martijnvwezel's profile - activity

cvToCloud(const cv::Mat_<cv::Point3f>& points3d, pcl::PointCloud<PointT>& cloud, const cv::Mat& mask = cv::Mat())   
{   //cv to cloud
    cloud.clear();
    cv::Mat_<cv::Point3f>::const_iterator point_it = points3d.begin(), point_end = points3d.end();
    const bool has_mask = !mask.empty();
    cv::Mat_<uchar>::const_iterator mask_it;
    if (has_mask)
             mask_it = mask.begin<uchar>();
    cloud.reserve(cloud.width*cloud.height);
    for (; point_it != point_end; ++point_it, (has_mask ? ++mask_it : mask_it))
        {
          if (has_mask && !*mask_it)
            continue;
          cv::Point3f p = *point_it;
          if (p.x != p.x || p.y != p.y || p.z != p.z) //throw out NANs
            continue;
          PointT cp;
          cp.x = p.x;
          cp.y = p.y;
          cp.z = p.z;
          cloud.push_back(cp);
        }   
 }

Here is my code for cv to cloud, soon I will update this for the otherway conversie.

what is the type when you use point3DMat.depth() . I use the datatype CV_32FC1 instead of CV_16UC1 without errors. If you get a number I can check for you

How to use extern modules from OpenCV in c#. I want to use c#, but I don't know how to implement the extern modules. I tried to rebuild emgu, but no function call directed to a function from the extern modules.

1) http://stackoverflow.com/questions/25... 2,3) http://docs.opencv.org/doc/tutorials/... (I GOOGLE this...)

but is the tasks that the cpu needs to do. Old computer? put linux on it, because it's lighter etc. In windows you have normal one and embedded windows without fax and other stuff

The question is not understandable at all...

depend what you system is and how mutch cpu you are using, mostly linus is lighter

depend if the code is patented. Mostly yes, but it also depend what for kind code when you google for the functions you will find the licence. Most companies use OpenCV for commercial applications, but donate sometime a little.(what my company does)

'winStride' = argument (default 8x8)- controls how the window is slide across the input window'. 'padding' = adds a certain amount of extra pixels on each side of the input image. That way the detection window is placed a bit outside the input image. It's because of that padding that HOG can detect people who are very close to the edge of the input image. Setting it to (0,0) add no extra pixels on each side of the input image.

maybe you have something with this link try to understand what they do with other variables, but I think you will understand: http://scikit-image.org/docs/dev/auto...

or you use paint...

@berak error's with missing files, but i added the directories in cmake. Also missing cuda but I added the directoriers. OpenNI error with include folder. rgbd error with missing headers. If I start over a new project build sometime i don't get all those errors, but then in vc10 i get a lot of operators overload errors when i solve them all then he has problems with includes of headers

2^8 = 256 counting from 0 is 0-> 255 is in total 256, This is ==> a 8 bit memory need for each pixel, this is a commonly stored way of pixels. It could be more or less, but this depend what you want

I am busy for days with OpenCV 3.0.0 + external module + OpenNI + cuda, but I cannot make it work.

I tried it with vc10 + cmake and the git files.

I can somehow sometimes generate it, but when i try to compile it with visual studio and I have get the one error after the other error. I tried several tutorials, but nothing works for me. :(

I there someone build this already? (and willing to share) Or wants to skype with me?

CUDA 7.5 opencv 3.0.0 opencv extern modules (rgbd=>odometry I need) Win 8.1 OpenNI 2

solved: there was an error with the extern modules in opencv that is now solved! Solution: get the new opencv git version with git pull

If my program works at it would be slow if it takes more tan 250 ms. This is because the main program for section control is really big.

To detect a car trailer its very hard. There're a lot of car trailers that has the same license plate as the car itself. To detect if its a trailer I need to search in a area around the license plate. I already are making a detector for the license plate with Viola and Jones. Only to detect if there is a triangle you don't see very clearly on the image. The images coming from section control so a lot of different thresholds in the day time.

For the privacy I have to delete the license plate details

So my question are there special ways of image processing that would be help full. I thought about: - canny - adapting threshold - Image Gradients

But big difficulty is that the application needs to be real-time.

I know if I wanted to show you the picture with correct positives and negatives you have to wait 3,5 day. Can the negatives have other sizes. If I use the original negative images and add a gray rectangle on the number plate will this solving my problem?

I have data from the Dutch, Belgium and other countries section controls. I have data files with the correct position about the license plates. So I made a python program that crop out the correct license plate by a size of w 140 and h 40. To make the negatives I use the same program only add a random value (in a certain area). Yesterday I found out that in my data where 10 wrong sizes that I deleted. I checked if every image has the same size and there are now. After this I change all the files for training. I have 3309 positive pictures and 3297 negative files. I define –numNeg 3297. So I did not use createsamples utility for the rotations, translation and skewing deformations. My link to the screenshot of the training https://www.dropbox.com/s/6m0i5fzet3q...

I tried OpenCV in python stereo matching, but it's very slow. If it's important to make it faster try it with multi-threading. So divide your data calculate and add it together again.

@StevenPuttemans This is my question

I am trying to learn the cascade classifier to learn license plates. I'm running OpenCV 2.4.9 on a Windows 7 server with 16 GB RAM and i7 processor of 3,4 GHz. I want to train the classifier, but it never continuous stage 3.

This is how I made my .vec file on the xxxx there is personal information:

D:\xxxx\trainingBinairImages\build\x64\vc12\bin\opencv_createsamples.exe -info D:\xxxx\trainingBinairImages\positiveNew.dat -bg D:\xxxx\trainingBinairImages\NegativeNew.bg -vec D:\xxxx\trainingBinairImages\output_vec.vec -maxxangle 0 -maxyangle 0 -maxzangle 0 -w 140 -h 40 -num 3311

pause

Parameters:

D:\xxxx\trainingBinairImages\build\x64\vc12\bin\opencv_traincascade.exe -data D:\xxxx\trainingBinairImages\result -vec output_vec.vec -bg negativeNew.bg -numPos 2980 -numNeg 3311 -miniHitRate 0.995 -maxFalseAlarmRate 0.5 -w 140 -h 40 -featureType HAAR -precalcValBufSize 2048 -precalcIdxBufSize 2048
pause

From cmd:

 ===== TRAINING 3-stage =====
<BEGIN
POS count : consumed   2980 : 2981
Train dataset for temp stage can not be filled. Branch training terminated.

If needed to get a look in the files I will edit the extra's. I tried to change the positive and negative values the numStages. Change the directory paths etc. Online there are question about this, that I tried without a solved solutions.