├── Sans titre.png
└── README.md


/Sans titre.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/simonMadec/Wheat-Ears-Detection-Dataset/HEAD/Sans titre.png


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | ## Note / News
 2 | For people interested about this work/dataset I recommand to check the Global Wheat Dataset : A more large and diverse dataset for wheat head detection 
 3 | http://www.global-wheat.com/
 4 | https://zenodo.org/record/5092309#.YaPEh9DMJPY
 5 | The Wheat Ears Dection Dataset has been integrated in  Global Wheat Dataset
 6 | 
 7 | # Wheat-Ears-Detection-Dataset
 8 | Dataset from the Ear density estimation from high resolution RGB imagery using deep learning technique paper
 9 | 
10 | [Simon Madec], [Frederic Baret], [Benoit de Solan], [Shouyang Liu]
11 | The Wheat-Ears-Dection-Dataset (WEDD) is a dataset is a image dataset designed fo wheat ears detection in field condition.  
12 | ![Ex](https://github.com/simonMadec/Wheat-Ears-Dection-Dataset/blob/master/Sans%20titre.png?raw=true)
13 | ## Overview
14 | - [Highlights](#highlights)
15 | - [Research Paper](#research-paper)
16 | - [Downloads](#downloads)
17 | - [Labels](#labels)
18 | - [Results](#results)
19 | - [Annotation Tool](#annotation-tool)
20 | 
21 | ## Highlights
22 | - 236 high resolution images images (6000*4000)
23 | - Wheat ears annotated with a bounding box
24 | - 30729 ears identified
25 | - Spatial resolution (GSD) of 0.13mm/pixel
26 | - Two images for each microplots 
27 | - 20 contrasted genotype with 6 replicated growth in two environment 
28 | 
29 | ## Research Paper
30 | To cite the paper :
31 | 
32 | Madec, S., Jin, X., Lu, H., De Solan, B., Liu, S., Duyme, F., et al. (2019). Ear density estimation from high resolution RGB imagery using deep learning technique. Agric. For. Meteorol. 264, 225–234. doi:10.1016/j.agrformet.2018.10.013.
33 | 
34 | 
35 | ## Downloads
36 | Dataset avalaible [here](https://zenodo.org/record/1471626#.W9H_XFUzZjU)
37 | 
38 | ## Labels
39 | Please download replicate information along with images used for training and testing [here]
40 | ## Results
41 | Below we present results.
42 | 
43 | Method               | Date | Source| AP | rRMSE | R² 
44 | ---                  | ---  | ---   | ---                     | ---             | ---     
45 | Faster-RCNN [1]       |   21/10/2018  |[2]| 0.85              | 5.3%|0.91     
46 | 
47 | ### Annotation Tool
48 | The LabelIMG tool were used, please refer to [this repository](https://github.com/tzutalin/labelImg).
49 | 
50 | 


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