US 10,891,482 B2
Systems, devices, and methods for in-field diagnosis of growth stage and crop yield estimation in a plant area
Jose Angelo Gurzoni, Jr., Sao Paulo (BR); Plinio Thomaz Aquino, Jr., Sao Bernardo do Campo (BR); and Milton Perez Cortez, Jr., Sao Paulo (BR)
Assigned to Adroit Robotics, Sao Paulo (BR)
Filed by Adroit Robotics, Sao Paulo (BR)
Filed on Jul. 10, 2018, as Appl. No. 16/31,801.
Prior Publication US 2020/0019777 A1, Jan. 16, 2020
Int. Cl. A01B 79/00 (2006.01); G01N 21/00 (2006.01); G06F 3/0481 (2013.01); G06K 9/00 (2006.01); G06K 9/03 (2006.01); G06K 9/34 (2006.01); G06K 9/62 (2006.01); G06Q 50/02 (2012.01); G06T 17/05 (2011.01); G06T 7/00 (2017.01); G06T 7/90 (2017.01)
CPC G06K 9/00657 (2013.01) [G06T 7/0002 (2013.01); G06T 7/90 (2017.01); G06T 17/05 (2013.01); G06F 3/04817 (2013.01); G06T 2200/24 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/10048 (2013.01); G06T 2207/30168 (2013.01); G06T 2207/30188 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A crop yield estimation system for detecting one or more properties of a plant area, the crop yield estimate system comprising: an inspection system mountable to a transport device, the inspection system comprising:
a global positioning system;
at least two stereoscopic cameras for capturing image data from at least two focal points;
a communication system;
at least one processor; and
at least one non-transitory computer-readable storage medium storing instructions thereon that, when executed by the at least one processor, cause the inspection system to:
capture at least three-dimensional (3D) image data and two-dimensional (2D) image data of the plant area via the at least two stereoscopic cameras;
receive geolocational data via the global positioning system;
pre-process the captured at least three-dimensional (3D) image data and the two-dimensional (2D) image data of the plant area at the inspection system to determine color, brightness, and resolution of the at least three-dimensional (3D) image data and the two-dimensional (2D) image data of the plant area;
associate, at the inspection system, the at least three-dimensional (3D) image data and the two-dimensional (2D) image data with the received geolocational data; and
generate pre-processed data including the at least three-dimensional (3D) image data and the two-dimensional (2D) image data and the associated geolocational data;
a datacenter remote from the inspection system, the datacenter configured to:
receive the pre-processed data from the inspection system;
analyze the pre-processed data via one or more machine learning techniques to identify the one or more properties of the plant area and locations of the one or more properties;
responsive to the identified one or more properties, generate one or more automated recommendations for dosing of fertilizers and pesticides in one or more areas of the plant area; and
generate a map with icons indicating the one or more properties and the locations of the one or more properties; and
a dashboard configured to display the map and the generated one or more automated recommendations.