# Investigation of Initialization Strategies for the Multiple Instance Adaptive Cosine Estimator

## TL;DR

This paper explores new initialization methods for the MI-ACE algorithm in electromagnetic induction sensors, aiming to improve subsurface explosive hazard detection by enhancing detection accuracy and computational efficiency.

## Contribution

It proposes and evaluates novel initialization techniques for MI-ACE, demonstrating their impact on detection performance and processing speed.

## Key findings

- New initialization methods improve detection accuracy.
- Certain approaches increase processing speed.
- Performance varies with different initialization strategies.

## Abstract

Sensors which use electromagnetic induction (EMI) to excite a response in conducting bodies have long been investigated for subsurface explosive hazard detection. In particular, EMI sensors have been used to discriminate between different types of objects, and to detect objects with low metal content. One successful, previously investigated approach is the Multiple Instance Adaptive Cosine Estimator (MI-ACE). In this paper, a number of new initialization techniques for MI-ACE are proposed and evaluated using their respective performance and speed. The cross validated learned signatures, as well as learned background statistics, are used with Adaptive Cosine Estimator (ACE) to generate confidence maps, which are clustered into alarms. Alarms are scored against a ground truth and the initialization approaches are compared.