Project Details
Goal of the project
(Semi)Autonomous underwater vehicles, (S)AUVs, equipped with high-resolution (HR) sonar systems have shown a great potential for detecting and classifying mines. The level of detail recorded by these HR sonars is typically on the order of hundreds of pixels on an underwater object, valuable for improving computer-aided detection/classification (CAD, CAC) performance. This will lead to the introduction of sensors equipped with Automatic target Recognition (ATR). Therefore, the trend goes towards little or no human interaction during mine-searching operations with (S)AUVs. This implies that the vehicle must be intelligently operated and the mission must be adapted to the environment and the task in hand in order to select the optimal path such that the best possible data are gathered. In this context, this study aims to:
- Determine a confidence level map for CAC performance prediction by analysing some of the elements that affect image quality (e.g., seafloor roughness, sea state, navigation accuracy).
- Improve the classification performance by adaptive mission re-planning, with the aim of maximizing the classification accuracy while minimizing classification time/number of manoeuvres.
- Analyse when human intervention is required and study the added value of CAC algorithms.
HR data are obtained through a JRP with CMRE-STO and by measurement campaigns on board of the RV Belgica.
Acronym | MRN/17 |
---|---|
Status | Finished |
Effective start/end date | 1/01/15 → 31/12/18 |
Collaborative partners
- Royal Military Academy (lead)
- NATO STO Center for Maritime Research and Experimentation
RHID domain
- Data acquisition and processing
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.