Patents by Inventor James H. Crain
James H. Crain has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11048938Abstract: A method for recognizing individual plants of a selected type growing in a field, wherein the method comprises capturing color NIR image data of an entire field having plants of a selected type growing therein utilizing an automated plant counting system and calculating a ratio value between each pixel of the color image data and the corresponding pixel of the NIR image data utilizing a plant recognition algorithm executed via a data processing system of the plant counting system. The method additionally comprises generating, via execution of the plant recognition algorithm, a false color image of the field based on the calculated ratios for each pixel, and identifying, via execution of the plant recognition algorithm, all plants of the selected type in the false color image based on a plant distinguishing characteristic uniquely rendered for each individual plant of the selected type in the false color image.Type: GrantFiled: April 16, 2019Date of Patent: June 29, 2021Assignee: Monsanto Technology LLCInventors: Christopher A. Hundley, James H. Crain
-
Publication number: 20190244022Abstract: A method for recognizing individual plants of a selected type growing in a field, wherein the method comprises capturing color NIR image data of an entire field having plants of a selected type growing therein utilizing an automated plant counting system and calculating a ratio value between each pixel of the color image data and the corresponding pixel of the NIR image data utilizing a plant recognition algorithm executed via a data processing system of the plant counting system. The method additionally comprises generating, via execution of the plant recognition algorithm, a false color image of the field based on the calculated ratios for each pixel, and identifying, via execution of the plant recognition algorithm, all plants of the selected type in the false color image based on a plant distinguishing characteristic uniquely rendered for each individual plant of the selected type in the false color image.Type: ApplicationFiled: April 16, 2019Publication date: August 8, 2019Inventors: Christopher A. Hundley, James H. Crain
-
Patent number: 10303944Abstract: A method for recognizing individual plants of a selected type growing in a field, wherein the method comprises capturing color NIR image data of an entire field having plants of a selected type growing therein utilizing an automated plant counting system and calculating a ratio value between each pixel of the color image data and the corresponding pixel of the NIR image data utilizing a plant recognition algorithm executed via a data processing system of the plant counting system. The method additionally comprises generating, via execution of the plant recognition algorithm, a false color image of the field based on the calculated ratios for each pixel, and identifying, via execution of the plant recognition algorithm, all plants of the selected type in the false color image based on a plant distinguishing characteristic uniquely rendered for each individual plant of the selected type in the false color image.Type: GrantFiled: November 14, 2016Date of Patent: May 28, 2019Assignee: Monsanto Technology LLCInventors: Christopher A. Hundley, James H. Crain
-
Publication number: 20170061211Abstract: A method for recognizing individual plants of a selected type growing in a field, wherein the method comprises capturing color NIR image data of an entire field having plants of a selected type growing therein utilizing an automated plant counting system and calculating a ratio value between each pixel of the color image data and the corresponding pixel of the NIR image data utilizing a plant recognition algorithm executed via a data processing system of the plant counting system. The method additionally comprises generating, via execution of the plant recognition algorithm, a false color image of the field based on the calculated ratios for each pixel, and identifying, via execution of the plant recognition algorithm, all plants of the selected type in the false color image based on a plant distinguishing characteristic uniquely rendered for each individual plant of the selected type in the false color image.Type: ApplicationFiled: November 14, 2016Publication date: March 2, 2017Inventors: Christopher A. Hundley, James H. Crain
-
Patent number: 9495597Abstract: A method for recognizing individual plants of a selected type growing in a field, wherein the method comprises capturing color NIR image data of an entire field having plants of a selected type growing therein utilizing an automated plant counting system and calculating a ratio value between each pixel of the color image data and the corresponding pixel of the NIR image data utilizing a plant recognition algorithm executed via a data processing system of the plant counting system. The method additionally comprises generating, via execution of the plant recognition algorithm, a false color image of the field based on the calculated ratios for each pixel, and identifying, via execution of the plant recognition algorithm, all plants of the selected type in the false color image based on a plant distinguishing characteristic uniquely rendered for each individual plant of the selected type in the false color image.Type: GrantFiled: October 18, 2012Date of Patent: November 15, 2016Assignee: Monsanto Technology LLCInventors: Christopher A. Hundley, James H. Crain
-
Publication number: 20140285673Abstract: A method for recognizing individual plants of a selected type growing in a field, wherein the method comprises capturing color NIR image data of an entire field having plants of a selected type growing therein utilizing an automated plant counting system and calculating a ratio value between each pixel of the color image data and the corresponding pixel of the NIR image data utilizing a plant recognition algorithm executed via a data processing system of the plant counting system. The method additionally comprises generating, via execution of the plant recognition algorithm, a false color image of the field based on the calculated ratios for each pixel, and identifying, via execution of the plant recognition algorithm, all plants of the selected type in the false color image based on a plant distinguishing characteristic uniquely rendered for each individual plant of the selected type in the false color image.Type: ApplicationFiled: October 18, 2012Publication date: September 25, 2014Inventors: Christopher A. Hundley, James H. Crain
-
Publication number: 20110320229Abstract: Generating a crop prescription using a computer coupled to a memory area includes receiving yield data for a plurality of crop population trials, wherein each trial is varied by at least one of a hybrid line, a population density, and a row spacing. At least one statistical model is generated based on the yield data to obtain a plurality of coefficients, which are stored in the memory area. A predicted yield for at least one selected hybrid line is determined based on the coefficients and a selected row spacing, and a predicted profit is determined for the at least one selected hybrid line based on the coefficients and the selected row spacing. A crop prescription is presented that includes a recommended hybrid line and population density for use by a grower.Type: ApplicationFiled: October 14, 2009Publication date: December 29, 2011Applicant: MONSANTO TECHNOLOGY LLCInventors: Sammy J. Stehling, James H. Crain, Timothy D. Perez