Detection and classification of mine-like objects from sonar data

Automatic Target Recognition (ATR) provides a sophisticated PMA (Post-Mission Analysis) workflow for analysing side scan sonar data from uncrewed maritime systems. It uses rapid and reliable machine learning techniques to detect and classify mine-like objects, reducing operator fatigue and improving mission tempo.

The system provides operators with a measure of how ‘mine-like’ each potential target is, helping to distinguish between threats and false alarms. It features an intuitive workflow for common tasks such as fusing duplicate contacts.

ATR provides a variety of powerful features, including:

• Contact Fusion Module – merge detections generated by the ATR algorithms, removing the need to manually review detections to combine or remove duplicates
• Software Development Kit (SDK) – provides easy integration of third-party ATR or fusion algorithms