matlab中findpeak 函数用法

matlab中findpeak 函数用法

help findpeaks
findpeaks Find local peaks in data
PKS = findpeaks(Y) finds local peaks in the data vector Y. A local peak
is defined as a data sample which is either larger than the two
neighboring samples or is equal to Inf.

[PKS,LOCS]= findpeaks(Y) also returns the indices LOCS at which the
peaks occur.

[PKS,LOCS] = findpeaks(Y,X) specifies X as the location vector of data
vector Y. X must be a strictly increasing vector of the same length as
Y. LOCS returns the corresponding value of X for each peak detected.
If X is omitted, then X will correspond to the indices of Y.

[PKS,LOCS] = findpeaks(Y,Fs) specifies the sample rate, Fs, as a
positive scalar, where the first sample instant of Y corresponds to a
time of zero.

[...] = findpeaks(...,'MinPeakHeight',MPH) finds only those peaks that
are greater than the minimum peak height, MPH. MPH is a real valued
scalar. The default value of MPH is -Inf.

[...] = findpeaks(...,'MinPeakProminence',MPP) finds peaks guaranteed
to have a vertical drop of more than MPP from the peak on both sides
without encountering either the end of the signal or a larger
intervening peak. The default value of MPP is zero.

[...] = findpeaks(...,'Threshold',TH) finds peaks that are at least
greater than both adjacent samples by the threshold, TH. TH is real
valued scalar greater than or equal to zero. The default value of TH is
zero.

findpeaks(...,'WidthReference',WR) estimates the width of the peak as
the distance between the points where the signal intercepts a
horizontal reference line. The points are found by linear
interpolation. The height of the line is selected using the criterion
specified in WR:

 'halfprom' - the reference line is positioned beneath the peak at a
    vertical distance equal to half the peak prominence.

 'halfheight' - the reference line is positioned at one-half the peak 
    height. The line is truncated if any of its intercept points lie
    beyond the borders of the peaks selected by the 'MinPeakHeight',
    'MinPeakProminence' and 'Threshold' parameters. The border between
    peaks is defined by the horizontal position of the lowest valley
    between them. Peaks with heights less than zero are discarded.

 The default value of WR is 'halfprom'.

[...] = findpeaks(...,'MinPeakWidth',MINW) finds peaks whose width is
at least MINW. The default value of MINW is zero.

[...] = findpeaks(...,'MaxPeakWidth',MAXW) finds peaks whose width is
at most MAXW. The default value of MAXW is Inf.

[...] = findpeaks(...,'MinPeakDistance',MPD) finds peaks separated by
more than the minimum peak distance, MPD. This parameter may be
specified to ignore smaller peaks that may occur in close proximity to
a large local peak. For example, if a large local peak occurs at LOC,
then all smaller peaks in the range [N-MPD, N+MPD] are ignored. If not
specified, MPD is assigned a value of zero.

[...] = findpeaks(...,'SortStr',DIR) specifies the direction of sorting
of peaks. DIR can take values of 'ascend', 'descend' or 'none'. If not
specified, DIR takes the value of 'none' and the peaks are returned in
the order of their occurrence.

[...] = findpeaks(...,'NPeaks',NP) specifies the maximum number of peaks
to be found. NP is an integer greater than zero. If not specified, all
peaks are returned. Use this parameter in conjunction with setting the
sort direction to 'descend' to return the NP largest peaks. (see
'SortStr')

[PKS,LOCS,W] = findpeaks(...) returns the width, W, of each peak by
linear interpolation of the left- and right- intercept points to the
reference defined by 'WidthReference'.

[PKS,LOCS,W,P] = findpeaks(...) returns the prominence, P, of each
peak.

findpeaks(...) without output arguments plots the signal and the peak
values it finds

findpeaks(...,'Annotate',PLOTSTYLE) will annotate a plot of the
signal with PLOTSTYLE. If PLOTSTYLE is 'peaks' the peaks will be
plotted. If PLOTSTYLE is 'extents' the signal, peak values, widths,
prominences of each peak will be annotated. 'Annotate' will be ignored
if called with output arguments. The default value of PLOTSTYLE is
'peaks'.

% Example 1:
%   Plot the Zurich numbers of sunspot activity from years 1700-1987
%   and identify all local maxima at least six years apart
load sunspot.dat
findpeaks(sunspot(:,2),sunspot(:,1),'MinPeakDistance',6)
xlabel('Year');
ylabel('Zurich number');

% Example 2: 
%   Plot peak values of an audio signal that drop at least 1V on either
%   side without encountering values larger than the peak.
load mtlb
findpeaks(mtlb,Fs,'MinPeakProminence',1)

% Example 3:
%   Plot all peaks of a chirp signal whose widths are between .5 and 1 
%   milliseconds.
Fs = 44.1e3; N = 1000;
x = sin(2*pi*(1:N)/N + (10*(1:N)/N).^2);
findpeaks(x,Fs,'MinPeakWidth',.5e-3,'MaxPeakWidth',1e-3, ...
          'Annotate','extents')