Trying to identify certain peaks in a plot

Question

Shruthimol 2022-8-22

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1783430-trying-to-identify-certain-peaks-in-a-plot

评论： Mathieu NOE 2022-8-23

I have managed to plot 2 different plots as shown in the below figures.

In this figure, I have added 2 types of data. The yellow corresponds to a reference undamaged specimen, indicated by the presence of the peaks between 10 and 12 (is called backwall echo). The one in red is indicating presence of damage because of the absense of backwall echo and the increased amplitude between 3 and 8.

In this figure the tested specimen has no damage as there is presence of backwall echo.

I'm ot able to figure out a method to categorize and identify the presence of backwall echo from a given data. Please suggest me some methods so that my output correctly determines if there is damage or not (presence of backwall echo or not).

Thanks in Advance.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Mathieu NOE 2022-8-23

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1783430-trying-to-identify-certain-peaks-in-a-plot#answer_1031210

编辑：Mathieu NOE 2022-8-23

在 MATLAB Online 中打开

hello

I have reworked your code

check this :

%% Plot of Undamaged Specimen
% na=readmatrix('Default Dataset (4).csv');
% na=readmatrix('Default Dataset (3).csv');
na=readmatrix('Default Dataset (5).csv');
%time
x= na(:,1);
%amplitude
y=na(:,2);
t=0:0.006:12;
D=interp1(x,y,t);
% plot (x,y,t,D);
plot (t,D);
% [v,idx]=findpeaks(D,MinPeakProminence=0.5);
[v,idx]=findpeaks(D,'MinPeakProminence',0.25);
y_threshold = 10;
idx(v>y_threshold)=[];
v(v>y_threshold)=[];
x_threshold=500;
v(idx<x_threshold)=[];
idx(idx<x_threshold)=[];
hold on
scatter(x(idx),v,"y","filled")
%% Plot of Damaged Specimen 
nb=readmatrix('Default Dataset (6).csv');
x1= nb(:,1);
y1=nb(:,2);
E=interp1(x1,y1,t); % no need to use u instead of t 
hold on
plot (t,E);
% [w,idx]=findpeaks(E,MinPeakProminence=0.5);
[w,idy]=findpeaks(E,'MinPeakProminence',0.25);
y_threshold = 10;
idy(w>y_threshold)=[];
w(w>y_threshold)=[];
x_threshold=500;
w(idy<x_threshold)=[];
idy(idy<x_threshold)=[];
scatter(x(idy),w,"r","filled")
legend('Undamaged Specimen','Peaks of Undamaged Specimen','Tested Specimen','Peaks of Tested Specimen');
hold off
%% Detecting peaks
% Thresholds
xmin_threshold = 10 ;
xmax_threshold = 12 ;
% Remove below min threshold
v(t(idx)<xmin_threshold )=[];
idx(t(idx)<xmin_threshold )=[];
w(t(idy)<xmin_threshold )=[];
idy(t(idy)<xmin_threshold )=[];
% Remove above max threshold
v(t(idx)>xmax_threshold )=[];
idx(t(idx)>xmax_threshold)=[];
w(t(idy)>xmax_threshold )=[];
idy(t(idy)>xmax_threshold)=[];
n=numel(v); % <= here
m=numel(w); % <= here
% if (n>m)
%    disp('The specimen is showing characteristics of damage at this site.');
% else
%    disp('The specimen is showing no characteristics of damage at this site.');
% end
%% simpler alternative  : if m>0 the tested specimen is OK (whatever the n value is)
if (m>0)
   disp('The specimen is showing no characteristics of damage at this site.');
else
   disp('The specimen is showing characteristics of damage at this site.');
end