【DL】读取VOC下的JPG&XML并显示

最新推荐文章于 2022-09-17 17:23:06 发布

张林克

最新推荐文章于 2022-09-17 17:23:06 发布

阅读量616

点赞数

分类专栏：【DeepL】

本文链接：https://blog.csdn.net/weixin_45292794/article/details/109667687

版权

PASCAL VOC XML解析图像标注 OpenCV 目标检测

关键词由CSDN通过智能技术生成

VOC2007数据集下载

官方下载地址： https://pjreddie.com/projects/pascal-voc-dataset-mirror/

classes = ["aeroplane", "bicycle", "bird", "boat", "bottle", "bus", "car", "cat", "chair", "cow", "diningtable", "dog", "horse", "motorbike", "person", "pottedplant", "sheep", "sofa", "train", "tvmonitor"]

beta v1.0

仅能显示单目标XML
*.jpg *.xml 必须是数字的名字

读取的XML如：

<Annotation>
    <filename>1</filename>
    <size>
        <width>416</width>
        <height>416</height>
        <depth>3</depth>
    </size>
    <object>
        <name>OKKK</name>
        <pose>Unspecified</pose>
        <truncated>0</truncated>
        <difficult>0</difficult>
        <bndbox>
            <xmin>83</xmin>
            <ymin>71</ymin>
            <xmax>95</xmax>
            <ymax>396</ymax>
        </bndbox>
    </object>
</Annotation>

CODE

import xml.etree.ElementTree as ET
import matplotlib.pyplot as plt
import cv2


def xyxyFromXML(path):
    # "./voc/1.xml"
    tree = ET.parse(path)
    # 文档根元素
    root = tree.getroot()   

    for element in root.findall('object'):
        label = element.find('name').text   # 访问Element文本
        # print(name)
        for xywh in element.findall('bndbox'):
            xmin = xywh.find('xmin').text
            ymin = xywh.find('ymin').text
            xmax = xywh.find('xmax').text
            ymax = xywh.find('ymax').text
            # print(xmin, ymin, xmax, ymax)
            
            return [xmin, ymin, xmax, ymax], label


def drawImg(img, xyxy):
    
    xmin = int(xyxy[0])
    ymin =