FIELD: electronic data processing.
SUBSTANCE: invention relates in particular to image processing methods and systems for supporting the detection of plant diseases. The system stores a convolutional neural network (120) trained on a multi-crop dataset. The convolutional neural network (120) has an extended topology including an image branch (121) based on a classification convolutional neural network for classifying input images according to certain plant disease traits, a crop identification branch (122) for adding species information plants and a branch integrator to integrate plant species information with each input image. Plant species information (20) identifies the crop in the corresponding input image (10). The system receives test input data containing an image (10) of a specific crop (1) showing one or more specific plant disease symptoms, and further receives the corresponding crop identifier (20) associated with the test input data through an interface (110). The system classifier module (130) applies the trained convolutional network (120) to the received test input data and provides a classification result (CR1) according to the output vector of the convolutional neural network (120). The classification result (CR1) indicates one or more plant diseases associated with one or more specific plant disease symptoms.
EFFECT: increase in the accuracy and information content of the obtained data.
15 cl, 8 dwg
Title | Year | Author | Number |
---|---|---|---|
AUTOMATIC DETERMINATION OF SET OF CATEGORIES FOR DOCUMENT CLASSIFICATION | 2018 |
|
RU2701995C2 |
FUEL COMBUSTION MODES MONITORING SYSTEM BY MEANS OF TORCH IMAGES ANALYSIS USING CLASSIFIER BASED ON CONVOLUTIONAL NEURAL NETWORK | 2018 |
|
RU2713850C1 |
DETECTION OF PLANTS DISEASES BY CLASSIFICATION OF PLANTS PHOTOGRAPHY | 2019 |
|
RU2805670C2 |
SYSTEM AND METHOD FOR ARTIFICIAL NEURAL NETWORK INVARIANT TO TRANSFERRING | 2017 |
|
RU2656990C1 |
METHOD OF RECOGNIZING SPEECH EMOTIONS USING 3D CONVOLUTIONAL NEURAL NETWORK | 2023 |
|
RU2816680C1 |
CHARACTER RECOGNITION USING A HIERARCHICAL CLASSIFICATION | 2018 |
|
RU2693916C1 |
METHOD AND APPARATUS FOR IDENTIFYING PORTIONS OF FRAGMENTED MATERIAL WITHIN AN IMAGE | 2016 |
|
RU2694021C1 |
NEURAL NETWORK FOR GENERATING SYNTHETIC MEDICAL IMAGES | 2017 |
|
RU2698997C1 |
GAN-CNN FOR PREDICTING MHC-PEPTIDE BINDING | 2019 |
|
RU2836823C2 |
GAN-CNN FOR PREDICTION OF MHC-PEPTIDE BINDING | 2019 |
|
RU2777926C2 |
Authors
Dates
2024-03-18—Published
2020-05-14—Filed