Device and method for evaluating the composition of a stream of harvested material

Patent metadata

Field	Value
Publication number	US-11880970-B2
Application number	US-202117472083-A
Country	US
Kind code	B2
Filing date	Sep 10, 2021
Priority date	Sep 16, 2020
Publication date	Jan 23, 2024
Grant date	Jan 23, 2024

Abstract

A device and a method for determining a portion of broken grain and/or non-grain components in a stream of harvested material is disclosed. The device includes a camera to generate images of the stream of harvested material and an evaluation unit to estimate the portion in an image supplied by the camera. The evaluation unit includes a first-order classifier to estimate a first parameter of the stream of harvested material, and a plurality of second-order classifiers assigned to various values of the first parameter to estimate the portion in a stream of harvested material that has the assigned parameter value. The evaluation unit, with assistance of the first-order classifier, estimates a value of the first parameter using a first number of images of the camera, and then selects the second-order classifier assigned to the value of the first parameter to estimate the portion with assistance of the selected second-order classifier.

First claim

The invention claimed is:

1. A device configured to determine a portion of at least one of broken grain or non-grain components in a stream of harvested material, the device comprising:
a camera configured to generate images of the stream of harvested material; and
an evaluation unit configured to estimate the portion in an image generated by the camera;
wherein the evaluation unit comprises a first-order classifier configured to estimate a first parameter of the stream of harvested material, and a plurality of second-order classifiers assigned to various values of the first parameter to estimate the portion in a stream of harvested material that has the assigned parameter value;
wherein the evaluation unit is configured to:
estimate, using the first-order classifier, at least one value of the first parameter using one or more images of the camera in order to identify a specific variety of crop;
select the second-order classifier, from the plurality of second-order classifiers, using the specific variety of the crop, wherein each of the plurality of second-order classifiers are trained for a respective variety of crop, and wherein the second-order classifier is trained for the specific variety of the crop in order to estimate the portion of the at least one of the broken grain or the non-grain components; and
estimate, using the selected second-order classifier, the portion of the at least one of the broken grain or the non-grain components.

2. The device of claim 1 , wherein the evaluation unit is configured to estimate, using the first-order classifier, the value of the first parameter using a first set of images; and
wherein evaluation unit is configured to estimate, using the selected second-order classifier, the portion of one or both of broken grain or non-grain components using a second set of images different from the first set of images.

3. The device of claim 1 , wherein the first parameter comprises a discrete number of values; and
wherein each respective value in the discrete number of values has associated at most one second-order classifier such that a number of the second-order classifiers are at most the discrete number.

4. The device of claim 1 , wherein the second-order classifier is configured to assign a confidence level to the estimation of the portion of one or both of broken grain or non-grain components.

5. The device of claim 4 , wherein the evaluation unit is further configured to:
determine whether the confidence level is less than a threshold; and
responsive to determining that the confidence level is less than the threshold, invoke the first-order classifier.

6. The device of claim 1 , wherein the first parameter designates at least one of specific type of the harvested material or specific variety of the harvested material as the value of the first parameter; and
wherein the evaluation unit is configured to select the second-order classifier correlated to at least one of the specific type of the harvested material or the specific variety of the harvested material.

7. The device of claim 1 , wherein the first parameter designates a specific type of the harvested material and a specific variety of the harvested material as the value of the first parameter; and
wherein the evaluation unit is configured to select the second-order classifier correlated to the specific type of the harvested material and the specific variety of the harvested material.

8. The device of claim 1 , wherein the first-order classifier comprises a first-order sub-classifier and a plurality of second-order sub-classifiers;
wherein the first-order sub-classifier is configured to identify a type, from a plurality of potential types, of the harvested material; and
wherein the second-order sub-classifiers are configured to identify a variety, from a plurality of potential varieties, of the type of the harvested material.

9. The device of claim 1 , wherein the first-order classifier and the second-order classifier are trained by using images of streams of harvested material and known portions of the imaged streams of harvested material.

10. The device of claim 1 , wherein the device is configured for installation in a self-propelled harvester; and
wherein the evaluation unit is further configured to:
cause modification of at least one aspect of the self-propelled harvester thereby modifying a process of the self-propelled harvester with respect to at least one of throughput, grain loss or the broken grain portion; or
cause display of the determined broken grain portions in order to allow a driver of the self-propelled harvester to make adaptations or modifications to operating parameters of the process of the self-propelled harvester.

11. A self-propelled harvester comprising:
at least one of a threshing units, a separating device, or a cleaning device;
a display; and
a device configured to determine a portion of one or both of broken grain or non-grain components in a stream of harvested material, the device include:
a camera configured to generate images of the stream of harvested material; and
an evaluation unit configured to estimate the portion in an image generated by the camera;
wherein the evaluation unit comprises a first-order classifier configured to estimate a first parameter of the stream of harvested material, and a plurality of second-order classifiers assigned to various values of the first parameter to estimate the portion in a stream of harvested material that has the assigned parameter value;
wherein the evaluation unit is configured to:
estimate, using the first-order classifier, at least one value of the first parameter using one or more images of the camera; and
select the second-order classifier assigned to the at least one value of the first parameter;
estimate, using the selected second-order classifier, the portion of one or both of broken grain or non-grain components; and
perform at least one of:
cause modification of at least one aspect of the self-propelled harvester thereby modifying a process of the self-propelled harvester with respect to at least one of throughput, grain loss or the broken grain portion; or
cause display of the determined broken grain portions in order to allow a driver of the self-propelled harvester to make adaptations or modifications to operating parameters of the process of the self-propelled harvester.

12. A method for determining a portion of one or both of broken grain or non-grain components in a stream of harvested material, the method comprising:
accessing a number of images of a harvested material;
accessing a first-order classifier that is configured to estimate a first parameter of a stream of harvested material in order to identify one of a plurality of varieties of crop by using at least one image of the stream of harvested material, and a plurality of second-order classifiers, each of the plurality of second-order classifiers are trained for a respective variety of crop and are configured to estimate the portion;
using the first-order classifier to estimate a value of the first parameter in order to identify a specific variety of crop;
by using the specific variety of crop, selecting a second-order classifier from the plurality of second-order classifiers, wherein the second-order classifier that is selected is trained for the specific variety of the crop in order to estimate the portion of the at least one of the broken grain or the non-grain components; and
using the selected second-order classifier in order to obtain an estimated value of the portion of one or both of broken grain or non-grain components.

13. The method of claim 12 , wherein

Assignees

• Claas Selbstfahrende Erntemaschinen Gmbh

Inventors

• Fischer Frédéric
• Witte Johann
• Kettelhoit Boris
• Töniges Torben

Classifications

• G06T7/0004 Primary

  Industrial image inspection · CPC title

• A01D41/1277

  for measuring grain quality · CPC title

• G06F18/2431

  Multiple classes · CPC title

• G06T2207/30128

  Food products · CPC title

• G06V20/56 Primary

  exterior to a vehicle by using sensors mounted on the vehicle · CPC title