Data Augmentation From RGB to Chlorophyll Fluorescence Imaging Application to Leaf Segmentation of Arabidopsis Thaliana From Top View Images

In this report we investigate various strategies to boost the performance for leaf segmentation of Arabidopsis thaliana in chlorophyll fluorescent imaging without any manual annotation. Direct conversion of RGB images to gray levels picked from CVPPP challenge or from a virtual Arabidopsis thaliana simulator are tested together with synthetic noisy versions of these. Segmentation performed with a state of the art U-Net convolutional neural network is shown to benefit from these approaches with a Dice coefficient between 0.95 and 0.97 on the segmentation of the border of the leaves. A new annotated dataset of fluorescent images is made available.

[1]  Hanno Scharr,et al.  Image Analysis: The New Bottleneck in Plant Phenotyping [Applications Corner] , 2015, IEEE Signal Processing Magazine.

[2]  Etienne Belin,et al.  Multiscale imaging of plants: current approaches and challenges , 2015, Plant Methods.

[3]  D. Rousseau,et al.  High throughput quantitative phenotyping of plant resistance using chlorophyll fluorescence image analysis , 2013, Plant Methods.

[4]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[5]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[6]  Hanno Scharr,et al.  Machine Learning for Plant Phenotyping Needs Image Processing. , 2016, Trends in plant science.

[7]  Hanno Scharr,et al.  Finely-grained annotated datasets for image-based plant phenotyping , 2016, Pattern Recognit. Lett..

[8]  Jochen Hemming,et al.  Data synthesis methods for semantic segmentation in agriculture: A Capsicum annuum dataset , 2018, Comput. Electron. Agric..

[9]  Alexey Shvets,et al.  TernausNetV2: Fully Convolutional Network for Instance Segmentation , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[10]  Alexandr A. Kalinin,et al.  Albumentations: fast and flexible image augmentations , 2018, Inf..

[11]  Ayan Chaudhury.  Computer Vision Problems in 3D Plant Phenotyping , 2017 .

[12]  Jeffrey C. Berry,et al.  Quantitative, Image-Based Phenotyping Methods Provide Insight into Spatial and Temporal Dimensions of Plant Disease1[OPEN] , 2016, Plant Physiology.

[13]  Ciro Potena,et al.  Automatic model based dataset generation for fast and accurate crop and weeds detection , 2016, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[14]  Hanno Scharr,et al.  ARIGAN: Synthetic Arabidopsis Plants using Generative Adversarial Network , 2017, bioRxiv.

[15]  Nicolas Courty,et al.  Optimal Transport for Domain Adaptation , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  C. Ura,et al.  Quantitative Mapping of Leaf Photosynthesis using Chlorophyll Fluorescence Imaging , 1994 .

[17]  T. Pridmore,et al.  Computer Vision Problems in Plant Phenotyping, CVPPP 2017: Introduction to the CVPPP 2017 Workshop Papers , 2017, 2017 IEEE International Conference on Computer Vision Workshops (ICCVW).

[18]  Kobus Barnard,et al.  Gaussian Process Shape Models for Bayesian Segmentation of Plant Leaves , 2015 .

[19]  Thomas Walter,et al.  Segmentation of Nuclei in Histopathology Images by Deep Regression of the Distance Map , 2019, IEEE Transactions on Medical Imaging.

[20]  Gregory R. Johnson,et al.  Label-free prediction of three-dimensional fluorescence images from transmitted light microscopy , 2018, Nature Methods.

[21]  Serge Beucher,et al.  THE WATERSHED TRANSFORMATION APPLIED TO IMAGE SEGMENTATION , 2009 .

[22]  Jerry L. Prince,et al.  Cross-modality image synthesis from unpaired data using CycleGAN: Effects of gradient consistency loss and training data size , 2018, SASHIMI@MICCAI.

[23]  C. Granier,et al.  Quantifying spatial heterogeneity of chlorophyll fluorescence during plant growth and in response to water stress , 2015, Plant Methods.

[24]  S. Rolfe,et al.  Chlorophyll fluorescence imaging of plant–pathogen interactions , 2010, Protoplasma.

[25]  Warren L. Butler,et al.  Energy Distribution in the Photochemical Apparatus of Photosynthesis , 1978 .

[26]  Sotirios A. Tsaftaris,et al.  An interactive tool for semi-automated leaf annotation , 2015 .

[27]  Jean-Michel Pape,et al.  Utilizing machine learning approaches to improve the prediction of leaf counts and individual leaf segmentation of rosette plant images , 2015 .

[28]  Christian Klukas,et al.  3-D Histogram-Based Segmentation and Leaf Detection for Rosette Plants , 2014, ECCV Workshops.

[29]  Joaquin Vanschoren,et al.  Data Augmentation using Conditional Generative Adversarial Networks for Leaf Counting in Arabidopsis Plants , 2018, BMVC.

[30]  Serge Beucher,et al.  Use of watersheds in contour detection , 1979 .

[31]  Julien Cohen-Adad,et al.  AxonDeepSeg: automatic axon and myelin segmentation from microscopy data using convolutional neural networks , 2017, Scientific Reports.

[32]  Sébastien Ourselin,et al.  Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations , 2017, DLMIA/ML-CDS@MICCAI.

[33]  Peyman Moghadam,et al.  Deep Leaf Segmentation Using Synthetic Data , 2018, BMVC.