Sonar Noise Classifier
We invite you to with the first CARIS Mira AI offering, the Sonar Noise Classifier!
CARIS seeks to significantly reduce the need for manual cleaning and move your data swiftly from acquisition to review. The Sonar Noise Classifier automatically identifies the vast majority of sonar noise, resulting in a reduction of manual cleaning effort by a factor of up to 10x at an accuracy of 95%.
The Sonar Noise Classifier is trained to identify a variety of noise patterns from acoustic sensors. The algorithm assigns confidence values for all soundings on a scale of 0 to 100% with 100% indicating the classifier's highest confidence that the sounding is noise. Soundings can be rejected while running the Sonar Noise Classifier by setting your confidence threshold value.
By leveraging the Sonar Noise Classifier during the survey through CARIS Onboard, raw sonar data is fully processed, cleaned and put into a final product, autonomously and in realtime. The powerful combination of Onboard and the Sonar Noise Classifier allows the possibility of having a processed and cleaned dataset by the time your survey is finished, allowing the final quality review efforts to be substantially reduced.
Allow us to help you tackle your data backlog save your personnel for the tasks a machine cannot do.
Choose to purchase access to the Sonar Noise Classifier service on a yearly, or month by month subscription basis for a customized software solution to meet processing needs. CARIS invites you to try it for yourself with a free 30 day trial.