A blog narrative principle is as follows:
In describing the adaptive median filter to use the following symbols:
\ [Z_ {min}=S_ {x, y} \) the minimum gray value
Max \ (Z_ {}=S_ {x, y} \) the biggest grey value
\ (Z_ {med}=S_ \ {x, y}) of the grey value of the median
\ \ (Z_ {x, y}) said coordinates \ ((x, y) \) grey value of the
\ [S_ Max {}=S_ {x, y} \) allow maximum window size
The adaptive median filter has two processing, respectively is: remember A and B, A: A1=\ [Z_ {med} - Z_ {min} \) A2=\ [Z_ {med} - Z_ {Max} \) if A1 & gt; 0 and A2 & lt; 0, jump to B; Otherwise, the size of the window if increase the dimension of the window after \ (\ leq S_ {Max} \), then repeat A process, otherwise, the output \ (Z_ {med} \)

B: B1=\ (Z_ {x, y} - Z_ {min} \) B2=\ [Z_ {x, y} - Z_ {Max} \) if B1 & gt; 0 and B2 & lt; 0, the output \ (Z_ {x, y} \) or output \ (Z_ {med} \)

Adaptive median filtering principle description
Has A purpose is to determine the current window median \ (Z_ {med} \) whether noise, if \ (Z_ {min} & lt; Z_ {med} & lt; Z_ {Max} \), the median \ (Z_ {med} \) is not noise, then transferred to the process B test, the center of the current window pixels \ (Z_ {x, y} \) is a noise point, if \ (Z_ {min} & lt; Z_ {x, y} & lt; Z_ {Max} \], then \ (Z_ {x, y} \) is not a noise, the output filter \ (Z_ {x, y} \); If you do not meet the above conditions, can determine \ (Z_ {x, y} \) is the noise, this is the output value \ (Z_ {med} \) (in A already tell \ (Z_ {med} \) not noise),

If, in the process of A get \ (Z_ {med} \) does not comply with the conditions \ (Z_ {min} & lt; Z_ {med} & lt; Z_ {Max} \), can determine the median \ (Z_ {med} \) is a noise, and in this case, the need to increase the window size of the filter, in a larger range of looking for a non noise value, until you find the median of a non noise, jump to B; Or, the size of the window at the upper end of the then returned to find the median, quit,

is : if \ (Z_ {min} & lt; Z_ {med} & lt; Z_ {Max} \), the median \ (Z_ {med} \) is not noise, then transferred to the process B test, but,,, value must be between the maximum and minimum?? What is the meaning of this sentence?

CodePudding user response:

literal meaning

CodePudding user response:

Maximum minimum value is not adaptive?