Noise in imaging systems: Fixed pattern noise, electronic and interference noise José Manuel López-Alonso, Rubén González-Moreno, Javier Alda Optics Department. University Complutense of Madrid. School of Optics. Av. Arcos del Jalón, s/n, 28037 Madrid, Spain Phone: +34 91 394 68 72, Fax: +34 91 394 68 80 e-mail: jmlopez@opt.ucm.es, j.alda@fis.ucm.es Abstract Imaging digital systems are widely used nowadays. CCD and CMOS sensors are embedded in a lot of metrologic devices for metrology in a lot of devices. One of these applications is the characterization of laser beams. For these kinds of applications, it is necessary to use cameras with high dynamic range. Some algorithms have been proposed in the past for this purpose. But, normally they enhance not only the dynamic range but the noise sensor too. In this paper we have applied an automatic algorithm to classify the different noise processes appearing in a CCD matrix. The method is based on a principal component expansion of the covariance matrix of some frames taken with the camera. It is possible to classify not only non-uniformity noise of the detector matrix, but also those contributions due to electronics and electronic interference and vibrations. Some of these noise processes represent only a very low amount of the total noise. A method to filter these noises is also presented. 1. Introduction. The measurement of the irradiance profile is an important issue in the characterization of a laser beam, because it provides information about the energy density, its spatial distribution, and the collimation of the beam. Therefore, there have been developed several devices to perform these measurements. 1,2 Among these systems, we can find mechanical scanning devices (such those based on knife-edge test) or electronic ones (such vidicons, solid-state pyroelectric arrays, CID and CCD cameras). In order to achieve fast an reliable measurements, electronic systems are better preferred than mechanical ones. In particular, solid-state devices, like CID and CCD cameras, could be specially useful for these purposes because they have some advantages: low cost, high spatial resolution, and linearity. But they have the disadvantage of a low dynamic range. As example, an image acquisition system composed of a standard CCD camera and an 8-bit frame-grabber has a typical dynamic range of about 60. To extend the dynamic range of a CCD, the system can be cooled. Recently, some algorithms have appeared to increase the dynamic range by means of software instead of hardware. 3,4,5 In these techniques some frames are taken each one with a different SPIE USE, V. 2 5468-52 (p.1 of 9) / Color: No / Format: A4/ AF: A4 / Date: 2004-03-15 09:16:11 Please verify that (1) all pages are present, (2) all figures are acceptable, (3) all fonts and special characters are correct, and (4) all text and figures fit within the margin lines shown on this review document. Return to your MySPIE ToDo list and approve or disapprove this submission.