US 11,874,407 B2
Technologies for dynamic, real-time, four-dimensional volumetric multi-object underwater scene segmentation
Alam Abbas Syed, Orlando, FL (US); and Blair G. Cunningham, Orlando, FL (US)
Assigned to CODA OCTOPUS GROUP INC., Orlando, FL (US)
Filed by CODA OCTOPUS GROUP, INC., Orlando, FL (US)
Filed on Feb. 19, 2020, as Appl. No. 16/795,201.
Prior Publication US 2021/0256704 A1, Aug. 19, 2021
Int. Cl. G01S 7/539 (2006.01); G06T 7/12 (2017.01); G01S 15/89 (2006.01); G06V 20/00 (2022.01); G06V 20/64 (2022.01); G06F 18/24 (2023.01)
CPC G01S 7/539 (2013.01) [G01S 15/89 (2013.01); G06F 18/24 (2023.01); G06T 7/12 (2017.01); G06V 20/00 (2022.01); G06V 20/64 (2022.01); G06T 2207/10028 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for processing a three-dimensional (3D) underwater scene, the method comprising:
receiving, by a computing device, a data set output by a sonar transducer assembly, wherein the data set represents an underwater environment, wherein the data set corresponds to sonar data generated by the sonar transducer assembly, and wherein the data set includes a plurality of 3D volumetric points representing at least one 3D volumetric data set at one or more time points;
receiving a specification of one or more tasks to perform on the received data set;
identifying, based on the specified one or more tasks to perform on the received data set and based on a known shape descriptor, two or more segmentation modes from a plurality of segmentation modes and a sequence for applying the identified segmentation modes to the received data set, wherein the two or more segmentation modes include a mode for extracting 3D geometric primitives from the received data set and wherein the known shape descriptor is indicative of at least one of a 3D mathematical shape or a physical 3D volumetric data set from an a priori model;
applying, by the computing device, the identified segmentation modes on the data set according to the identified sequence, wherein the application of each identified segmentation mode generates a segmented subset of the plurality of 3D volumetric points representing an interpretation of the underwater environment, each segmented subset including one or more 3D volumetric objects; and
classifying, by the computing device, the one or more 3D volumetric objects based on a combination of the plurality of segmented subsets.