What is the signal-to-noise ratio?

The signal-to-noise ratio is the ratio between the amount of light converted into an electrical signal by a sensor and the noise of the sensor itself.
In theory, an image sensor converts a given amount of light in a precise tension value: however, in reality, many factors affect sensor performances, such as working temperature or EMI. The difference between the ideal and the real signals is defined as noise.
Usually, this value is given in dB, according to this formula:

SNR = 20 logs (signal/noise)

You can improve the signal-to-noise ratio by reducing the causes of the noise: for example preventing electronic interference and using only shielded cables.