Median Filter

2-D median filtering

expand all in page

Library:
Vision HDL Toolbox / Analysis & Enhancement
Vision HDL Toolbox / Filtering

Description

TheMedian Filterblock replaces each input pixel with the median value of a specified surroundingN-by-Nneighborhood. The median is less sensitive to extreme values than the mean. You can use this block to remove salt-and-pepper noise from an image without significantly reducing the sharpness of the image. You can specify the neighborhood size and padding values for edges of the input image.

Ports

This block uses a streaming pixel interface with a bus for frame control signals. This interface enables the block to operate independently of image size and format. Thepixelports on this block support single pixel streaming or multipixel streaming. Single pixel streaming accepts and returns a single pixel value each clock cycle. Multipixel streaming accepts and returns a vector ofMpixels per clock cycle to support high-frame-rate or high-resolution formats. TheMvalue corresponds to theNumber of pixelsparameter of theFrame To Pixelsblock. Along with the pixel, the block accepts and returns apixelcontrolbus that contains five control signals. The control signals indicate the validity of each pixel and their location in the frame. For multipixel streaming, one set of control signals applies to all pixels in the vector. To convert a frame (pixel matrix) into a serial pixel stream and control signals, use theFrame To Pixelsblock. For a full description of the interface, seeStreaming Pixel Interface.

Input

expand all

`pixel`— Input pixel stream
scalar | vector

This block supports single pixel streaming or multipixel streaming. For single pixel streaming, specify a single input pixel as a scalar intensity value. For multipixel streaming, specify a vector of four or eight pixel intensity values. For details of how to set up your model for multipixel streaming, seeFilter Multipixel Video Streams.

This block does not support multicomponent streaming. To process multicomponent streams, replicate the block for each component. Thepixelcontrolbus for all components is identical, so you can connect a single bus to multiple replicated blocks.

doubleandsingledata types are supported for simulation, but not for HDL code generation.

`ctrl`— Control signals associated with pixel stream
`pixelcontrol`bus

Thepixelcontrolbus contains five signals. The signals describe the validity of the pixel and its location in the frame. For more information, seePixel Control Bus.

For multipixel streaming, each vector of pixel values has one set of control signals. Because the vector has only onevalidsignal, the pixels in the vector must be either all valid or all invalid. ThehStartandvStartsignals apply to the pixel with the lowest index in the vector. ThehEndandvEndsignals apply to the pixel with the highest index in the vector.

Data Types:bus

Output

expand all

`pixel`— Output pixel stream
scalar | vector

This block supports single pixel streaming or multipixel streaming. When using single pixel streaming, the block returns a single pixel as a scalar intensity value. When using multipixel streaming, the block returns a vector of intensity values. This vector is the same size as the inputpixelvector. For details of how to set up your model for multipixel streaming, seeFilter Multipixel Video Streams.

The output pixel data type is the same as the data type of the input pixels.

doubleandsingledata types are supported for simulation, but not for HDL code generation.

`ctrl`— Control signals associated with pixel stream
`pixelcontrol`bus

Thepixelcontrolbus contains five signals. The signals describe the validity of the pixel and its location in the frame. For more information, seePixel Control Bus.

Data Types:bus

Parameters

expand all

`Neighborhood size`— Size of image region used to compute median
`3×3`(default) |`5×5`|`7×7`

Select the neighborhood size, in pixels, as3×3,5×5, or7×7.

You cannot use a neighborhood size of7×7when you use multipixel streaming and setPadding methodtoNone.

`Padding method`— Method for padding boundary of input image
`Symmetric`(default) |`Constant`|`Replicate`|`Reflection`|`None`

Select one of these methods for padding the boundary of the input image. For more information about these methods, seeEdge Padding.

Constant— Interpret pixels outside the image frame as having a constant value.
Replicate— Repeat the value of pixels at the edge of the image.
Symmetric— Set the value of the padding pixels to mirror the edge of the image.
Reflection——设置填充像素的值反映了round the pixel at the edge of the image.
None——排除填充逻辑。块不设置the pixels outside the image frame to any particular value. This option reduces the hardware resources used by the block and the blanking required between frames but affects the accuracy of the output pixels at the edges of the frame. To maintain pixel stream timing, the output frame is the same size as the input frame. However, to avoid using pixels calculated from undefined padding values, mask off theKernelSize/2 pixels around the edge of the frame for downstream operations. For details, seeIncrease Throughput with Padding None.

You cannot use a neighborhood size of7×7when you setPadding methodtoNone.

`Padding value`— Value used to pad boundary of input image
`0`(default) | integer

Specify an integer to pad the boundary of the input image. The block casts this value to the same data type as the input pixel.

Dependencies

To enable this parameter, set thePadding methodparameter toConstant.

`Line buffer size`— Size of line buffer
`2048`(default) | integer

Size of line memory buffer, specified as a positive integer. Choose a power of two that accommodates the number of active pixels in a horizontal line. If you specify a value that is not a power of two, the buffer uses the next largest power of two.

The block allocates (N— 1) -by-Line buffer sizememory locations to store the pixels used to compute the median value.Nis the dimension of the square region specified by theNeighborhood sizeparameter.

Model Examples

Noise Removal and Image Sharpening

Implement a front-end module of an image processing design. This front-end module removes noise and sharpens the image to provide a better initial condition for the subsequent processing.

Open Model

Filter Multipixel Video Streams

Design filters that operate on a multipixel input video stream. Use multipixel streaming to process high-resolution or high-frame-rate video with the same synthesized clock frequency as a single-pixel streaming interface. Multipixel streaming also improves simulation speed and throughput because fewer iterations are required to process each frame, while maintaining the hardware benefits of a streaming interface.

Open Model

Tips

When you use a block with an internal line buffer inside anEnabled Subsystem(Simulink), the enable signal pattern must maintain the timing of the pixel stream, including the minimum blanking intervals. If the enable pattern corrupts the timing of the pixel stream, you might see partial output frames, corrupted pixel stream control signals, or mismatches between Simulink^®and HDL simulation results. You may need to extend the blanking intervals to accommodate for cycles when the enable is low. For more information, seeConfigure Blanking Intervals.

Algorithms

expand all

When you use multipixel streaming, the block uses a single line memory and implements one filter kernel for each of theMinput pixels, in parallel. This increase in hardware resources is a trade off for increasing throughput compared to single-pixel streaming.

Latency

The latency of the block is the line buffer latency plus the latency of the kernel calculation. The line buffer latency includes edge padding by default. The latency of the padding operation depends on the size of the kernel. If edge padding is not necessary for your design, you can reduce the latency by setting thePadding methodparameter toNone. When you use this option, the block latency does not depend on your kernel size. To determine the exact latency for any configuration of the block, measure the number of time steps between the input and output control signals.

To find the median value of each neighborhood, the kernel logic compares the pixel values to each other. The latency of the filter kernel depends on the neighborhood size. The table shows the number of comparisons needed to find the median value of each size of neighborhood.

Neighborhood Size	Number of Comparisons Performed
`3`×`3`	11
`5`×`5`	75
`7`×`7`	230

Note

When you use edge padding, use a horizontal blanking interval greater than twice the kernel width. This interval lets the block finish processing one line before it starts processing the next one, including adding padding pixels before and after the active pixels in the line. Standard streaming video formats use a horizontal blanking interval of about 25% of the frame width. This interval is much larger than the filters applied to each frame. When you disable edge padding, the horizontal blanking interval must be at least 12 cycles and is independent of the kernel size. If you are using a custom video format, set the horizontal blanking interval by using theFrame To Pixelsblock parameters. The horizontal blanking interval is equal toTotal pixels per line–Active pixels per lineor, equivalently,Front porch+Back porch. For more information, seeConfigure Blanking Intervals.

Performance

For theMedian Filterblock, the comparison logic that finds the median pixel value of each neighborhood is typically the critical path that determines the achievable clock frequency after synthesis. As the size of a neighborhood increases, the length of this path increases.

Extended Capabilities

C/C++ Code Generation
Generate C and C++ code using Simulink® Coder™.

This block supports C/C++ code generation for Simulink accelerator and rapid accelerator modes and for DPI component generation.

HDL Code Generation
Generate Verilog and VHDL code for FPGA and ASIC designs using HDL Coder™.

HDL Coder™ provides additional configuration options that affect HDL implementation and synthesized logic.

HDL Architecture

This block has a single, default HDL architecture.

HDL Block Properties

ConstrainedOutputPipeline	Number of registers to place at the outputs by moving existing delays within your design. Distributed pipelining does not redistribute these registers. The default is`0`. For more details, seeConstrainedOutputPipeline(HDL Coder).
InputPipeline	Number of input pipeline stages to insert in the generated code. Distributed pipelining and constrained output pipelining can move these registers. The default is`0`. For more details, seeInputPipeline(HDL Coder).
OutputPipeline	Number of output pipeline stages to insert in the generated code. Distributed pipelining and constrained output pipelining can move these registers. The default is`0`. For more details, seeOutputPipeline(HDL Coder).

Restrictions

You cannot generate HDL code for this block if it is inside aResettable Synchronous Subsystem(HDL Coder).

Version History

Introduced in R2015a

Median Filter

Description

Ports

Input

pixel— Input pixel streamscalar | vector

ctrl— Control signals associated with pixel streampixelcontrolbus

Output

pixel— Output pixel streamscalar | vector

ctrl— Control signals associated with pixel streampixelcontrolbus

Parameters

Neighborhood size— Size of image region used to compute median3×3(default) |5×5|7×7

Padding method— Method for padding boundary of input imageSymmetric(default) |Constant|Replicate|Reflection|None

Padding value— Value used to pad boundary of input image0(default) | integer

Dependencies

Line buffer size— Size of line buffer2048(default) | integer

Model Examples

Noise Removal and Image Sharpening

Filter Multipixel Video Streams

Tips

Algorithms

Latency

Performance

Extended Capabilities

C/C++ Code GenerationGenerate C and C++ code using Simulink® Coder™.

HDL Code GenerationGenerate Verilog and VHDL code for FPGA and ASIC designs using HDL Coder™.

Version History

See Also

`pixel`— Input pixel stream
scalar | vector

`ctrl`— Control signals associated with pixel stream
`pixelcontrol`bus

`pixel`— Output pixel stream
scalar | vector

`ctrl`— Control signals associated with pixel stream
`pixelcontrol`bus

`Neighborhood size`— Size of image region used to compute median
`3×3`(default) |`5×5`|`7×7`

`Padding method`— Method for padding boundary of input image
`Symmetric`(default) |`Constant`|`Replicate`|`Reflection`|`None`

`Padding value`— Value used to pad boundary of input image
`0`(default) | integer

`Line buffer size`— Size of line buffer
`2048`(default) | integer

C/C++ Code Generation
Generate C and C++ code using Simulink® Coder™.

HDL Code Generation
Generate Verilog and VHDL code for FPGA and ASIC designs using HDL Coder™.