In multimedia applications of video transmission there is usually the need to account the receiver’s capabilities, for example for different color formats and video frequencies before initiating the transmission. Consequently, in situations where the content is not available in a color space supported by the receiver there is the need to convert that content to one color space that is supported.

The representation of color in digital systems is done using mathematical models called color spaces. Each color space defines a set of coordinates to be used as primaries in the reference CIE XYZ color space, which contains all visible colors. These primaries are usually red, green and blue (RGB) although their exact location varies. The other representable colors are derived from this primaries whose coordinates in the XYZ color space define the range of colors that each particular color space is capable of represent – its color gamut. Moreover, the use of different color spaces is also related with the capability of the display systems (televisions, computer monitors, projectors, printers…) and their technology (CRT, LCD, LED, plasma, OLED…) to display colors, providing different experiences to the viewers.

Besides the RGB model of representation of color, the luma/color-difference system (YCrCb or YCC) is also broadly used. Luma is a quantity related to the luminance of the picture which together with the color difference signals (Cr and Cb) generates a representation of color that can be derived from the RGB signals. YCrCb model takes advantage from the fact that the human vision is less sensitive to the color variations than to luma variations by providing chroma subsampling schemes. In this schemes, the number of chroma samples per luma sample is horizontally and/or vertically reduced allowing for a bandwidth reduction of the signal without a visible loss of image quality.

The development of image systems not only provided new color spaces for data encoding but also demanded the development of transmission and processing systems capable of supporting the increase of image resolutions from standard television to high definition (1920×1080) and ultra high definition television (4k and 8k). In addition, the introduction of 3D broadcasting formats and the necessity of keeping backward compatibility with legacy contents also imposes new challenges for video systems. This improvements require hardware designs that are compliant with the various systems standards and provide the necessary performance capabilities.

When one of the systems involved in the video transmission (for example between a DVD player and a LCD television) is not compatible with the data encoding format, one possibility is to convert the content to a format that is mutually supported.

This MSc Dissertation will be developed in the context of the Master in Electrical and Computers Engineering at Faculty of Engineering of University of Porto. This thesis was proposed by Synopsys Portugal and will be developed at the company’s offices in Maia, Portugal.

The steps involved in the color space conversion are presented in the following figure. Depending on the input and output color spaces, some of the steps may be unnecessary.