ROI Selector

Select a region of interest (ROI) from pixel stream

Libraries:
Vision HDL Toolbox / Utilities

Description

The ROI Selector block selects a portion of the active frame from a video stream. The size of the inactive frame remains the same. The output control signals indicate a new active region of the frame. You can use this block with a single pixel, multipixel, or multipixel-multicomponent pixel stream. This diagram shows the inactive pixel regions in blue and the requested output region outlined in orange.

Specified region of the input frame returned as a pixel stream with pixels outside of the region marked as inactive.

You can specify a fixed size and location for the ROI, or you can select the region location dynamically by using an input port. You can select more than one region. Define each region by specifying its upper-left corner coordinates and dimensions. By default, the block returns one set of pixels and control signals for each region you specify. The block sets the inactive pixels in the output frame to zero.

Regions are independent from each other, so they can overlap. If you specify a region that includes the edge of the active frame, the block returns only the active portion of the region. This diagram shows the output frames for three requested regions. The second output region does not include the inactive area above the image.

Three regions (one partially falling outside of the active frame, and one overlapping another), returned as three pixel streams.

The block also provides a mode for vertical reuse. In this mode, you must specify regions that have no vertical overlap and are aligned in columns. Each column of regions shares one output pixel stream. The control signals define each region in the stream. This arrangement enables parallel processing of each column, and the reuse of downstream processing logic for each region in a column. To use this mode, select the Reuse output ports for vertically aligned regions parameter. Use the visionhdlframetoregions function to divide a frame into tiled regions for vertical reuse.

Eight tiled regions returned as two output pixel streams that each contain four vertically-aligned regions.

Examples

Select Region of Interest

Select a region of active frame from a video stream by using the ROI Selector block from the Vision HDL Toolbox™.

Open Model

Select Regions for Vertical Reuse

Divide a frame into tiled regions of interest (ROIs) and use those regions to configure the ROI Selector block for vertical reuse.

Open Script

Ports

This image shows the additional ports on the block when you configure three regions of interest from input ports.

ROI Selector block with optional 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. The pixel ports 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 of M pixels per clock cycle to support high-frame-rate or high-resolution formats. The M value corresponds to the Number of pixels parameter of the Frame To Pixels block. Along with the pixel, the block accepts and returns a pixelcontrol bus 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 the Frame To Pixels block. For a full description of the interface, see Streaming Pixel Interface.

This block also supports multipixel-multicomponent streams, where the pixel input is a matrix of M-by-N values. M is number of pixels and N is number of components. These values correspond to the Number of pixels and Number of components parameters of the Frame To Pixels block.

Input

expand all

pixel — Input pixel stream
scalar | vector | matrix

For single pixel streams, specify pixel as a scalar or a vector of 1-by-Number of components values. For multipixel streams, specify pixel as a matrix of Number of pixels-by-Number of components pixel intensity values. Number of pixels can be two, four, or eight.

The software supports double and single data types for simulation, but not for HDL code generation.

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

The pixelcontrol bus contains five signals. The signals describe the validity of the pixel and its location in the frame. For more information, see Pixel Control Bus.

For multipixel streaming, each vector of pixel values has one set of control signals. Because the vector has only one valid signal, the pixels in the vector must be either all valid or all invalid. The hStart and vStart signals apply to the pixel with the lowest index in the vector. The hEnd and vEnd signals apply to the pixel with the highest index in the vector.

Data Types: bus

region1,...,regionR — Region of interest
four-element row vector

Region of interest, specified as a row vector of four positive integers that define the coordinates of the top-left corner and dimensions of each desired output frame. The vector must have the form [hPos vPos hSize vSize]. The block has R region ports, where R is the Number of regions parameter value.

For multipixel streaming the regions must meet these conditions, where M is Number of pixels.

mod(hPos, M) = 1 — For example, if your pixel stream has four pixels per cycle, 129 is a valid value of hPos (mod(129,4) = 1), while 128 is invalid (mod(128,4) = 0).
mod(hSize, M) = 0 — For example, if your pixel stream has four pixels per cycle, 120 is a valid hSize (mod(120,4) = 0), while 118 is invalid (mod(118,4) = 2).

Dependencies

To enable this port, clear the Reuse output ports for vertically aligned regions parameter and set the Regions source parameter to Input port.

Data Types: int | uint

Output

expand all

pixel1,...,pixelR — Output pixel stream
scalar | vector | matrix

Output pixel, returned as a single pixel or multipixel stream. The output dimensions and data type match those of the input pixel port. By default, the block has R output pixel ports, where R is the Number of regions parameter, or the number of rows in the Regions parameter matrix.

When you select Reuse output ports for vertically aligned regions, the block has one output pixel port for each column of vertically-aligned regions.

ctrl1,...,ctrlR — Control signals
`pixelcontrol` bus

Control signals, returned as a pixelcontrol bus that contains five signals. By default, the block has R output ctrl ports, where R is the Number of regions parameter, or the number of rows in the Regions parameter matrix.

When you select Reuse output ports for vertically aligned regions, the block has one output ctrl port for each column of aligned regions.

Data Types: bus

Parameters

expand all

Reuse output ports for vertically aligned regions — Option to return one pixel stream for each column of regions
`off` (default) | `on`

When you select this parameter, you must specify regions that have no vertical overlap and that are aligned in columns. The block has one output pixel stream for each column of regions. This arrangement enables parallel processing of each column, and the reuse of downstream processing logic for each region in a column. Use the visionhdlframetoregions function to divide a frame into tiled regions for vertical reuse.

Dependencies

To enable this parameter, set the Regions source parameter to Property.

Regions source — Location of output region definitions
`Property` (default) | `Input port`

Specify the location of the output region definitions. You can set Regions source to one of these values:

Property — Specify the regions by using the Regions parameter.
Input port — Specify the regions by using input ports. Each input port corresponds to one region. The block samples the region input ports when vStart is true in the input control bus.

Dependencies

To enable this parameter, clear the Reuse output ports for vertically aligned regions parameter.

Regions — Rectangular ROIs to select from input frame
R-by-4 matrix

Specify rectangular ROIs to select from the input frame as an R-by-4 matrix.

R is the number of regions. The four elements that define each region are the top-left starting coordinates and the dimensions of the region and must be of the form [hPos vPos hSize vSize]. The coordinates count from the upper-left corner of the active frame, defined as [1,1]. hSize must be greater than 1.

For multipixel streaming the regions must meet these conditions, where M is Number of pixels.

mod(hPos, M) = 1 — For example, if your pixel stream has four pixels per cycle, 129 is a valid value of hPos (mod(129,4) = 1), while 128 is invalid (mod(128,4) = 0).
mod(hSize, M) = 0 — For example, if your pixel stream has four pixels per cycle, 120 is a valid hSize (mod(120,4) = 0), while 118 is invalid (mod(118,4) = 2).

When you clear the Reuse output ports for vertically aligned regions parameter, the regions can overlap, and R must not be greater than 16.

When you select Reuse output ports for vertically aligned regions, the regions in this matrix must have no vertical overlap and be aligned in columns. This diagram shows two examples of invalid regions outlined with a blue dashed line. The region on the left overlaps another region vertically. The two regions at the bottom do not align with the other regions in their columns. The regions do not have to cover the entire frame. Pixels outside of the tiled regions are marked as inactive pixels. The number of columns must not be greater than 16.

Dependencies

To enable this parameter, set the Regions source parameter to Property.

Number of regions — Number of region input ports
integer in the range [1, 16]

Specify the number of region input ports as an integer in the range [1, 16].

Dependencies

To enable this parameter, set the Regions source parameter to Input port.

Algorithms

expand all

The generated HDL code for the ROI Selector block uses two 32-bit counters. The block does not use additional counters for additional regions.

Latency

The block has a latency of three cycles.

Performance

This table shows the resource use after synthesis of the block for the Xilinx^® Zynq^®-7000 SoC ZC706 Evaluation Kit with single-pixel uint8 input and the default parameter settings. The design achieves a clock frequency of 357 MHz.

Resource	Usage
Slice LUTs	103
Slice Registers	96
DSP48	0
Block RAM	0

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 VHDL, Verilog and SystemVerilog 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 one 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, see ConstrainedOutputPipeline (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, see InputPipeline (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, see OutputPipeline (HDL Coder).

Version History

Introduced in R2016a

expand all

R2023a: Two pixels-per-clock streaming

The block now supports multipixel streams that have 2 pixels per clock cycle.

R2021a: Multipixel-multicomponent streaming

The block now supports streams that are both multipixel and multicomponent. The HDL implementation replicates the algorithm for each pixel and component in parallel.

The block supports input and output matrices of NumPixels-by-NumComponents pixels. The ctrl ports remain scalar, and the control signals in the pixelcontrol bus apply to all pixels in the matrix.

R2021a: Resource sharing

The block now provides an option to share hardware resources when you select vertically aligned regions. Regions in the same column share the same pixelcontrol bus output.

Select the Reuse output ports for vertically aligned regions checkbox, and provide a set of regions that are aligned in columns and do not overlap vertically within each column. You can specify up to 1024 regions per column. To divide a frame into tiled regions that are compatible with vertical reuse, use the visionhdlframetoregions function.

ROI Selector

Description