"Converts xml annotation file into required text format training data"

xml_to_text.py

@@ -0,0 +1,44 @@
+import xml.etree.ElementTree as ET
+import os
+
+
+images_path="./Images/" # Add path of labelled images
+xml_path='/Annotations/' # add images xml file path
+
+
+
+print("Converting images")
+for file in os.listdir(xml_path):
+
+
+    root = ET.parse(xml_path+'/'+file)
+    #image_name=root.find("filename").text
+    for elem in root.iter():
+
+        if 'filename' in elem.tag:
+            image_name=elem.text
+        if 'object' in elem.tag:
+            for attr in list(elem):
+                #print(attr)
+                if 'name' in attr.tag:
+                    object_name=attr.text
+
+                if 'bndbox' in attr.tag:
+                    for dim in list(attr):
+                        if 'xmin' in dim.tag:
+                            xmin=dim.text
+
+                        if 'ymin' in dim.tag:
+                            ymin=dim.text
+
+                        if 'xmax' in dim.tag:
+                            xmax=dim.text
+
+                        if 'ymax' in dim.tag:
+                            ymax=dim.text
+            data=images_path+image_name+"," +xmin+','+ ymin +',' + xmax + ","+ ymax + ","+ object_name
+
+            with open("test.txt", "a") as myfile:
+                myfile.write(data+"\n")
+print("finished")
+