Peach Tree Disease Detection Dataset

Citation Author(s):
Christos
Chaschatzis
Ilias
Siniosoglou
Anna
Triantafyllou
Chrysoula
Karaiskou
Athanasios
Liatifis
Panagiotis
Radoglou-Grammatikis
Dimitrios
Pliatsios
Vasiliki
Kelli
Thomas
Lagkas
Vasileios
Argyriou
Panagiotis
Sarigiannidis
Submitted by:
Panagiotis Sari...
Last updated:
Wed, 11/23/2022 - 14:34
DOI:
10.21227/w67n-0q72
Data Format:
Link to Paper:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

This peach tree disease detection dataset is a multimodal, multi-angle dataset which was constructed for monitoring the growth of peach trees, including stress analysis and prediction. An orchard of peach trees is considered in the area of Thessaly, where 889 peach trees were recorded in a full crop season starting from Jul. 2021 to Sep. 2022. The dataset includes a) aerial / Unmanned Aerial Vehicle (UAV) images, b) ground RGB images/photos, and c) ground multispectral images/photos. Two experts (agronomists) annotated the dataset by identifying whether the trees are healthy or stressed, where the identified stress is either Anarsia lineatella or Grapholita molesta.

Instructions: 

Please cite the following papers when using this dataset:

C. Chaschatzis, C. Karaiskou, E. Mouratidis, E. Karagiannis, and P. Sarigiannidis, “Detection and Characterization of Stressed Sweet Cherry Tissues Using Machine Learning”, Drones, vol. 6, no. 1, 2022.

P. Radoglou-Grammatikis, P. Sarigiannidis, T. Lagkas, & I. Moscholios, “A compilation of UAV applications for precision agriculture,” Computer Networks, vol. 172, no. 107148, 2020.

A. Lytos, T. Lagkas, P. Sarigiannidis, M. Zervakis, & G. Livanos, “Towards smart farming: Systems, frameworks and exploitation of multiple sources,” Computer Networks, vol. 172, no. 107147, 2020.

In this dataset, an orchard of peach trees is considered in the area of Thessaly, where 889 peach trees were recorded in a full crop season starting from Jul. 2021 to Sep. 2022. The tree mapping within the orchard is depicted in Fig. 1., where each circle represents a peach tree. The color annotation on the circles (green, red, blue, yellow) will be elaborated in the following Sections. The six time periods, where the orchard was recorded are: 27th of Jul. 2021, 14th of Sep. 2021, 4th of Nov. 2021, 25th of May 2022, 15th of Jul. 2022, and 14th of Sep. 2022, providing data to a full year of trees life cycle. Each tree is identified by two numbers, the row and the column as the tree is mapped in the orchard. For example, the tree (3-26) is located in the 3rd row, 26th column of the orchard, as depicted in the following mapping images. Within the dataset files, each image of this tree (3-26) is identified as 3-26.jpg, subject to the corresponding folder. For example, if this file, 3-26.jpg, is under the folder Ground_RGB_Photos, the photo 3-26.jpg is referred to the RGB ground image of this tree.

This dataset includes a) aerial / UAV images, b) ground RGB images/photos, and c) ground multispectral images/photos. Two experts (agronomists) annotated this dataset by identifying stress, which in this case are two common diseases in peach trees: Grapholita molesta [1][2] and Anarsia lineatella [3]. In particular, the following modalities are featured in this dataset:

  • Ground RGB images
  • Ground multispectral images
  • UAV/Aerial images (RGB, multispectral, and NDVI).

These modalities represent the peach tree cultivation in many levels. Each modality describes the same object (peach tree) within the dataset, i.e., for each tree within.

This dataset was annotated by two experts (agronomists) in terms of two diseases (Anarsia lineatella and Grapholita molesta). In particular, they annotated each peach tree, one by one, in four states:

  • Healthy: the peach tree is completely healthy;
  • Anarsia lineatella: the Anarsia lineatella is presented on withered tops on the branches of trees;
  • Grapholita molesta: the Grapholita molesta located in the tree trunks and branches of trees;
  • Dead Trees: the peach tree is killed due to Anarsia lineatella or Grapholita molesta.

The annotation process was considered by each one of the underlying modalities (RGB, multispectral and UAV/aerial).

The image collection is depicted in the following figure in terms of the three modalities (aerial / Unmanned Aerial Vehicle (UAV) images, ground RGB images/photos, and ground multispectral images/photos).

Please note that some trees were replaced in the middle of the monitoring period (25/5/2022 and 04/10/2022) since these trees were wither killed or heavily infected by Anarsia lineatella or Grapholita molesta.

[1] W. Dalmorra de Souza, T. Bystronski Remboski, M. Sanchotene de Aguiar, and P. R. Ferreira Júnior, ‘A Model for Pest Infestation Prediction in Crops Based on Local Meteorological Monitoring Stations’, in 2017 Sixteenth Mexican International Conference on Artificial Intelligence (MICAI), Oct. 2017, pp. 39–45. doi: 10.1109/MICAI-2017.2017.00015.

[2] V. A. M. Martins, L. C. Freitas, M. S. de Aguiar, L. B. de Brisolara, and P. R. Ferreira, ‘Deep Learning applied to the Identification of Fruit Fly in Intelligent Traps’, in 2019 IX Brazilian Symposium on Computing Systems Engineering (SBESC), Nov. 2019, pp. 1–8. doi: 10.1109/SBESC49506.2019.9046088.

[3] P. Damos et al., ‘Degree-day risk thresholds for predicting the occurrence of Anarsia lineatella, Grapholita molesta and Adoxophyes orana in northern Greece peach orchards’, Plant Protect. Sci., vol. 58, no. No. 3, May 2022, pp. 234–244 doi: 10.17221/137/2021-PPS.

Funding Agency: 
H2020 & Operational Program Competitiveness, Entrepreneurship and Innovation
Grant Number: 
957406 & Τ1EDK-04759

Dataset Files

LOGIN TO ACCESS DATASET FILES

Documentation

AttachmentSize
File Read_Me.pdf1.38 MB