yolotxt文件转为sddxml

### 将YOLO格式的TXT文件转换为SDD XML格式 #### 转换过程概述 为了实现从YOLO格式到SDD XML格式的转换，需要编写脚本解析原始YOLO标签文件中的边界框坐标，并按照目标检测标准（Pascal VOC）创建对应的XML描述文件。YOLO格式通常保存在`.txt`文件中，每行代表一个对象及其位置；而SDD XML则是一种更详细的标注方式，适用于多种计算机视觉任务。 #### 解析YOLO TXT文件 YOLO格式下的每一行记录如下所示： - 类别索引 - 边界框中心点相对于图像宽度的比例 (x_center) - 边界框中心点相对于图像高度的比例 (y_center) - 边界框宽高分别占整个图片尺寸比例 (width, height) 这些数值均位于0至1之间[^1]。 #### 创建相应的XML结构 对于每一个YOLO `.txt`文件关联的对象，都需要构建一段特定模式的XML片段来表示该物体的位置和其他属性。典型的XML条目可能看起来像这样： ```xml <object> <name>person</name> <pose>Unspecified</pose> <truncated>0</truncated> <difficult>0</difficult> <bndbox> <xmin>x_min</xmin> <ymin>y_min</ymin> <xmax>x_max</xmax> <ymax>y_max</ymax> </bndbox> </object> ``` 其中 `<name>` 是指类别名称，其余字段用于指定边界的最小和最大坐标值[xmin,ymin,xmax,ymax][^3]。 #### Python代码示例 下面是一个简单的Python函数例子，它读取YOLO格式的文本文件并将它们转化为上述提到的标准XML格式: ```python import xml.etree.ElementTree as ET from pathlib import Path def yolo_to_xml(yolo_file_path: str, image_width: int, image_height: int, output_dir: str): """ Convert a single YOLO format .txt file to Pascal VOC style XML. :param yolo_file_path: path of the input YOLO annotation text file :param image_width: width of corresponding image in pixels :param image_height: height of corresponding image in pixels :param output_dir: directory where generated XML files will be saved """ def create_object_element(class_name, xmin, ymin, xmax, ymax): obj = ET.SubElement(root, 'object') name = ET.SubElement(obj, 'name') pose = ET.SubElement(obj, 'pose') truncated = ET.SubElement(obj, 'truncated') difficult = ET.SubElement(obj, 'difficult') bndbox = ET.SubElement(obj, 'bndbox') name.text = class_names[int(class_name)] pose.text = "Unspecified" truncated.text = "0" difficult.text = "0" ET.SubElement(bndbox, 'xmin').text = str(xmin) ET.SubElement(bndbox, 'ymin').text = str(ymin) ET.SubElement(bndbox, 'xmax').text = str(xmax) ET.SubElement(bndbox, 'ymax').text = str(ymax) with open(yolo_file_path) as f: lines = f.readlines() root = ET.Element('annotation') folder = ET.SubElement(root, 'folder') filename = ET.SubElement(root, 'filename') size_part = ET.SubElement(root, 'size') depth = ET.SubElement(size_part, 'depth') # Assuming images are stored under './images/' relative to this script's location, # and have extension '.jpg'. Adjust accordingly based on actual setup. img_folder = "./images/" ext = ".jpg" base_name = Path(yolo_file_path).stem folder.text = img_folder filename.text = base_name + ext ET.SubElement(size_part, 'width').text = str(image_width) ET.SubElement(size_part, 'height').text = str(image_height) depth.text = "3" # assuming RGB color space for line in lines: parts = list(map(float, line.strip().split())) cls_id = int(parts[0]) center_x = float(parts[1]) * image_width center_y = float(parts[2]) * image_height bbox_w = float(parts[3]) * image_width bbox_h = float(parts[4]) * image_height x_min = max(int(center_x - bbox_w / 2), 0) y_min = max(int(center_y - bbox_h / 2), 0) x_max = min(int(center_x + bbox_w / 2), image_width - 1) y_max = min(int(center_y + bbox_h / 2), image_height - 1) create_object_element(cls_id, x_min, y_min, x_max, y_max) tree = ET.ElementTree(root) out_file = Path(output_dir) / (base_name + ".xml") tree.write(out_file, encoding='utf8', method="xml") if __name__ == "__main__": # Example usage classes = ["class_0", "class_1"] # Replace these names according to your dataset labels global class_names class_names = {i: n for i, n in enumerate(classes)} yolo_to_xml("path/to/your/yolo.txt", 640, 480, "output/xmls/") ```