US 11,925,173 B2
Data collection system and method for feeding aquatic animals
Hemang Ravi Rishi, Richmond (GB); Pieter Jan Fabry, London (GB); Ivan Makeev, London (GB); and Charchris Sloan, London (GB)
Assigned to Observe Technologies Limited, Richmond Greater London (GB)
Appl. No. 16/627,203
Filed by OBSERVE TECHNOLOGIES LIMITED, Richmond (GB)
PCT Filed Jun. 28, 2018, PCT No. PCT/GB2018/051824
§ 371(c)(1), (2) Date Dec. 27, 2019,
PCT Pub. No. WO2019/002880, PCT Pub. Date Jan. 3, 2019.
Claims priority of application No. 1710372 (GB), filed on Jun. 28, 2017.
Prior Publication US 2020/0113158 A1, Apr. 16, 2020
Int. Cl. G06T 7/20 (2017.01); A01K 61/10 (2017.01); A01K 61/13 (2017.01); A01K 61/60 (2017.01); A01K 61/80 (2017.01); G06F 18/21 (2023.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01); G06N 20/00 (2019.01); G06V 40/20 (2022.01)
CPC A01K 61/80 (2017.01) [A01K 61/10 (2017.01); A01K 61/13 (2017.01); A01K 61/60 (2017.01); G06F 18/217 (2023.01); G06N 3/04 (2013.01); G06N 3/08 (2013.01); G06N 20/00 (2019.01); G06T 7/20 (2013.01); G06V 40/20 (2022.01); G06T 2207/10016 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 25 Claims
OG exemplary drawing
 
1. A computer-implemented method for detecting motion in relation to one or more aquatic animals, the method comprising:
receiving sensor data;
determining from the sensor data one or more moving objects using one or more learned functions, wherein the one or more learned functions comprise one or more neural networks, wherein the one or more neural networks perform a task of defining images obtained through real time image analysis of the sensor data as either feed or not feed, and wherein the one or more neural networks comprise one or more feedback loops; and
generating output data in relation to the determined one or more moving objects, wherein the output data is based on temporal information provided by the one or more feedback loops.