The main element of any digital camera is either CCD or CMOS sensor. Either works by converting photons of visible light into electrons detectable by the sensor circuitry, which forms raw picture.
Both types of sensors are sensitive not only to the visible light. Detectable electrons can come from ionization caused by sufficiently high energy particles (like those in alpha and beta types of radiation) and quanta of short wave electromagnetic radiation (x-rays and gamma rays). This is a well known phenomenon. In particular it is a nuisance in astronomical CCD imaging: conspicuous spurious bright dots on astronomical images may be caused by cosmic rays.
Such sensitivity to sources of radiation other than visible light is not a problem in consumer digital photography because the camera usually deals with strong visible light signal, which overpowers any unwanted sources. However, giving the proper setup, the effect of ionizing radiation can be detected. Normally the signal from radiation sources around us is very weak and its detection requires some image analysis.
The results of radiation detection with CCD or CMOS may vary greatly since there are way too many unknowns involved, which may skew the method by orders of magnitude. Here is a very incomplete list of some of the unknowns:
- CCD chips characteristics vary noticeable even among the same batch of cameras, and variations are much greater between different models.
- Unless various sources of noise are properly accounted for under controlled conditions, thermal and other internal electronic noise unrelated to radiation can throw the results off.
- Presence of lens or metal case greatly reduces sensitivity of the method making it by far less accurate to weak signal which becomes indistinguishable from background noise.
- Sensitivity to different kinds of radiation can vary (e.g. alpha particles do not penetrate solids too far and can be mostly or completely blocked by the lens and the device casing. Beta particles can penetrate farther but may not have enough energy to cause detectable effect). Thus, piercing radiation (which includes x-rays and gamma rays) is best candidate for detection.
Not surprisingly, calibration of the method presents a challenge. The output of trueGeiger app, when it is executed on low-end consumer digital cameras, like those found in mobile devices, may be viewed only as a rough ballpark estimation of the piercing radiation. A more detailed discussion can be found in the section “What does it measure?”