Bookmark and Share

Image Color Conversion and Formatting Using FPGAs

Converting images from one color domain to another is a common and often misunderstood operation in embedded vision processing. Image formats and color spaces vary when information is interchanged among processing domains. FPGAs provide a flexible, fast and scalable way of converting images using readily available IP cores. This video tutorial from José Alvarez, Video Technology Engineering Director at Xilinx Corporation, provides a quick reference to the available IP to perform this task efficiently and conveniently.

jralvarez
Offline
Last seen: 3 weeks 2 days ago
Level 1: Prestidigitator
Joined: 2011-08-18
Points: 1

The best way is to make sure you are working in the YUV or YCrCb domain. If you would like to reduce chroma noise you can apply a low pass FIR filter to the Cr Cb components independently. There are more complex techniques, but this will get you started. In the suggestion above, you have to take into account the length of the filter (number of taps) to properly align with the Y signal (due to processing delay of the FIR).

pravindahale
Offline
Last seen: 3 years 10 weeks ago
Level 1: Prestidigitator
Joined: 2012-04-03
Points: 1

How to apply chrominance filtering onimage?

I am trying to implement in opencv.