IP Library Granted Patent US 12,499,667
Granted Patent B2
US 12,499,667 · App. 18/587,243 · Granted Dec 16, 2025

Systems and methods for worksite imagery selection

Inventors: Alexander C. Reiners (Thibodaux, LA); John P. Just (Ames, IA); Eric M. Heien (Davis, CA); Marcos Antonio de Jesus Filho (Napa, CA); Matthew J. Burke (Des Moines, IA)
Assignee: DEERE & COMPANY
G06V10/776A01D41/127G06V10/40G06V10/774G06V10/94G06V20/188
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,499,667
App. No.
18/587,243
Granted
Dec 16, 2025
Kind
B2
Abstract

A computer implemented method includes obtaining, from one or more image providers, a set of one or more overhead images of a worksite, each overhead image including values indicative of a characteristic at different geographic locations across the worksite; identifying an image performance quality metric for each overhead image based on an image performance quality model; determining if at least one overhead image can be selected, as a selected image, based on the identified image performance quality metrics for the one or more overhead images; selecting, in response to determining that at least one overhead image can be selected, as a selected image, the at least one overhead image and outputting the at least one selected image for use in control of the mobile work machine and generating control signals to control the mobile work machine based, at least, on the at least one selected image.

Claims (62)

1 . A worksite operation system comprising:

one or more processors; and

memory storing computer executable instructions, executable by the one or more processors, the computer executable instructions, when executed by the one or more processors, causing the one or more processors perform steps comprising:

obtaining, from one or more image providers, a first set of one or more overhead images of a worksite, each overhead image of the first set of one or more overhead images of the worksite including values indicative of a characteristic of the worksite at different geographic locations across the worksite;

identifying an image performance quality metric for each overhead image of the first set of one or more overhead images of the worksite based on an image performance quality model, wherein the image performance quality model models a relationship between overhead image values and characteristic values of the characteristic of the worksite;

determining if at least one overhead image of the first set of one or more overhead images of the worksite can be selected, based on the identified image performance quality metrics for the at least one overhead image of the first set of one or more overhead images of the worksite;

selecting, in response to determining that at least one overhead image of the first set of one or more overhead images can be selected, as a selected at least one image, the at least one overhead image of the first set of one or more overhead images of the of the worksite; and

controlling a controllable subsystem of a mobile work machine based, at least, on the selected at least one image.

2 . The worksite operation system of claim 1 , wherein the computer executable instructions, when executed by the one or more processors, further cause the one or more processors to perform a step comprising:

obtaining, in response to determining that none of the one or more overhead images of the first set of one or more overhead images of the worksite can be selected, as a selected image, based on the identified image performance quality metrics for the at least one overhead image of the first set of one or more overhead images of the worksite, a second set of one or more overhead images of the worksite, from the one or more image providers or a different one or more image providers, each overhead image of the second set of one or more overhead images of the worksite including values indicative of the characteristic of the worksite at the different geographic locations across the worksite.

3 . The worksite operation system of claim 2 , wherein the computer executable instructions, when executed by the one or more processors, further cause the one or more processors to perform a step comprising:

selecting, in response to determining that none of the one or more overhead images of the first set of one or more overhead images of the worksite can be selected, as a selected image, based on the identified image performance quality metrics for the one or more overhead images of the first set of overhead images of the worksite, as a selected image, at least one of the one or more overhead images of the second set of one or more overhead images of the worksite.

4 . The worksite operation system of claim 2 , wherein the one or more image providers are one or more public image providers and wherein the one or more different image providers include at least one different public image provider or include one or more private image providers, or both.

5 . The worksite operation system of claim 1 , wherein the computer executable instructions, when executed by the one or more processors, further cause the one or more processors to perform a step comprising:

obtaining, in response to determining that more than the at least one overhead image of the first set of one or more overhead images of the worksite are needed, from the one or more image providers or a different one or more image providers, a second set of one or more overhead images of the worksite, each overhead image of the second set of one or more overhead images of the worksite including values indicative of the characteristic of the worksite at the different geographic locations across the worksite.

6 . The worksite operation system of claim 5 , wherein the computer executable instructions, when executed by the one or more processors, further configure the one or more processors to perform the step of:

selecting, in addition to the selected at least one image of the first set of one or more overhead images, at least one of the overhead images of the second set of one or more overhead images of the worksite.

7 . The worksite operation system of claim 1 , wherein identifying the image performance quality metric for each overhead image of the first set of one or more overhead images of the worksite based on the image performance quality model comprises:

identifying one or more image features of each overhead image of the first set of one or more overhead images of the worksite; and

providing the one or more image features of each overhead image of the first set of one or more overhead images of the worksite to the image performance quality model to obtain, as a model output, the image performance quality metric for each overhead image of the first set of one or more overhead images of the worksite.

8 . The worksite operation system of claim 7 , wherein the one or more image features comprise: (i) image resolution, (ii) image source sensor technology, (iii) timing information, or (iv) image value statistics or a combination of (i) image resolution, (ii) image source sensor technology, (iii) timing information, and (iv) image value statistics.

9 . The worksite operation system of claim 1 , wherein the first set of one or more overhead images of the worksite comprise one or more overhead multi-spectral images of the worksite, each overhead multi-spectral image of the one or more overhead multi-spectral images of the worksite including, as the values indicative of the characteristic of the worksite at different geographic locations across the worksite, multi-spectral values indicative of the characteristic of the worksite at different geographic locations across the worksite.

10 . The worksite operation system of claim 1 , wherein the computer executable instructions, when executed by the one or more processors, further cause the one or more processors to perform steps comprising:

obtaining one or more historical overhead images, each historical overhead image of the one or more historical overhead images including values indicative of the characteristic;

identifying one or more historical image features of each historical overhead image of the one or more historical overhead images;

providing the one or more historical image features of each historical overhead image of the one or more historical overhead images to an in-training image performance quality model to obtain, as an in-training model output, an in-training image performance quality metric for each historical overhead image of the one or more historical overhead images;

obtaining a set of historical values of the characteristic corresponding to each historical overhead image of the one or more historical overhead images;

modeling the values of each historical overhead image of the one or more historical overhead images and the corresponding set of historical values of the characteristic to identify, for each historical overhead image of the one or more historical overhead images, a historical overhead image performance quality metric; and

providing the historical image performance quality metrics for the one or more historical overhead images to the in-training image performance quality model to generate the image performance quality model.

11 . A computer implemented method of controlling a mobile work machine, the computer implemented method comprising;

obtaining, from one or more image providers, a first set of one or more overhead images of a worksite, each overhead image of the first set of one or more overhead images of the worksite including values indicative of a characteristic of the worksite at different geographic locations across the worksite;

identifying an image performance quality metric for each overhead image of the first set of one or more overhead images of the worksite based on an image performance quality model;

determining, based on the image performance quality metrics for the first set of one or more overhead images, a number of images in the first set of one or more overhead images that satisfy a threshold image performance quality;

in response to a determination that the number of images is below a threshold number, obtaining a second set of one or more overhead images of the worksite, each overhead image of the second set of one or more overhead images of the worksite including values indicative of the characteristic of the worksite at the different geographic locations across the worksite;

determining that at least one overhead image of the second set of one or more overhead images of the worksite can be selected, based on image performance quality metrics for the at least one overhead image of the second set of one or more overhead images of the worksite; and

selecting, in response to determining that at least one overhead image of the second set of one or more overhead images of the worksite can be selected, as a selected at least one image, the at least one overhead image of the second set of one or more overhead images of the worksite; and

generating one or more control signals to control the mobile work machine based, at least, on the selected at least one image.

12 . The computer implemented method of claim 11 wherein the number of images is zero.

13 . The computer implemented method of claim 11 and further comprising:

selecting, in response to determining that none of the one or more overhead images of the first set of one or more overhead images of the worksite can be selected, as a selected image, based on the identified image performance quality metrics for the one or more overhead images of the first set of overhead images of the worksite, as a selected image, at least one of the one or more overhead images of the second set of one or more overhead images of the worksite.

14 . The computer implemented method of claim 11 and further comprising:

obtaining, in response to determining that more than the at least one overhead image of the first set of one or more overhead images of the worksite are needed, from the one or more image providers or a different one or more image providers, a second set of one or more overhead images of the worksite, each overhead image of the second set of one or more overhead images of the worksite including values indicative of the characteristic of the worksite at the different geographic locations across the worksite.

15 . The computer implemented method of claim 14 and further comprising:

selecting, in addition to the selected at least one image of the first set of one or more overhead images, at least one of the overhead images of the second set of one or more overhead images of the worksite.

16 . The computer implemented method of claim 11 , wherein the image performance quality model models a relationship between overhead image values and characteristic values of the characteristic of the worksite.

17 . A mobile work machine comprising:

one or more processors;

memory storing computer executable instructions, executable by the one or more processors, the computer executable instructions, when executed by the one or more processors, causing the one or more processors perform steps comprising:

obtaining, from one or more image providers, a first set of one or more overhead images of a worksite, each overhead image of the first set of one or more overhead images of the worksite including values indicative of a characteristic of the worksite at different geographic locations across the worksite;

identifying an image performance quality metric for each overhead image of the first set of one or more overhead images of the worksite based on an image performance quality model, wherein identifying the image performance quality metric for each overhead image of the first set of one or more overhead images of the worksite based on the image performance quality model comprises:

identifying two or more image features of each overhead image of the first set of one or more overhead images of the worksite; and

providing the two or more image features of each overhead image of the first set of one or more overhead images of the worksite to the image performance quality model to obtain, as a model output, the image performance quality metric for each overhead image of the first set of one or more overhead images of the worksite; and

wherein the two or more image features comprise two or more of: (i) image resolution, (ii) image source sensor technology, (iii) timing information, and (iv) image value statistics;

determining if at least one overhead image of the first set of one or more overhead images of the worksite can be selected, as a selected image, based on the identified image performance quality metrics for the at least one overhead image of the first set of one or more overhead images of the worksite; and

selecting, in response to determining that at least one overhead image of the first set of one or more overhead images can be selected, as a selected image, the at least one overhead image of the first set of one or more overhead images of the of the worksite and controlling the mobile work machine based on the at least one selected image.

18 . The mobile work machine of claim 17 , wherein the computer executable instructions, when executed by the one or more processors, further cause the one or more processors to perform steps comprising:

obtaining, in response to determining that none of the one or more overhead images of the first set of one or more overhead images of the worksite can be selected, as a selected image, based on the identified image performance quality metrics for the one or more overhead images of the first set of one or more overhead images of the worksite, a second set of one or more overhead images of the worksite, from the one or more image providers or a different one or more image providers, each overhead image of the second set of one or more overhead images of the worksite including values indicative of the characteristic of the worksite at the different geographic locations across the worksite; and

selecting, in response to determining that none of the one or more overhead images of the first set of one or more overhead images of the worksite can be selected, as a selected image, based on the identified image performance quality metrics for the one or more overhead images of the first set of overhead images of the worksite, as a selected image, at least one of the one or more overhead images of the second set of one or more overhead images of the worksite.

19 . The mobile work machine of claim 17 , wherein the computer executable instructions, when executed by the one or more processors, further cause the one or more processors to perform steps comprising:

obtaining, in response to determining that more than the at least one overhead image of the first set of one or more overhead images of the worksite are needed, from the one or more image providers or a different one or more image providers, a second set of one or more overhead images of the worksite, each overhead image of the second set of one or more overhead images of the worksite including values indicative of the characteristic of the worksite at the different geographic locations across the worksite; and

selecting, in addition to the at least one selected image of the first set of one or more overhead images, as part of the one or more selected images, at least one of the overhead images of the second set of one or more overhead images of the worksite.

20 . The mobile work machine of claim 17 , wherein the image performance quality model models a relationship between overhead image values and characteristic values of the characteristic of the worksite.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 28, 2024
From: REINERS, ALEXANDER C.; JUST, JOHN P.; HEIEN, ERIC M.; DE JESUS FILHO, MARCOS ANTONIO; BURKE, MATTHEW J.
To: DEERE & COMPANY
Reel/Frame 066588/0313 →
Continuity (1)
Related Publication 20250272963A1 · Aug 28, 2025
References Cited (27)
US 5995895A · Watt · 1999 [cited by examiner]
US 8527157B2 · Imhof et al. · 2013 [cited by applicant]
US 10568316B2 · Gall et al. · 2020 [cited by applicant]
US 10684137B2 · Kean · 2020 [cited by examiner]
US 10866109B2 · Madsen et al. · 2020 [cited by applicant]
US 10909368B2 · Guo et al. · 2021 [cited by applicant]
US 20030040971A1 · Freedenberg et al. · 2003 [cited by applicant]
US 20050149235A1 · Seal et al. · 2005 [cited by applicant]
US 20110056178A1 · Sauerwein et al. · 2011 [cited by applicant]
US 20110270494A1 · Imhof et al. · 2011 [cited by applicant]
US 20110270495A1 · Knapp · 2011 [cited by applicant]
US 20150254800A1 · Johnson et al. · 2015 [cited by applicant]
US 20170115862A1 · Stratton et al. · 2017 [cited by applicant]
US 20170223947A1 · Gall et al. · 2017 [cited by applicant]
US 20170318743A1 · Sauder et al. · 2017 [cited by applicant]
US 20180101728A1 · Zhong · 2018 [cited by examiner]
US 20180129879A1 · Achtelik et al. · 2018 [cited by applicant]
US 20190050948A1 · Perry et al. · 2019 [cited by applicant]
US 20190114847A1 · Wagner et al. · 2019 [cited by applicant]
US 20190129430A1 · Madsen et al. · 2019 [cited by applicant]
US 20200323134A1 · Dan et al. · 2020 [cited by applicant]
US 20210321566A1 · Darr et al. · 2021 [cited by applicant]
US 20220406055A1 · Rooney · 2022 [cited by applicant]
EP 3469878A1 · 2019 [cited by applicant]
WO 2023282787A1 · 2023 [cited by applicant]
Prosecution History of U.S. Appl. No. 16/360,691 including: Application and Drawings filed on Apr. 10, 2019 and Notice of Allowance dated Aug. 18, 2021, 47 pages. [cited by applicant]
U.S. Appl. No. 17/360,747 Application and Drawings filed on Jun. 28, 2021, 42 pages. [cited by applicant]