error "matrix.cpp, line 322" or bug (Bug #2411)

Added by angel angel over 12 years ago. Updated over 12 years ago.

Status:Cancelled Start date:2012-10-04
Priority:Normal Due date:
Assignee:- % Done:

0%
Category:android
Target version:2.4.3
Affected version: Operating System:
Difficulty: HW Platform:
Pull request:

Description

Hi I am developping using opencv in Android. My program works correctly but sometimes I get this error:

Stitching module's ROI issue (Bug #2385) 10-04 16:03:18.989: E/cv::error()(1150): OpenCV Error: Assertion failed (0 <= roi.x && 0 <= roi.width && roi.x + roi.width <= m.cols && 0 <= roi.y && 0 <= roi.height && roi.y + roi.height <= m.rows) in cv::Mat::Mat(const cv::Mat&, const cv::Rect&), file /home/andreyk/OpenCV/opencv/modules/core/src/matrix.cpp, line 322
10-04 16:03:18.999: E/SurfaceTexture(1150): [unnamed-1150-1] dequeueBuffer: no available buffer slots
10-04 16:03:18.999: E/AndroidRuntime(1150): FATAL EXCEPTION: Thread-133
10-04 16:03:18.999: E/AndroidRuntime(1150): CvException [org.opencv.core.CvException: /home/andreyk/OpenCV/opencv/modules/core/src/matrix.cpp:322: error: (-215) 0 <= roi.x && 0 <= roi.width && roi.x + roi.width <= m.cols && 0 <= roi.y && 0 <= roi.height && roi.y + roi.height <= m.rows in function cv::Mat::Mat(const cv::Mat&, const cv::Rect&)

My error seems to be the same king of error than "Stitching module's ROI issue (Bug #2385)". I dont know exactly what part of my code is the origine of this error. I think it is related avec Imgproc.resize but I am not sure. Related to the bug 2385 it seems now it works. But I dont understand if it was a mistake of the code or this is an Opencv error. Anyway is there a known solution for this kind of error? Thanks of your answer

Associated revisions

Revision 7dc549f1
Added by Roman Donchenko about 11 years ago

Merge pull request #2411 from asmorkalov:android_mat_leak_fix

History

#1
Updated by Alexander Smorkalov over 12 years ago

It looks like your ROI is out of Mat size. Please attach some piece of code to reproduce the problem.

  • Category set to android
  • Target version set to 2.4.3

#2
Updated by angel angel over 12 years ago

Hi Alexander
I think the error could be in the next lines of code. My code works correctly during a time. I am not sure if dimlado du code has to be always bigger than 160... etc. I have to debug deeply.

*Mat matrixSelec= matrixRgba.submat(0, dimlado,0, dimlado );//A sub matrix: dimlado is changing
Mat matrixSelec_matrixRgba = matrixRgba.submat(0, 160,0, 160 );
            Imgproc.resize(matrixSelec, matrixSelec_mRgba, matrixSelec_matrixRgba.size());//New size: to 160x160
Mat RmatrixSeleccionada = matrixRgba.submat(0, dimlado,0, dimlado );
            Mat CRmatrixSeleccionada= new Mat(RmatrixSeleccionada.size(), RmatrixSeleccionada.type());
            RMatrixSeleccionada.copyTo(CRMatrixSeleccionada);*__
There is a big matrix. I take a submatrix matrixSelec. Then I reside matrixSelec to a matrix 160x160. And then I copy the result to CRMatrixSeleccionada.

regards

#3
Updated by angel angel over 12 years ago

a mistake typing:

Mat RmatrixSeleccionada = matrixRgba.submat(0, 160,0, 160 );
Mat CRmatrixSeleccionada= new Mat(RmatrixSeleccionada.size(), RmatrixSeleccionada.type());
RMatrixSeleccionada.copyTo(CRMatrixSeleccionada);*

#4
Updated by angel angel over 12 years ago

I think I have found the problemm... it was before the code I have shown... I use also a submat function and you are right the size of the submat sometimes is out the big matrix... and consecuently I have a Mat size error... Anyway, taking advantage of the code above,do you think the code to resize the submatrix could be done in a better way? Thanks for your support...

#5
Updated by Vadim Pisarevsky over 12 years ago

the code is fine, just make sure that ROI is within the matrix. The ticket is cancelled, since this is not a bug

  • Status changed from Open to Cancelled

Also available in: Atom PDF