WJS's profile - activity

Tetragramm, Your suggestion is elegant in its simplicity! I will use it. WJS

I’m using the following matrix comparison to see if two 3 channel 8 bit color images are identical:

Mat result = image1 != image2;

Your documentation states: The result of comparison is an 8-bit single channel mask whose elements are set to 255 (if the particular element or pair of elements satisfy the condition) or 0.

So, I was expecting the result matrix to be a single channel binary image. However, I find that result is actually a three channel matrix. If I split the result into three planes and count non-zero for each plane, and if each of the counts is zero then I know that the two images are identical. Do I misunderstand your documentation? Can someone clarify this for me?

Thank you, WJS

Are there any plans to support the VP8 video codec? I searched the OpenCV Forum and found only: this. Thank you, WJS.

Berak, I'm not quite clear on what these number mean. Can you explain a little more? Thanks, WJS.

I’m using the LBPHFaceRecognizer and I understand from reviewing bug 3514 that the correct Chi-Square distance should be CV_COMP_CHISQR_ALT. However, the only Chi-Square distance available in the opencv version 2.4.11 I’m using is CV_COMP_CHISQR. Has anyone compared recognition results for these two Chi-Square distance measures in a controlled way? What prediction errors can I expect if I don’t use the new Chi-Square distance? Finally, is there a way to specify which distance measure to use for the LBPHFaceRecognizer model? WJS

Stan, Thank you. It's never too late! WJS

Eduardo, your link to the code I used is good. I also used your files box.png and box_in_scene.png obtained from .../sources/samples/c from my 2.4.10 install. WJS

I have successfully built [debug build] and tested most of the tutorials prior to “6.9 Feature Matching with FLANN”. This one compiles ok but crashes with “unhandled exception” when executing the second extractor.compute(img_2, keypoints_2, descriptors_2); statement under step 2 of the demo code. The first extractor.compute executes ok. The debugger stack trace shows the problem is somewhere inside the opencv_features2d2410d.dll. I can provide more info through some screen shots if anyone is interested. Has anyone else had this problem and what is the fix? I took a fast look at the version 3.0 change list but didn’t see any current bug fixes directly related to this particular problem. I’m using Microsoft visual studio 2010 and my project is set up locally. I’m using the opencv pre-built debug libs from ..\x86\vc10\lib and the debug DLLs from ..\x86\vc10\bin. WJS

Hello, I'm currently using the opencv library version 2.4.10 and I'm a new user. Can anyone give me a C++ example for creating signatures [i.e., convert histograms into signatures] for the emd function? I have read through the tutorial and reference manuals that came with the package I have but I couldn't find any examples for this. I found an example for the old 1.0 version library but had no success trying to convert it over to the new version I have. Thank you in advance for your help, WJS