US 2023/0267731 A1 · MULTI-MODAL AQUATIC BIOMASS ESTIMATION

Abstract

Computer-implemented techniques for multi-modal aquatic biomass estimation includes determining a fish mass measurement, a fish count, or a direct fish biomass estimate from a sample of one or more digital images or digital video captured by a digital camera immersed underwater in a fish farm enclosure. A technique further includes determining a plurality of fish densities for a plurality of water volumes within the fish farm enclosure based on analysis of signals received from a plurality of bias estimation sensors immersed underwater in the fish farm enclosure. The technique also includes computing a bias-adjusted fish mass measurement, fish count, or direct fish biomass estimate for the fish farm enclosure based on (i) the fish mass measurement, the fish count, or the direct fish biomass estimate determined from the sample of one or more digital images or digital video captured by the digital camera immersed underwater in the fish farm enclosure and (ii) the plurality of fish densities determined for the plurality of water volumes.

Claims20

Dependent claims

A method for multi-modal aquatic biomass estimation, comprising:

determining a fish mass measurement, a fish count, or a direct fish biomass estimate from a sample of one or more digital images or digital video captured by a digital camera immersed underwater in a fish farm enclosure;
determining a fish density for a water volume within the fish farm enclosure based on an analysis of signals received from a bias estimation sensor immersed underwater in the fish farm enclosure;
and
computing a bias-adjusted fish mass measurement, fish count, or direct fish biomass estimate for the fish farm enclosure based on:
(i) the fish mass measurement, the fish count, or the direct fish biomass estimate determined from the sample of one or more digital images or digital video captured by the digital camera immersed underwater in the fish farm enclosure, and
(ii) the fish density determined for the water volume;
wherein the method is performed by one or more computing devices.

4 more elements

A method for multi-modal aquatic biomass estimation, comprising:

determining a fish mass measurement, a fish count, or a direct fish biomass estimate from a sample of one or more digital images or digital video captured by a digital camera immersed underwater in a fish farm enclosure;
determining a position of the digital camera in the fish farm enclosure at a time the sample of the one or more digital images or digital video is captured by the digital camera;
and
computing a bias-adjusted fish mass measurement, fish count, or direct fish biomass estimate for the fish farm enclosure based on:
(i) the fish mass measurement, the fish count, or the direct fish biomass estimate determined from the sample of the one or more digital images or digital video captured by the digital camera immersed underwater in the fish farm enclosure, and
(ii) the position of the digital camera in the fish farm enclosure at the time the sample of the one or more digital images or digital video is captured by the digital camera.

3 more elements

A method for multi-modal aquatic biomass estimation, comprising:

determining a fish mass measurement, a fish count, or a direct fish biomass estimate from a sample of one or more digital images or digital video captured by a digital camera immersed underwater in a fish farm enclosure;
determining a fish density for a water volume within the fish farm enclosure based on an analysis of signals received from a bias estimation sensor immersed underwater in the fish farm enclosure;
determining a position of the digital camera in the fish farm enclosure at a time the sample of the one or more digital images or digital video is captured by the digital camera;
and
computing a bias-adjusted fish mass measurement, fish count, or direct fish biomass estimate for the fish farm enclosure based on:
(i) the fish mass measurement, the fish count, or the direct fish biomass estimate determined from the sample of the one or more digital images or digital video captured by the digital camera immersed underwater in the fish farm enclosure,
(ii) the fish density for the water volume, and
(iii) the position of the digital camera in the fish farm enclosure at the time the sample of the one or more digital images or digital video is captured by the digital camera.

5 more elements

Dependencies are inferred from the claim text.

Multi-modal aquatic biomass estimation

Life of this patent

Multi-modal aquatic biomass estimation

Life of this patent

Abstract

Claims20