三、目标检测xml文件（VOC格式标注文件）的可视化

614b阿智

已于 2023-01-04 19:34:44 修改

阅读量1k

点赞数

分类专栏： # 目标检测标注文件处理文章标签： xml python 人工智能

于 2022-12-21 13:15:50 首次发布

本文链接：https://blog.youkuaiyun.com/m0_51321564/article/details/128388519

版权

该博客介绍了一个用于VOC格式的目标检测XML文件的可视化工具。通过解析XML文件，该工具能够读取标注信息，并在图像上以不同颜色矩形框显示各个目标类别，同时在框内标注类别名称。博客提供了详细代码，适用于图像分析和机器学习项目的标注验证。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

目标检测xml文件（VOC格式标注文件）的可视化

from xml.dom.minidom import parse
import matplotlib.pyplot as plt
import xml.dom.minidom
import os, shutil
import matplotlib
import numpy as np
import cv2
from PIL import Image, ImageDraw, ImageFont

##########################################################
root = '/home/jojo/VSST/zqz/Object_detection/xml_file_visualization'
# only need to change these
##########################################################

annroot = '/home/jojo/VSST/zqz/Object_detection/xml_file_visualization/groundtruth/Annotations/'
picroot = '/home/jojo/VSST/zqz/Object_detection/xml_file_visualization/groundtruth/JPEGImages/'
anns = os.listdir(annroot)
imgs = os.listdir(picroot)
line_thickness = 6
# labelmap = ["person", "car", "cat", "dog"]
labelmap = ['person', 'rider', 'car', 'truck', 'bus', 'train', 'motorcycle', 'bicycle']

colormap = ['#FF3838', '#FF9D97', '#FF701F', '#FFB21D', '#CFD231', '#48F90A', '#92CC17', '#3DDB86']
# colormap = ["#48F90A", "#FFB21D", "#FF701F", "#CFD231", "pink", "olive", "deeppink", "darkorange", "purple", "cyan", "red",
#             "green", "blue", "yellow", "pink", "olive", "deeppink", "darkorange", "purple", "cyan", "red"]


def mkdir(path):
    folder = os.path.exists(path)
    if not folder:
        os.makedirs(path)

fontPath = "/home/jojo/VSST/zqz/Object_detection/xml_file_visualization/ARIAL.TTF"  # 字体路径
font = ImageFont.truetype(fontPath, 65)
# font = ImageFont.truetype(fontPath, 36)
txt_color=(0, 0, 0)

number = 0
nn = 0
for ann in anns:
    number += 1
    print(number)

最低0.47元/天解锁文章