update dataset 操作

Author: jadehh

dataset_tools/jade_create_paddle_text_detection_datasets.py

@@ -13,6 +13,7 @@
 import shutil
 import cv2
 import random
+from dataset_tools import *
 
 
 def sortPoints(points, label):
@@ -200,9 +201,32 @@ def removeNolabelDatasets(root_path):
                 os.remove(os.path.join(root_path, GetLastDir(image_path)[:-4] + ".json"))
                 os.remove(image_path)
         else:
+            print("清除图片")
             os.remove(image_path)
         progressBar.update()
 
+def removeNolabelVocDatasets(root_path):
+    for day in os.listdir(root_path):
+        if os.path.isdir(os.path.join(root_path,day)):
+            image_path_list = GetAllImagesPath(os.path.join(root_path, day, DIRECTORY_IMAGES))
+            progressBar = ProgressBar(len(image_path_list))
+            for image_path in image_path_list:
+                if os.path.exists(
+                        os.path.join(root_path, day, DIRECTORY_ANNOTATIONS, GetLastDir(image_path)[:-4] + ".xml")):
+                    imagename, shape, bboxes, labels_text, labels, difficult, truncated = ProcessXml(
+                        os.path.join(root_path, day, DIRECTORY_ANNOTATIONS, GetLastDir(image_path)[:-4] + ".xml"))
+                    if len(bboxes) == 0:
+                        print("清除{},清除{}".format(image_path, os.path.join(root_path, day, DIRECTORY_ANNOTATIONS,
+                                                                          GetLastDir(image_path)[:-4] + ".xml")))
+                        os.remove(
+                            os.path.join(root_path, day, DIRECTORY_ANNOTATIONS, GetLastDir(image_path)[:-4] + ".xml"))
+                        os.remove(image_path)
+                else:
+                    print("清除{}".format(image_path))
+                    os.remove(image_path)
+                progressBar.update()
+
+
 
 def GetContaNumberPath(image_path_list):
     ContaNumber_list = []

main.py

@@ -12,6 +12,7 @@
 if __name__ == '__main__':
 
     #removeNolabelDatasets(r"F:\数据集\关键点检测数据集\定制版箱号关键点数据集\2022-03-09")
-    create_text_detection_datasets(r"F:\数据集\关键点检测数据集\定制版箱号关键点数据集",r"E:\Data\字符检测识别数据集\定制版箱号关键点数据集",0.95)
+    #create_text_detection_datasets(r"F:\数据集\关键点检测数据集\定制版箱号关键点数据集",r"E:\Data\字符检测识别数据集\定制版箱号关键点数据集",0.95)
     #CreatePaddleOCRDatasets(root_path="E:\Data\字符检测识别数据集\镇江大港厂内车牌关键点检测数据集", save_path="E:\Data\OCR\镇江大港厂内车牌识别数据集",dataset_type="镇江厂内车牌数据集")
-    # CreateYearsDatasets(r"F:\数据集\VOC数据集\定制版顶相机箱号检测数据集")
+    removeNolabelVocDatasets(r"F:\数据集\VOC数据集\验残集装箱检测数据集")
+    CreateYearsDatasets(r"F:\数据集\VOC数据集\验残集装箱检测数据集")