Andrey Dmitriev
2008-02-03 16:40:07 UTC
Hi, Goju-Ryu,
Its not extremely complicated. All what you needed is to calling OpenCV functions. Theoretically you can call these functions directly from LabVIEW, but then you should figure out how to transfer structures from call to call. I prefer usually to create DLL wrapper, which will be called from LabVIEW, and inside of the wrapper you have to place OpenCV calls.
For example how to use Convolution (5x5 fixed kernel - gradient) from OpenCV:
We need to call cvFilter2D in this case.
I will create following code:
<img src="Loading Image...
">
In the code above we will send to DLL the pointers of the Src/Dst, LineWidth of each image and Width/Height.
This DLL will be called from LabVIEW like this:
<img src="Loading Image...
">
In the attachment you can found complete source code of the example.
Best regards,
Andrey
ConvolutionTester.zip:
http://forums.ni.com/attachments/ni/200/16713/1/ConvolutionTester.zip
Its not extremely complicated. All what you needed is to calling OpenCV functions. Theoretically you can call these functions directly from LabVIEW, but then you should figure out how to transfer structures from call to call. I prefer usually to create DLL wrapper, which will be called from LabVIEW, and inside of the wrapper you have to place OpenCV calls.
For example how to use Convolution (5x5 fixed kernel - gradient) from OpenCV:
We need to call cvFilter2D in this case.
I will create following code:
<img src="Loading Image...
">In the code above we will send to DLL the pointers of the Src/Dst, LineWidth of each image and Width/Height.
This DLL will be called from LabVIEW like this:
<img src="Loading Image...
">
In the attachment you can found complete source code of the example.
Best regards,
Andrey
ConvolutionTester.zip:
http://forums.ni.com/attachments/ni/200/16713/1/ConvolutionTester.zip
