{"id":24322,"date":"2019-02-15T15:55:11","date_gmt":"2019-02-15T20:55:11","guid":{"rendered":"https:\/\/mjtsai.com\/blog\/?p=24322"},"modified":"2019-02-15T15:55:11","modified_gmt":"2019-02-15T20:55:11","slug":"color-spaces","status":"publish","type":"post","link":"https:\/\/mjtsai.com\/blog\/2019\/02\/15\/color-spaces\/","title":{"rendered":"Color Spaces"},"content":{"rendered":"

Bartosz Ciechanowski<\/a>:<\/p>\n

\n

A color space can specify how the numeric values of the red, green, and blue components map to intensity of the corresponding light source. In other words, the position of a slider may not be equal to intensity of the light the slider controls.<\/p>\n

[…]<\/p>\n

This may seem all like a pointless transformation, but there is a good reason for doing all this nonlinear mapping. The human eye is not a simple detector of the power of the incoming light – its response is nonlinear. A two-fold increase in emitted number of photons per second will not<\/em> be perceived as twice as bright light.<\/p>\n

If we were to encode the colors using floating point numbers<\/a> the need for a nonlinear encoding function would be diminished. However, the numeric values of color are often encoded using the familiar 8 bits per component, e.g. in the most common configurations of JPEG<\/a> and PNG<\/a> files. Using a nonlinear tone response curve<\/em>, or TRC for short, lets us maintain more or less perceptual<\/em> uniformity and use the chunky, quantized range to keep the detail in the darker parts.<\/p>\n<\/blockquote>\n\n

Previously:<\/p>\n