從pdf文件中提取的圖像是反轉/負圖像。（C＃）
[英]Images extracted from pdf file is inverted/negative image.(C#)
問題描述
嗨，我正在使用pdf小丑庫（C＃）進行pdf到xhtml的轉換。我正在提取pdf內容，我提取的圖像是倒置圖像。我想要真實的圖像，請提出任何建議嗎？
該代碼粘貼在下面，
using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Windows.Forms;
using it.stefanochizzolini.clown.documents;
using it.stefanochizzolini.clown.files;
using it.stefanochizzolini.clown.documents.contents;
using it.stefanochizzolini.clown.documents.contents.objects;
using it.stefanochizzolini.clown.tools;
using it.stefanochizzolini.clown.documents.contents.composition;
using it.stefanochizzolini.clown.documents.contents.fonts;
using System.IO;
using it.stefanochizzolini.clown.bytes;
using it.stefanochizzolini.clown.objects;
namespace Test1
{
public partial class Form1 : Form
{
public int index;
public Form1()
{
InitializeComponent();
}
private void btnBrowse_Click(object sender, EventArgs e)
{
openFileDialog1.Filter = "PDF Files|*.pdf";
openFileDialog1.ShowDialog();
txtFileName.Text = openFileDialog1.FileName;
}
private void txtFileName_TextChanged(object sender, EventArgs e)
{
if (txtFileName.Text.Trim().Length > 0 && txtOutputPath.Text.Trim().Length > 0)
{
btnProcess.Enabled = true;
}
else { btnProcess.Enabled = false; }
}
private void btnProcess_Click(object sender, EventArgs e)
{
InitiateProcess(txtFileName.Text.Trim());
}
private void InitiateProcess(string FileName)
{
it.stefanochizzolini.clown.files.File file;
Document document;
try
{
// Open the PDF file!
file = new it.stefanochizzolini.clown.files.File(FileName);
// Get the PDF document!
document = file.Document;
//file.Dispose();
}
catch
{
MessageBox.Show("Sorry. Error in File");
return;
}
//Page stamper is used to draw contents on existing pages.
PageStamper stamper = new PageStamper();
foreach (Page page in document.Pages)
{
txtOutput.Text = txtOutput.Text + Environment.NewLine + Environment.NewLine + "\nScanning page " + (page.Index + 1) + "...\n";
stamper.Page = page;
// Wraps the page contents into a scanner.
Extract(new ContentScanner(page), stamper.Foreground, page );
stamper.Flush();
this.Refresh();
}
}
private void Extract(ContentScanner level, PrimitiveFilter builder, Page page)
{
string ctype = string.Empty;
if (level == null)
return;
while (level.MoveNext())
{
ContentObject content = level.Current;
//TextExtractor extractor = new TextExtractor();
string aa = content.GetType().ToString();
/*if(content is Text)
{
txtOutput.Text = txtOutput.Text + aa + Environment.NewLine;
List<ITextString> textStrings = extractor.Extract(page)[TextExtractor.DefaultArea];
foreach(ITextString textstring in textStrings)
{
RectangleF textstringbox=textstring.Box.Value;
txtOutput.Text = txtOutput.Text + "x:" + Math.Round(textstringbox.X);
txtOutput.Text = txtOutput.Text + "y:" + Math.Round(textstringbox.Y) + ",";
txtOutput.Text = txtOutput.Text + "w:" + Math.Round(textstringbox.Width) + ",";
txtOutput.Text = txtOutput.Text + "h:" + Math.Round(textstringbox.Height);
txtOutput.Text=txtOutput.Text+ "]: " + textstring.Text.ToString();
}
}*/
if (content is Text)
{
ContentScanner.TextWrapper text = (ContentScanner.TextWrapper)level.CurrentWrapper;
//ContentScanner.GraphicsState test = level.getState();
foreach (ContentScanner.TextStringWrapper textString in text.TextStrings)
{
RectangleF rf = (RectangleF)textString.Box;
/*txtOutput.Text = txtOutput.Text + Environment.NewLine + "Text [font size: " + textString.Style.FontSize + " ],[font Name: " +
textString.Style.Font.Name + " ]: " + textString.Text + "[position = left :" + rf.Left.ToString() + " & Top: " + rf.Top.ToString() + "X:" + rf.X.ToString() + "Y:" + rf.Y.ToString();*/
txtOutput.Text = txtOutput.Text + Environment.NewLine + textString.Text;
}
}
//else if (content is MarkedContent)
//{
// //MarkedContent mc = (MarkedContent) content
//}
else if (content is ShowText)
{
it.stefanochizzolini.clown.documents.contents.fonts.Font font = level.State.Font;
txtOutput.Text = txtOutput.Text + (font.Decode(((ShowText)content).Text));
}
else if (content is ContainerObject)
{
// Scan the inner level!
Extract(level.ChildLevel, builder, page);
}
else if (content is it.stefanochizzolini.clown.documents.contents.objects.InlineImage)
{
ContentScanner.InlineImageWrapper img = (ContentScanner.InlineImageWrapper)level.CurrentWrapper;
ExportImage(
img.InlineImage.Body.Value,
txtOutputPath.Text.Trim() + System.IO.Path.DirectorySeparatorChar + "ImageExtractionSample_" + (index++) + ".jpg"
);
}
else if (content is it.stefanochizzolini.clown.documents.contents.objects.GraphicsObject)
{
/*=============================================================================
* TO EXTRACT THE GRAPHICAL OBJECT WITHIN THE PDF
=============================================================================*/
ContentScanner.GraphicsObjectWrapper objectWrapper = level.CurrentWrapper;
if (objectWrapper == null)
{
continue;
}
/*
NOTE: Images can be represented on a page either as
external objects (XObject) or inline objects.
*/
SizeF? imageSize = null; // Image native size.
/*if (objectWrapper is PdfDataObject)
{
ContentScanner.GraphicsObjectWrapper gobjectwrapper=(ContentScanner.GraphicsObjectWrapper)gobjectwrapper;
it.stefanochizzolini.clown.objects.PdfDataObject pdobjt=gobjectwrapper
}*/
//if(objectWrapper is Image)
//{
// }
if (objectWrapper is ContentScanner.XObjectWrapper)
{
ContentScanner.XObjectWrapper xObjectWrapper = (ContentScanner.XObjectWrapper)objectWrapper;
it.stefanochizzolini.clown.documents.contents.xObjects.XObject Xobject = xObjectWrapper.XObject;
// Is the external object an image?
if (Xobject is it.stefanochizzolini.clown.documents.contents.xObjects.ImageXObject)
{
txtOutput.Text = txtOutput.Text + Environment.NewLine +
"External Image '" + xObjectWrapper.Name + "' (" + Xobject.BaseObject + ")"; // Image key and indirect reference.
imageSize = Xobject.Size; // Image native size.
PdfDataObject dataObject = Xobject.BaseDataObject;
;
PdfDictionary header = ((PdfStream)dataObject).Header;
if (header.ContainsKey(PdfName.Type) && header[PdfName.Type].Equals(PdfName.XObject) && header[PdfName.Subtype].Equals(PdfName.Image))
{
if (header[PdfName.Filter].Equals(PdfName.Image)) // JPEG image.
{
// Get the image data (keeping it encoded)!
IBuffer body1 = ((PdfStream)dataObject).GetBody(false);
//object h1 = PdfName.ColorSpace;
// Export the image!
ExportImage(
body1,
txtOutputPath.Text + System.IO.Path.DirectorySeparatorChar + "Image_" + (index++) + ".png"
);
}
}
}
else if (content is it.stefanochizzolini.clown.documents.interaction.annotations.Link)
{
Dictionary<RectangleF?, List<ITextString>> textStrings = null;
PageAnnotations annotations = page.Annotations;
TextExtractor extractor = new TextExtractor();
if (annotations == null)
{
Console.WriteLine("No annotations here.");
continue;
}
foreach (it.stefanochizzolini.clown.documents.interaction.annotations.Annotation annotation in annotations)
{
if (annotation is it.stefanochizzolini.clown.documents.interaction.annotations.Link)
{
if (textStrings == null)
{ textStrings = extractor.Extract(page); }
it.stefanochizzolini.clown.documents.interaction.annotations.Link link = (it.stefanochizzolini.clown.documents.interaction.annotations.Link)annotation;
RectangleF linkBox = link.Box;
StringBuilder linkTextBuilder = new StringBuilder();
foreach (ITextString linkTextString in extractor.Filter(textStrings, linkBox))
{ linkTextBuilder.Append(linkTextString.Text); }
string bb = linkTextBuilder.ToString();
txtOutput.Text = txtOutput.Text + "Link '" + linkTextBuilder.ToString();
txtOutput.Text = txtOutput.Text + " Position: "
+ "x:" + Math.Round(linkBox.X) + ","
+ "y:" + Math.Round(linkBox.Y) + ","
+ "w:" + Math.Round(linkBox.Width) + ","
+ "h:" + Math.Round(linkBox.Height);
}
}
}
else if (objectWrapper is ContentScanner.InlineImageWrapper)
{
txtOutput.Text = txtOutput.Text + Environment.NewLine + "Inline Image";
InlineImage inlineImage = ((ContentScanner.InlineImageWrapper)objectWrapper).InlineImage;
imageSize = inlineImage.Size; // Image native size.
}
if (imageSize.HasValue)
{
RectangleF box = objectWrapper.Box.Value; // Image position (location and size) on the page.
txtOutput.Text = txtOutput.Text + Environment.NewLine +
" on page " + (page.Index + 1) + " (" + page.BaseObject + ")"; // Page index and indirect reference.
txtOutput.Text = txtOutput.Text + Environment.NewLine + " Coordinates:";
txtOutput.Text = txtOutput.Text + Environment.NewLine + " x: " + Math.Round(box.X);
txtOutput.Text = txtOutput.Text + Environment.NewLine + " y: " + Math.Round(box.Y);
txtOutput.Text = txtOutput.Text + Environment.NewLine + " width: " + Math.Round(box.Width) + " (native: " + Math.Round(imageSize.Value.Width) + ")";
txtOutput.Text = txtOutput.Text + Environment.NewLine + " height: " + Math.Round(box.Height) + " (native: " + Math.Round(imageSize.Value.Height) + ")";
}
/*=============================================================================*/
}
}
}
}
private void ExportImage(IBuffer data, string outputPath)
{
FileStream outputStream;
try
{ outputStream = new FileStream(outputPath, FileMode.CreateNew); }
catch (Exception e)
{ throw new Exception(outputPath + " file couldn't be created.", e); }
try
{
BinaryWriter writer = new BinaryWriter(outputStream);
//ImageConverter ic = new ImageConverter();
//Image img = (Image)ic.ConvertFrom(data);
writer.Write(data.ToByteArray());
writer.Close();
outputStream.Close();
}
catch (Exception e)
{ throw new Exception(outputPath + " file writing has failed.", e); }
Console.WriteLine("Output: " + outputPath);
}
private void txtOutputPath_TextChanged(object sender, EventArgs e)
{
if (txtFileName.Text.Trim().Length > 0 && txtOutputPath.Text.Trim().Length > 0)
{
btnProcess.Enabled = true;
}
else { btnProcess.Enabled = false; }
}
private void btnBrowseOutput_Click(object sender, EventArgs e)
{
folderBrowserDialog1.ShowDialog();
txtOutputPath.Text = folderBrowserDialog1.SelectedPath;
}
private void btnsave_Click(object sender, EventArgs e)
{
if(txtOutput.Text=="")
{
MessageBox.Show("text box is empty");
}
else
{
saveFileDialog1.Filter = "Text Files (*.txt)|*.txt|RTF Files (*.rtf)|*.rtf";
if (saveFileDialog1.ShowDialog() == DialogResult.OK)
{
StreamWriter yaz = new StreamWriter(saveFileDialog1.FileName.ToString());
string yasoutput = txtOutput.Text;
yaz.Write(yasoutput);
}
}
}
private void btncancel_Click(object sender, EventArgs e)
{
this.Close();
}
private void btnclear_Click(object sender, EventArgs e)
{
txtFileName.Text = "";
txtOutput.Text = "";
txtOutputPath.Text = "";
}
}
}
2 個解決方案
解決方案1
1 2011-02-01 04:41:37
您需要反轉圖像緩沖區，即，將接收到的圖像緩沖區復制到新緩沖區，從圖像緩沖區的最后一個字節開始到圖像緩沖區的第一個字節。 這應該注意反轉的圖像。 對於每個像素數據，假定24位RGB值，則交換R和B值，以使RGB變為BGR。 這應該擺脫“負”效應並保存此新緩沖區。
解決方案2
0 已采納 2013-01-24 02:13:08
將您的位圖圖像傳遞給以下功能...
private byte[] BmpToBytes_Unsafe (Bitmap bmp)
{
BitmapData bData = bmp.LockBits(new Rectangle(0,0,IMAGE_WIDTH,IMAGE_HEIGHT),
ImageLockMode.ReadOnly,
PixelFormat.Format24bppRgb);
int lineSize = bData.Width * 3;
int byteCount = lineSize * bData.Height;
byte[] bmpBytes = new byte[byteCount];
byte[] tempLine = new byte[lineSize];
int bmpIndex = 0;
IntPtr scan = new IntPtr(bData.Scan0.ToInt32() + (lineSize * (bData.Height-1)));
for (int i = 0; i < bData.Height; i++)
{
Marshal.Copy(scan, tempLine, 0, lineSize);
scan = new IntPtr(scan.ToInt32()-bData.Stride);
tempLine.CopyTo(bmpBytes, bmpIndex);
bmpIndex += lineSize;
}
bmp.UnlockBits(bData);
return bmpBytes;
}
快樂的編碼... :)
Tesseract OCR 未按預期工作，無法從圖像中獲取完整文本。 C#
[英]Tesseract OCR not working as expected, unable to get the complete text from image. c#
聲明:本站的技術帖子網頁，遵循CC BY-SA 4.0協議，如果您需要轉載，請注明本站網址或者原文地址。任何問題請咨詢:yoyou2525@163.com.