Mercurial > hg > aim92
comparison man/man1/smooth.1 @ 0:5242703e91d3 tip
Initial checkin for AIM92 aimR8.2 (last updated May 1997).
author | tomwalters |
---|---|
date | Fri, 20 May 2011 15:19:45 +0100 |
parents | |
children |
comparison
equal
deleted
inserted
replaced
-1:000000000000 | 0:5242703e91d3 |
---|---|
1 .TH SMOOTH 1 "1 September 1993" | |
2 | |
3 .SH NAME | |
4 smooth \- Data smoothing by convolution with Gaussian window. | |
5 | |
6 .SH SYNTAX | |
7 smooth [options] [file] | |
8 | |
9 .SH DESCRIPTION | |
10 Low-pass filter by convolving the input with a unit area Gaussian window | |
11 in the time or frequency domain. | |
12 The transform between domains uses the FFT. | |
13 | |
14 | |
15 .SH OPTIONS | |
16 | |
17 1. length | |
18 | |
19 The amount of input to smooth in time units (ms or s) or samples (no units). | |
20 Time units are converted to samples using the given `samplerate' option. | |
21 The string "max" is recognised as the remainder of input. | |
22 | |
23 2. domain | |
24 | |
25 Select algorithm for smoothing. | |
26 In the time domain ("domain=time") the discrete convolution formula is applied | |
27 directly as a local averaging operation on the input with weights obtained | |
28 by time-reversing and shifting a Gaussian window. | |
29 In the frequency domain ("domain=frequency") the FFT of the whole input is | |
30 multiplied by a Gaussian frequency response function (the FFT of the | |
31 Gaussian window), and the result inverse FFT'd. | |
32 | |
33 3. variance, range | |
34 | |
35 The variance of the Gaussian window and its | |
36 ordinal range in standard deviations either side of the mode. | |
37 The window is assumed to be zero for all time outside the specified range. | |
38 These parameters control the size of the window and consequently the amount | |
39 of smoothing. The larger the variance, the more smoothing. | |
40 The size of the window is given by: | |
41 | |
42 .nf | |
43 [ 2 * range * sqrt( variance ) ] samples + 1 sample | |
44 .fi | |
45 | |
46 The variance can be given with time units (ms or s) or samples (no units). | |
47 Time units are converted to samples using the given `samplerate' option. | |
48 | |
49 4. type | |
50 | |
51 The input and output data type. Recognised types are "short" and "float". | |
52 | |
53 .SH "SEE ALSO" | |
54 options gauss filt1 conv | |
55 | |
56 .SH COPYRIGHT | |
57 .LP | |
58 Copyright (c) Applied Psychology Unit, Medical Research Council, 1995 | |
59 .LP | |
60 Permission to use, copy, modify, and distribute this software without fee | |
61 is hereby granted for research purposes, provided that this copyright | |
62 notice appears in all copies and in all supporting documentation, and that | |
63 the software is not redistributed for any fee (except for a nominal | |
64 shipping charge). Anyone wanting to incorporate all or part of this | |
65 software in a commercial product must obtain a license from the Medical | |
66 Research Council. | |
67 .LP | |
68 The MRC makes no representations about the suitability of this | |
69 software for any purpose. It is provided "as is" without express or | |
70 implied warranty. | |
71 .LP | |
72 THE MRC DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING | |
73 ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO EVENT SHALL | |
74 THE A.P.U. BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES | |
75 OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, | |
76 WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, | |
77 ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS | |
78 SOFTWARE. | |
79 |