HDL-Optimized Algorithm Design

Select blocks or System objects for streaming video processing

The Vision HDL Toolbox™ blocks and System objects in this category implement hardware-friendly architectures and support HDL code generation when used with HDL Coder™.

Blocks

expand all

Birds-Eye ViewTransform front-facing camera image into top-down view
Chroma ResamplerDownsample or upsample chrominance component
Color Space ConverterConvert color information between color spaces
Demosaic InterpolatorConstruct RGB pixel data from Bayer pattern pixels
Gamma CorrectorApply or remove gamma correction of pixel stream
Lookup TableMap input pixel to output pixel using custom rule
Bilateral Filter2-D bilateral filtering
Edge DetectorFind edges of objects in grayscale pixel stream
Image Filter2-D FIR filtering
Median Filter2-D median filtering
Corner DetectorFind corners using FAST algorithm
ClosingMorphological closing of binary pixel data
DilationMorphological dilation of binary pixel data
ErosionMorphological erosion of binary pixel data
OpeningMorphological opening of binary pixel data
Grayscale ClosingMorphological closing of grayscale pixel data
Grayscale DilationMorphological dilation of grayscale pixel data
Grayscale ErosionMorphological erosion of grayscale pixel data
Grayscale OpeningMorphological opening of grayscale pixel data
HistogramFrequency distribution of pixel values in video stream
Image StatisticsCalculate mean, variance, and standard deviation
Line BufferStore video lines and return neighborhood pixels
Pixel Stream AlignerAlign two streams of pixel data
ROI SelectorSelect a region of interest (ROI) from pixel stream
HV CounterCount active dimensions of pixel stream

Objects

expand all

visionhdl.BirdsEyeViewTransform front-facing camera image into top-down view
visionhdl.ChromaResamplerDownsample or upsample chrominance component
visionhdl.ColorSpaceConverterConvert color information between color spaces
visionhdl.CornerDetectorFinds corners using FAST algorithm
visionhdl.DemosaicInterpolatorConstruct full RGB pixel data from Bayer pattern pixels
visionhdl.GammaCorrectorApply or remove gamma correction
visionhdl.EdgeDetectorFind edges of objects
visionhdl.BilateralFilterPerform 2-D filtering of a pixel stream
visionhdl.ImageFilter2-D FIR filtering
visionhdl.MedianFilter2-D median filtering
visionhdl.ClosingMorphological closing of binary pixel data
visionhdl.DilationMorphological dilation of binary pixel data
visionhdl.ErosionMorphological erosion of binary pixel data
visionhdl.OpeningMorphological opening of binary pixel data
visionhdl.GrayscaleClosingMorphological closing of grayscale pixel data
visionhdl.GrayscaleDilationMorphological dilation of grayscale pixel data
visionhdl.GrayscaleErosionMorphological erosion of grayscale pixel data
visionhdl.GrayscaleOpeningMorphological opening of grayscale pixel data
visionhdl.HistogramFrequency distribution
visionhdl.ImageStatisticsMean, variance, and standard deviation
visionhdl.LookupTableMap input pixel to output pixel using custom rule
visionhdl.LineBufferStore video lines and return neighborhood pixels
visionhdl.ROISelectorSelect region of interest (ROI) from pixel stream
visionhdl.PixelStreamAlignerAlign two streams of pixel data
visionhdl.HVCounterCounts active pixel dimensions of streaming video

Topics

Video Modeling

Edge Padding

Padding algorithms used to filter edge pixels of a frame.

Increase Throughput with Padding None

This example shows how to reduce latency and save hardware resources by not adding padding pixels at the edge of each frame.

Configure the Simulink Environment for HDL Video Processing

Set up a Simulink® model for HDL image and video processing.

Modeling External Memory

You can model external memory using features from Computer Vision Toolbox™ Support Package for Xilinx® Zynq®-Based Hardware or SoC Blockset™.

In MATLAB

Accelerate a MATLAB Design With MATLAB Coder

Compile MATLAB® designs to C code for faster simulation.

Pixel-Streaming Design in MATLAB

This example demonstrates a workflow for designing pixel-stream video processing algorithms using Vision HDL Toolbox™ in the MATLAB® environment and generating HDL code from the design.

Enhanced Edge Detection from Noisy Color Video

This example demonstrates how to develop a complex pixel-stream video processing algorithm, accelerate its simulation using MATLAB Coder™, and generate HDL code from the design.

Featured Examples

Lane Detection

Lane Detection

FPGA acceleration of lane marking detection.

Open Script
Pothole Detection

Pothole Detection

Extends the cartooning example to include calculating a centroid and overlaying a centroid marker and text label on detected potholes.

Open Script
Image Pyramid for FPGA

Image Pyramid for FPGA

Generate multi-level image pyramid pixel streams from an input stream. This model derives multiple pixel streams by downsampling the original image in both the horizontal and vertical directions, using Gaussian filtering. This type of filter avoids aliasing artifacts.

Open Script
Image Resize

Image Resize

Downsample an image by using the bilinear, bicubic, or Lanczos-2 interpolation algorithm.

Open Model
FPGA Implementation of Stereo Disparity using Semi-Global Block Matching

FPGA Implementation of Stereo Disparity using Semi-Global Block Matching

Compute disparity between left and right stereo camera images using the Semi-Global Block Matching algorithm. This algorithm is suitable for implementation on an FPGA.

Open Model
Stereo Image Rectification

Stereo Image Rectification

Implement stereo image rectification for a calibrated stereo camera pair. The example model is FPGA-hardware compatible and provides real-time performance. This example compares its results with the Computer Vision Toolbox™ rectifyStereoImages function.

Open Model
Fog Rectification

Fog Rectification

Remove fog from images captured under foggy conditions and implement the algorithm for FPGAs.

Open Model
FPGA Implementation of Low Light Enhancement

FPGA Implementation of Low Light Enhancement

Enhance low-light images on an FPGA.

Open Model
FPGA Implementation of Contrast Limited Adaptive Histogram Equalization

FPGA Implementation of Contrast Limited Adaptive Histogram Equalization

Implement a contrast-limited adaptive histogram equalization (CLAHE) algorithm using Simulink blocks. It uses the adapthisteq function from the Image Processing Toolbox™ as reference. The example model is FPGA-hardware compatible, and can provide real-time performance.

Open Model
FAST Corner Detection

FAST Corner Detection

Perform corner detection using the features-from-accelerated-segment test (FAST) algorithm. This algorithm is an alternative to the Harris Corner Detection example. The FAST algorithm determines if a corner is present by testing a circular area around the potential center of the corner. The test detects a corner if a contiguous section of pixels are either brighter than the center plus a threshold or darker than the center minus a threshold.

Open Script
Harris Corner Detection

Harris Corner Detection

Use edge detection as the first step in corner detection.

Open Model
Generate Cartoon Images Using Bilateral Filtering

Generate Cartoon Images Using Bilateral Filtering

Generate cartoon lines and overlay them onto an image.

Open Script
Histogram Equalization

Histogram Equalization

Use the Vision HDL Toolbox Histogram library block to implement histogram equalization.

Open Model
Using the Line Buffer to Create Efficient Separable Filters

Using the Line Buffer to Create Efficient Separable Filters

Design and implement a separable image filter, which uses fewer hardware resources than a traditional 2-D image filter.

Open Script
Multi-Zone Metering

Multi-Zone Metering

Use the Image Statistics block to perform multi-zone metering to extract a region of interest (ROI).

Open Model
Edge Detection and Image Overlay

Edge Detection and Image Overlay

Demonstrates how to detect and highlight object edges in a video stream. The behavior of the pixel-stream Sobel Edge Detector, video stream alignment, and overlay, is verified by comparing the results with the same algorithm calculated by the full-frame blocks from the Computer Vision Toolbox™.

Open Model
Noise Removal and Image Sharpening

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
Gamma Correction

Gamma Correction

Model pixel-streaming gamma correction for hardware designs. The model compares the results from the Vision HDL Toolbox™ Gamma Corrector block with the results generated by the full-frame Gamma block from Computer Vision System Toolbox™.

Open Model
Vision HDL Toolbox Documentation
Support