ABSTRACT
The bilateral filter is a nonlinear filter proposed by Tomasi and Manduchi [201] that keeps strong edges while smoothing internal values. The filter for an image I is defined as
I ′(x) = B(I, f, g) = 1
k(x)
∑ y∈Ω
I(y)f(‖x− y‖)g(‖I(y) − I(x)‖),
k(x) = ∑ y∈Ω
f(‖x− y‖)g(‖I(y) − I(x)‖), (A.1)
where I ′ is the filtered image, f is the smoothing function for the range term, g is the smoothing function for the intensity term, and k(x) is the normalization term. In the standard notation, parameters for f are indicated with s, and the ones for g are indicated with r. An example of the application of the filter is shown in Figure A.1.