An Overview of FPGA in Image Processing

BMW

FPGA in image processing is ideal for applications that rely on low latency between acquisition of the image and processing the image.

Image processing involves operations on pixel, lines, and regions of interest. Image processing functions include the following features:

Preprocessing, where FPGA performs bit-level processing:
Image transforms
Image operators
Shading correction
Color space conversion
Filtering (smooth/sharpen)

Feature extraction:
Edge, line and corners
Binary objects
Color

Measurements:
Centroid
Area measurement

One of the benefits of relying on FPGA for image processing is the hardware enables low latencies. The FPGA is used to process the image and CPU intervention is not necessary. That means the system will not slow down.

Image processing becomes more computer intensive at a higher resolution, but with an FPGA, this process can be accelerated.

Intel’s Stratix 10 SoCs and Xilinx’s Zynq-7000 line are ideal FPGA for image processing applications. The high-end FPGA are specifically designed for data-intensive applications.

Applications:

1) Land boundary detection in an ADAS system
Automatically detecting a lane or a boundary while the vehicle is in motion requires a hardware system that can process information almost instantaneously.

2) Thermal Imaging in infrared cameras
Thermal imaging infrared cameras are used in a number of commercial settings. Their applications include security, firefighting, gas leak detection, and test and measurement.

3) Corner detection in a surveillance system
This is used to extract certain information from a video stream. Applications include motion detection, image registration, and video tracking. Ideal for applications in a security-related setting.
Other applications include high-speed sorting, eye-tracking, and laser alignment.

Once an image has been processed with an FPGA, the CPU can step in for further image analysis such as pattern matching and classification. This is defined as co-processing. In some cases, the FPGA can be relied onto process the image from start to finish, but in other cases, the CPU may be called into share the burden. This approach is still preferable to relying solely on the CPU since the burden placed on the CPU is minimal.

Luxe Electronics is an independent distributor of electronic components based in Amesbury, MA. Luxe is always open to new RFQs and excess opportunities. Contact us to learn more.

click for rfq

Sources:

http://www.ni.com/white-paper/11417/en/
https://www.mathworks.com/discovery/fpga-image-processing.html

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s