lsm/PdfPig
1
0
Fork 0
mirror of https://github.com/UglyToad/PdfPig.git synced 2025-09-20 03:17:57 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Make clustering algos public and use shorter names

Browse Source
This commit is contained in:
BobLd
2020-04-23 11:32:05 +01:00
committed by Eliot Jones
parent 00b9d416df
commit d4210cd5d1
3 changed files with 9 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
src/UglyToad.PdfPig.DocumentLayoutAnalysis/ClusteringAlgorithms.cs → src/UglyToad.PdfPig.DocumentLayoutAnalysis/Clustering.cs
View File

@@ -9,7 +9,7 @@
/// <summary> /// <summary>
/// Clustering Algorithms. /// Clustering Algorithms.
/// </summary> /// </summary>
internal class ClusteringAlgorithms public static class Clustering
{ {
/// <summary> /// <summary>
/// Algorithm to group elements using nearest neighbours. /// Algorithm to group elements using nearest neighbours.
@@ -26,7 +26,7 @@
/// <param name="maxDegreeOfParallelism">Sets the maximum number of concurrent tasks enabled. /// <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. /// <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> /// If it is -1, there is no limit on the number of concurrently running operations.</para></param>
internal static IEnumerable<HashSet<int>> ClusterNearestNeighbours<T>(IReadOnlyList<T> elements, internal static IEnumerable<HashSet<int>> NearestNeighbours<T>(IReadOnlyList<T> elements,
Func<PdfPoint, PdfPoint, double> distMeasure, Func<PdfPoint, PdfPoint, double> distMeasure,
Func<T, T, double> maxDistanceFunction, Func<T, T, double> maxDistanceFunction,
Func<T, PdfPoint> pivotPoint, Func<T, PdfPoint> candidatesPoint, Func<T, PdfPoint> pivotPoint, Func<T, PdfPoint> candidatesPoint,
@@ -94,7 +94,7 @@
/// <param name="maxDegreeOfParallelism">Sets the maximum number of concurrent tasks enabled. /// <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. /// <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> /// If it is -1, there is no limit on the number of concurrently running operations.</para></param>
internal static IEnumerable<HashSet<int>> ClusterNearestNeighbours<T>(IReadOnlyList<T> elements, int k, internal static IEnumerable<HashSet<int>> NearestNeighbours<T>(IReadOnlyList<T> elements, int k,
Func<PdfPoint, PdfPoint, double> distMeasure, Func<PdfPoint, PdfPoint, double> distMeasure,
Func<T, T, double> maxDistanceFunction, Func<T, T, double> maxDistanceFunction,
Func<T, PdfPoint> pivotPoint, Func<T, PdfPoint> candidatesPoint, Func<T, PdfPoint> pivotPoint, Func<T, PdfPoint> candidatesPoint,
@@ -161,7 +161,7 @@
/// <param name="maxDegreeOfParallelism">Sets the maximum number of concurrent tasks enabled. /// <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. /// <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> /// If it is -1, there is no limit on the number of concurrently running operations.</para></param>
internal static IEnumerable<HashSet<int>> ClusterNearestNeighbours<T>(T[] elements, internal static IEnumerable<HashSet<int>> NearestNeighbours<T>(IReadOnlyList<T> elements,
Func<PdfLine, PdfLine, double> distMeasure, Func<PdfLine, PdfLine, double> distMeasure,
Func<T, T, double> maxDistanceFunction, Func<T, T, double> maxDistanceFunction,
Func<T, PdfLine> pivotLine, Func<T, PdfLine> candidatesLine, Func<T, PdfLine> pivotLine, Func<T, PdfLine> candidatesLine,
@@ -185,12 +185,12 @@
* (i,j,k) will form a group and (m,n) will form another group. * (i,j,k) will form a group and (m,n) will form another group.
*************************************************************************************/ *************************************************************************************/
int[] indexes = Enumerable.Repeat(-1, elements.Length).ToArray(); int[] indexes = Enumerable.Repeat(-1, elements.Count).ToArray();
ParallelOptions parallelOptions = new ParallelOptions() { MaxDegreeOfParallelism = maxDegreeOfParallelism }; ParallelOptions parallelOptions = new ParallelOptions() { MaxDegreeOfParallelism = maxDegreeOfParallelism };
// 1. Find nearest neighbours indexes // 1. Find nearest neighbours indexes
Parallel.For(0, elements.Length, parallelOptions, e => Parallel.For(0, elements.Count, parallelOptions, e =>
{ {
var pivot = elements[e]; var pivot = elements[e];

4
src/UglyToad.PdfPig.DocumentLayoutAnalysis/PageSegmenter/DocstrumBoundingBoxes.cs
View File

@@ -197,7 +197,7 @@
private static IEnumerable<TextLine> GetLines(List<Word> words, double maxDist, AngleBounds withinLine, int maxDegreeOfParallelism) private static IEnumerable<TextLine> GetLines(List<Word> words, double maxDist, AngleBounds withinLine, int maxDegreeOfParallelism)
{ {
TextDirection textDirection = words[0].TextDirection; TextDirection textDirection = words[0].TextDirection;
var groupedIndexes = ClusteringAlgorithms.ClusterNearestNeighbours(words, 2, Distances.Euclidean, var groupedIndexes = Clustering.NearestNeighbours(words, 2, Distances.Euclidean,
(pivot, candidate) => maxDist, (pivot, candidate) => maxDist,
pivot => pivot.BoundingBox.BottomRight, candidate => candidate.BoundingBox.BottomLeft, pivot => pivot.BoundingBox.BottomRight, candidate => candidate.BoundingBox.BottomLeft,
pivot => true, pivot => true,
@@ -246,7 +246,7 @@
new PdfPoint(left + d / 2, l2.Point1.Y)); new PdfPoint(left + d / 2, l2.Point1.Y));
} }
var groupedIndexes = ClusteringAlgorithms.ClusterNearestNeighbours(lines, var groupedIndexes = Clustering.NearestNeighbours(lines,
euclidianOverlappingMiddleDistance, euclidianOverlappingMiddleDistance,
(pivot, candidate) => maxDist, (pivot, candidate) => maxDist,
pivot => new PdfLine(pivot.BoundingBox.BottomLeft, pivot.BoundingBox.BottomRight), pivot => new PdfLine(pivot.BoundingBox.BottomLeft, pivot.BoundingBox.BottomRight),

2
src/UglyToad.PdfPig.DocumentLayoutAnalysis/WordExtractor/NearestNeighbourWordExtractor.cs
View File

@@ -106,7 +106,7 @@
throw new ArgumentException("NearestNeighbourWordExtractor.GetWords(): Mixed Text Direction."); throw new ArgumentException("NearestNeighbourWordExtractor.GetWords(): Mixed Text Direction.");
} }
var groupedIndexes = ClusteringAlgorithms.ClusterNearestNeighbours(pageLetters, var groupedIndexes = Clustering.NearestNeighbours(pageLetters,
distMeasure, maxDistanceFunction, distMeasure, maxDistanceFunction,
l => l.EndBaseLine, l => l.StartBaseLine, l => l.EndBaseLine, l => l.StartBaseLine,
l => !string.IsNullOrWhiteSpace(l.Value), l => !string.IsNullOrWhiteSpace(l.Value),