The accuracy of length measurements made using imaging SONAR is inversely proportional to the beam width

Multibeam imaging SONARs have been used for a range of measurement applications, such as measurements of fish lengths. This study aimed to quantify the accuracy of imaging SONAR systems, that varied in frequency and beam geometry, to measure the length of synthetic targets positioned perpendicularly. Blueprint Oculus imaging SONAR systems, with four different (centre) frequencies (750 kHz, 1.2 MHz, 2.1 MHz, and 3 MHz), were used to measure the length of three targets of nominal lengths: 10 cm, 20 cm, and at ranges between 1 m and 15.5 m. The effect of beam geometry on measurement error was then examined using regression analysis. This study found that there was an overestimation of the actual length of the target for all measurements that was inversely proportional to the horizontal beam width. The measurement error can be reduced by normalising for beam width. However, the variation of measurements (i.e. the precision), was found to also increase with range, which was attributed to the increasing beam separation. It is important that the effect of beam width on the accuracy of target length measurements, by imaging SONARs is acknowledged in future studies. One approach to mitigating this problem, is to limit the range at which length measurements are made to a beam width that produces an estimated level of error that is acceptable for that study. Keywords: acoustic camera, imaging SONAR, underwater acoustics, underwater measurements.

Methods
Data collection

To investigate measurement accuracy, three different Blueprint Oculus imaging SONAR systems were used in this study, models: M750d, M1200d, and M3000d. Between them, they operated at four different (centre) frequencies: 750 kHz, 1.2 MHz, 2.1 MHz, and 3 MHz. These imaging SONAR systems were used to measure the length of three rectangular targets of nominal lengths: 10 cm, 20 cm, and 1 m. The targets were designed to present consistent, uniform acoustic scattering. The measurements were carried out in a (freshwater) swimming pool on 16th September 2021. The targets were created from layers of hollow polycarbonate sheets, with the edges sealed to maintain the airspaces and backed with a 1 mm thick aluminium sheet. The targets were completely wrapped in black PVC duct tape (Figure 3). The actual final length of each nominal target size was 1 m = 1005 mm, 20 cm = 195 mm, 10 cm = 105 mm, but are referred to by their nominal lengths. The height of each target was 1 m = 255 mm, 20 cm = 115 mm, 10 cm = 55 mm. For the 1 m and the 20 cm targets, the thickness was 33 mm, and 25 mm for the 10 cm target. The targets were attached to fishing line and orientated so that the length was perpendicular to the face of the imaging SONAR (Figure 3). Surfactant was applied to the fishing line immediately prior to measurements, to mitigate any contribution bubbles on the line could have to “elongating the target” as per standard underwater acoustic protocols. 

The Blueprint Oculus imaging SONAR systems were connected through a RJE Oceanbotics™ SRV-8 ROV and recorded using Oculus Viewpoint software. The sound velocity of the freshwater pool was calculated through a temperature probe integrated into the Oculus systems, and was also independently verified. Imaging SONAR data of the targets were collected at ranges between 1 m and 15.5 m, by moving the targets either closer to or further away from the ROV, which remained at a fixed location. At each range step (nominally 1 m steps), the ROV would pan horizontally to collect measurements at different bearings over the full swath width. The ROV was paused when necessary to allow targets to be perpendicular to the SONAR. This was repeated for the four different system frequencies. The swimming pool depth was approximately 1.8 m, and there was minimal reverberation noise. 

Length measurements

The imaging SONAR data were analyzed by someone who was not present during the data collection and did not know the length of the targets, so as not to introduce bias. The measurements were made using the measurement tool in Oculus Viewpoint software, and the co-registered video from the ROV was used to confirm the targets were present. For each length measurement, the frequency, range setting, and measured range of the target were recorded. For the 750 kHz and 1.2 MHz systems, the bearing of the target was recorded if it was measured at greater than +/-32.5o, as beyond this, the beam width was more than 20% of that of the centre beam for those systems and the effects of larger beam widths on the imagery could be seen at those bearings. These bearings were displayed on the software, making it easy for the user to distinguish this area. The measurements were repeated for consistency, and the mean length of the repeated measurements was used in the analysis.