Learn more about Video and Image Processing Blockset

Median Filter - 2-D median filtering

Library

Filtering / Analysis & Enhancement

Description

The Median Filter block replaces the central value of an M-by-N neighborhood with its median value. If the neighborhood has a center element, the block places the median value there, as illustrated in the following figure.

The block has a bias toward the upper-left corner when the neighborhood does not have an exact center. See the median value placement in the following figure.

The block pads the edge of the input image, which sometimes causes the pixels within [M/2 N/2] of the edges to appear distorted. The median value is less sensitive than the mean to extreme values. As a result, the Median Filter block can remove salt-and-pepper noise from an image without significantly reducing the sharpness of the image.

Port	Input/Output	Supported Data Types	Complex Values Supported
I	Matrix of intensity values	Double-precision floating point Single-precision floating point Fixed point Boolean 8-, 16-, 32-bit signed integer 8-, 16-, 32-bit unsigned integer	No
Val	Scalar value that represents the constant pad value	Same as I port	No
Output	Matrix of intensity values	Same as I port	No

Port

Input/Output

Supported Data Types

Complex Values Supported

Matrix of intensity values

Double-precision floating point
Single-precision floating point
Fixed point
Boolean
8-, 16-, 32-bit signed integer
8-, 16-, 32-bit unsigned integer

Val

Scalar value that represents the constant pad value

Same as I port

Output

Matrix of intensity values

Same as I port

If the data type of the input signal is floating point, the output has the same data type. The data types of the signals input to the I and Val ports must be the same.

Fixed-Point Data Types

The information in this section is applicable only when the dimensions of the neighborhood are even.

For fixed-point inputs, you can specify accumulator and output data types as discussed in Dialog Box. Not all these fixed-point parameters apply to all types of fixed-point inputs. The following table shows the output and accumulator data type used for each fixed-point input.

Fixed-Point Input	Output Data Type	Accumulator Data Type
Even `M`	X	X
Odd `M`	X
Odd `M` and complex	X	X
Even `M` and complex	X	X

When M is even, fixed-point signals use the accumulator and output data types. The accumulator data type store the result of the sum performed while calculating the average of the two central rows of the input matrix. The output data type stores the total result of the average.

Complex fixed-point inputs use the accumulator parameters. The calculation for the sum of the squares of the real and imaginary parts of the input occur, before sorting input elements. The accumulator data type stores the result of the sum of the squares.

Dialog Box

The Main pane of the Median Filter dialog box appears as shown in the following figure.

Neighborhood size

Specify the size of the neighborhood over which the block computes the median.

Enter a scalar value that represents the number of rows and columns in a square matrix.
Enter a vector that represents the number of rows and columns in a rectangular matrix.

Output size

This parameter controls the size of the output matrix.

If you choose Same as input port I, the output has the same dimensions as the input to port I. The Padding options parameter appears in the dialog box. Use the Padding options parameter to specify how to pad the boundary of your input matrix.
If you select Valid, the block only computes the median where the neighborhood fits entirely within the input image, with no need for padding. The dimensions of the output image are, output rows = input rows - neighborhood rows + 1,
and
output columns = input columns - neighborhood columns + 1.

Padding options

Specify how to pad the boundary of your input matrix.

Select Constant to pad your matrix with a constant value. The Pad value source parameter appears in the dialog box
Select Replicate to pad your input matrix by repeating its border values.
Select Symmetric to pad your input matrix with its mirror image.
Select Circular to pad your input matrix using a circular repetition of its elements. This parameter appears if, for the Output size parameter, you select Same as input port I.

For more information on padding, see the Image Pad block reference page.

Pad value source

Use this parameter to specify how to define your constant boundary value.

Select Specify via dialog to enter your value in the block parameters dialog box. The Pad value parameter appears in the dialog box.
Select Input port to specify your constant value using the Val port. This parameter appears if, for the Padding options parameter, you select Constant.

Pad value

Enter the constant value with which to pad your matrix. This parameter appears if, for the Pad value source parameter, you select Specify via dialog. Tunable.

The Fixed-point pane of the Median Filter dialog box appears as follows. The parameters on this dialog box becomes visible only when the dimensions of the neighborhood are even.

Rounding mode

Select the rounding mode for fixed-point operations.

Overflow mode

Select the overflow mode for fixed-point operations.

Note Only certain cases require the use of the accumulator and output parameters. Refer to Fixed-Point Data Types for more information.

Accumulator

Use this parameter to specify the accumulator word and fraction lengths resulting from a complex-complex multiplication in the block:

When you select Same as input, these characteristics match the related input to the block.
When you select Binary point scaling, you can enter the word length and the fraction length of the accumulator, in bits.
When you select Slope and bias scaling, you can enter the word length, in bits, and the slope of the accumulator. This block requires power-of-two slope and a bias of 0.

Output

Choose how to specify the output word length and fraction length:

When you select Same as input, these characteristics match the related input to the block.
When you select Binary point scaling, you can enter the word length and the fraction length of the output, in bits.
When you select Slope and bias scaling, you can enter the word length, in bits, and the slope of the output. This block requires power-of-two slope and a bias of 0.

Lock scaling against changes by the autoscaling tool

Select this parameter to prevent the autoscaling tool in the Fixed-Point Tool from overriding any fixed-point scaling you specify in this block mask. For more information, see fxptdlg, a reference page on the Fixed-Point Tool in the Simulink documentation.

References

[1] Gonzales, Rafael C. and Richard E. Woods. Digital Image Processing. 2nd ed. Englewood Cliffs, NJ: Prentice-Hall, 2002.

2-D Convolution	Video and Image Processing Blockset
2-D FIR Filter	Video and Image Processing Blockset
`medfilt2`	Image Processing Toolbox