【基于WPF+OneNote+Oracle的中文图片识别系统阶段总结】之篇四:关于OneNote入库处理以及审核

时间:2021-01-16 23:00:32

篇一:WPF常用知识以及本项目设计总结:http://www.cnblogs.com/baiboy/p/wpf.html

篇二:基于OneNote难点突破和批量识别:http://www.cnblogs.com/baiboy/p/wpf1.html

篇三:批量处理后的txt文件入库处理:http://www.cnblogs.com/baiboy/p/wpf2.html

篇四:关于OneNote入库处理以及审核:http://www.cnblogs.com/baiboy/p/wpf3.html


【小记】:大胆尝试才能突破,某个中医药大学有一批图片需要处理(OCR),然后进行数据挖掘。之前没有接触过OCR这个东西,但是还是应允了。在网上搜索一番,关于中文图片识别,最终敲定为基于微软的OneNote,其识别率相对较高。网上这个技术点的资料真心不多,后来于博客园找到一篇博文,但是那个程序还是bug百出,而且只是单处理。后来经过一番摸索逐个突破,批处理完成。然后进行界面设计,这些零碎工作完成后,便是入库处理。由于OneNote生成的xml文件封装好的,即不可视的。便将其代码处理生成txt文件,再进行Oracle入库处理。入库前需要文件内容审核,并且在WPF开发中数据绑定和分页中做了独特处理。现在经过半个月的工作,本项目做个阶段总结。一则知识总结便于二次开发,尽量保持程序流畅性,核心知识做以梳理;另外,相关WPFOneNote常用技术共享,便于部分园友所需。本人技术有限,欢迎交流。项目还未结束,暂作阶段文章发布,随后相继发布。


篇四:关于OneNote入库处理以及审核

【开篇概述】:文本作为系列文章的最后一篇,主要对页面进行设计布局和审核页面的完善,其中主页面用户可以查询审核通过的信息,浏览详细信息。也可以作为导航到orc页面,对批量图片进行文字处理,处理后的文件,可以通过入库页面对文件信息提取。达到半自动填写表格的状态,由于文件信息不一定符合主观要求或者格式不一致,所以设置了审核页面,也是信息修改页面,信息无误后,选择下面审核通过复选框即可成功入库。后面就是交付做数据挖掘处理了。效果如图所示

主页面:

【基于WPF+OneNote+Oracle的中文图片识别系统阶段总结】之篇四:关于OneNote入库处理以及审核

完整代码:

namespace OnenoteOCRDemo
{
/// <summary>
/// Index.xaml 的交互逻辑
/// </summary>
public partial class Index : Window
{
public Index()
{
InitializeComponent();
this.Loaded += new RoutedEventHandler(MainWindow_Loaded);
}
page page1 = new page();
void MainWindow_Loaded(object sender, RoutedEventArgs e)
{
this.c1.Children.Add(page1);
this.Loaded += delegate
{
InitData();
};
dataGrid1.LoadingRow += new EventHandler<DataGridRowEventArgs>(dataGrid1_LoadingRow);
}
void dataGrid1_LoadingRow(object sender, DataGridRowEventArgs e)
{
e.Row.Header = e.Row.GetIndex() + ; //设置行表头的内容值
}
//查询
private void InitData()
{
string sql = "select * from T_TCM_THESIS where auditflag=1 order by ID desc";
DataSet ds = MYHelper.SQLHelper.GetOrlData(sql);
if (ds.Tables[].Rows.Count > )
{
page1.ShowPages(this.dataGrid1, ds, );
}
else
{
System.Windows.Forms.MessageBox.Show("Erroy");
}
} private void Window_Loaded(object sender, RoutedEventArgs e)
{
InitData();
} private void hlorc_Click(object sender, RoutedEventArgs e)
{
Main m = new Main();
m.Show();
} private void hltext_Click(object sender, RoutedEventArgs e)
{
TextData td = new TextData();
td.Show();
}
//刷新
private void hlnew_Click(object sender, RoutedEventArgs e)
{
InitData();
} private void hyper2_Click(object sender, RoutedEventArgs e)
{
this.Hide();
DataRowView mySelectedElement = (DataRowView)dataGrid1.SelectedItem;
int ID = Convert.ToInt32((mySelectedElement.Row[])); //获取该行的FID
falg f = new falg(ID,);
f.Show();
} }
}

审核页面:

【基于WPF+OneNote+Oracle的中文图片识别系统阶段总结】之篇四:关于OneNote入库处理以及审核

完整代码:

namespace OnenoteOCRDemo
{
/// <summary>
/// falg.xaml 的交互逻辑
/// </summary>
public partial class falg : Window
{
int i;
public falg(int id,int n)
{
InitializeComponent();
txt.Text = id.ToString();
THESIS(id);
i = n;
if (i == )
{
btnupdate.Visibility = System.Windows.Visibility.Hidden;
cbflag.Visibility = System.Windows.Visibility.Hidden;
txtflag.Text="审核通过";
}
}
private void THESIS(int id)
{
string sql = "select * from T_TCM_THESIS where id="+id; DataTable dt = MYHelper.SQLHelper.GetTabOrlData(sql);
if (dt.Rows.Count>)
{
txtdocname.Text = dt.Rows[]["docname"].ToString();
txtauthor.Text = dt.Rows[]["author"].ToString();
txtcompany.Text = dt.Rows[]["authorcompany"].ToString();
txtcontent.Text = dt.Rows[]["content"].ToString();
txtdigest.Text = dt.Rows[]["digest"].ToString();
txtdisease.Text = dt.Rows[]["disease"].ToString();
txtdoctype.Text = dt.Rows[]["doctype"].ToString();
txtenglishtitle.Text = dt.Rows[]["englishtitle"].ToString();
txtkeyword.Text = dt.Rows[]["keyword"].ToString();
txtmemo.Text = dt.Rows[]["memo"].ToString();
txtpath.Text = dt.Rows[]["path"].ToString();
txtsource.Text = dt.Rows[]["source"].ToString();
txtsourcetype.Text = dt.Rows[]["sourcetype"].ToString();
txtsummary.Text = dt.Rows[]["summary"].ToString();
txttitle.Text = dt.Rows[]["title"].ToString();
txtyears.Text = dt.Rows[]["years"].ToString();
int flag =Convert.ToInt32(dt.Rows[]["auditflag"].ToString());
if (flag == )
{
cbflag.IsChecked = true;
}
else
{
cbflag.IsChecked = false;
}
}
}
public void TextIsNull()
{
txtdocname.Text = "";
txtyears.Text = "";
txtdoctype.Text = "";
txtsummary.Text = "";
txtauthor.Text = "";
txtsource.Text = "";
txtsourcetype.Text = "";
txtmemo.Text = "";
txttitle.Text = "";
txtenglishtitle.Text = "";
txtcompany.Text = "";
txtkeyword.Text = "";
txtcontent.Text = "";
txtdigest.Text = "";
txtpath.Text = "";
txtdisease.Text = "";
cbflag.IsChecked = false;
}
//审核信息(更新信息)
private void btnupdate_Click(object sender, RoutedEventArgs e)
{
int id=Convert.ToInt32( txt.Text);
string docname = txtdocname.Text.ToString();
string years = txtyears.Text.ToString();
string doctype = txtdoctype.Text.ToString();
string summary = txtsummary.Text.ToString();
string author = txtauthor.Text.ToString();
string source = txtsource.Text.ToString();
string sourcetype = txtsourcetype.Text.ToString();
string memo = txtmemo.Text.ToString();
string title = txttitle.Text.ToString();
string englishtitle = txtenglishtitle.Text.ToString();
string authorcompany = txtcompany.Text.ToString();
string keyword = txtkeyword.Text.ToString();
string content = txtcontent.Text.ToString().Trim();
string digest = txtdigest.Text.ToString();
string path = txtpath.Text.ToString();
string disease = txtdisease.Text.ToString();
int auditflag;
if (cbflag.IsChecked == true)
{
auditflag = ;
}
else
{
auditflag =;
}
try
{
string sql = "update T_TCM_THESIS set docname='" + docname + "',years='" + years + "',doctype='" + doctype + "',summary='" + summary + "',";
sql += "author='" + author + "',source='" + source + "',sourcetype='" + sourcetype + "',memo='" + memo + "',title='" + title + "',";
sql += "englishtitle='" + englishtitle + "',authorcompany='" + authorcompany + "',keyword='" + keyword + "',content='" + content + "',digest='" + digest + "',";
sql += "path='" + path + "',disease='" + disease + "',auditflag=" + auditflag + " where ID= "+id;
int i = SQLHelper.OrlExecuteQuery(sql);
if (i > )
{
System.Windows.Forms.MessageBox.Show("审核通过", "审核信息");
TextIsNull();
this.Close();
TextData td = new TextData();
td.Show();
}
else
{
System.Windows.Forms.MessageBox.Show("审核失败", "审核信息");
}
}
catch (Exception ex)
{
System.Windows.Forms.MessageBox.Show(ex.Message);
}
} private void btnback_Click(object sender, RoutedEventArgs e)
{
this.Close();
if (i == )
{
TextData td = new TextData();
td.Show();
}
else {
Index index = new Index();
index.Show();
}
}
}
}

【功能以及操作描述】

  1. 打开主页面可以看到,每条信息前序号的生成,通过后台对事件重写完成,代码如下。
            void MainWindow_Loaded(object sender, RoutedEventArgs e)
    {
    this.c1.Children.Add(page1);
    this.Loaded += delegate
    {
    InitData();
    };
    dataGrid1.LoadingRow += new EventHandler<DataGridRowEventArgs>(dataGrid1_LoadingRow);
    }
    void dataGrid1_LoadingRow(object sender, DataGridRowEventArgs e)
    {
    e.Row.Header = e.Row.GetIndex() + ; //设置行表头的内容值
    }

      

  2. 页面自动显示为分页显示,通过用户控件进行分页,然后对数据控件进行绑定,详细操作见系列二文章。本功能代码如下:
     //查询
    private void InitData()
    {
    string sql = "select * from T_TCM_THESIS where auditflag=1 order by ID desc";
    DataSet ds = MYHelper.SQLHelper.GetOrlData(sql);
    if (ds.Tables[].Rows.Count > )
    {
    page1.ShowPages(this.dataGrid1, ds, );
    }
    else
    {
    System.Windows.Forms.MessageBox.Show("Erroy");
    }
    }
  3. 点击详细信息按钮,跳转的详细信息页面。如何获取DataGrid主键是关键,具体实现为:
                this.Hide();
    DataRowView mySelectedElement = (DataRowView)dataGrid1.SelectedItem;
    int ID = Convert.ToInt32((mySelectedElement.Row[])); //获取该行的FID
    falg f = new falg(ID,);
    f.Show();
  4. 在目标页面,通过页面接受页面传值,显示到详细信息。
     private void THESIS(int id)
    {
    string sql = "select * from T_TCM_THESIS where id="+id; DataTable dt = MYHelper.SQLHelper.GetTabOrlData(sql);
    if (dt.Rows.Count>)
    {
    txtdocname.Text = dt.Rows[]["docname"].ToString();
    txtauthor.Text = dt.Rows[]["author"].ToString();
    txtcompany.Text = dt.Rows[]["authorcompany"].ToString();
    txtcontent.Text = dt.Rows[]["content"].ToString();
    txtdigest.Text = dt.Rows[]["digest"].ToString();
    txtdisease.Text = dt.Rows[]["disease"].ToString();
    txtdoctype.Text = dt.Rows[]["doctype"].ToString();
    txtenglishtitle.Text = dt.Rows[]["englishtitle"].ToString();
    txtkeyword.Text = dt.Rows[]["keyword"].ToString();
    txtmemo.Text = dt.Rows[]["memo"].ToString();
    txtpath.Text = dt.Rows[]["path"].ToString();
    txtsource.Text = dt.Rows[]["source"].ToString();
    txtsourcetype.Text = dt.Rows[]["sourcetype"].ToString();
    txtsummary.Text = dt.Rows[]["summary"].ToString();
    txttitle.Text = dt.Rows[]["title"].ToString();
    txtyears.Text = dt.Rows[]["years"].ToString();
    int flag =Convert.ToInt32(dt.Rows[]["auditflag"].ToString());
    if (flag == )
    {
    cbflag.IsChecked = true;
    }
    else
    {
    cbflag.IsChecked = false;
    }
    }
    }
  5. 审核页面的具体操作也是类似, 不同之处在于对传值页面进行逻辑判断,如果是审核则允许页面信息的更新,如果查看详细信息,则更新按钮隐藏。

【篇末】:截止到此,本程序完全结束,本章涉及页面传值和数据绑定等基本信息,这方面大体与winfrom操作一致。本程序采用讲涉及的知识点数十个,均一一列出整理。数据库sql文件在项目的db文件夹中,附上完整源码如下:

注:源码地址链接总是不正确,后来打开管理--〉文件,查看源码,找到本文件的链接地址粘贴即可。