Detecting maximum with threshold
12 次查看(过去 30 天)
显示 更早的评论
%example threshold detection code
function [MAXES] = identifyMaximums(Avg)
X = Avg;
threshold = 30;
[Xon, Xoff] = detectThreshold(X,threshold);
nSteps = length(Xon);
firstOff = 1;
while Xoff(firstOff) > Xon(1)
firstOff = firstOff+1;
end
lastOn = nSteps;
while Xon(lastOn) > Xoff(end)
lastOn = lastOn-1;
end
%Adjust ons and offs
Xon_align = Xon(1:lastOn);
Xoff_align = Xoff(firstOff:end);
%Count number of steps after aligning
nStepsFinal = length(Xon_align);
%Initialize matrix of maximums
MAXES = NaN(nStepsFinal,1);
for i = 1:nStepsFinal
MAXES(i) = max(X(Xon_align(i):Xoff_align(i)));
end
function [Ons,Offs] = detectThreshold(signal,threshold)
overThreshold = signal>threshold;
onOffs = diff(overThreshold);
Offs = find(onOffs==-1)+1;
Ons = find(onOffs==1);
end
end
Error:
Index exceeds the number of array elements. Index must
not exceed 0.
Error in identifyMaximums (line 14)
while Xoff(firstOff) > Xon(1)
Error in load_data_10_15 (line 119)
[MAXES] = mean(identifyMaximums(Avg));
Data (Avg) that is being processed:attached image
Hello I hope all is well. I am currently working on a side project where I am trying to match data from a sensor to a force plate. At the moment I am trying to linearize the data. To do so, I must collect the peak values. For some reason I am getting this error and I am not sure why. If you have the time, please look over my work and provide some guidence.
Thank you!
0 个评论
采纳的回答
更多回答(1 个)
Matt J
2022-11-28
编辑:Matt J
2022-11-28
If you have the time, please look over my work and provide some guidence.
In detectThreshold(), you need to add control logic to handle the case where find() returns empty [].
Offs = find(onOffs==-1)+1;
Ons = find(onOffs==1);
2 个评论
Matt J
2022-11-28
I left it to you because only you knowwhat's supposed to happen when no peaks are detected.
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Numeric Types 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!