Merge branch 'main' of github.com:weecology/MillionTrees into main

data_prep/collect_tasks.py

@@ -34,7 +34,8 @@
     "/orange/ewhite/DeepForest/Alejandro_Chile/alejandro/annotations.csv",
     "/orange/ewhite/DeepForest/UrbanLondon/annotations.csv",
     "/orange/ewhite/DeepForest/OliveTrees_spain/Dataset_RGB/annotations.csv",
-    "/orange/ewhite/DeepForest/Araujo_2020/annotations.csv"
+    "/orange/ewhite/DeepForest/Araujo_2020/annotations.csv",
+    "/orange/ewhite/DeepForest/justdiggit-drone/label_sample/annotations.csv"
     ]
 
 # Current errors

data_prep/justdiggit.py

@@ -3,30 +3,46 @@
 from deepforest.utilities import read_file
 from deepforest.visualize import plot_results
 from matplotlib import pyplot as plt
+import rasterio as rio
 
 with open("/orange/ewhite/DeepForest/justdiggit-drone/label_sample/Annotations_trees_only.json") as jsonfile:
     data = json.load(jsonfile)    
     ids = [x["id"] for x in data["images"]]
     image_paths = [x["file_name"] for x in data["images"]]
-    id_df = pd.DataFrame({"id":ids,"image_path":image_paths})
+    id_dict = {key:value for key, value in zip(ids,image_paths)}
     annotation_df = []
     for row in data["annotations"]:
-        b = {"id":row["id"],"xmin":row["bbox"][0],"ymin":row["bbox"][1],"xmax":row["bbox"][2],"ymax":row["bbox"][3]}
+        polygon = row["segmentation"][0]
+
+        # split each pair into a tuple
+        as_tuples = [(polygon[i],polygon[i+1]) for i in range(0,len(polygon)-1,2)]
+
+        # close the polygon
+        as_tuples.append(as_tuples[0])
+
+        # convert polygon to wkt format
+        polygon_coordinates = ",".join(["{} {}".format(i[0],i[1]) for i in as_tuples])
+        polygon = "POLYGON ((" + polygon_coordinates + "))"
+
+        b = {"id":row["id"],"polygon":polygon}
+        b["image_path"] = id_dict[row["image_id"]]
         annotation_df.append(b)
 annotation_df = pd.DataFrame(annotation_df)
-annotations = annotation_df.merge(id_df)
-annotations["label"] = "Tree"
-annotations["source"] = "Justdiggit et al. 2023"
+annotation_df["label"] = "Tree"
+annotation_df["source"] = "Justdiggit 2023"
 
-print("There are {} annotations in {} images".format(annotations.shape[0], len(annotations.image_path.unique())))
+print("There are {} annotations in {} images".format(annotation_df.shape[0], len(annotation_df.image_path.unique())))
 
-annotations["image_path"] = "/orange/ewhite/DeepForest/justdiggit-drone/label_sample/" + annotations["image_path"]
-annotations = read_file(annotations, root_dir="/orange/ewhite/DeepForest/justdiggit-drone/label_sample/")
+annotations = read_file(annotation_df, root_dir="/orange/ewhite/DeepForest/justdiggit-drone/label_sample/")
 
 for image in annotations.image_path.unique():
     print(image)
     gdf = annotations[annotations.image_path == image]
     gdf.root_dir = "/orange/ewhite/DeepForest/justdiggit-drone/label_sample/"
-    plot_results(gdf)
-    plt.savefig("current.png")
-annotations.to_csv("/orange/ewhite/DeepForest/justdiggit-drone/label_sample/train.csv")
+    width, height = rio.open(gdf.root_dir + image).shape
+
+    #plot_results(gdf, height=height, width=width)
+    #plt.savefig("current.png")
+
+annotations["image_path"] = "/orange/ewhite/DeepForest/justdiggit-drone/label_sample/" + annotations["image_path"]
+annotations.to_csv("/orange/ewhite/DeepForest/justdiggit-drone/label_sample/annotations.csv")

tests/test_TreeBoxes.py

@@ -4,7 +4,6 @@
 import torch
 import pytest
 import os
-import pandas as pd
 import numpy as np
 
 # Check if running on hipergator