公开(公告)号：US12073327B2

公开(公告)日：2024-08-27

申请号：US17761849

申请日：2020-09-29

Applicant: BASF SE

Inventor： Aitor Alvarez Gila ,  Amaia Maria Ortiz Barredo ,  David Roldan Lopez ,  Javier Romero Rodriguez ,  Corinna Maria Spangler ,  Christian Klukas ,  Till Eggers ,  Jone Echazarra Huguet ,  Ramon Navarra Mestre ,  Artzai Picon Ruiz ,  Aranzazu Bereciartua Perez

IPC: G06T7/00 ,  G06N3/082 ,  G06T7/11 ,  G06V10/44 ,  G06V10/56 ,  G06V10/762 ,  G06V10/764 ,  G06V10/82 ,  G06V20/10

CPC classification number: G06N3/082 ,  G06T7/0012 ,  G06T7/11 ,  G06V10/454 ,  G06V10/56 ,  G06V10/762 ,  G06V10/764 ,  G06V10/82 ,  G06V20/10 ,  G06V20/188 ,  G06T2207/20081 ,  G06T2207/20084

Abstract: Quantifying plant infestation is performed by estimating the number of biological objects (132) on parts (122) of a plant (112). A computer (202) receives a plant-image (412) taken from a particular plant (112). The computer (202) uses a first convolutional neural network (262/272) to derive a part-image (422) that shows a part of the plant. The computer (202) splits the part-image into tiles and uses a second network to process the tiles to density maps. The computer (202) combines the density maps to a combined density map in the dimension of the part-image and integrates the pixel values to an estimate number of objects for the part. Object classes (132(1), 132(2)) can be differentiated to fine-tune the quantification to identify class-specific countermeasures.

公开(公告)号：US12148197B2

公开(公告)日：2024-11-19

申请号：US17611517

申请日：2020-05-14

Applicant: BASF SE

Inventor： Artzai Picon Ruiz ,  Matthias Nachtmann ,  Maximilian Seitz ,  Patrick Mohnke ,  Ramon Navarra-Mestre ,  Alexander Johannes ,  Till Eggers ,  Amaia Maria Ortiz Barredo ,  Aitor Alvarez-Gila ,  Jone Echazarra Huguet

IPC: G06V10/764 ,  G06N3/048 ,  G06T7/00 ,  G06T7/10 ,  G06V10/82 ,  G06V20/10

Abstract: A computer-implemented method, computer program product and computer system (100) for detecting plant diseases. The system stores a convolutional neural network (120) trained with a multi-crop dataset. The convolutional neural network (120) has an extended topology comprising an image branch (121) based on a classification convolutional neural network for classifying the input images according to plant disease specific features, a crop identification branch (122) for adding plant species information, and a branch integrator for integrating the plant species information with each input image. The plant species information (20) specifies the crop on the respective input image (10). The system receives a test input comprising an image (10) of a particular crop (1) showing one or more particular plant disease symptoms, and further receives a respective crop identifier (20) associated with the test input via an interface (110). A classifier module (130) of the system applies the trained convolutional network (120) to the received test input, and provides a classification result (CR1) according to the output vector of the convolutional neural network (120). The classification result (CR1) indicates the one or more plant diseases associated with the one or more particular plant disease symptoms.

公开(公告)号：US20230351743A1

公开(公告)日：2023-11-02

申请号：US17761849

申请日：2020-09-29

Applicant: BASF SE

Inventor： Aitor ALVAREZ GILA ,  Amaia Maria Ortiz Barredo ,  David Roldan Lopez ,  Javier Romero Rodriguez ,  Corinna Maria Spangler ,  Christian Klukas ,  Till Eggers ,  Jone Echazarra Huguet ,  Ramon Navarra Mestre ,  Artzai Picon Ruiz ,  Aranzazu Bereciartua Perez

IPC: G06V20/10 ,  G06V10/44 ,  G06V10/764 ,  G06T7/00 ,  G06T7/11

CPC classification number: G06V20/188 ,  G06V10/454 ,  G06V10/764 ,  G06T7/0012 ,  G06T7/11 ,  G06T2207/20081 ,  G06T2207/20084

Abstract: Quantifying plant infestation is performed by estimating the number of biological objects (132) on parts (122) of a plant (112). A computer (202) receives a plant-image (412) taken from a particular plant (112). The computer (202) uses a first convolutional neural network (262/272) to derive a part-image (422) that shows a part of the plant. The computer (202) splits the part-image into tiles and uses a second network to process the tiles to density maps. The computer (202) combines the density maps to a combined density map in the dimension of the part-image and integrates the pixel values to an estimate number of objects for the part. Object classes (132(1), 132(2)) can be differentiated to fine-tune the quantification to identify class-specific countermeasures.