Add dataset split

README.md

@@ -30,7 +30,8 @@ segmentation mask.
 The multispectral images were captured by the Micasense RedEdge-P multispectral sensor embedded in a 
 DJI M210 V2 UAV. Radiometric calibration was performed for all images based on the known reflectance
 values of a calibration panel. All images were used to process a Digital Ortophoto Map, which was then 
-sliced into the 256x256x6 image patches that are contained in this dataset.
+sliced into the 256x256x6 image patches that are contained in this dataset. Regions from each of the 
+ortophotos were assigned for training (70%), validation (15%) and testing (15%). 
 
 Each image file is composed of six channels:
 
@@ -49,12 +50,21 @@ were manually annotated using the CVAT software.
 The dataset files are organized as following:
 
 ```
-📁 dam-segmentation
-├── 📁 images                # Multispectral images
-├── 📁 index_ndre            # NDRE index values
-├── 📁 labels_binary         # Binary Segmentation Masks
-├── 📁 labels_multiclass     # Multiclass Segmentation Masks
-└── 📝 README.md
+  dam-segmentation
+├──   test                         # Test dataset
+│   ├──   images
+│   ├──   mask_binary
+│   └──   mask_multiclass
+├──   train                       # Training dataset
+│   ├──   images
+│   ├──   mask_binary
+│   └──   mask_multiclass
+├──   val                         # Validation dataset
+│   ├──   images
+│   ├──   mask_binary
+│   └──   mask_multiclass
+└──  README.md
+
 ```
 
 ## Segmentation Classes