Icon

Plant Disease Prediction

The workflow demonstrates how to do transfer-learning. We are using the pre-trained deep learning network Resnet50 and are repurposing it to predict two apple tree diseases, rust and scab, based on images of apple leaves. To do so, we are freezing the weights of the first third of the layers and only retrained the weights of the remaining two thirds of the layers. Additionally, we are using the Keras Imaga Data Generator for image augmentation.

Plant Disease PredictionThis workflow reads and preprocesses images of apple tree leaves to predict the two apple tree diseases, rust and scab, using the pretrained Resnet50 network. ground truthload imagebest_model.h5Load Resnet50multiple disease to scabmultiple disease to rustextract file name List Files/Folders Path to String CSV Reader RowID Image Reader(Table) DL PythonNetwork Learner Keras Freeze Layers Partitioning DL PythonNetwork Creator DL PythonNetwork Editor Column Filter Rule Engine Rule Engine String Manipulation Renderer to Image View Results Image Preprocessing Joiner Joiner Column Expressions DL Python NetworkExecutor Conda EnvironmentPropagation Plant Disease PredictionThis workflow reads and preprocesses images of apple tree leaves to predict the two apple tree diseases, rust and scab, using the pretrained Resnet50 network. ground truthload imagebest_model.h5Load Resnet50multiple disease to scabmultiple disease to rustextract file nameList Files/Folders Path to String CSV Reader RowID Image Reader(Table) DL PythonNetwork Learner Keras Freeze Layers Partitioning DL PythonNetwork Creator DL PythonNetwork Editor Column Filter Rule Engine Rule Engine String Manipulation Renderer to Image View Results Image Preprocessing Joiner Joiner Column Expressions DL Python NetworkExecutor Conda EnvironmentPropagation

Nodes

Extensions

Links