DOTA数据集转VOC数据集，模仿DIOR数据集类型

DOTA数据集转VOC数据集，模仿DIOR数据集类型

DOTA文本数据集转XML格式

说明：由于本人比较喜欢DIOR数据集格式，所以仿照DIOR数据集转的相同格式

DOTA数据集下载链接：
提取码：7vwv

DOIR数据集下载链接：

提取码：jcpg

代码如下（改了网上的代码，此处贴上链接）

import os
from xml.dom.minidom import Document
from xml.dom.minidom import parse
import xml.dom.minidom
import numpy as np
import csv
import cv2
import stringdef WriterXMLFiles(filename,img_name, path, box_list, label_list, w, h, d):# dict_box[filename]=json_dict[filename]doc = xml.dom.minidom.Document()root = ateElement('annotation')doc.appendChild(root)# foldername = ateElement("folder")# foldername.ateTextNode("JPEGImages"))# root.appendChild(foldername)nodeFilename = ateElement('filename')nodeFilename.ateTextNode(img_name))root.appendChild(nodeFilename)# pathname = ateElement("path")# pathname.ateTextNode("xxxx"))# root.appendChild(pathname)sourcename&#ateElement("source")databasename = ateElement("database")databasename.ateTextNode("DOTA"))sourcename.appendChild(databasename)# annotationname = ateElement("annotation")# annotationname.ateTextNode("xxx"))# sourcename.appendChild(annotationname)# imagename = ateElement("image")# imagename.ateTextNode("xxx"))# sourcename.appendChild(imagename)# flickridname = ateElement("flickrid")# flickridname.ateTextNode("0"))# sourcename.appendChild(flickridname)root.appendChild(sourcename)nodesize = ateElement('size')nodewidth = ateElement('width')nodewidth.ateTextNode(str(w)))nodesize.appendChild(nodewidth)nodeheight = ateElement('height')nodeheight.ateTextNode(str(h)))nodesize.appendChild(nodeheight)nodedepth = ateElement('depth')nodedepth.ateTextNode(str(d)))nodesize.appendChild(nodedepth)root.appendChild(nodesize)segname = ateElement("segmented")segname.ateTextNode("0"))root.appendChild(segname)for (box, label) in zip(box_list, label_list):nodeobject = ateElement('object')nodename = ateElement('name')nodename.ateTextNode(str(label)))nodeobject.appendChild(nodename)nodebndbox = ateElement('bndbox')nodex1 = ateElement('x1')nodex1.ateTextNode(str(box[0])))nodebndbox.appendChild(nodex1)nodey1 = ateElement('y1')nodey1.ateTextNode(str(box[1])))nodebndbox.appendChild(nodey1)nodex2 = ateElement('x2')nodex2.ateTextNode(str(box[2])))nodebndbox.appendChild(nodex2)nodey2 = ateElement('y2')nodey2.ateTextNode(str(box[3])))nodebndbox.appendChild(nodey2)nodex3 = ateElement('x3')nodex3.ateTextNode(str(box[4])))nodebndbox.appendChild(nodex3)nodey3 = ateElement('y3')nodey3.ateTextNode(str(box[5])))nodebndbox.appendChild(nodey3)nodex4 = ateElement('x4')nodex4.ateTextNode(str(box[6])))nodebndbox.appendChild(nodex4)nodey4 = ateElement('y4')nodey4.ateTextNode(str(box[7])))nodebndbox.appendChild(nodey4)# ang = ateElement('angle')# ang.ateTextNode(str(angle)))# nodebndbox.appendChild(ang)nodeobject.appendChild(nodebndbox)root.appendChild(nodeobject)fp = open(path + filename, 'w')doc.writexml(fp, indent='n')fp.close()def load_annoataion(p):'''load annotation from the text file:param p::return:'''text_polys = []text_tags = []if not ists(p):return np.array(text_polys, dtype=np.float32)with open(p, 'r') as f:for line adlines()[2:]:label = 'text'# strip BOM. ufeff for python3,  xefxbbbf for python2#line = [i.strip('ufeff').strip('xefxbbxbf') for i in line]#print(line)x1, y1, x2, y2, x3, y3, x4, y4 ,label= line.split(' ')[0:9]#print(label)text_polys.append([x1, y1, x2, y2, x3, y3, x4, y4])text_tags.append(label)return np.array(text_polys, dtype=np.float), np.array(text_tags, dtype=np.str)txt_path = r'E:/baidudisk/DOTA/train/labelTxt-v1.5/DOTA-v1.5_train/'
xml_path = r'E:/baidudisk/DOTA/train/labelTxt-v1.5-voc/DOTA-v1.5_train/'
img_path = r'E:/baidudisk/DOTA/train/images/images/'
print(ists(txt_path))
txts = os.listdir(txt_path)
for count, t in enumerate(txts):path = os.path.join(txt_path,t)print(path)# boxes存储的是八个坐标# labels存储的是标签boxes, labels = load_annoataion(os.path.join(txt_path, t))# lxml_name = t.replace('.txt', '.xml')img_name = t.replace('.txt', '.png')# P0000.pngimg = cv2.imread(os.path.join(img_path, img_name))h, w, d = img.shape#print(xml_name, xml_path, boxes, labels, w, h, d)WriterXMLFiles(xml_name,img_name, xml_path, boxes, labels, w, h, d)if count % 1000 == 0:print(count)

说明：
1：请务必将自己的文件路径进行修改
2：导包出现如下图横线，感觉也没有影响程序运行

3：文件转换最终如下图：