Many factors contribute to the technical quality of a digital image. The quality and sharpness of the camera lens are critical, the photographer must ensure that the image is properly recorded using an appropriate exposure, camera movement should generally be avoided, and the quality of the sensor's pixels must be excellent.
Image quality is also dependent upon resolution. When everything else is correct the definition, tonal gradation, sharpness and overall quality of an image should increase with the number of pixels used to create it. However, it is important to note that a greater number of lower-quality pixels may not improve resolution. Indeed an image consisting of 10 million high-quality pixels may have better resolution than a comparable image consisting of 12 million lower-quality pixels. Put simply, the quality of individual pixels may be more important than the total number of pixels. Pixels of larger physical dimensions generally have a better signal-to-noise ratio (ie less noise).
The three images below show the same portrait recorded using a decreasing number of pixels, and the consequent and obvious degradation. However, once the number of pixels in an image of a particular size is such that the human eye cannot differentiate between them, a further increase in the number of pixels makes no difference.
If an image having a particular number of pixels is enlarged, the information becomes more thinly spread and the eye may once again find the result unacceptable. This problem cannot be alleviated by up-sampling using image-manipulation software. The re-sampling process may actually result in visible deterioration.