[
https://issues.apache.org/jira/browse/PDFBOX-4126?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16375700#comment-16375700
]
Alexandre edited comment on PDFBOX-4126 at 2/24/18 6:06 PM:
------------------------------------------------------------
More generally...
{code:java}
import org.apache.mahout.math.QRDecomposition;
import org.apache.mahout.math.DenseMatrix;
import org.apache.mahout.math.*;
import org.apache.pdfbox.text.TextPosition;
public class TextPositionAdvanced {
TextPosition _textPosition;
DenseMatrix _denseMatrix;
QRDecomposition _qr;
public TextPositionAdvanced(TextPosition textPosition) {
_textPosition = textPosition;
org.apache.pdfbox.util.Matrix matrix =
_textPosition.getTextMatrix();
_denseMatrix = new DenseMatrix(3,3);
_denseMatrix.set(0, 0, matrix.getScaleX()); // a
_denseMatrix.set(0, 1, matrix.getShearY()); // b
_denseMatrix.set(0, 2, 0);
_denseMatrix.set(1, 0, matrix.getShearX()); // c
_denseMatrix.set(1, 1, matrix.getScaleY()); // d
_denseMatrix.set(1, 2, 0);
_denseMatrix.set(2, 0, matrix.getTranslateX()); // tx
_denseMatrix.set(2, 1, matrix.getTranslateY()); // ty
_denseMatrix.set(2, 2, 1);
_qr = new QRDecomposition(_denseMatrix.transpose());
}
public String toString() {
return _textPosition.toString() + "angle: " + this.getAngle();
}
public double getAngle() {
return this.getEulerAngles(_qr.getQ());
}
public double getEulerAngles(Matrix R) {
return Math.toDegrees(Math.atan2(-R.get(0, 1), R.get(0, 0)));
}
}
{code}
was (Author: arelaxend):
More generally...
{code:java}
import org.apache.mahout.math.QRDecomposition;
import org.apache.mahout.math.DenseMatrix;
import org.apache.mahout.math.*;
import org.apache.pdfbox.text.TextPosition;
public class TextPositionAdvanced {
TextPosition _textPosition;
DenseMatrix _denseMatrix;
QRDecomposition _qr;
public TextPositionAdvanced(TextPosition textPosition) {
_textPosition = textPosition;
org.apache.pdfbox.util.Matrix matrix = _textPosition.getTextMatrix();
_denseMatrix = new DenseMatrix(3,3);
_denseMatrix.set(0, 0, matrix.getScaleX()); // a
_denseMatrix.set(0, 1, matrix.getShearY()); // b
_denseMatrix.set(0, 2, 0);
_denseMatrix.set(1, 0, matrix.getShearX()); // c
_denseMatrix.set(1, 1, matrix.getScaleY()); // d
_denseMatrix.set(1, 2, 0);
_denseMatrix.set(2, 0, matrix.getTranslateX()); // tx
_denseMatrix.set(2, 1, matrix.getTranslateY()); // ty
_denseMatrix.set(2, 2, 1);
_qr = new QRDecomposition(_denseMatrix.transpose());
}
public String toString() {
return _textPosition.toString() + "angle: " + this.getAngle();
}
public double getAngle() {
return this.getEulerAngles(_qr.getQ());
}
public double getEulerAngles(Matrix R) {
return Math.toDegrees(Math.atan2(-R.get(0, 1), R.get(0, 0)));
}
}
{code}
> TextPosition exact angle position
> ---------------------------------
>
> Key: PDFBOX-4126
> URL: https://issues.apache.org/jira/browse/PDFBOX-4126
> Project: PDFBox
> Issue Type: Improvement
> Components: Text extraction
> Affects Versions: 2.0.8
> Reporter: Alexandre
> Priority: Minor
> Attachments: test_rotation.pdf
>
>
> Dear Apache contributors,
> Is there an efficient way to get the exact angle in radians or degrees of a
> TextPosition, not only the ranges (0, 90, 180, 270). Basically, what we have
> currently uses the following information.
>
> {code:java}
> float a = textMatrix.getScaleY();
> float b = textMatrix.getShearY();
> float c = textMatrix.getShearX();
> float d = textMatrix.getScaleX();
> {code}
> Hagd, A.
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
