Deep Learning for Multi-Task Plant Phenotyping

Michael P. Pound, Jonathan A. Atkinson, Darren M. Wells, Tony P. Pridmore, Andrew P. French; The IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2055-2063


There is a particular phenotyping demand to accurately quantify images of crops, and the natural variability and structure of these plants presents unique difficulties. Recently, machine learning approaches have shown impressive results in many areas of computer vision, but these rely on large datasets that are at present not available for crops. We present a new dataset, called ACID, that provides hundreds of accurately annotated images of wheat spikes and spikelets, along with image level class annotation. We then present a deep learning approach capable of accurately localising wheat spikes and spikelets, despite the varied nature of this dataset. As well as locating features, our network offers near perfect counting accuracy for spikes (95.91%) and spikelets (99.66%). We also extend the network to perform simultaneous classification of images, demonstrating the power of multi-task deep architectures for plant phenotyping.

Related Material

author = {Pound, Michael P. and Atkinson, Jonathan A. and Wells, Darren M. and Pridmore, Tony P. and French, Andrew P.},
title = {Deep Learning for Multi-Task Plant Phenotyping},
booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
month = {Oct},
year = {2017}