Leaf Vein Network CNN Images

This download site contains the CNN vein network predictions and set of Matlab programs  that were used for the analyses in Xu et al., (2020) and Blonder et al., (2020). These require Matlab 2020a or later. They may work on earlier versions of MatLab, but this has not been tested and cannot be guaranteed.

The files are as follows:

Zip files (e.g. BEL_downsampled_images.zip) containing a complete set of images of leaf vein predictions from a fully trained convolutional neural network (CNN), along with the ground truth data. Each folder in the unzipped file contains a sample represented by a CODE with format X-TY-BZ. X represents the name of a plot in the Global Ecosystems Monitoring network database (e.g. 'BEL'). Tree (T) Y indicates the number of a tree within a plot (e.g. '101') and Z represents the light stratum of the canopy where the leaf was collected (either 'S' for 'sunlit' or 'SH' for 'shaded').

A set of Matlab programs (Matlab files.zip) to compare the CNN predictions against other vein extraction approaches.

A Matlab Readme file with instructions on how to run the analyses.

References

Software GUI:

Xu, H., Blonder, B., Jodra, M., Malhi, Y. and Fricker, M.D. (2020) Automated and accurate segmentation of leaf venation networks via deep learning. New Phytol. (In press).

Analysis of trait data:

Blonder, B., S. Both, M. Jodra, H. Xu, M. Fricker, I. S. Matos, N. Majalap, D. F. R. P. Burslem, Y. Teh and Y. Malhi (2020) Linking functional traits to multiscale statistics of leaf venation networks. New Phytol. (In press).

Original image data set and ground truths

Blonder, B., Both, S., Jodra, M., Majalap, N., Burslem, D., Teh, Y. A., and Malhi, Y. (2019) Leaf venation networks of Bornean trees: images and hand‐traced segmentations. Ecology 100: e02844.10.1002/ecy.2844.

Available from: https://ora.ox.ac.uk/objects/uuid:de65fc07-4b8f-4277-a6c4-82836afbdeb3