Unformatted text preview: © B § ¦ 5
")
5
¡ (3.34) ¡ The 2D DFT is computed by calling two times the 1D DFT. This is very simple: you ﬁrst
compute the DFT of all the columns of
, then you compute the DFT to rows of the
previous result. In fact, 2D DFT codes are just 1D FFT’s codes working on rows and
columns. The 2D DFT is important at the time of ﬁltering 2D images (i.e., gravity maps,
seismic records). Notice that in the 2D DFT we need to consider 2D symmetries.
¦¥ 3.3. ON THE DESIGN OF FINITE IMPULSE RESPONSE FILTERS 3.3 73 On the Design of Finite Impulse Response ﬁlters So far we have studied operators (ﬁlters) that are capable of collapsing a wavelet into a
spike. These ﬁlters are often called spiking ﬁlters or Wiener ﬁlters. In this section we
will examine the problem of designing FIR (Finite Impulse Response) ﬁlters. These are
ﬁlters that are used to eliminate undesired spectral components from our data. 3.3.1 Low Pass FIR ﬁlters In this case we want to design a ﬁlter that operates in the time domain with a amplitude
spectrum with the following characteristics: ...
View
Full Document
 Winter '12
 Sacchi
 Digital Signal Processing, DFT

Click to edit the document details