The identification of unexploded ordnance (UXO) is time-consuming and dangerous, requiring a thorough canvas of test ranges to support disposal or removal activities. Current UXO characterization technologies lack the ability to reliably discriminate UXO from nonhazardous materials, leading to a large number of false alarms. Technologies that have high detection rates and low false alarm rates are required to facilitate cleanup and in so doing minimize pollution. When UXO remain on a range, hazardous materials in the casing have the long-term potential to enter the ecosystem and create sources of solid waste or leached hazardous material.

 The objective of this project was to develop optical augmentation for use in identifying UXO on the surface of a test range. Optical augmentation makes the surface munitions as conspicuous as possible even when obscured by vegetation or partially buried.

Technical Approach

Two optical augmentation approaches were examined in this project: one active, using a laser to illuminate a modified munition, and the other passive, using the spectral characteristics of unmodified munitions to discriminate. The active concept requires a laser illuminator to produce a conspicuous flash of light from retroreflective material added to the munition. This allows the operator to quickly identify objects of interest as those treated with retroreflective material. The passive concept uses the spectroscopic properties of the munition’s existing paint for identification. It requires signal processing of at least two filtered images of the range. The filtering methods are specific to a particular paint, precluding its use on more than one type of munition. However, different filters could be applied to various paint schemes in a similar manner, allowing a serial search to be performed on each munition type expected during a range clearing operation.


Sensors were developed for both optical augmentation approaches, and field tests were performed. Experimental results have shown that the two approaches are feasible. Both provide adequate marking of the objects in question at usable distances. However, the spectroscopy-based or passive concept is a more practical and timely solution to identifying UXO on the surface of test ranges since it does not require modification of the munition and covers a larger area in less time. This project was completed in FY 2002.


Through enhanced discrimination capabilities, optical augmentation will facilitate the timely removal of hazardous UXO material in support of ongoing range operations. Reductions in the amount of UXO unaccounted for on a test range decrease the long-term safety hazards. Removal of UXO also mitigates the long-term potential for entry of hazardous materials into the ecosystem. Pollution prevention is served by removing munitions that may over time leach hazardous material.

  • Classifier ,

  • Machine Learning ,

  • Optical ,

  • Sensors