IP Library Granted Patent US 12,568,876
Granted Patent B2
US 12,568,876 · App. 18/006,893 · Granted Mar 10, 2026

Method for classifying plants for agricultural purposes

Inventors: Andreas Weimer (Gaertringen, DE); Farid Khani (Boennigheim, DE)
Assignee: ROBERT BOSCH GMBH
A01C23/007A01G13/00A01G25/09A01M7/0089G05B19/416G06V10/764G06V10/98G06V20/188G05B2219/45013
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 12,568,876
App. No.
18/006,893
Filed
Jan 26, 2023
Granted
Mar 10, 2026
Kind
B2
Art Unit
2668
USPC
382/100
Abstract

A method for classifying plants for agricultural purposes. The method includes: detecting a field section having plants using an optical detection unit inclined toward the field section to obtain image information having a rear image edge having an image width and a front image edge having a larger image width; classifying the plants in the image information into first and second plant classes using an algorithm to treat the plants of the second plant class with a liquid spray agent, in accordance with the classification. The step of classifying comprises a step of reclassifying, in which classified plants are reclassified from the second plant class to the first plant class using the algorithm if they extend in a rear image section of the image information defined by a defined image separation line from a lateral image edge of the image information maximally up to a defined reclassification boundary line.

Claims (39)

1 . A method for classifying plants for agricultural purposes, the method comprising the following steps:

detecting a field section having plants using an optical detection unit inclined toward the field section to obtain image information having a rear image edge having an image width and a front image edge having a larger image width; and

classifying the plants in the obtained image information into at least one first plant class and into at least one second plant class using an algorithm in order to treat the second plant class plants with a medium in accordance with the classification, the medium being a liquid spray agent;

wherein the step of classifying includes a step of reclassifying, in which classified plants are reclassified from the second plant class into the first plant class by the algorithm, when the classified plants extend in a rear image section of the image information defined by a defined image separation line from a lateral image edge of the image information maximally up to a defined reclassification boundary line.

2 . The method according to claim 1 , wherein the image information includes an upper overlapping area with forward image information situated spatially and/or temporally upstream, wherein a position of the image separation line in the image information is selected depending on the upper overlapping area, the image separation line being aligned with a rear overlapping area of the upper overlapping area.

3 . The method according to claim 1 , wherein the image information has a lateral overlapping area having at least one laterally arranged and/or simultaneously detected item of lateral image information, wherein a position of the reclassification boundary line in the image information is selected depending on the lateral overlapping area, and is selected to intersect a lateral overlapping edge of the lateral overlapping area.

4 . The method according to claim 3 , wherein the image separation line, the reclassification boundary line, and the lateral overlapping edge have a common intersection point.

5 . The method according to claim 1 , wherein, in the image information:

the image separation line is arranged horizontally, and/or

the reclassification boundary line is arranged vertically in the image information.

6 . The method according to claim 1 , wherein both lateral image edges of the image information are each assigned a reclassification boundary line.

7 . The method according to claim 1 , further comprising identifying plant rows in the image information using classified plants of the first plant class in the image information using an algorithm.

8 . The method according to claim 1 , wherein agricultural crops and/or cultivated plants are classified into the first plant class, and/or weeds are classified into the second plant class.

9 . A method for classifying plants for agricultural purposes, the method comprising the following steps:

detecting a field section having plants using an optical detection unit inclined toward the field section to obtain image information having a rear image edge having an image width and a front image edge having a larger image width; and

classifying the plants in the obtained image information into at least one first plant class and into at least one second plant class using an algorithm in order to treat the second plant class plants with a medium in accordance with the classification, the medium being a liquid spray agent;

wherein the step of classifying includes a step of reclassifying, in which classified plants are reclassified from the second plant class into the first plant class by the algorithm, when the classified plants extend in a rear image section of the image information defined by a defined image separation line from a lateral image edge of the image information maximally up to a defined reclassification boundary line,

wherein in the step of classifying, plants classified into the second plant class remain in the second plant class, when they are not found in the rear image section of the image information and/or are not abutting the lateral image edge of the image information and/or extend from the lateral image edge of the image information across the reclassification boundary.

10 . A method for classifying plants for agricultural purposes, the method comprising the following steps:

detecting a field section having plants using an optical detection unit inclined toward the field section to obtain image information having a rear image edge having an image width and a front image edge having a larger image width;

classifying the plants in the obtained image information into at least one first plant class and into at least one second plant class using an algorithm in order to treat the second plant class plants with a medium in accordance with the classification, the medium being a liquid spray agent; and

determining a plant identification number for a defined image evaluation area in the obtained image information using classified plants of the second plant class in the defined image evaluation area using an algorithm,

wherein the step of classifying includes a step of reclassifying, in which classified plants are reclassified from the second plant class into the first plant class by the algorithm, when the classified plants extend in a rear image section of the image information defined by a defined image separation line from a lateral image edge of the image information maximally up to a defined reclassification boundary line.

11 . The method according to claim 10 , wherein the image evaluation area has a width greater than or equal to the rear image width and less than the front image width of the image information.

12 . The method according to claim 10 , wherein the plant identification number represents a degree of coverage of the image evaluation area by plant material and/or an amount of a plant material in the image evaluation area and/or a number of classified plants of the second plant class in the image evaluation area.

13 . The method according to claim 10 , further comprising applying the medium to the classified plants of the second plant class depending on the determined plant identification number, using an agricultural spray device.

14 . A classification unit, configured to:

classify plants in an item of obtained image information into at least one first plant class and into at least one second plant class using an algorithm; and

reclassify classified plants from the second plant class into the first plant class using the algorithm when the classified plants extend in a rear image section of the image information defined by a defined image separation line from a lateral image edge of the image information maximally up to a defined reclassification boundary line.

15 . An agricultural spray device for applying a medium to plants of an agricultural area, the medium being a liquid spray agent, the device comprising:

at least one spray nozzle unit configured to apply the medium;

at least one optical detection unit configured to detect the plants; and

a classifier configured to classify the plants, the classifier configured to:

classify plants in an item of obtained image information into at least one first plant class and into at least one second plant class using an algorithm; and

reclassify classified plants from the second plant class into the first plant class using the algorithm when the classified plants extend in a rear image section of the image information defined by a defined image separation line from a lateral image edge of the image information maximally up to a defined reclassification boundary line.

16 . A non-transitory machine-readable storage medium on which is stored a computer program for classifying plants for agricultural purposes, the computer program, when executed by a computer, causing the computer to perform the following steps:

detecting a field section having plants using an optical detection unit inclined toward the field section to obtain image information having a rear image edge having an image width and a front image edge having a larger image width; and

classifying the plants in the obtained image information into at least one first plant class and into at least one second plant class using an algorithm in order to treat the second plant class plants with a medium in accordance with the classification, the medium being a liquid spray agent;

wherein the step of classifying includes a step of reclassifying, in which classified plants are reclassified from the second plant class into the first plant class by the algorithm, when the classified plants extend in a rear image section of the image information defined by a defined image separation line from a lateral image edge of the image information maximally up to a defined reclassification boundary line.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 16, 2023
From: WEIMER, ANDREAS; KHANI, FARID
To: ROBERT BOSCH GMBH
Reel/Frame 063003/0950 →
Priority Claims (1)
DE 10 2020 215 876.2 · Dec 15, 2020 · national
Continuity (1)
Related Publication 20230301228A1 · Sep 28, 2023
References Cited (7)
US 10255670B1 · Wu · 2019 [cited by examiner]
DE 102017210804A1 · 2018 [cited by applicant]
DE 102017217258A1 · 2019 [cited by applicant]
DE 102018217742A1 · 2020 [cited by applicant]
DE 102018222428A1 · 2020 [cited by applicant]
Shrestha, D. S., Brian L. Steward, and Stuart J. Birrell. “Video processing for early stage maize plant detection.” Biosystems engineering 89.2 (2004): 119-129. (Year: 2004). [cited by examiner]
International Search Report for PCT/EP2021/082920, Issued Mar. 17, 2022. [cited by applicant]