This project examined the capability of an ultrasonic three dimensional (3-D) visualization system to provide the undisturbed characterization and identification of submerged shallow-buried objects. Analysis of a location by this system would provide a visual image of individual buried unexploded ordnance (UXO) not visible on the surface.

The 3-D Ultrasonic system was originally developed at University of California, Los Angeles’ (UCLA) School of Civil and Environmental Engineering by Dr. Scott Brandenberg. The goal was to study the potential of using p-wave reflection imaging for geotechnical engineering. The demonstration system was adapted for underwater use by SPAWAR Systems Center, Pacific. The ultrasonic system components are off-the-shelf hardware consisting of two ultrasonic transducers, source pulser, receiver amplifier, receiver analog filter, terminal block, data acquisition cards mounded in a PXI chassis, a laptop computer running the control and data processing software, and the underwater delivery vehicle with the X-Y positioning system. The vehicle is the benthic flux sampling device (BFSD) with the X-Y positioning system attached. It is composed of a X-Y gantry system operated by underwater servo motors controlled by the operator’s computer. For this project, the flux sampling equipment was removed and replaced by the X-Y positioning system.

The results for each performance objective are provided below:

Detection Depth: objective was to determine the ability of the system to generate a useful sonar return signal at the target depth. 

• Metric: Evaluate the strength of the ping return signal to determine if it is above the noise floor of the system.
• Data Requirements: Fix the transducer and receiver over four buckets filled to the top with each of the standardized sediment types. Place the system in continuous ping and return mode. Evaluate the 5000 count discrete root mean square (RMS) average of signal return displayed in the user interface and compare it to the no-signal level 0.1 msec before the signal return.
• Success Criteria: The objective will be met if the signal-to-noise ratio is above 0.1. Where the signal-to-noise is the square of the RMS of the signal amplitude divided by RMS of the noise amplitude.
• Performance Assessment: The signal-to-noise ratio was above 0.1 for all four sediment types at a depth of 0.5 m.

Scan Time: objective was to evaluate the length of time it takes the system to cover a 1m² area. The test was setup by placing the system over a tank of water containing know objects. The system was set to scan a full 1m² area and the render times were noted and recorded.

• Metric: The metric is the length of time it takes the system to cover 1m² using a scanning step size of 3 mm.
• Data Requirements: Length of time required for the scan as recorded in the data log.
• Success Criteria: A scan time of 20 minutes or less.
• Performance Assessment: Scan time for a 1m² area was 19.3 minutes.

Render Time: objective was to evaluate the length of time it takes the software to process the data once the surface scan is complete. The test was setup by placing the system over a tank of water containing known objects. The system was set to scan a full 1m² area and the start and stop times were noted and recorded.

• Metric: The metric is the length of time it takes to process a 1m² area data set.
• Data Requirements: The data required is the length of time to process the data as measured by a digital watch.
• Success Criteria: A rendering time of 5 minutes or less will be considered a success.
• Performance Assessment: The rendering time for a 1m² area data set using a personal computer (PC) with a dual core processor and 8 GB of RAM averages 2.2 minutes.

Software User Interface Errors: objective was to evaluate the number of user interface errors in the software. The test was setup by placing the system over a tank of water containing known objects. The system was set to scan a full 1m² area. Software bugs were noted by the test team and corrected by the software development team.

• Metric: The metric is the number of software bugs identified and not corrected during the quality assurance/quality control (QA/QC) of the software before delivery to the users.
• Data Requirements: Evaluation of all the user interface controls and outputs.
• Success Criteria: 100% error free, all identified bugs have been corrected.
• Performance Assessment: Three software bugs were identified during the QA/QC of the software. Software modifications were completed to address these errors and the new version of software passed QA/QC without errors. 

Object Recognition: objective was to evaluate the ability of the system to create identifiable 3-D models of objects buried in the sample sediment.

• Metric: This metric is the identification of a 60mm mortar round and 5 lb Danforth anchor scanned at an average depth of 0.5m in sandy clay and a 60mm mortar buried in sandy clay at a depth of 0.25m.
• Data Requirements: U3D format data file of the object.
• Success Criteria: The data model looks like the object.
• Performance Assessment: The two items were rendered and identified. All users were able to correctly identify the objects from the 3-D Adobe Acrobat Reader files created by the software.

After wide area assessment of underwater UXO, there is often insufficient information to classify UXO from clutter or non-UXO items. Before remediation planning can begin, these debris fields must be classified as hazardous (UXO-containing) or nonhazardous (debris only). This is traditionally done by digging up shallow buried potential munitions/UXO at spot locations by Navy-trained and certified underwater explosive ordnance disposal (EOD) teams. Traditional removal, identification, and disposal of these buried objects by trained EOD teams is slow, expensive, and, for benign objects, unnecessary. Undisturbed identification of these buried objects either by a single diver using a handheld sensor, by a remotely operated vehicle (ROV)-mounted sensor, or in shallow water, by a pole-mounted sensor would greatly speed up spot identification, paving the way for more cost-effective remediation and cleanup planning.