RGB versus HSV color spaces

WHAT THIS TESTS The interviewer checks whether you understand that the same pixel can be represented in coordinate systems chosen to make a task easier, and why intensity-invariance matters for color work.

A GOOD ANSWER COVERS RGB describes a color by the additive intensities of red, green, and blue light. Its drawback is that brightness is spread across all three channels, so a change in lighting shifts every value and there is no single axis that means the color of the object. HSV reorganizes the same information into hue, the dominant wavelength or which color it is; saturation, how vivid or washed-out it is; and value, how bright it is. The win is that hue stays relatively stable when illumination changes, separating what color from how bright.

COMMON WRONG ANSWERS Saying HSV can represent colors RGB cannot; both span the same gamut. Claiming RGB thresholding handles lighting variation just as well. Confusing saturation with brightness.

LIKELY FOLLOW-UPS Why does hue wrap around and cause issues near red. How would you threshold a color robustly in HSV. What is HSL and how does it differ.

ONE CONCRETE EXAMPLE Suppose you must detect a red ball moving across a room with patches of sunlight and shadow. In RGB the red pixels under bright light might be (220, 30, 30) while the same ball in shadow is (90, 15, 15); a fixed threshold on R, G, B struggles because all three channels dropped with the lighting. In HSV the hue of red stays in roughly the same narrow band in both regions, and you simply allow a wide range on value and saturation. So you threshold tightly on hue and loosely on value, and the ball is reliably segmented despite the shadows. This intensity-invariance of hue is exactly why color-based segmentation, skin detection, and object tracking by color are typically done in HSV rather than RGB.