Where can find the illustration for detectSingleScale function

asked 2015-09-02 07:25:51 -0500

learner gravatar image

updated 2015-09-02 09:39:41 -0500

Suppose there is image of 200x200, and the detecting window is 20x20, which is equal to trained sample. What is detecting procedure to detect the face in the image?

the features are determinded by harr in the lib, but why do we use integral image as feature matching in detecting? This function is in detectMultiScale,

for( double factor = 1;; factor * = scaleFactor)//detect under each size
{
    SizeoriginalWindowSize =getOriginalWindowSize();//minWndsize
  ...
    SizescaledImageSize( cvRound(grayImage.cols/factor ), cvRound ( grayImage.rows/factor ) );//change image size
    SizeprocessingRectSize(scaledImageSize.width -originalWindowSize.width + 1,scaledImageSize.height -originalWindowSize.height + 1 );//slide distance
    ...       

    if( !**detectSingleScale**( scaledImage, stripCount, processingRectSize, stripSize,yStep,factor,candidates,rejectLevels,levelWeights,outputRejectLevels) )//invoke detectSingleScale
        break;
}
objects.resize(candidates.size())
edit retag flag offensive close merge delete

Comments

1

hi @learner, - it is probably not about eigenfaces(your tag), but about object-detection ?

also, there is no detectSingleScale() method in opencv's api, so what do you mean ?

can you try to read the docs again (or even the paper mentionend there) ?

berak gravatar imageberak ( 2015-09-02 07:38:38 -0500 )edit

I find no suitable tag except this one. well, object detection maybe more suitable. detectSingleScale is invoked by detectMultiScale function.

learner gravatar imagelearner ( 2015-09-02 08:26:19 -0500 )edit

You should start reading the original paper to even get the slightest idea on why you should use integral images. It has everything to do with efficient calculations.

StevenPuttemans gravatar imageStevenPuttemans ( 2015-09-02 08:39:13 -0500 )edit
1

@StevenPuttemans, To get the feature value, but I don;t know what relationship betweent it and harr features? harr feature value is got from sum(white-black), but integral image is only a distributon funcon of gray

learner gravatar imagelearner ( 2015-09-02 08:57:10 -0500 )edit
1

I add a picture, but I was told it has to wait until tomorrow

learner gravatar imagelearner ( 2015-09-02 09:32:52 -0500 )edit
2

Again read the paper! It explains what the relation between integral image and HAAR wavelet is exactly.

StevenPuttemans gravatar imageStevenPuttemans ( 2015-09-02 10:06:11 -0500 )edit
learner gravatar imagelearner ( 2015-09-02 22:02:56 -0500 )edit