Crop residue based field operation adjustment

Patent metadata

Field	Value
Publication number	US-11510364-B2
Application number	US-201916517482-A
Country	US
Kind code	B2
Filing date	Jul 19, 2019
Priority date	Jul 19, 2019
Publication date	Nov 29, 2022
Grant date	Nov 29, 2022

Abstract

A crop residue monitoring system may include a harvester, a camera to capture an image of crop residue generated by the harvester, an analytical unit to derive a value for a crop residue parameter of the crop residue based upon an optical analysis of the image and a control unit to adjust a subsequent field operation based upon the value of the crop residue parameter.

First claim

What is claimed is:

1. A crop residue monitoring system comprising:
a harvester;
a camera to capture an image of crop residue generated by the harvester;
an analytical unit to derive a value for a crop residue parameter of the crop residue based upon an analysis of the image; and
a controller to adjust a subsequent field operation based upon the value of the crop residue parameter;
wherein the value for the crop residue parameter comprises a crop residue parameter category;
wherein the analytical unit comprises a neural network;
wherein the neural network derives category criteria and compares characteristics of the image to the category criteria to assign a category to the image of the crop residue.

2. The system of claim 1 , wherein the camera is carried by the harvester.

3. The system of claim 1 , wherein the adjustment of the subsequent field operation by the controller comprises adjusting an operational parameter of the harvester.

4. The system of claim 1 further comprising a non-transitory machine-readable medium storing a crop residue parameter field map comprising the crop residue parameter, wherein the adjustment of the subsequent field operation comprises adjusting an operational setting of an agricultural machine different than the harvester based upon the crop residue parameter field map.

5. The system of claim 1 further comprising a non-transitory machine-readable medium storing a crop residue parameter field map comprising the crop residue parameter, wherein the adjustment of the subsequent field operation comprises adjusting a parameter of the subsequently applied material based upon the crop residue parameter field map.

6. The system of claim 1 , wherein the harvester is to discharge the crop residue across a row having a width and wherein the crop residue parameter is derived for a first portion of the width of the row and wherein the analytical unit is to further derive a second value for the crop residue parameter for a second portion of the width of the row.

7. The system of claim 1 , wherein the controller is to display the value for the crop residue parameter and the second value for the crop residue parameter along a length of the row.

8. The system of claim 1 , wherein the analytical unit is to compare the value for the crop residue parameter and the second value for the second crop residue parameter to at least one threshold and categorize the first portion of the row as a first category based on the comparison and the second portion of the row as a second category, different than the first category, based on the comparison.

9. The system of claim 1 , wherein the subsequent field operation has a first characteristic for a first geo-region corresponding to the first portion of the row and a second characteristic, different than the first characteristic, for a second geo-region corresponding to the second portion of the row.

10. The system of claim 1 , wherein the neural network derives the category for the crop residue parameter by:
receiving first images of crop residue (CR) that have ground truth category labels;
identifying category criteria for different ground truth category labels based upon analysis of the first images and their ground truth category labels;
receiving second images of CR that have ground truth category labels;
applying the identified category criteria to label the second images with analytical unit-based category labels;
comparing the ground truth category labels of the second images to the analytical unit-based category labels of the second images; and
adjusting the identified category criteria based upon the comparison.

11. The system of claim 1 ,
wherein the analysis of the image comprises an optical analysis.

12. A crop residue monitoring system comprising:
a harvester;
a camera to capture an image of crop residue generated by the harvester;
an analytical unit to derive a value for a crop residue parameter of the crop residue based upon an analysis of the image; and
a controller to adjust a subsequent field operation based upon the value of the crop residue parameter,
wherein the adjustment of the subsequent field operation by the controller comprises adjusting an operational parameter of the harvester following discharge of the crop residue, the operational parameter selected from a group of operational parameters consisting of: chopper speed, chopper power; harvester speed; harvester feed rate; chopper counter knife position, header height, spreader speeds, spreader vane positions, threshing speed, cleaning speed, threshing clearance and separator discharge vanes.

13. The system of claim 12 , wherein the analytical unit is to derive the crop residue parameter by:
optically identifying individual pieces of crop residue; and
measuring a length of each of the pieces of crop residue, the value of crop residue parameter being based upon a count of a number of the pieces having each of a plurality of lengths.

14. The system of claim 12 , wherein the crop residue parameter comprises a parameter selected from a group of parameters consisting of chopped size, crop residue moisture, crop residue constituents and crop residue dispersion.

15. The crop residue monitoring system of claim 12 , wherein the analysis of the image comprises an optical analysis.

16. A method comprising:
capturing an image of crop residue generated by a harvester;
deriving, with a neural network, a value for a crop residue parameter of the crop residue discharged from the harvester based upon an analysis of the image, wherein the value for the crop residue parameter comprises a crop residue parameter category and the neural network derives category criteria and compares characteristics of the image to the category criteria to assign the crop residue parameter category to the image of the crop residue; and
adjusting a subsequent field operation based upon the value of the crop residue parameter.

17. The method of claim 16 , wherein the adjustment of the subsequent field operation comprises adjusting an operational parameter of the harvester.

18. The method of claim 16 , wherein the adjustment of the subsequent field operation comprises adjustment of an operational parameter of an agricultural machine other than the harvester.

19. The method of claim 16 , wherein the adjustment of the subject field operation comprises adjustment of an applied material parameter.

20. The method of claim 16 , wherein the analysis of the image comprises an optical analysis.

21. A non-transitory machine-readable medium containing instructions to direct a processing unit to:
capture an image of crop residue generated by a harvester;
derive a value, with a neural network, for a crop residue parameter of the crop residue based upon an analysis of the image, wherein the value for the crop residue parameter comprises a crop residue parameter category and the neural network derives category criteria and compares characteristics of the image to the category criteria to assign the crop residue parameter category to the image of the crop residue; and
generate a crop residue field map based upon the derived value, the crop residue field map for use in subsequently applied field operations.

22. The machine-readable medium of claim 21 , wherein the instructions are to further direct the processing unit to adjust an operational setting of the harvester based upon the derived value for the crop residue parameter.

23. The non-transitory machine-

Assignees

• Deere & Co
• Deere & Co

Inventors

• Vandike Nathan R
• Gilmore Brian J
• Phatak Dinesh V
• Wells David L
• Bothe Mahesh S

Classifications

• G06T7/60 Primary

  Analysis of geometric attributes · CPC title

• G06V10/454

  Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN] · CPC title

• A01D41/127 Primary

  Control or measuring arrangements specially adapted for combines · CPC title

• G06V20/68

  Food, e.g. fruit or vegetables · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title