从乐谱中提取每行的小节[关闭]
Posted
技术标签:
【中文标题】从乐谱中提取每行的小节[关闭]【英文标题】：Extract measures on each line from sheet music [closed] 【发布时间】：2013-01-29 15:11:27 【问题描述】：我想知道一种提取单个度量值的方法。我不确定是否已经存在这种算法，所以我考虑从左到右扫描乐谱，从一行小节的上方和下方提取所有空白。
我不是在寻找将乐谱转换为 MusicXML 或提取其他有用信息的方法。不，基本上我正在处理的是一份常规文件。我需要将段落分开。我对段落所传达的信息不感兴趣，而只是将它们与文档的区域分开。在这种情况下，一个段落将是一系列措施。我不需要单独的小节，而是每行乐谱上的所有小节。
这是我想要的完整乐谱的输出之一，但没有标题、作曲家等。
【问题讨论】：
五线谱上方/下方的笔记也需要吗？有时这些可能会有点纠结。 “提取单个度量行”是什么意思？不是音乐术语。 不是直接答案，而是在谷歌上搜索“python 乐谱 ocr”，带我到另一个堆栈问题 :: ***.com/questions/675077/ocr-for-sheet-music :: 这可能会帮助你 这个问题实际上有点类似于描述如何在文本中找到“河流”（空白的线性流）的问题。我建议你也看看这个答案：dsp.stackexchange.com/questions/374/river-detection-in-text 【参考方案1】：假设您在 PDF 文件中有乐谱，我将使用 Apache PDFBox 从包含乐谱的输入 PDF 文件中获取图像，然后找到您需要的整个条形图的坐标，并使用选定的图像定义坐标来裁剪图像并对其进行操作，直到获得所需的结果。
PDDocument document = null;
document = PDDocument.load(inFile);
List pages = document.getDocumentCatalog().getAllPages();
Iterator iter = pages.iterator();
while (iter.hasNext())
PDPage page = (PDPage) iter.next();
PDResources resources = page.getResources();
Map pageImages = resources.getImages();
if (pageImages != null)
Iterator imageIter = pageImages.keySet().iterator();
while (imageIter.hasNext())
String key = (String) imageIter.next();
PDXObjectImage image = (PDXObjectImage) pageImages.get(key);
image.write2OutputStream(/* some output stream */);
这是 Apache PDFBox 中可用的示例代码。
import java.io.File;
import java.io.IOException;
import java.util.Iterator;
import java.util.List;
import java.util.Map;
import org.apache.pdfbox.pdmodel.PDDocument;
import org.apache.pdfbox.pdmodel.PDPage;
import org.apache.pdfbox.pdmodel.PDResources;
import org.apache.pdfbox.pdmodel.encryption.AccessPermission;
import org.apache.pdfbox.pdmodel.encryption.StandardDecryptionMaterial;
import org.apache.pdfbox.pdmodel.graphics.xobject.PDXObject;
import org.apache.pdfbox.pdmodel.graphics.xobject.PDXObjectForm;
import org.apache.pdfbox.pdmodel.graphics.xobject.PDXObjectImage;
/**
* This will read a read pdf and extract images. <br/><br/>
*
* usage: java org.apache.pdfbox.ExtractImages <pdffile> <password> [imageprefix]
*
* @author <a href="mailto:ben@benlitchfield.com">Ben Litchfield</a>
* @version $Revision: 1.7 $
*/
public class ExtractImages
private int imageCounter = 1;
private static final String PASSWORD = "-password";
private static final String PREFIX = "-prefix";
private static final String ADDKEY = "-addkey";
private static final String NONSEQ = "-nonSeq";
private ExtractImages()
/**
* This is the entry point for the application.
*
* @param args The command-line arguments.
*
* @throws Exception If there is an error decrypting the document.
*/
public static void main( String[] args ) throws Exception
ExtractImages extractor = new ExtractImages();
extractor.extractImages( args );
private void extractImages( String[] args ) throws Exception
if( args.length < 1 || args.length > 4 )
usage();
else
String pdfFile = null;
String password = "";
String prefix = null;
boolean addKey = false;
boolean useNonSeqParser = false;
for( int i=0; i<args.length; i++ )
if( args[i].equals( PASSWORD ) )
i++;
if( i >= args.length )
usage();
password = args[i];
else if( args[i].equals( PREFIX ) )
i++;
if( i >= args.length )
usage();
prefix = args[i];
else if( args[i].equals( ADDKEY ) )
addKey = true;
else if( args[i].equals( NONSEQ ) )
useNonSeqParser = true;
else
if( pdfFile == null )
pdfFile = args[i];
if(pdfFile == null)
usage();
else
if( prefix == null && pdfFile.length() >4 )
prefix = pdfFile.substring( 0, pdfFile.length() -4 );
PDDocument document = null;
try
if (useNonSeqParser)
document = PDDocument.loadNonSeq(new File(pdfFile), null, password);
else
document = PDDocument.load( pdfFile );
if( document.isEncrypted() )
StandardDecryptionMaterial spm = new StandardDecryptionMaterial(password);
document.openProtection(spm);
AccessPermission ap = document.getCurrentAccessPermission();
if( ! ap.canExtractContent() )
throw new IOException(
"Error: You do not have permission to extract images." );
List pages = document.getDocumentCatalog().getAllPages();
Iterator iter = pages.iterator();
while( iter.hasNext() )
PDPage page = (PDPage)iter.next();
PDResources resources = page.getResources();
// extract all XObjectImages which are part of the page resources
processResources(resources, prefix, addKey);
finally
if( document != null )
document.close();
private void processResources(PDResources resources, String prefix, boolean addKey) throws IOException
if (resources == null)
return;
Map<String, PDXObject> xobjects = resources.getXObjects();
if( xobjects != null )
Iterator<String> xobjectIter = xobjects.keySet().iterator();
while( xobjectIter.hasNext() )
String key = xobjectIter.next();
PDXObject xobject = xobjects.get( key );
// write the images
if (xobject instanceof PDXObjectImage)
PDXObjectImage image = (PDXObjectImage)xobject;
String name = null;
if (addKey)
name = getUniqueFileName( prefix + "_" + key, image.getSuffix() );
else
name = getUniqueFileName( prefix, image.getSuffix() );
System.out.println( "Writing image:" + name );
image.write2file( name );
// maybe there are more images embedded in a form object
else if (xobject instanceof PDXObjectForm)
PDXObjectForm xObjectForm = (PDXObjectForm)xobject;
PDResources formResources = xObjectForm.getResources();
processResources(formResources, prefix, addKey);
private String getUniqueFileName( String prefix, String suffix )
String uniqueName = null;
File f = null;
while( f == null || f.exists() )
uniqueName = prefix + "-" + imageCounter;
f = new File( uniqueName + "." + suffix );
imageCounter++;
return uniqueName;
/**
* This will print the usage requirements and exit.
*/
private static void usage()
System.err.println( "Usage: java org.apache.pdfbox.ExtractImages [OPTIONS] <PDF file>\n" +
" -password <password> Password to decrypt document\n" +
" -prefix <image-prefix> Image prefix(default to pdf name)\n" +
" -addkey add the internal image key to the file name\n" +
" -nonSeq Enables the new non-sequential parser\n" +
" <PDF file> The PDF document to use\n"
);
System.exit( 1 );
现在可以使用裁剪图像：
/**
* Crop the main image according to this rectangle, and scale it to the
* correct size for a thumbnail.
*/
public InputStream cropAndScale(InputStream mainImageStream,
CropRectangle crop)
try
RenderedOp mainImage = loadImage(mainImageStream);
RenderedOp opaqueImage = makeImageOpaque(mainImage);
RenderedOp croppedImage = cropImage(opaqueImage, crop);
RenderedOp scaledImage = scaleImage(croppedImage);
byte[] jpegBytes = encodeAsJpeg(scaledImage);
return new ByteArrayInputStream(jpegBytes);
catch (Exception e)
throw new IllegalStateException("Failed to scale the image", e);
在this page 和project 中可用
还有其他选项可以解析pdf文件中的图像，看看at this code特别是this
【讨论】：
不确定为什么选择此答案，因为它完全无法回答用户的问题。以上是关于从乐谱中提取每行的小节[关闭]的主要内容，如果未能解决你的问题，请参考以下文章