Feature extraction from a signal and classification

Question

Harry 2017-2-19

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/325824-feature-extraction-from-a-signal-and-classification

评论： Liyifei LiYiFei 2023-4-24

I was wondering if anyone could help me with a few steps or even code to get started on feature extraction from a signal. I would like to extract the features of a signal and then classify them in the classification learner app.

The data I have is a vibration signal that varies between a set of healthy bearings and bearings that are faulty.

The aim is to extract the features and then compare them in the classification app with each other to identify when bearings are faulty.

Any help is appreciated!

Thanks

3 个评论
显示 1更早的评论隐藏 1更早的评论

Harry 2017-2-26

I uploaded the signals and the code.

Thank you

Liyifei LiYiFei 2023-4-24

你好，我也有和你相似的问题，你解决了吗，如何对信号进行特征提取并进行故障分类

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Abel Babu 2017-2-22

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/325824-feature-extraction-from-a-signal-and-classification#answer_255823

编辑：Abel Babu 2017-2-22

Hi Harry,

This MATLAB example would be a good starting point:

http://in.mathworks.com/examples/signal/mw/signal-ex86966978-extracting-classification-features-from-physiological-signals

The example deals with the classification of physiological signals but the features used here can be applied to classification of signals in general. The extracted features can then be fed as features for the classification app.

As suggested by Image Analyst, spectral analysis can be used to generate more features as well.

Abel

3 个评论
显示 1更早的评论隐藏 1更早的评论

Jan 2017-2-22

+1. For Matlab all signals are a list of numbers only and it does not matter what the signal means physically.

Aditya Baru 2018-5-2

编辑：Aditya Baru 2018-5-2

Here's an example for feature extraction for bearing signals :) https://www.mathworks.com/help/predmaint/examples/wind-turbine-high-speed-bearing-prognosis.html https://www.mathworks.com/help/predmaint/examples/Rolling-Element-Bearing-Fault-Diagnosis.html

The new Predictive Maintenance Toolbox has a lot of capabilities that support this kind of workflow, so you could check out its documentation for more information.

请先登录，再进行评论。

Answer 2

Zhao Lu 2021-3-22

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/325824-feature-extraction-from-a-signal-and-classification#answer_654187

MATLAB has EMD function, for EEMD, the code can be downloaded from

http://perso.ens-lyon.fr/patrick.flandrin/emd.html

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Francisco Navarro 2021-3-22

thanks!

请先登录，再进行评论。

Answer 3

Rahmawati Rahmawati 2018-5-2

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/325824-feature-extraction-from-a-signal-and-classification#answer_318299

halo everyone, I am rahma and i am totally newbie in EEG data analysis. I got an assignment to make a classification between two conditions using spectral powers based on Raw EEG data which has been given by my Professor. but to be honest i don't know how to start with this. and here are the state: Sampling rate: 512 HZ Channel position: POz, PO1, PO2, PO3, PO4, Oz, O1, O2

any help and hints are totally helpful for me.

Thanks in advance.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Smith Khare 2019-9-25

One need complete dataset to process and understand, then after the main work starts

请先登录，再进行评论。

Answer 4

karthikeyan chandrasekar 2019-1-8

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/325824-feature-extraction-from-a-signal-and-classification#answer_355487

hi everyone can anyone tell me how to extract features using PCA for a signal ,i.e 8190x2 signal which is in text[matrix] format.

Thankyou in advance

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

John Navarro 2021-2-2

By PCA did you mean Principal component analysis?

If so, PCA does not extract features, it evaluates their correlation and indicates the more useful ones. PCA is employed for feature selection, no feature extraction. It should be done according the expertise, the case of study, and the features of interest.

请先登录，再进行评论。