Ask Your Question

Revision history [back]

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*511

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*511511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Here a some working resolutions:

1713 x 1217 537 x 562 1089 x 563 527 x 563 1099 x 563 522 x 542

Not working resolutions:

511 x 532 1662 x 562 1661 x 562 1662 x 563 1099 x 563

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Here a some working resolutions:

1713 x 1217 1217; 537 x 562 562; 1089 x 563 563; 527 x 563 563; 1099 x 563 563; 522 x 542542;

Not working resolutions:

511 x 532 532; 1662 x 562 562; 1661 x 562 562; 1662 x 563 563; 1099 x 563563,

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Here a some working resolutions:

1713 x 1217; 537 x 562; 1089 x 563; 527 x 563; 1099 x 563; 522 x 542;

Not working resolutions:

511 x 532; 1662 x 562; 1661 x 562; 1662 x 563; 1099 1089 x 563,

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Here a some working resolutions:

1713 x 1217; 537 x 562; 1089 x 563; 527 x 563; 1099 x 563; 522 x 542;

Not working resolutions:

511 x 532; 1662 x 562; 1661 x 562; 1662 x 563; 1089 x 563,563;

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Here a some working resolutions:

1713 x 1217; 537 x 562; 1089 x 563; 527 x 563; 522 x 542;

Not working resolutions:

511 x 532; 1662 x 562; 1661 x 562; 1662 x 563;

Test results! Maybe somebody sees a logic behind it:

1659 x 562: works 1660 x 562: works 1661 x 562: Application crash 1662 x 562: Application crash 1663 x 562: Application crash 1664 x 562: works 1665 x 562: works 1666 x 562: works 1667 x 562: works 1668 x 562: works 1669 x 562: works

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Here a some working resolutions:

1713 x 1217; 537 x 562; 1089 x 563; 527 x 563; 522 x 542;

Not working resolutions:

511 x 532; 1662 x 562; 1661 x 562; 1662 x 563;

Test results! Maybe somebody sees a logic behind it:

1659 x 562: works works

1660 x 562: works works

1661 x 562: Application crash crash

1662 x 562: Application crash crash

1663 x 562: Application crash crash

1664 x 562: works works

1665 x 562: works works

1666 x 562: works works

1667 x 562: works works

1668 x 562: works works

1669 x 562: works

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Here a some working resolutions:

1713 x 1217; 537 x 562; 1089 x 563; 527 x 563; 522 x 542;

Not working resolutions:

511 x 532; 1662 x 562; 1661 x 562; 1662 x 563;

Test results! Maybe somebody sees a logic behind it:

1659 x 562: works

1660 x 562: works

1661 x 562: Application crash

1662 x 562: Application crash

1663 x 562: Application crash

1664 x 562: works

1665 x 562: works

1666 x 562: works

1667 x 562: works

1668 x 562: works

1669 x 562: works

I have a photo app where my users can crop images in different ratios like 4:3, 16:9 etc. Different ratios means different resulting image resolutions. The resolution is not only dependent by the ratio, but also by the zoom factor. The ratio is always the same, no matter how much the users zoomed in or out. The resolution is dependet on the zoom. The above resolutions represent the maximum resolution for the ratios I use for my app e.g. 1660x562 = 1:2.95 ratio. The cropped image with a ratio of 1:2.95 could have a resolution of 580x281(zoomed out). I then need to resize that image so I take the maxium supported size of 1660x562. I just adjusted the resolution until it did not crash. So instead of 1661x562 I took 1660x562 and it worked. I have 12 different resolutions/ratios for my app and 4 of them crashed. I just adjusted the resolution one pixel +/- until it worked. I still do not know what the problem is, but it works now.

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Here a some working resolutions:

1713 x 1217; 537 x 562; 1089 x 563; 527 x 563; 522 x 542;

Not working resolutions:

511 x 532; 1662 x 562; 1661 x 562; 1662 x 563;

Test results! Maybe somebody sees a logic behind it:

1659 x 562: works

1660 x 562: works

1661 x 562: Application crash

1662 x 562: Application crash

1663 x 562: Application crash

1664 x 562: works

1665 x 562: works

1666 x 562: works

1667 x 562: works

1668 x 562: works

1669 x 562: works

I have a photo app where my users can crop images in different ratios like 4:3, 16:9 etc. Different ratios means different resulting image resolutions. The resolution is not only dependent by the ratio, but also by the zoom factor. The ratio is always the same, no matter how much the users zoomed in or out. The resolution is dependet on the zoom. The above resolutions represent the maximum resolution for the ratios I use for my app e.g. 1660x562 = 1:2.95 ratio. The cropped image with a ratio of 1:2.95 could have a resolution of 580x281(zoomed out). in). I then need to resize that image so I take the maxium supported size of 1660x562. I just adjusted the resolution until it did not crash. So instead of 1661x562 I took 1660x562 and it worked. I have 12 different resolutions/ratios for my app and 4 of them crashed. I just adjusted the resolution one pixel +/- until it worked. I still do not know what the problem is, but it works now.

Thanks!

OpenCV on Android crash if resolution of image for laplace is too small

Hello!

I am working on a blur detection with OpenCV for Android and used following code found here which seemed to work correctly:

http://answers.opencv.org/question/111798/android-blur-image-detection-with-opencv-results-into-heavy-apk-size/

Now, my problem is that it crashs, if the resolution of the image is too small. In the sample code the transformed image gets resized to 2000*2000, but only if it is bigger than that.

It happens here:

Imgproc.Laplacian(matImageGrey, laplacianImage, CvType.CV_8U);

I have tested it serveral times and no crash with big images!

No crash image resolution: 1048 * 1048

Crash with resolution: 511*532

Does anybody now why this happens? It seems that there is a minimum image resolution size needed, but I can not find any info in the OpenCV documentation or in the internet.

UPDATE: Maybe someone can provide me a link to the C++ implementation of the laplacian funciton...can't find the code...should be open source?

Here a some working resolutions:

1713 x 1217; 537 x 562; 1089 x 563; 527 x 563; 522 x 542;

Not working resolutions:

511 x 532; 1662 x 562; 1661 x 562; 1662 x 563;

Test results! Maybe somebody sees a logic behind it:

1659 x 562: works

1660 x 562: works

1661 x 562: Application crash

1662 x 562: Application crash

1663 x 562: Application crash

1664 x 562: works

1665 x 562: works

1666 x 562: works

1667 x 562: works

1668 x 562: works

1669 x 562: works

I have a photo app where my users can crop images in different ratios like 4:3, 16:9 etc. Different ratios means different resulting image resolutions. The resolution is not only dependent by the ratio, but also by the zoom factor. The ratio is always the same, no matter how much the users zoomed in or out. The resolution is dependet on the zoom. The above resolutions represent the maximum resolution for the ratios I use for my app e.g. 1660x562 = 1:2.95 ratio. The cropped image with a ratio of 1:2.95 could have a resolution of 580x281(zoomed in). in) or a bigger one like 2320x1024. I then need to resize that image so I take the maxium supported size of 1660x562. I just adjusted the resolution until it did not crash. So instead of 1661x562 I took 1660x562 and it worked. I have 12 different resolutions/ratios for my app and 4 of them crashed. I just adjusted the resolution one pixel +/- until it worked. I still do not know what the problem is, but it works now.

Thanks!