TreeAI Global Initiative - Advancing tree species identification from aerial images with deep learning

TreeAI - Advancing Tree Species Identification from Aerial Images with Deep Learning

Data Structure for the TreeAI Database Used in the TreeAI4Species Competition

The data are in the COCO format, each folder contains training and validation subfolders with images and labels with the tree species ID.

Training: Images (.png) and Labels (.txt)

Validation: Images (.png) and Labels (.txt)

Images: RGB bands, 8-bit, chip size 640 x 640 pixels = 32 x 32 m, 5 cm pixel spatial resolution.

Labels: labels are prepared for object detection tasks, the number of classes varies per dataset, e.g. dataset 12_RGB_all_L has 53 classes, and the Latin name of the species is given for each class ID in the file named classDatasetName.xlsx.

Species class: classDatasetName.xlsx contains 3 columns Species_ID, Labels (number of labels), and Species_Class (Latin name of the species).

Masked images: The data set with partial labels was masked, i.e. a buffer of 30 pixels was created around a label, and the image was masked based on these buffers, e.g. 34_RGB_all_L_PascalVoc_640Mask.

Additional filters to clean up the data:Labels at the edge: only images with labels at the edge were removed.Valid labels: images with labels that were completely within an image have been retained. 

Table 1. Description of the datasets included in the TreeAI database.

a) Fully labeled images (i.e. the image has all the trees delineated and each polygon has species information)

b) Partially labeled images (i.e. the image has only some trees delineated, and each polygon has species information)

No.
Dataset name
Training images
Validation images
Fully labeled
Partially labeled

1
12_RGB5cm_FullyLabeled
1066
304
x

2
ObjectDetection_TreeSpecies
422
84
x

3
34_RGB_all_L_PascalVoc_640Mask
951
272

x

4
34_RGB_PartiallyLabeled640
917
262

x

Steps to access the dataset and participate in the TreeAI4Species competition:

Register: Access to the data will be granted upon registering for the competition, see the registration form: https://form.ethz.ch/research/tree-ai-global-database/treeai-competition.html 

Request the dataset: Download the competition record after registration by requesting it. Enter your full name, purpose e.g. accept the TreeAI4Species data license, affiliation, and the country of affiliation in the request. This allows us to check whether you are already registered.

Test dataset: Only the participants registered for the competition will receive the test dataset.

Submit your DL models for evaluation by June 2025.

Award: The best models win a prize.

Publication: All participants in the competition who submit the required files for evaluation will be included in the subsequent publication.

License

== CC BY-NC-ND (Attribution-NonCommercial-NoDerivatives) ==

Dear user,

We appreciate your interest in the TreeAI4Species Competition:  https://form.ethz.ch/research/tree-ai-global-database.html

DATA ANALYSIS AND PUBLICATION

The TreeAI database is released under a variant of the CC BY-NC-ND license. This database is confidential and can be used only for the TreeAI4Species data science competition. It is not permitted to pass on the data or the characteristics directly derived from it to third parties. Written consent from the data supplier is required for use for any other purpose. 

LIABILITY

The data are based on the current state of existing scientific knowledge. However, there is no liability for the completeness. This is the first version of the database, and we plan to improve the tree annotations and include new tree species. Therefore, another version will be released in the future.

The data can only be used for the purpose described by the user when requesting the data.

------------------------------------------------------

ETH Zürich

Dr. Mirela Beloiu Schwenke

Institute of Terrestrial Ecosystems 

Department of Environmental Systems Science, CHN K75

Universitätstrasse 16, 8092 Zürich, Schweiz

[e-mail]