lsm/PdfPig
1
0
Fork 0
mirror of https://github.com/UglyToad/PdfPig.git synced 2025-09-19 19:07:56 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Improve default max function

Browse Source
This commit is contained in:
BobLd
2020-02-28 11:41:32 +00:00
committed by Eliot Jones
parent c864fa512c
commit 5ae38f1bad
1 changed files with 31 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

55
src/UglyToad.PdfPig.DocumentLayoutAnalysis/WordExtractor/NearestNeighbourWordExtractor.cs
View File

@@ -31,43 +31,52 @@
/// <param name="letters">The letters in the page.</param>
public IEnumerable<Word> GetWords(IReadOnlyList<Letter> letters)
{
Func<Letter, Letter, double> baseMaxFunc = (l1, l2) =>
{
return Math.Max(Math.Max(Math.Max(
Math.Abs(l1.GlyphRectangle.Width),
Math.Abs(l2.GlyphRectangle.Width)),
Math.Abs(l1.Width)),
Math.Abs(l2.Width));
};
List<Word> wordsH = GetWords(
letters.Where(l => l.TextDirection == TextDirection.Horizontal),
(l1, l2) => Math.Max(l1.GlyphRectangle.Width, l2.GlyphRectangle.Width) * 0.2,
letters.Where(l => l.TextDirection == TextDirection.Horizontal).ToList(),
(l1, l2) => baseMaxFunc(l1, l2) * 0.2,
Distances.Manhattan, MaxDegreeOfParallelism)
.OrderByDescending(x => x.BoundingBox.Bottom)
.ThenBy(x => x.BoundingBox.Left).ToList();
List<Word> words270 = GetWords(
letters.Where(l => l.TextDirection == TextDirection.Rotate270),
(l1, l2) => Math.Max(l1.GlyphRectangle.Width, l2.GlyphRectangle.Width) * 0.2,
var words270 = GetWords(
letters.Where(l => l.TextDirection == TextDirection.Rotate270).ToList(),
(l1, l2) => baseMaxFunc(l1, l2) * 0.2,
Distances.Manhattan, MaxDegreeOfParallelism)
.OrderBy(x => x.BoundingBox.Right)
.ThenByDescending(x => x.BoundingBox.Bottom).ToList();
.ThenByDescending(x => x.BoundingBox.Bottom);
wordsH.AddRange(words270);
List<Word> words180 = GetWords(
letters.Where(l => l.TextDirection == TextDirection.Rotate180),
(l1, l2) => Math.Max(Math.Abs(l1.GlyphRectangle.Width), Math.Abs(l2.GlyphRectangle.Width)) * 0.2,
var words180 = GetWords(
letters.Where(l => l.TextDirection == TextDirection.Rotate180).ToList(),
(l1, l2) => baseMaxFunc(l1, l2) * 0.2,
Distances.Manhattan, MaxDegreeOfParallelism)
.OrderBy(x => x.BoundingBox.Top)
.ThenByDescending(x => x.BoundingBox.Right).ToList();
.ThenByDescending(x => x.BoundingBox.Right);
wordsH.AddRange(words180);
List<Word> words90 = GetWords(
letters.Where(l => l.TextDirection == TextDirection.Rotate90),
(l1, l2) => Math.Max(Math.Abs(l1.GlyphRectangle.Width), Math.Abs(l2.GlyphRectangle.Width)) * 0.2,
var words90 = GetWords(
letters.Where(l => l.TextDirection == TextDirection.Rotate90).ToList(),
(l1, l2) => baseMaxFunc(l1, l2) * 0.2,
Distances.Manhattan, MaxDegreeOfParallelism)
.OrderByDescending(x => x.BoundingBox.Left)
.ThenBy(x => x.BoundingBox.Top).ToList();
.ThenBy(x => x.BoundingBox.Top);
wordsH.AddRange(words90);
List<Word> wordsU = GetWords(
letters.Where(l => l.TextDirection == TextDirection.Other),
(l1, l2) => Math.Max(Math.Abs(l1.GlyphRectangle.Width), Math.Abs(l2.GlyphRectangle.Width)) * 0.5,
var wordsU = GetWords(
letters.Where(l => l.TextDirection == TextDirection.Other).ToList(),
(l1, l2) => baseMaxFunc(l1, l2) * 0.5,
Distances.Euclidean, MaxDegreeOfParallelism)
.OrderByDescending(x => x.BoundingBox.Bottom)
.ThenBy(x => x.BoundingBox.Left).ToList();
.ThenBy(x => x.BoundingBox.Left);
wordsH.AddRange(wordsU);
return wordsH;
@@ -85,7 +94,7 @@
/// <param name="maxDegreeOfParallelism">Sets the maximum number of concurrent tasks enabled.
/// <para>A positive property value limits the number of concurrent operations to the set value.
/// If it is -1, there is no limit on the number of concurrently running operations.</para></param>
public List<Word> GetWords(IEnumerable<Letter> pageLetters,
public List<Word> GetWords(IReadOnlyList<Letter> pageLetters,
Func<Letter, Letter, double> maxDistanceFunction, Func<PdfPoint, PdfPoint, double> distMeasure,
int maxDegreeOfParallelism)
{
@@ -97,9 +106,7 @@
throw new ArgumentException("NearestNeighbourWordExtractor.GetWords(): Mixed Text Direction.");
}
Letter[] letters = pageLetters.ToArray();
var groupedIndexes = ClusteringAlgorithms.ClusterNearestNeighbours(letters,
var groupedIndexes = ClusteringAlgorithms.ClusterNearestNeighbours(pageLetters,
distMeasure, maxDistanceFunction,
l => l.EndBaseLine, l => l.StartBaseLine,
l => !string.IsNullOrWhiteSpace(l.Value),
@@ -107,9 +114,9 @@
maxDegreeOfParallelism).ToList();
List<Word> words = new List<Word>();
for (int a = 0; a < groupedIndexes.Count(); a++)
for (int a = 0; a < groupedIndexes.Count; a++)
{
words.Add(new Word(groupedIndexes[a].Select(i => letters[i]).ToList()));
words.Add(new Word(groupedIndexes[a].Select(i => pageLetters[i]).ToList()));
}
return words;