DEPRECATED

Deep Plant Phenomics is no longer actively maintained. It is available here for historical purposes - however, it is provided as-is with no updates or bug fixes planned.

See this thread for discussion.

Deep Plant Phenomics

Deep Plant Phenomics (DPP) is a platform for plant phenotyping using deep learning. Think of it as Keras for plant scientists.

DPP integrates Tensorflow for learning. This means that it is able to run on both CPUs and GPUs, and scale easily across devices.

Read the doumentation for tutorials, or see the included examples. You can also read the paper.

DPP is maintained at the Plant Phenotyping and Imaging Research Center (P2IRC) at the University of Saskatchewan.

What's Deep Learning?

Principally, DPP provides deep learning functionality for plant phenotyping and related applications. Deep learning is a category of techniques which encompasses many different types of neural networks. Deep learning techniques lead the state of the art in many image-based tasks, including image classification, object detection and localization, image segmentation, and others.

What Can I Do With This?

This package provides two things:

1. Useful tools made possible using pre-trained neural networks

For example, calling tools.predict_rosette_leaf_count(my_files) will use a pre-trained convolutional neural network to estimate the number of leaves on each rosette plant.

2. An easy way to train your own models

For example, using a few lines of code you can easily use your data to train a convolutional neural network to rate plants for biotic stress. See the tutorial for how the leaf counting model was built.

Features

• Several trained networks for common plant phenotyping tasks.
• Easy ways to load data.
  • Loaders for some popular plant phenotyping datasets.
  • Plenty of different loaders for your own data, however it exists.
• Support for semantic segmentation.
• Support for object detection.
• Support for object counting via density estimation, including Countception networks.
• Support for classification and regression tasks.
• Tensorboard integration for visualization.
• Easy-to-use API for building new models.
  • Pre-defined neural network architectures so you don't have to make your own.
  • Several data augmentation options.
  • Many ready-to-use neural network layers.
• Easy to deploy your own models as a Python function!

Example Usage

Train a simple regression model:

```python import deepplantphenomics as dpp

model = dpp.RegressionModel(debug=True)

3 channels for colour, 1 channel for greyscale

channels = 3

Setup and hyperparameters

model.setbatchsize(64) model.setimagedimensions(256, 256, channels) model.setmaximumtrainingepochs(25) model.settestsplit(0.2) model.setvalidation_split(0.0)

Load dataset of images and ground-truth labels

model.loadmultiplelabelsfromcsv('./data/mylabels.csv') model.loadimageswithidsfromdirectory('./data')

Use a predefined model

model.usepredefinedmodel('vgg-16')

Train!

model.begin_training() ```

Installation

1. git clone https://github.com/p2irc/deepplantphenomics.git
2. pip install ./deepplantphenomics

Note: The package now requires Python 3.6 or greater. Python 2.7 is no longer supported.