Exploring the use of optical domain information for automated target detection and recognition

Duncan L. Hickman

doi:10.1117/12.2584983

12 April 2021 Exploring the use of optical domain information for automated target detection and recognition

Duncan L. Hickman

Proceedings Volume 11759, Virtual, Augmented, and Mixed Reality (XR) Technology for Multi-Domain Operations II; 1175903 (2021) https://doi.org/10.1117/12.2584983
Event: SPIE Defense + Commercial Sensing, 2021, Online Only

Abstract

Automated Target Detection and Recognition (ATD/R) of targets in a cluttered background presents major challenges for the designers of military sensor systems. To achieve acceptable levels of detection and recognition performance, extensive use is often made of complex image processing techniques such as neural network architectures which attempt to replicate the capabilities of the human vision system. Although such methods generate good levels of performance, they are often not suitable for those applications where smaller and lower cost sensor systems are used across a more diverse range of scene content and imaging conditions. A new approach is proposed here where system performance is achieved through a more effective balance between optical domain and post-detection processing. Specifically, the Signal to Clutter Ratio (SCR) is maximised by using broadband spectral and polarisation information to offset performance deficiencies associated with simple image and data processing functions. It is shown that this approach offers a basis for introducing ATD/R functionality in low cost imaging systems such as those flown on drones. A simple imaging system is used to demonstrate the concept, which compromises two broadband cameras operating in the visible and near infrared bands, and with one of the sensors additionally providing polarimetric information. The concept of a joint spectral polarisation weight map is proposed and the potential performance gain is illustrated using targets in moderate to high clutter situations. The results obtained indicate potential benefits for future ATD/R systems and it is hoped that this will encourage future design engineers to consider the wider use of optical domain information.

Conference Presentation

Citation Download Citation

Duncan L. Hickman "Exploring the use of optical domain information for automated target detection and recognition", Proc. SPIE 11759, Virtual, Augmented, and Mixed Reality (XR) Technology for Multi-Domain Operations II, 1175903 (12 April 2021); https://doi.org/10.1117/12.2584983

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available