Photoelectron Noise Limitations in High Performance Imaging Sensors
Abstract
In the conventional signal-to-noise ratio analysis of image forming sensors, the imaged area is broken into fictional 'resolution elements' and the noise calculated as the square root of the total number of counts or events, n, falling within a single element. Since the signal is proportional to the total number of counts within the element, the signal-to-noise ratio is square root of n. By postulating a minimum required, or threshold, signal-to-noise ratio, these calculations have led to expressions for the limiting resolution of the device as a function of incident light level. The effects of a finite aperture have been considered by Schade and more recently by Rosell. Schade's analysis led him to define the noise equivalent sampling area as the reciprocal of the noise equivalent passband. The concept of a noise equivalent sampling area has been used by Rosell to calculate the display signal-to-noise ratio and limiting resolution of a variety of television camera tubes. Analyses of this type have been extensively used to calculate the performance of military television systems. In the following analysis some of the ambiguities of the Rosell-Schade approach are resolved. The apriori assumption of the existence of fictitious resolution elements is eliminated and reintroduced only after considerable development. Two equivalent square sampling apertures or areas are defined.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1976
- Accession Number
- ADA026005
Entities
People
- Ronald D. Graft