quality Assessment and Residual Risk Mitigation in Real Time
Abstract
This project demonstrated the capability of advanced electromagnetic induction (EMI) models to perform discrimination of unexploded ordnance (UXO) at live sites; to achieve a high probability of UXO discrimination in situations involving widespread clutter; to minimize the number of false positives in such cases; to identify all UXO with high confidence; to assess the quality of the data; and to provide a robust dig threshold that will minimize the risk to regulators. Specific technical objectives were to: 1. Use advanced physics-based EMI models to extract robust features that will allow reliable classification when starting from dynamic or cued EMI sensor data. Establish the validity and limitations of these advanced models, taking into account the number of objects in a given cell, their size and material heterogeneity, the geology, and the level of background noise. 2. Combine the advanced models with a statistical model-based approach to select robust classification feature vectors for a specific live UXO site that can reliably and effectively discriminate hazardous targets of interest (TOI) from nonhazardous items. 3. Deploy advanced EMI and statistical signal processing tools to assess EMI data quality onsite and provide a robust dig-threshold point; as a last step, use the different targets' extracted extrinsic parameters to mitigate the residual risk due to UXO and to increase the confidence of regulators at the site.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 2018
- Accession Number
- AD1084270
Entities
People
- Fridon Shubitidze