Meaning
Spatial distributions of light in the image plane resulting from a single point source in the object plane describe the fundamental blurring of an optical system. The point spread function acts as the basic building block of any image. If the function is narrow, the system is sharp.
As it widens, the image becomes more blurred. The boundary of this measurement is defined by the sensor pixel size and the wavelength of the light.
Impulse Response
Mathematically, the image of an object is the convolution of the object with the point spread function. Aberrations like spherical distortion or astigmatism change the shape of this function from a perfect circle to a complex pattern. Measurements are often taken at the center and the edges of the field.
A consistent shape across the sensor is preferred.
Diffraction Limit
Physics prevents a lens from focusing light into an infinitely small point. Even a perfect lens has a point spread function governed by the wavelength of light and the aperture size. This central spot is known as the Airy disk.
It sets the absolute maximum resolution of the system.
Deconvolution Process
Digital signal processing can sometimes reverse the effects of a known blur. If the point spread function is stable and well characterized, mathematical algorithms can sharpen the resulting image. This technique is often used in microscopy and satellite imaging.
Processing software uses a measured or modeled version of the function to recover lost detail. It requires high signal to noise ratios to be effective.