Guided filtering of images
filters the imageB
= imguidedfilter(___,Name,Value
)A
using name-value pairs to control aspects of guided filtering.
TheDegreeOfSmoothing
参数指定了一个软阈值方差for the given neighborhood. If a pixel's neighborhood has variance much lower than the threshold, it will see some amount of smoothing. If a pixel's neighborhood has variance much higher than the threshold it will have little to no smoothing.
Input imagesA
andG
can be of different classes. If eitherA
orG
is of class integer or logical, thenimguidedfilter
converts them to floating-point precision for internal computation.
Input imagesA
andG
can have different number of channels.
If bothA
andG
are RGB images, thenimguidedfilter
filters each channel ofA
independently using the corresponding channel ofG
.
IfA
is an RGB image andG
is a single-channel image, thenimguidedfilter
filters each channel ofA
independently using the same guidance image,G
.
IfA
is a single-channel image andG
is an RGB image, thenimguidedfilter
filtersA
using the combined color statistics of all the three channels ofG
.
[1] Kaiming He, Jian Sun, Xiaoou Tang.Guided Image Filtering. IEEE®Transactions on Pattern Analysis and Machine Intelligence, Volume 35, Issue 6, pp. 1397-1409, June 2013.