Friday, June 26, 2009

White Bump On The Gums

Il rumore digitale in una compatta

Molti e molte principianti si chiedono cosa sia e come sia il rumore digitale (digital noise) di cui sentono parlare.
Semplice;la vostra voce ha certe tonalità e frequenze precise,ma,se dovete amplificarle,cioè urlare,la voce cambia e si creano "disturbi" cioè alterazioni nella voce.Stessa cosa nel sensore della macchina.
Di solito i sensori hanno una sensibilità originaria,intorno ai 100 iso,ma in the absence of light the camera will set a higher ISO value, to boost the signal and give an image still readable.
amplifying the signal, ie from 100 ISO to 200,400,800, and so on, the image is altered, filled with "noise" made up of dots or marks.
Here is an example;


Picture on the basis of 100 ISO sensitivity.

The amplified image at 200 ISO.

The amplified image at 400 ISO

The amplified image at 800 ISO


The amplified image at 1600 ISO

As shown, the image is degraded, it is "dirty" as they increase the ISO.Questo applies to all macchine.Ma there are a couple of clarifications to make.

1) Not all machines have the same sensore.Il sensor is made of pixels, the single-cell basis, which may have different misure.Se the pixel is larger, has less rumore.Se is smaller will have more noise . Thus, everything else being equal, if a sensor has 10 million pixels of 1.5 microns and another has 10 million pixels, 2.5 microns, the second will produce less noise than the first.

2) Not all machines have the same software noise reduction, and therefore not have the same effect estetico.Le machines have adopted the software to reduce the rumore.In truth is an algorithm, an operation that decides which parts of the image - output from the sensor - or cancel modificare.E decide whether to do it with dots , stains, ecc.Cioè, there are several ways to lower the noise that give different results depending on the type of final macchina.Ecco an article where we examples of different types of noise reduction , which may make it more enjoyable or disturbing image.

0 comments:

Post a Comment