commiting to fix merge conflict

.gitignore

@@ -0,0 +1,5 @@
+**/train/
+**/val/
+_*/
+**/_*.py
+**.json

README.md

@@ -0,0 +1,89 @@
+---
+annotations_creators:
+  - CIAT (International Center for Tropical Agriculture)
+  - Producers Direct
+task_categories:
+  - object-detection
+size_categories:
+- 1K<n<10K
+pretty_name: Croppie coffee uganda
+tags:
+  - yield estimates
+  - cherry count
+  - coffee cherries
+  - coffee trees
+  - arabica
+  - robusta
+  - digital agriculture
+language:
+  - en
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: "data/train.zip"
+  - split: val
+    path: "data/val.zip"
+---
+
+# Croppie training datasets
+## General information
+Croppie dataset for machine-vision assisted coffee cherry detection. The dataset is made of a mix of Arabica and Robusta coffee tree parts (with and without a background isolation element) with individual bounding boxes around all coffee cherries.
+
+The original dataset is composed of 633 images with about 61 050 unique bounding boxes over coffee cherries in YOLO format. This original dataset has been processed to cut-down all images into 480 x 640 size pieces and the full original image downscaled to 480 x 640. We provide the processed dataset with Python scripts that allow easy visualization of the annotated dataset.
+
+Coffee cherries of more than 10mm (following the longitudinal axis) are annotated according to their color:
+- green
+- yellow
+- red
+- dark brown (overripe/dry cherries)
+- an extra class indicates low visibility/unsure label appreciation.
+
+Here is an example of an annotated image:
+![plot](./images/annotated_1688033955437.jpg)
+
+## Data structure
+This repository has the following structure:
+```
+.
+├── annotation_guide.html # original annotation instructions
+├── classes.json  # json to convert numerical classes into the cherry type
+├── data
+│   ├── train.zip
+│   └── val.zip
+├── images
+│   ├── annotated_1688033955437.jpg
+│   ├── train_counts.png
+│   └── val_counts.png
+├── README.md
+└── scripts  # script for easy visualization of the annotated data
+    ├── label_training_images.py
+    └── requirements.txt
+```
+### Dataset information
+Each numerical class corresponds to the following cherry type:
+```
+{0: "dark_brown_cherry", 1: "green_cherry", 2: "low_visibility_unsure", 3: "red_cherry", 4: "yellow_cherry"}
+```
+
+* ```train```: 
+  * Training dataset
+  * 5 836 annotated images
+  * YOLO format
+
+![plot](./images/train_counts.png)
+
+* ```val```: 
+  * Validation dataset
+  * 2 497 annotated images
+  * YOLO format
+
+![plot](./images/val_counts.png)
+
+* ```annotation_guide.html```: instructions provided to label the images for cherry detection
+
+## Scripts
+The script ```label_training_images.py``` allows to label the images of the datasets and saves them in a folder  ```./_labelled_dataset_images```.
+Assuming you are in the ```scripts``` folder, first
+ run ```pip3 install -r requirements.txt``` if required package are not installed. After that, simply run 
+```python3 label_training_images.py```

_classes.txt

@@ -0,0 +1,5 @@
+dark_brown_cherry
+green_cherry
+low_visibility_unsure
+red_cherry
+yellow_cherry

_pictures_to_exclude.txt

@@ -0,0 +1,2 @@
+1688117174257
+1688718305705

annotation_guide.html

@@ -0,0 +1,13 @@
+<ul>
+<li><b>EVERY CHERRY</b> should be annotated with an <b>INDIVIDUAL bounding box</b>. Try not to miss any cherry, <b>even if they are located in the BACKGROUND</b>.</li>
+<li>Each label should correspond to the <b>color of the cherry</b> (green, yellow, red or dark brown). Do not confond exposure and color: a green cherry with low exposure can appear dark</li>
+<li>If you can see <b>AT LEAST 10% OF A CHERRY, annotate it</b>. For  <b>all partially visible cherries</b>, guess the occulted part and <b>match BOTH the occulted and visible part with the same box</b>.</li>
+<li>If you feel like it is a cherry, but you're <b>unsure</b>, use a "low_visibility_unsure" [blue] bounding box. <b>CAUTION</b>: a cherry can be partially occulted but still having good visibility.</li>
+<li>Each bounding box should fit the cherry <b>AS CLOSELY AS POSSIBLE</b>.</li>
+</ul>
+TIPS:
+<ul>
+<li>using the sidebar, you can zoom, change the contrast or staturation of the image to ease the annotation</li>
+<li>you can select several labels keep pressing "ctrl" and then delete label(s) using the "backspace" key</li>
+</ul>
+<img src="https://i.ibb.co/KNYsbSF/annotation-example-2.png" alt="annotation-example-2" height="350px">

data/train.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e881328f07950cfbe23b30bd9e7e3f661003589b3b30d256e176d0b15b2b86ba
+size 651895390

data/val.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b517afde1dd0f5cbe6322ec38665b5d4cbdcb89d54c1c7f36f87fe995d8d35c
+size 278569732

scripts/label_training_images.py

@@ -0,0 +1,110 @@
+import cv2
+import matplotlib.pyplot as plt
+from matplotlib.path import Path
+plt.rcParams['figure.dpi'] = 100
+from PIL import ImageColor
+from pathlib import Path
+import glob
+import os
+import json
+
+def annotate_images_dataset(image_folder, label_folder, class_file_path, saving_folder, hex_class_colors=None, show=False):
+    """
+    Allows to visualize a set of images and corresponding YOLO labels.
+    Args:
+        image_folder (str): path of the folder containing the images for object detection
+        label_folder (str): path of the folder containing the labels corresponding to the images for object detection
+        class_file_path (str): path of the json file containing the labels of the object classes
+        saving_folder (str): path of the folder 
+        hex_class_colors (dict, optional): dictionary with HEX color for each label
+        show (bool, optional): if True, a prompt with the labelled image opens
+    """    
+    class_dic = get_class_dic(class_file_path)
+    Path(saving_folder).mkdir(parents=True, exist_ok=True)
+    if not hex_class_colors:
+        hex_class_colors = {class_name: (255, 0, 0) for class_name in class_dic.values()}
+    color_map = {key: ImageColor.getcolor(hex_class_colors[class_dic[key]], 'RGB') for key in [*class_dic]}
+    label_paths = sorted(glob.glob(os.path.join(label_folder, '*')))
+    n_labels = len(label_paths)
+    for i, label_path in enumerate(label_paths):
+        i += 1
+        if i % 100 == 0:
+            progress = i / n_labels
+            print(f'{progress: .0%} -> image {i} out of {n_labels}')
+        file_name = str(label_path).split('/')[-1].split('.')[0]
+        image_file = file_name + '.jpg'
+        image_path = os.path.join(image_folder, image_file)
+        if os.path.isfile(image_path):
+            img = cv2.imread(image_path)
+            img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+            dh, dw, _ = img.shape
+
+            with open(label_path, 'r') as f:
+                data = f.readlines()
+
+            for yolo_box in data:
+                yolo_box = yolo_box.strip()
+                c, x, y, w, h = map(float, yolo_box.split(' '))
+                l = int((x - w / 2) * dw)
+                r = int((x + w / 2) * dw)
+                t = int((y - h / 2) * dh)
+                b = int((y + h / 2) * dh)
+
+                if l < 0:
+                    l = 0
+                if r > dw - 1:
+                    r = dw - 1
+                if t < 0:
+                    t = 0
+                if b > dh - 1:
+                    b = dh - 1
+                cv2.rectangle(img, (l, t), (r, b), color_map[c], 3)
+            if show:
+                plt.imshow(img)
+                plt.show()
+            plt.imsave(os.path.join(saving_folder, f'annotated_{image_file}'), img)
+        else:
+            print(f'WARNING: {image_path} does not exists')
+
+
+def get_class_dic(classe_file_path):
+    """
+    Turns a label list txt file into a dict with numerical class as key and corresponding label as value
+    Args:
+        classe_file_path (str): path to the json file listing the labels
+
+    Returns:
+        dict: dictionary of numerical class as key and corresponding label as value
+    """    
+    class_dic = {}
+    with open(classe_file_path) as f:
+        class_dic = json.load(f) 
+    class_dic = {int(k):v for k,v in class_dic.items()}
+    return class_dic
+
+
+if __name__ == '__main__':
+    dataset_folder_names = ['train', 'val']
+    dataset_prefix_folder = '../data'
+    saving_prefix_folder = '../_labelled_dataset_images'
+    show = not True
+    hex_class_colors = {'green_cherry': '#9CF09A',
+                        'yellow_cherry': '#F3C63D',
+                        'red_cherry': '#F44336',
+                        'dark_brown_cherry': '#C36105',
+                        'low_visibility_unsure': '#02D5FA'}
+    for dataset_folder_name in dataset_folder_names:
+        print(f'dataset: {dataset_folder_name}:\n')
+        full_saving_folder = os.path.join(saving_prefix_folder, dataset_folder_name)
+        full_dataset_folder = os.path.join(dataset_prefix_folder, dataset_folder_name)
+        class_file_path = os.path.join('../', 'classes.json')
+        image_folder = os.path.join(full_dataset_folder, 'images')
+        label_folder = os.path.join(full_dataset_folder, 'labels')
+        annotate_images_dataset(
+            image_folder=image_folder,
+            label_folder=label_folder,
+            class_file_path=class_file_path,
+            saving_folder=full_saving_folder,
+            hex_class_colors=hex_class_colors,
+            show=show,
+        )

scripts/requirements.txt

@@ -0,0 +1,3 @@
+Pillow==9.3.0
+matplotlib==3.7.1
+opencv-python==4.7.0.72