In some projects, it is usually necessary to sense the color of the object. The most straightforward way for our project to "see" things is, of course, to install CMOS image sensors on them. However, using CMOS image sensors can complicate things if we simply need to recognize an object's color, rather than actually performing advanced functions such as image recognition. At this point, a simple color sensor can help us solve the problem. This time, we'll discuss how color sensors work and how to use them.

  Humans are so-called "trichromatic" animals, which means that our color vision is dominated by three colors. By now, many readers have probably guessed that these three colors are the R, G, and B that we've been talking about.

  In the retina of the human eye, there are two types of cells: cone-shaped cells that can sense color, and rod-shaped cells that cannot sense color but are sensitive to light and can work in low illumination. There are three kinds of cones, called S, M and L cones. S, M and L represent cones with different sensitivity to different wavelengths of light.

  Light of different wavelengths enters the eye and produces signals of different intensities in different cones. These signals travel through the nervous system to the brain, where they are processed into the colours we see.

  The human eye really can't tell whether yellow is a single wavelength of yellow or a mixture of red and green. This phenomenon is called "homophilia", meaning that different spectral components appear to be the same colour to the human eye. Although human eyes look the same color, their spectral composition is still different when measured with a spectrophotometer.

  One of the most common components for machines to see light is the so-called photoelectric conversion module.

  The photoelectric effect tells us that the energy carried by each photon depends only on its wavelength, not on the intensity of the light. The wavelength range of signals emitted by different substances under the photoelectric effect depends only on their electron configuration.

  After the explosion of semiconductor technology in the 20th century, the most common semiconductor material was silicon, and the electronic structure of silicon enabled the photodiodes made of it to sense light at wavelengths between 400 and 1100 nm, covering the entire spectrum of visible light. The wavelength of the human eye has a bit of near-infrared light. This wonderful coincidence allows us to make all kinds of photosensitive devices using silicon semiconductors without having to look for other special materials and processes.