c++ - OpenCV Dense feature detector -

i using opencv dense feature extraction. example, code

densefeaturedetector detector(12.f, 1, 0.1f, 10); 

i don't understand parameters in above constructor. mean ? reading opencv documentation not either. in documentation arguments are:

densefeaturedetector( float initfeaturescale=1.f, int featurescalelevels=1,                           float featurescalemul=0.1f,                           int initxystep=6, int initimgbound=0,                           bool varyxystepwithscale=true,                           bool varyimgboundwithscale=false ); 

what supposed ? i.e. meaning of scale, initfeaturescale, featurescalelevels etc ? how know grid or grid spacing etc dense sampling.

i'm using opencv dense detector , think can something. i'm not sure i'm going experience learnt me that.

when use dense detector pass there gray scale image. detector makes threshold filters opencv uses gray minimum value used transform image. píxels have more gray level threshold made black points , others white point. action repeated in loop threshold bigger , bigger. parameter initfeaturescale determine first threshold put loop, featurescalelevels parameter indicates how threshold bigger between 1 loop iteration , next 1 , featurescalemul multiply factor calculate next threshold.

anyway if looking optimal parameters use dense detector detect particular points offer program made that. liberated in github. program can test detectors (dense detector 1 of them) , check how works if change parameters user interface let change detectors parameters long executing program. see how detected points change. try click on link, , download files. might need files execute program.