VC实现基于Office 2007 MODI 12.0实现图形文字识别(OCR)的程序
1.安装Microsoft Office Document Imaging
1）查看是否安装：开始->所有程序->Microsoft Office->工具-> Microsoft Office Document Imaging
或者
C:/Program Files/Common Files/Microsoft Shared/MODI/12.0
2）如未安装，则打开：控制面板->添加或删除程序->Microsoft Office 2007->更改->添加或删除功能->继续->office工具->Microsoft Office Document Imaging ->从本机运行全部程序->继续
需要指定office2007安装盘路径，按照提示即可；
2.开发：
VC6.0建个MFC的Dialog based工程“OCR”，然后"Project"->"add to project"->"Components and controls"，在Registered ActiveX Controls中找到Microsoft Office Document Imaging Viewer Control 12.0，INSERT建立一个类。
然后从控件工具箱中拖一个到窗口上，右键classWizard，Member Variables找到IDC_MIDOCVIEW1然后AddVariable变量名为m_MIDOCtrl。然后在窗口上拖三个按钮控件"Brower"、"OCR"和"CLOSE"、一个静态文本控件“IDC_STATIC_PATH”和一个可编辑文本框控件“IDC_EDIT_TEXT”。
然后在你的文件里面找到OCRDlg.cpp，在最上面写的#import "C:/Program Files/Common Files/Microsoft Shared/MODI/12.0/MDIVWCTL.DLL"，路径需要你自己改一下，大多数就是这个路径，然后编译一下，在Debug下能找到MDIVWCTL.tlh，复制到工程文件夹中，在OCRDlg.cpp中写入#include "MDIVWCTL.tlh"，然后删除刚刚写的#import. "C:/Program Files/Common Files/Microsoft Shared/MODI/12.0/MDIVWCTL.DLL"。
1）实现打开TIF图片并显示在 ActiveX Control 中，Browser按钮函数：
void COCRDlg::OnBtnBrowser()
{
// TODO: Add extra validation here
CString csFilePath;
CFileDialog FileDlg(TRUE,"tif",NULL,OFN_HIDEREADONLY|OFN_FILEMUSTEXIST|OFN_ALLOWMULTISELECT,"Image Files (*.tif;*.tif)|*.tif;*.tif|All Files(*.*)|*.*||");
if(FileDlg.DoModal()==IDOK)
{
csFilePath = FileDlg.GetPathName();
SetDlgItemText(IDC_STATIC_PATH,csFilePath);
((CMiDocView *) GetDlgItem( IDC_MIDOCVIEW1 ))->SetFileName(csFilePath) ;
((CMiDocView *) GetDlgItem( IDC_MIDOCVIEW1 ))->SetFitMode(1);
}
}
2）COCRDlg类添加成员函数bReadOCRByMODIAXCtrl(CString csFilePath, CString &csText)，主实现识别
BOOL COCRDlg::bReadOCRByMODIAXCtrl(CString csFilePath, CString &csText)
{
BOOL bRet = TRUE;
HRESULT hr = 0;
csText.Empty();
IUnknown *pVal = NULL;
MODI::IDocument *IDobj = NULL;
MODI::ILayout *ILayout = NULL;
MODI::IImages *IImages = NULL;
MODI::IImage *IImage = NULL;
MODI::IWords *IWords = NULL;
MODI::IWord *IWord = NULL;
try
{
pVal = (IUnknown *) m_MIDOCtrl.GetDocument();
if ( pVal != NULL )
{
//Already has image in it, Don't need to create again
//Just get IDocument interface
hr = pVal->QueryInterface(IID_IDocument,(void**) &IDobj);
if ( SUCCEEDED(hr) )
{
hr = IDobj->OCR(MODI::miLANG_ENGLISH/*miLANG_CHINESE_SIMPLIFIED*/,1,1);
if ( SUCCEEDED(hr) )
{
IDobj->get_Images(&IImages);
long iImageCount=0;
IImages->get_Count(&iImageCount);
for ( int img =0; img<iImageCount;img++)
{
IImages->get_Item(img,(IDispatch**)&IImage);
IImage->get_Layout(&ILayout);
long numWord=0;
ILayout->get_NumWords(&numWord);
ILayout->get_Words(&IWords);
IWords->get_Count(&numWord);
for ( long i=0; i<numWord;i++)
{
IWords->get_Item(i,(IDispatch**)&IWord);
CString csTemp;
BSTR result;
IWord->get_Text(&result);
char buf[256];
sprintf(buf,"%S",result);
csTemp.Format("%s",buf);
csText += csTemp;
csText +=" ";
}
//Release all objects
IWord->Release();
IWords->Release();
ILayout->Release();
IImage->Release();
}
IImages->Release();
} else {
bRet = FALSE;
}
} else {
bRet = FALSE;
}
IDobj->Close(0);
IDobj->Release();
pVal->Release();
} else {
bRet = FALSE;
}
pVal = NULL;
IDobj = NULL;
ILayout = NULL;
IImages = NULL;
IImage = NULL;
IWords = NULL;
IWord = NULL;
}
catch(...)
{
}
return bRet;
}
此时VC编译器会显示IID_IDocument未声明，后面编辑MDIVWCTL.tlh即可。
3）双击OCR按钮，写入响应函数，调用识别函数：
void COCRDlg::OnBtnOCR()
{
// TODO: Add your control notification handler code here
CString csText;
CString csFilePath;
GetDlgItemText(IDC_STATIC_PATH,csFilePath);
if ( bReadOCRByMODIAXCtrl(csFilePath,csText) == TRUE )
{
SetDlgItemText(IDC_EDIT_TEXT,csText);
}else
SetDlgItemText(IDC_EDIT_TEXT,"ERROR");
}
4）解决IID_IDocument未声明的问题，编辑MDIVWCTL.tlh，在在最后#pragma pack(pop)之前加入如下：
//
// Named GUID constants initializations
//
extern "C" const GUID __declspec(selectany) LIBID_MODI =
{0xa5ededf4,0x2bbc,0x45f3,{0x82,0x2b,0xe6,0x0c,0x27,0x8a,0x1a,0x79}};
extern "C" const GUID __declspec(selectany) IID_IMiFont =
{0x1ffc9a7b,0x7606,0x4c97,{0x81,0x19,0x78,0x67,0x3e,0x9d,0x48,0x21}};
extern "C" const GUID __declspec(selectany) IID_IMiRect =
{0x87d6fb47,0x7cfa,0x4db3,{0x82,0xcd,0xe8,0xdc,0xc5,0xbd,0xa7,0xab}};
extern "C" const GUID __declspec(selectany) IID_IMiRects =
{0x895c88a9,0x8598,0x44cc,{0x91,0xc4,0x80,0x10,0xae,0x24,0x75,0xeb}};
extern "C" const GUID __declspec(selectany) IID_IWord =
{0x5612275b,0xb4f1,0x42af,{0xb6,0x96,0x36,0x0d,0x40,0xdf,0x04,0x1d}};
extern "C" const GUID __declspec(selectany) IID_IWords =
{0xd8363824,0x4b07,0x423b,{0x88,0x03,0x49,0xb4,0x70,0x43,0x4f,0x3d}};
extern "C" const GUID __declspec(selectany) IID_ILayout =
{0xc300c846,0xa3fd,0x4a5b,{0xad,0x65,0x4a,0x6a,0xb4,0x6b,0x78,0x21}};
extern "C" const GUID __declspec(selectany) IID_IImage =
{0xac0d48a6,0x886d,0x4eb5,{0xa8,0xa1,0x09,0x3d,0x60,0xb9,0xa8,0x4a}};
extern "C" const GUID __declspec(selectany) IID_IImages =
{0x77096557,0x54ca,0x45c8,{0x81,0xc9,0xe7,0x0c,0x2a,0xba,0xa0,0xfb}};
extern "C" const GUID __declspec(selectany) IID_IDocument =
{0xd4073843,0xa58a,0x469a,{0xa8,0xe2,0xcf,0xf3,0xff,0x77,0xee,0x4e}};
extern "C" const GUID __declspec(selectany) DIID__IDocumentEvents =
{0x0cf31963,0x5e4d,0x4772,{0xae,0x8d,0x82,0xfe,0x5a,0x08,0x2b,0x26}};
extern "C" const GUID __declspec(selectany) CLSID_Document =
{0x40942a6c,0x1520,0x4132,{0xbd,0xf8,0xbd,0xc1,0xf7,0x1f,0x54,0x7b}};
extern "C" const GUID __declspec(selectany) CLSID_Layout =
{0x1afc4117,0x128e,0x4314,{0x9d,0x53,0x64,0xcb,0xda,0x5c,0x7e,0x02}};
extern "C" const GUID __declspec(selectany) DIID__IImageEvents =
{0x99f3a122,0x448f,0x4d36,{0x82,0x43,0x48,0xbf,0xe3,0x2d,0x50,0xb2}};
extern "C" const GUID __declspec(selectany) CLSID_Image =
{0x56f963ec,0x6efc,0x4a6b,{0x9a,0x1e,0x5b,0xfe,0x54,0x5c,0x89,0xd0}};
extern "C" const GUID __declspec(selectany) CLSID_Word =
{0x5995e30f,0x59eb,0x42ae,{0xbd,0xcf,0x89,0x08,0x27,0x1c,0x0b,0x32}};
extern "C" const GUID __declspec(selectany) CLSID_MiFont =
{0xc5eae79b,0xaccc,0x4e51,{0x8e,0xd5,0xa1,0x6f,0x42,0xfb,0x7b,0x1f}};
extern "C" const GUID __declspec(selectany) CLSID_Words =
{0x72acbce3,0x9067,0x4d5e,{0x95,0x39,0x34,0x97,0xfe,0xcd,0x03,0x2d}};
extern "C" const GUID __declspec(selectany) CLSID_Images =
{0xf961d185,0x2cc1,0x43db,{0x80,0xab,0x56,0x7f,0xa1,0x02,0x51,0xec}};
extern "C" const GUID __declspec(selectany) CLSID_MiRect =
{0x00607e44,0x15df,0x49bc,{0xaf,0x0c,0xa9,0x02,0x3e,0x06,0x07,0x91}};
extern "C" const GUID __declspec(selectany) CLSID_MiRects =
{0xcb2fbd52,0x56c1,0x4dd8,{0x89,0x7f,0x96,0x1d,0x15,0x43,0xb8,0x77}};
extern "C" const GUID __declspec(selectany) IID_IMiSelectRect =
{0xea280c3c,0xe4b7,0x42bf,{0xac,0xc8,0xfe,0x3a,0xd3,0x58,0x16,0x38}};
extern "C" const GUID __declspec(selectany) IID_IMiSelectRects =
{0x3a1e1b7a,0xc041,0x4ddc,{0xbf,0x3b,0x04,0x2a,0x0b,0x95,0xb8,0x2b}};
extern "C" const GUID __declspec(selectany) IID_IMiSelectableItem =
{0x01c4414a,0xd123,0x4bc7,{0xa1,0xfa,0x64,0xe3,0x76,0xc0,0x16,0x55}};
extern "C" const GUID __declspec(selectany) IID_IMiSelectableImage =
{0xf6379198,0x3b20,0x461a,{0xb3,0xa9,0x19,0x19,0x45,0x75,0x25,0x57}};
extern "C" const GUID __declspec(selectany) IID_IMiSearchCallback =
{0x8376d508,0x78ca,0x416d,{0xa9,0x03,0x2e,0xd6,0x2e,0x91,0xe2,0x9b}};
extern "C" const GUID __declspec(selectany) IID_IMiDocView =
{0xf958524a,0x8422,0x4b07,{0xb6,0x9e,0x19,0x9f,0x24,0x21,0xed,0x89}};
extern "C" const GUID __declspec(selectany) DIID__IMiDocViewEvents =
{0x9028b775,0xec59,0x4118,{0x86,0x2a,0xef,0xde,0xaf,0x59,0x55,0xa4}};
extern "C" const GUID __declspec(selectany) IID_IMiDocSearch =
{0xbc06ed64,0xa40c,0x4fb4,{0xa6,0x60,0xac,0x6d,0xbc,0x75,0x22,0x92}};
extern "C" const GUID __declspec(selectany) CLSID_MiDocView =
{0xef347a62,0xba21,0x42e4,{0x94,0xa0,0x1c,0x0a,0x6d,0x7f,0xdf,0xe7}};
extern "C" const GUID __declspec(selectany) CLSID_MiDocSearch =
{0xfa554db9,0xc806,0x46a9,{0xae,0xf8,0xb5,0xaf,0x89,0x1a,0x94,0xf9}};
extern "C" const GUID __declspec(selectany) CLSID_MiSelectRect =
{0xa4d2b51f,0x809f,0x4c53,{0x91,0x47,0x55,0x16,0x87,0x18,0x16,0xbd}};
extern "C" const GUID __declspec(selectany) CLSID_MiSelectRects =
{0x23e9e94b,0xc202,0x4e9c,{0x8c,0x2e,0xf5,0x3b,0xc5,0x36,0xff,0x1d}};
这是office 2003生成的MDIVWCTL.tlh文件中存在的，但在office 2007生成的MDIVWCTL.tlh文件中缺失，因此要补上。
3.运行
运行程序，单击Browse导入测试图片，然后单击OCR进行光学字符识别，效果如图
特别说明：
1）只能识别TIF格式图片；
2）仅对英文识别较为准确；
VC实现基于Office 2007 MODI 12.0实现图形文字识别(OCR)的程序相关推荐
- Office 2007 SP1 V12.0.6207.1000 简/繁/英文版 迅雷高速下载
转自:x-beta 原文链接:http://www.x-beta.cn/software/Office-2007-SP1-v12.html 我们强烈推荐你在安装 Microsoft Office 2 ...
- 开源免费图片文字识别 OCR 工具 tesseract v4.1.0 的 Docker 镜像制作与使用
开源免费图片文字识别 OCR 工具 tesseract v4.1.0 的 Docker 镜像制作与使用 一 背景 在日常的一些工作中,偶尔也需要我们把图片转换为文字.目前大部分办公软件还无法实现类似的 ...
- 吴恩达《Machine Learning》精炼笔记 12：大规模机器学习和图片文字识别 OCR
作者 | Peter 编辑 | AI有道 系列文章: 吴恩达<Machine Learning>精炼笔记 1:监督学习与非监督学习 吴恩达<Machine Learning>精 ...
- python实现文字识别软件_文字识别(OCR)CRNN（基于pytorch、python3） 实现不定长中文字符识别...
文字识别(OCR)CRNN(基于pytorch.python3) 实现不定长中文字符识别 发布时间:2018-09-26 19:40, 浏览次数:1265 , 标签: OCR CRNN pytorch ...
- html识别文字转语音,万能君工具箱v1.8 - 基于百度AI实现语音转文字/文字转语音/文字识别OCR/翻译...
一.需求分析 1.音频转文字:目前市面上的音频转文字大多收费.音频转文字的需求是:上传一段音频,直接识别成文字,对于会议记录的比较好使,注意不是实时的语音识别. 2.文字转语音:转的语音不要太生硬. ...
- wps office 2007 10.1.0.7224
wps office 2007个人版是一款实用与家庭用户办公的文字处理办公软件.它包含三个板块,分别是:WPS文字.WPS表格.WPS演示,可以帮你处理文字,做表格.ppt的制作都是非常实用方便的.不 ...
- 文字识别(OCR)CRNN（基于pytorch、python3） 实现不定长中文字符识别
最近开源了一个之前做的人脸关键点检测的算法,欢迎star GitHub - Sierkinhane/TAB: Think about boundary: Fusing multi-level bou ...
- 基于笔画描述的文字识别OCR系统——技术说明
1. 已排除的方法 以下方法经过我们编写程序代码来实践验证了其不可行性. 1.1 游程统计法 思路:统计出点阵中的任意一点向六个角度0°.30°.60°.90°.120°.150°所能遍历的点数.如图 ...
- 基于Python实现的图像文字识别OCR工具
引言 最近在技术交流群里聊到一个关于图像文字识别的需求,在工作.生活中常常会用到,比如票据.漫画.扫描件.照片的文本提取. 博主基于 PyQt + labelme + PaddleOCR 写了一个桌面 ...
最新文章
- Python爬虫QQ说说并分析朋友状况
- [Leedcode][第十题][剑指offer]面试题第[19]题[正则表达式][动态规划][递归][JAVA]
- JavaSE09:String、StringBuffer、StringBuilder底层源码解析（纯干货）
- HTTP2.0，HTTP1.1，HTTP1.0三者在通性性能上的优化方法
- git log用法【转】
- 加一度分享：如何降低推广预算，增加KPI
- r语言html爬虫,如何用R语言爬取网页中的表格
- Exchange 2013反垃圾邮件功能
- GsonFormat的使用
- systemd 编写微信鱼虾蟹棋牌搭建服务管理脚本
- 中文图片验证码程序。
- 软件工程——软科中国大学专业排名
- 学习ui设计软件有哪些
- Kotlin高仿微信-第26篇-朋友圈-选择图片、小视频对话框
- 【Java中的构造器（构造方法）_基本语法_详细解读】
- 华为云计算之rainbow迁移实验
- 《当程序员的那些狗日日子》（五十五）另一种生存之道
- python数据导出excel_Python-将数据表中数据导出到excel
- 《图说VR入门》——DK2入门及其资源汇总
- 牛肉的营养价值高 晚上为什么不能吃牛肉