enxdtw's profile - activity

I think I know what's going on here. The smoothing operation (or the lightening operation) takes the darker pixels and l

cvtColor hls to rgb lightening image HI, I am trying to lighten an image by converting to HLS and then adjusting the li

Now I'm doing a blur to the Lightness image, I have modified the code accordingly. I also used my own HLS2RGB functions

cvtColor hls to rgb lightening image HI, I am trying to lighten an image by converting to HLS and then adjusting the li

cvtColor hls to rgb lightening image HI, I am trying to lighten an image by converting to HLS and then adjusting the li

Can you see the image link? Maybe the brightness effect is not so visible with +50, with +150 offset the effect is more

Can you see the image link?

Can you see the image?

cvtColor hls to rgb lightening image HI, I am trying to lighten an image by converting to HLS and then adjusting the li

cvtColor hls to rgb lightening image HI, I am trying to lighten an image by converting to HLS and then adjusting the li

cvtColor hls to rgb lightening image HI, I am trying to lighten an image by converting to HLS and then adjusting the li

cvtColor hls to rgb lightening image HI, I am trying to lighten an image by converting to HLS and then adjusting the li

Hi,

I have a camera capable of 41.5MP. When I try to save an image from the camera it gets created, only that the image is 0KB in size, there is no error caught. If I save a region from the image then the image is saved correctly (as in the code below). Is there some limit to image size above which the image cannot be saved? (I am using version 4.0.1 opencv on Windows 10 and VC++ 17)

cv::Mat im = grabCVImage();


int s = 1200;
roi.x = s;
roi.y = s;
roi.width = s;
roi.height = s;


try {
    imwrite("C:\\Users\\Watts\\Pictures\\alpha.png", im(roi));
}
catch (std::runtime_error& ex) {
    fprintf(stderr, "Exception converting image to PNG format: %s\n", ex.what());
    return 1;
}

The problem was windows MFC's parent document class overwriting the file!

Yes, strange, I can save a newly created image of the same dimensions OK. And when I call imshow("im", im); it shows

Yes, strange, I can save a newly created image of the same dimensions OK. And when I call imshow("im", im); it shows

But 2^30 or 1 Gigapixel is much larger than 41.5MP. It doesn't explain why the image does not save

opencv imwrite unable to save images >40MP Hi, I have a camera capable of 41.5MP. When I try to save an image from t

Sure, I did it myself, but with OpenCV there would be some optimisation, boundary handling, maybe GPU implementation, pl

The best way is to create a orientated band pass filter and filter the image with it. I made an online app where you can

The best way is to create a orientated band pass filter and filter the image with it. I made an online app where you can

cv::sepFilter2D in one dimension? Hi, Can sepFilter2D also just do a filter in X or Y but not in both. It would be usef

Seems my OpenCV 3 book had an error for the equation and the source code is fine.

Thanks for the link to the source. It seems to me that source code has some statements that really don't make sense, such as line 155 (the IF is redundant) and 156 (which doesn't sum the moments values), also the IF statement on line 167 is outside the loop so its also strange, I also don't get the logic. Anyway I asked a question on the github issues page.

I have been looking through the OpenCV documentation and I cant find any mention of the cv::matchShapes return value. There seem to be three methods cv::CONTOURS_MATCH_I1, cv::CONTOURS_MATCH_I2, cv::CONTOURS_MATCH_I3. They are a sum of distances so I guess a perfect match is 0, however what is the possible value range? maybe its not possible to give this, but cv::CONTOURS_MATCH_I3 seems to scale the moments at least. In either case the documentation should give information on at least the fact that a perfect match has a result of zero.

The thing is, if you want to look for matches better than a certain score, which score can you choose? IMHO the scores should be normalised somehow eg. 0-100 or 0-1.

(It seems that log moments are better for matching purposes)

You can try the drawMarker function (red cross, 10 width, 1 line width)

cv::drawMarker(myimage, cv::Point(x, y),  cv::Scalar(0, 0, 255), MARKER_CROSS, 10, 1);

seems the opencv_world prebuilt files are also usable (QT .pro example)

INCLUDEPATH += C:/opencv/build/include LIBS += -LC:/opencv/build/x64/vc14/lib

debug { LIBS += -lopencv_world310d

} release { LIBS += -lopencv_world310 }

add C:\opencv\build\x64\vc14\bin to the system path

Seems I had 2 problems with the CMAKE settings, first was that WITH_CUDA was selected (as default!), second was that to compile for 64 bits only change the CMAKE_*_LINKER_FLAGS to /machine:x64 dont mess with the WIN32 settings even though it looks like they are 32 bit, its legacy windows shit, setting them to WIN64 will also cause VS to fail to compile. Then you have to rebuild ALL followed by the INSTALL project, in both debug and release, add the path to where the DLL's are located, link and include the libraries. I also unchecked the WITH_WIN32UI, probably not a good idea since cvshow then doesn't work. Would be nice if doiing a 64 bit opencv build on windows is better documented

Just want to mention that when I try to build using MSVC++ 15 using the CMAKE generated solution file I get 319 errors! here a 3 examples

Severity Code Description Project File Line Suppression State Error C1083 Cannot open include file: 'pthread.h': No such file or directory opencv_core C:\OpenCV\modules\core\src\rand.cpp 58

Severity Code Description Project File Line Suppression State Error C2146 syntax error: missing ';' before identifier 'sysinfo' opencv_core C:\OpenCV\modules\core\src\parallel.cpp 523

Severity Code Description Project File Line Suppression State Error LNK1104 cannot open file '....\lib\Debug\opencv_core310d.lib' opencv_imgproc C:\opencv-msvc\modules\imgproc\LINK 1

Hi,

I had trouble compiling on windows for x64 so I thought I would try just using the binaries (using the following guide).

http://docs.opencv.org/3.0-last-rst/d...

I downloaded version 3.1.0 ran the setup then saw that there are directories

C:\OpenCV\sources\include

and

C:\OpenCV\build\x64\vc14\lib

The lib directory only includes one lib file (and the debug version)

opencv_world310.lib

I was expecting to use

-lopencv_shape310 -lopencv_core310 -lopencv_highgui310 -lopencv_imgproc310 -lopencv_imgcodecs310

So either the libraries have changed and there is only one file to link against or the distribution is incomplete or there are no pre-built binaries and you have to build them yourself. Either way the documentation is not correct.

Anyway so I tried to build using CMAKE everything goes ok, but everything is built for WIN32 which is not compatible with my QT version and is anyway not really the latest technology. I changed all the WIN32 flags in CMAKE to be WIN64, better but still compiles certain libraries as X86 (actually on the last attempt I got 319 errors - see comment below). Surely a simple install of the latest opencv version on the most popular OS should not be so difficult and poorly explained.

Thanks for any help

Cheers Dave