word中的图片如何保存为jpg格式（把word导出成图片的流程）

由于工作中需要从大量docx文档中提出图片，于是到网上搜索，找了一大堆都是分析xml文件并提取的，太过于复杂，实际上有更简单的方法，我们需要用到python-docx这个第三方库，但该库并未开发提取图片功能，不过可以通过其他方法出得到图片数据并保存为图片。

本文为原创，如需转载请注明出处（仅处理docx文档，不能处理doc文档，如果需要可执行文件的，可将代码文件通过pyinstaller进行编译）。

软件界面

下面这段代码是核心：

    for file in os.listdir(filepath):
        try:
            #跳过非docx文件
            if ".docx" not in file:
                continue
            # 创建imgpath
            subimgpath = imgpath + re.sub(".docx","",file)
            if not os.path.exists(subimgpath):
                os.makedirs(subimgpath)

            doc = docx.document(filepath + file)        #打开文件
            for rel in doc.part._rels:
                rel = doc.part._rels[rel]               #获得资源
                if "image" not in rel.target_ref:
                    continue
                imgname = re.findall("/(.*)",rel.target_ref)[0]
                with open(subimgpath + "/" + imgname,"wb") as f:
                    f.write(rel.target_part.blob)
            ui.currentfile.settext("当前文件：" + imgname)
        except:
            continue

后来经过改进，使用pyqt5制作了界面，下面为源代码：

import docx,re,os,sys,ui_imgextract
from pyqt5.qtwidgets import qapplication,qmainwindow,qwidget,qmessagebox
from pyqt5.qt import qfiledialog

def run():
    filepath = ui.filepath.text()
    imgpath = ui.imgpath.text()
    if not os.path.exists(filepath):
        qmessagebox.about(main, "错误", "请选择docx文件目录！")
        return
    if not os.path.exists(imgpath):
        os.makedirs(imgpath)

    for file in os.listdir(filepath):
        try:
            #跳过非docx文件
            if ".docx" not in file:
                continue
            # 创建imgpath
            subimgpath = imgpath + re.sub(".docx","",file)
            if not os.path.exists(subimgpath):
                os.makedirs(subimgpath)

            doc = docx.document(filepath + file)        #打开文件
            for rel in doc.part._rels:
                rel = doc.part._rels[rel]               #获得资源
                if "image" not in rel.target_ref:
                    continue
                imgname = re.findall("/(.*)",rel.target_ref)[0]
                with open(subimgpath + "/" + imgname,"wb") as f:
                    f.write(rel.target_part.blob)
            ui.currentfile.settext("当前文件：" + imgname)
        except:
            continue
    qmessagebox.about(main, "完成", "图片提取已完成！")

def init():
    ui.btnrun.clicked.connect(run)      #绑定开始提取按钮
    ui.btnfilepath.clicked.connect(choicefiledir)  # 绑定选择docx文件目录
    ui.btnimgpath.clicked.connect(choiceimgoutputdir)    #绑定选择图片保存目录

    # docx文件默认目录
    ui.filepath.settext(os.getcwd())
    #默认输出目录
    if not os.path.exists(os.getcwd() + "img\"):
        os.makedirs(os.getcwd() + "img\")
    ui.imgpath.settext(os.getcwd() + "img\")


#选择docx文件目录
def choicefiledir():
    dir = qfiledialog.getexistingdirectory(main, "选择docx文件目录", os.getcwd())
    ui.filepath.settext(dir + "/")

#选择图片保存目录
def choiceimgoutputdir():
    dir = qfiledialog.getexistingdirectory(main, "选择输出目录", os.getcwd())
    ui.imgpath.settext(dir + "/")

if __name__ == "__main__":
    app = qapplication(sys.argv)
    main = qwidget()
    ui = ui_imgextract.ui_form()
    ui.setupui(main)
    main.show()
    init()

    sys.exit(app.exec_())

下面是界面文件ui_imgextract.py：

# -*- coding: utf-8 -*-

# form implementation generated from reading ui file 'ui_iask.ui'
#
# created by: pyqt5 ui code generator 5.11.3
#
# warning! all changes made in this file will be lost!

from pyqt5 import qtcore, qtgui, qtwidgets

class ui_form(object):
    def setupui(self, form):
        form.setobjectname("form")
        form.resize(604, 100)
        self.layoutwidget = qtwidgets.qwidget(form)
        self.layoutwidget.setgeometry(qtcore.qrect(10, 10, 581, 83))
        self.layoutwidget.setobjectname("layoutwidget")
        self.gridlayout_4 = qtwidgets.qgridlayout(self.layoutwidget)
        self.gridlayout_4.setcontentsmargins(0, 0, 0, 0)
        self.gridlayout_4.setobjectname("gridlayout_4")
        self.label_8 = qtwidgets.qlabel(self.layoutwidget)
        self.label_8.setobjectname("label_8")
        self.gridlayout_4.addwidget(self.label_8, 0, 0, 1, 1)
        self.filepath = qtwidgets.qlineedit(self.layoutwidget)
        self.filepath.setobjectname("filepath")
        self.gridlayout_4.addwidget(self.filepath, 0, 1, 1, 1)
        self.btnfilepath = qtwidgets.qpushbutton(self.layoutwidget)
        self.btnfilepath.setobjectname("btnfilepath")
        self.gridlayout_4.addwidget(self.btnfilepath, 0, 2, 1, 1)
        self.label_9 = qtwidgets.qlabel(self.layoutwidget)
        self.label_9.setobjectname("label_9")
        self.gridlayout_4.addwidget(self.label_9, 1, 0, 1, 1)
        self.imgpath = qtwidgets.qlineedit(self.layoutwidget)
        self.imgpath.setobjectname("imgpath")
        self.gridlayout_4.addwidget(self.imgpath, 1, 1, 1, 1)
        self.btnimgpath = qtwidgets.qpushbutton(self.layoutwidget)
        self.btnimgpath.setobjectname("btnimgpath")
        self.gridlayout_4.addwidget(self.btnimgpath, 1, 2, 1, 1)
        self.btnrun = qtwidgets.qpushbutton(self.layoutwidget)
        self.btnrun.setobjectname("btnrun")
        self.gridlayout_4.addwidget(self.btnrun, 2, 2, 1, 1)
        self.currentfile = qtwidgets.qlabel(self.layoutwidget)
        self.currentfile.setobjectname("currentfile")
        self.gridlayout_4.addwidget(self.currentfile, 2, 0, 1, 2)

        self.retranslateui(form)
        qtcore.qmetaobject.connectslotsbyname(form)

    def retranslateui(self, form):
        _translate = qtcore.qcoreapplication.translate
        form.setwindowtitle(_translate("form", "docx图片批量提取"))
        self.label_8.settext(_translate("form", "docx文件目录："))
        self.btnfilepath.settext(_translate("form", "选择"))
        self.label_9.settext(_translate("form", "图片保存目录："))
        self.btnimgpath.settext(_translate("form", "选择"))
        self.btnrun.settext(_translate("form", "开始提取"))
        self.currentfile.settext(_translate("form", "当前文件："))