Method and device for classifying plants, and computer program product
A method for classifying plants. In the method, the plants or their plant components, in particular plant leaves, are detected in an evaluation region with the aid of an optical and/or infrared detection unit, and the detected image data of the detection unit are evaluated with the aid of an algorithm, via the evaluation a first plant type being distinguished from a second plant type, the second plant type being treated in particular with a medium, preferably a liquid spray, and the first plant type being planted in the form of a plant row in a tubular cultivation row tube.
1 . A method for classifying plants, the method comprising the following steps:
detecting the plants or plant components of the plants in an evaluation region using an optical and/or infrared detection unit;
evaluating detected image data of the detection unit using an algorithm, wherein, via the evaluation, a first plant type is distinguished from a second plant type; and
treating the second plant type with a medium;
wherein the first plant type is planted in the form of a plant row in a tubular cultivation row tube, the cultivation row tube having a first predetermined width, and the method further comprises:
generating, around the cultivation row tube, a neutral area that encloses the cultivation row tube in a shape of a tube at its longitudinal sides, the neutral area having a second predetermined width; and
generating, around the neutral area, a far area that encloses the neutral area at its longitudinal sides;
wherein those of the plants or plant components that are situated within the neutral area and that are associated with the plants or plant components situated within the cultivation row tube are classified as the first plant type, and those of the plants or plant components that extend from the cultivation row tube, across the neutral area, and into the far area are analyzed based on further features in the far area in order to classify the plants or plant components either as the first plant type or as the second plant type.
2 . The method as recited in claim 1 , wherein the plant components include leaves, and the medium is a liquid spray.
3 . The method as recited in claim 1 , wherein the further features include a geometric dimension of the plants in the far area, including a size of a surface area covered by the plants in the far area and/or a height, and the plants or plant components are assigned to the first plant type when the size of the surface area and/or the height of the plants or plant components is below a limiting value.
4 . The method as recited in claim 1 , wherein all of the plants or plant components that are associated with the plants or plant components that protrude into the far area are classified as the second plant type when the plants or plant components that protrude into the far area are classified as the second plant type.
5 . The method as recited in claim 1 , wherein all of the plants or plant components that are situated within the neutral area and that are not associated with the plants or plant components situated within the cultivation row tube are classified as the second plant type.
6 . The method as recited in claim 1 , wherein a width of the cultivation row tube and a width of the neutral area are adapted as a function of the first plant type, a growth size, and a growth level.
7 . The method as recited in claim 1 , wherein the detection unit is moved in a longitudinal direction of the cultivation row tube.
8 . The method as recited in claim 1 , wherein image data from two evaluation regions that are evaluated in succession partially overlap one another.
9 . A device which is part of an agricultural vehicle, comprising:
an optical and/or infrared detection unit; and
a control device including an algorithm, the control device configured to classify plants, the control device configured to:
detect the plants or plant components of the plants in an evaluation region using the detection unit;
evaluate detected image data of the detection unit using the algorithm, wherein, via the evaluation, a first plant type is distinguished from a second plant type; and
treat the second plant type with a medium;
wherein the first plant type is planted in the form of a plant row in a tubular cultivation row tube, the cultivation row tube having a first predetermined width, and the control device is further configure to:
generate, around the cultivation row tube, a neutral area that encloses the cultivation row tube in a shape of a tube at its longitudinal sides, the neutral area having a second predetermined width; and
generate, around the neutral area, a far area that encloses the neutral area at its longitudinal sides;
wherein those of the plants or plant components that are situated within the neutral area and that are associated with the plants or plant components situated within the cultivation row tube are classified as the first plant type, and those of the plants or plant components that extend from the cultivation row tube, across the neutral area, and up into the far area are analyzed based on further features in the far area in order to classify the plants or plant components either as the first plant type or as the second plant type.
10 . The device as recited in claim 9 , further comprising:
a spray unit including at least one selectively activatable spray nozzle for treating plants or plant components of the second plant type.
11 . The device as recited in claim 9 , wherein all of the plants or plant components that are situated within the neutral area and that are not associated with the plants or plant components situated within the cultivation row tube are classified as the second plant type.
12 . A non-transitory data medium on which is stored a computer program for classifying plants, the computer program, when executed by a computer, causing the computer to perform the following steps:
detecting the plants or plant components of the plants in an evaluation region using an optical and/or infrared detection unit;
evaluating detected image data of the detection unit using an algorithm, wherein, via the evaluation, a first plant type is distinguished from a second plant type; and
treating the second plant type with a medium;
wherein the first plant type is planted in the form of a plant row in a tubular cultivation row tube, the cultivation row tube having a first predetermined width, and wherein the computer program, when executed by the computer, further causes the computer to perform the following steps:
generating, around the cultivation row tube, a neutral area that encloses the cultivation row tube in a shape of a tube at its longitudinal sides, the neutral area having a second predetermined width; and
generating, around the neutral area, a far area that encloses the neutral area at its longitudinal sides;
wherein those of the plants or plant components that are situated within the neutral area and that are associated with the plants or plant components situated within the cultivation row tube are classified as the first plant type, and those of the plants or plant components that extend from the cultivation row tube, across the neutral area, and up into the far area are analyzed based on further features in the far area in order to classify the plants or plant components either as the first plant type or as the second plant type.
13 . The non-transitory data medium of claim 12 , wherein all of the plants or plant components that are situated within the neutral area and that are not associated with the plants or plant components situated within the cultivation row tube are classified as the second plant type.