txt转xml文件和txt文件内容读取

根据txt文件提供的内容，在图片上面画框

import cv2
import os

filePath = 'D:\study\WongKinYiu\\test\images'#图片所在文件夹路劲
names = os.listdir(filePath)
print(names)
for name in names:
    image = cv2.imread("D:\study\WongKinYiu\\test\\images\\"+ str(name))
    print(name[0:5])
    GrayImage = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

    with open("D:\study\WongKinYiu\\test\\txt\\"+str(name[0:5])+".txt", 'r') as f:#每个txt文件所在文件夹
        lines = f.readlines()
        # if(len(lines!=1)):
        for line in lines:
            line_split = line.split(';')
            draw_1 = cv2.rectangle(image, (int(line_split[1]), int(line_split[2])), (int(line_split[3]), int(line_split[4])), (0, 255, 0), 2)
    cv2.imwrite("D:\study\WongKinYiu\\test\\rectangle_image\\"+str(name), draw_1)
    # cv2.namedWindow("draw_0", 2)
    # cv2.imshow("draw_0", draw_1)  # 显示画过矩形框的图片
    # cv2.waitKey(0)
    # cv2.destroyWindow("draw_0")

从一个txt文件生成多个txt文件

# 将一个txt文件里面的内容写入多个txt文件，可循环写入

import xlwt  # 写入文件
import xlrd  # 打开excel文件

fopen = open('D:\study\WongKinYiu\\test\\annotations\gt3classes.txt', 'r')
lines = fopen.readlines()


def delt(st):
    if '\n' in st:
        p = st.replace('\n', '')
        return p
    else:
        return st


i = 0
txt_name_pre = 0
for line in lines:
    line = line.strip('\n')  # 去掉换行符
    txt_name = int(line[0:5])
    # 针对一个txt存有多个目标信息
    if txt_name == txt_name_pre:
        # http://www.cppcns.com/jiaoben/python/324611.html
        line_split = line.split(';')  # ['00700.png', '854', '431', '881', '458', '13']
        # print(line)
        # print(line_split)
        # print(line_split[1])
        pline = 'txt/%s.txt' % line[0:5]
        txt_name_pre = int(line[0:5])
        gg = open(pline, 'a')  # 在上一个txt文件中追加内容，不会覆盖掉
        gg.write("\n"+str(line_split[-1]) + ";" + str(line_split[1]) + ";" + str(line_split[2]) + ";" + str(line_split[3]) + ";" + str(line_split[4]))
        gg.close()
    else:
        # print(txt_name)
        # print(line[0:9])
        # print(line)
        # http://www.cppcns.com/jiaoben/python/324611.html
        line_split = line.split(';')  # ['00700.png', '854', '431', '881', '458', '13']
        # print(line)
        # print(line_split)
        # print(line_split[1])
        pline = 'txt/%s.txt' % line[0:5]
        txt_name_pre = int(line[0:5])
        gg = open(pline, 'w')
        gg.write(str(line_split[-1]) + ";" + str(line_split[1]) + ";" + str(line_split[2]) + ";" + str(
            line_split[3]) + ";" + str(line_split[4]))
        gg.close()

根据txt文件生成xml文件，一个xml文件可循环写入

# coding=utf-8
# makexml("txt所在文件夹","xml保存地址","图片所在地址")

from xml.dom.minidom import Document
import os
import cv2


def makexml(txtPath, xmlPath, picPath):  # 读取txt路径，xml保存路径，数据集图片所在路径
    dict = {'prohibitory': "prohibitory",  # 字典对类型进行转换
            'danger': "danger",
            'mandatory': "mandatory",
            }
    files = os.listdir(txtPath)
    for i, name in enumerate(files):
        xmlBuilder = Document()
        annotation = xmlBuilder.createElement("annotation")  # 创建annotation标签
        xmlBuilder.appendChild(annotation)
        txtFile = open(txtPath + name, 'r', encoding='iso8859-1')
        txtList = txtFile.readlines()
        img = cv2.imread(picPath + name[0:-4] + ".png")
        print(picPath + name[0:-4] + ".png")
        Pheight, Pwidth, Pdepth = img.shape
        # for i in txtList:
        # oneline = i.strip().split(" ")

        folder = xmlBuilder.createElement("folder")  # folder标签
        folderContent = xmlBuilder.createTextNode("VOC2007")
        folder.appendChild(folderContent)
        annotation.appendChild(folder)

        filename = xmlBuilder.createElement("filename")  # filename标签
        filenameContent = xmlBuilder.createTextNode(name[0:-4] + ".jpg")
        filename.appendChild(filenameContent)
        annotation.appendChild(filename)

        size = xmlBuilder.createElement("size")  # size标签
        width = xmlBuilder.createElement("width")  # size子标签width
        widthContent = xmlBuilder.createTextNode(str(Pwidth))
        width.appendChild(widthContent)
        size.appendChild(width)
        height = xmlBuilder.createElement("height")  # size子标签height
        heightContent = xmlBuilder.createTextNode(str(Pheight))
        height.appendChild(heightContent)
        size.appendChild(height)
        depth = xmlBuilder.createElement("depth")  # size子标签depth
        depthContent = xmlBuilder.createTextNode(str(Pdepth))
        depth.appendChild(depthContent)
        size.appendChild(depth)
        annotation.appendChild(size)

        for i in txtList:
            # oneline = i.strip().split(" ")
            oneline = i.split(";")
            print(oneline)
            object = xmlBuilder.createElement("object")
            picname = xmlBuilder.createElement("name")
            # if oneline[0] >= '4':
            #     continue
            nameContent = xmlBuilder.createTextNode(dict[oneline[0]])
            picname.appendChild(nameContent)
            object.appendChild(picname)
            pose = xmlBuilder.createElement("pose")
            poseContent = xmlBuilder.createTextNode("Unspecified")
            pose.appendChild(poseContent)
            object.appendChild(pose)
            truncated = xmlBuilder.createElement("truncated")
            truncatedContent = xmlBuilder.createTextNode("0")
            truncated.appendChild(truncatedContent)
            object.appendChild(truncated)
            difficult = xmlBuilder.createElement("difficult")
            difficultContent = xmlBuilder.createTextNode("0")
            difficult.appendChild(difficultContent)
            object.appendChild(difficult)
            bndbox = xmlBuilder.createElement("bndbox")
            xmin = xmlBuilder.createElement("xmin")
            # mathData = int(((float(oneline[1])) * Pwidth + 1) - (float(oneline[3])) * 0.5 * Pwidth)
            mathData =int(oneline[1])
            xminContent = xmlBuilder.createTextNode(str(mathData))
            # if xminContent < 0:
            #     xminContent = 0
            xmin.appendChild(xminContent)
            bndbox.appendChild(xmin)
            ymin = xmlBuilder.createElement("ymin")
            # mathData = int(((float(oneline[2])) * Pheight + 1) - (float(oneline[4])) * 0.5 * Pheight)
            mathData = int(oneline[2])
            yminContent = xmlBuilder.createTextNode(str(mathData))
            # if yminContent < 0:
            #     yminContent = 0
            ymin.appendChild(yminContent)
            bndbox.appendChild(ymin)
            xmax = xmlBuilder.createElement("xmax")
            # mathData = int(((float(oneline[1])) * Pwidth + 1) + (float(oneline[3])) * 0.5 * Pwidth)
            mathData = int(oneline[3])
            xmaxContent = xmlBuilder.createTextNode(str(mathData))
            # if xmaxContent > Pwidth:
            #     xmaxContent = Pwidth
            xmax.appendChild(xmaxContent)
            bndbox.appendChild(xmax)
            ymax = xmlBuilder.createElement("ymax")
            # mathData = int(((float(oneline[2])) * Pheight + 1) + (float(oneline[4])) * 0.5 * Pheight)
            mathData = int(oneline[4])
            ymaxContent = xmlBuilder.createTextNode(str(mathData))
            # if ymaxContent > Pheight:
            #     ymaxContent = Pheight
            ymax.appendChild(ymaxContent)
            bndbox.appendChild(ymax)
            object.appendChild(bndbox)

            annotation.appendChild(object)

        f = open(xmlPath + name[0:-4] + ".xml", 'w')
        xmlBuilder.writexml(f, indent='\t', newl='\n', addindent='\t', encoding='utf-8')
        f.close()


if __name__ == "__main__":
    txt_path = 'D:\study\WongKinYiu\\test\\txt\\'
    xml_path = 'D:\study\WongKinYiu\\test\\Annotations\\'
    image_path = 'D:\study\WongKinYiu\\test\images\\'

    makexml(txt_path, xml_path, image_path)

删除多余的image，根据txt的数量来删除，最终是为了使image和txt相互对应

import os
import numpy as np


def readname():
    filePath = 'D:\HZQ\jtrafficsigns\yolov3-point\data\测试\Annotations\\'
    name = os.listdir(filePath)
    return name


def get_filename(path, filetype):
    name = []
    final_name = []
    for root, dirs, files in os.walk(path):
        for i in files:
            if filetype in i:
                name.append(i.replace(filetype, ''))  # 生成不带‘.csv’后缀的文件名组成的列表
    final_name = [item for item in name]  # 生成‘.csv’后缀的文件名组成的列表
    return final_name  # 输出由有‘.csv’后缀的文件名组成的列表


def numpy_to_int(num):
    name = []
    for i in num:
        i = int(i)
        name.append(i)
    return name


if __name__ == "__main__":
    path1 = 'D:\HZQ\jtrafficsigns\yolov3-point\data\Annotations\\'  # 指定文件所在路径
    filetype1 = '.xml'  # 指定文件类型
    path2 = 'D:\HZQ\jtrafficsigns\yolov3-point\data\JPEGImages\\'
    filetype2 = '.jpg'

    name1 = get_filename(path1, filetype1)  # xml删
    name2 = get_filename(path2, filetype2)  # jpg

    result_name1 = numpy_to_int(name1)
    # print("1",result_name1)
    result_name2 = numpy_to_int(name2)
    # print("2",result_name2)
    result_name1 = np.sort(result_name1)
    result_name2 = np.sort(result_name2)
    # [2,4,6,7,8,10,33,67]
    # [2,7,10,33,67]
    # print(result_name1)
    # print(result_name1.shape)
    # print(result_name2)
    # print(result_name2.shape)
    length = len(result_name2)
    # print(length)
    i = 0
    j = 0
    while i < length:
        if result_name1[j] == result_name2[i]:
            i += 1
            j += 1
        else:
            join1 = os.path.join('D:\HZQ\jtrafficsigns\yolov3-point\data\Annotations\\', str(result_name1[j]) + '.xml')
            print(join1)
            os.remove(join1)
            j+=1