I am trying to do cwtft on earthquake signals in matlab without give Fs(frequency) and plot scalogram.
how to do it?
b=nextpow2(tt);
n=2^b;
s={win2,1/(n-1)};
cwtstruct=cwtft(s,wv);
det=str2double(cwtstruct)
MATLAB has some built-in examples, that are helpful to take a look, and luckily in your case, it has one for earthquake time and frequency localizations.
Example 1 Kobe data
load kobe
to load the time domain data (that I'm guessing).
[wt,f] = cwt(kobe,1);
Reconstructing the earthquake time data.
xrec = icwt(wt,f,[0.030 0.070],'SignalMean',mean(kobe));
Plotting and comparing the original time data and the data for frequencies in the range of [0.030, 0.070] Hz, which I'm guessing that it would be some technical frequency ranges for earthquakes.
subplot(2,1,1)
plot(kobe)
grid on
title('Original Data')
subplot(2,1,2)
plot(xrec)
grid on
title('Bandpass Filtered Reconstruction [0.030 0.070] Hz')
Example 2 El Nino data
Loading the El Nino (earthquake) data and obtaining its Continuous Wavelet Transformation of time data:
load ninoairdata
[cfs,period] = cwt(nino,years(1/12));
Obtain the inverse CWT for years 2 through 8.
xrec = icwt(cfs,period,[years(2) years(8)]);
and plotting the CWT of the reconstructed data using:
figure
cwt(xrec,years(1/12))
and finally comparing the original time data with the reconstructed data:
figure
subplot(2,1,1)
plot(nino)
grid on
title('Original Data')
subplot(2,1,2)
plot(xrec)
grid on
title('El Nino Data - Years 2-8')
You can simply load your own data into these models and see how you'd like to design your methods and analyze your data.
Related
I am trying to build a Keras model to implement to approach explained in this paper.
Context of my implementation:
I have two different kinds of data representing the same set of classes(labels) that needs to be classified. The 1st kind is Image data, and the second kind is EEG data (a time series sequence).
I know that to classify image data we can use CNN models like this:
model.add(Conv2D(filters=256, kernel_size=(11,11), strides=(1,1), padding='valid'))
model.add(Activation('relu'))
model.add(Dense(1000))
model.add(Activation('relu'))
model.add(Dropout(0.4))
# Batch Normalisation
model.add(BatchNormalization())
# Output Layer
model.add(Dense(40))
model.add(Activation('softmax'))
And to classify sequence data we can use LSTM models like this:
model.add(LSTM(units = 50, return_sequences = True))
model.add(Dropout(0.2))
model.add(Flatten())
model.add(Dense(32, activation='relu'))
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(40, activation='softmax'))
But the approach of the paper above shows that EEG feature vectors can be mapped with image vectors through regression like this:
The first approach is to train a CNN to map images to corresponding
EEG feature vectors. Typically, the first layers of CNN attempt to
learn the general (global) features of the images, which are common
between many tasks, thus we initialize the weights of these layers
using pre-trained models, and then learn the weights of the last
layers from scratch in an end-to-end setting. In particular, we used
the pre-trained AlexNet CNN, and modified it by replacing the
softmax classification layer with a regression layer (containing as
many neurons as the dimensionality of the EEG feature vectors),
using Euclidean loss as the objective function.
The second approach consists of extracting image features using
pre-trained CNN models and then employ regression methods to map
image features to EEG feature vectors. We used our fine-tuned
AlexNet as feature extractors by
reading the output of the last fully connected layer, and then
applied several regression methods (namely, k-NN regression, ridge
regression, random forest regression) to obtain the predicted
feature vectors
I am not able to comprehend how to code the above two approaches. I have never used a regressor for feature mapping and then do classification. Any leads on this are much appreciated.
In my understanding the training data consists of (eeg_signal,image,class_label) triplets.
Train the LSTM model with input=eeg_signal, output=class_label. Loss is crossentropy.
Peel off the last layer of the LSTM model. Let's say the pre-last layer's output is a vector of size 20. Let's call it eeg_representation.
Run this truncated model on all your eeg_signal inputs, save the output of eeg_representation. You will get a tensor of [batch, 20]
Take that AlexNet mentioned in the paper (or any other image classifier), peel off the last layer. Let's say the pre-last layer's output is a vector of size 30. Let's call it image_representation.
Stich a linear layer to the end of the previous layer. This layer will convert image_representation to eeg_representation. It has 20 x 30 weight.
Train the stiched model on (image, eeg_representation) pairs. Loss is the Euclidean distance.
And now the fun part: Stich together model trained in step 7. and the peeled off part of model trained in step 1. If you input an image, you will get class predictions.
This sound like not a big deal (because we do image classification all the time), but if this is really working, it means that this is a "prediction that is running through our brains" :)
Thank you bringing up this question and linking the paper.
I feel I just repeated what's in your question and in the the paper.
I would be beneficial to have some toy dataset to be able to provide code examples.
Here's a Tensorflow tutorial on how to "peel off" the last layer of a pretrained image classification model.
I‘m working on a visual data logger for my DMM, it writes every measurement to RS232 inteface. There I connect a Teensy 3.6 and collect the data points.
For each point I have the timestamp and the measured value. I will collect 10.000 readings.
I want to display the measured data on a display (800x480) in two ways. First as a rolling graph, that scrolls from right to left and shows the last minute or so. This is working fine.
Second, I want to display all collected measurements in total (max. 10k points). So I have to shrink or compress the data, but I want to preserve the shape of the curve.
To give you an idea how it should look like, please watch the video from Dave on EEV at YT (https://youtu.be/SObqPuUozNo) and skip to 41:20. There you see how another DMM is shrinking the incomming data and displays it. At about 1:01:05 10k measurements are shown on the display area of only 400px wide.
Question is, how is this done?
I’ve heard about Douglas-Pucker algorithm, but have no idea if this is the right way and how to use it on the Arduino/ Teensy platform.
Any help is very welcome, thank you....
I cannot just display all data points, because I‘m using an FT81x as display controller, and this can take only up to 2000 drawing commands per frame. And it takes more time.
Anyway, I solved the problem using the simple way.
I create bins and calculate the min and max values in this bin. Then simply draw a line between these points. Works fine!
BTW, I‘m the TO :-)
For cases where you got many more samples than pixels in x axis instead of LineTo like graph use vertical lines graph instead...
So depending on the number of samples per rendered time frame and x resolution you should compute ymin and ymax for eaxch x and render vertical line ...
something like:
xs=800;
for (x0=x,i=sample_left,y0=y1=sample[i],i<sample_right;i++)
{
x = (i-sample_left)*xs/(sample_right-sample_left);
y = sample[i]; // here add y scaling and offset
if (x0!=x) { line(x0,y0,x0,y1); x0=x; y0=y; y1=y; }
if (y0>y) y0=y;
if (y1<y) y1=y;
}
where sample[] are your stored values , sample_left,sample_right is the range to render and xs is graph x resolution. To speed up you can pre-compute the y0,y1 for each x and render that (recompute only on range or samples change) ... So as you can see you will use just xs line commands which shoul dbe fast enough. The x linear interpolation can be done without multiplication nor division if you rewrite it to integer DDA style ...
These QAs might interest you:
plotting real time Data on (qwt )Oscillocope
i don't really understand FFT and sample rates
[note]
After a second look The deleted answer is with the same approach as this (got deleted by review probably before the edit which transformed it from not an answer (comment) to the correct answer) so I voted for undelete even if it is considerably lower quality than mine but was posted sooner.
I'm working on a project in which a rod is attached at one end to a rotating shaft. So, as the shaft rotates from 0 to ~100 degrees back-and-forth (in the xy plane), so does the rod. I mounted a 3-axis accelerometer at the end of the moving rod, and I measured the distance of the accelerometer from the center of rotation (i.e., the length of the rod) to be about 38 cm. I have collected a lot of data, but I'm in need of help to find the best method to filter it. First, here's a plot of the raw data:
I think the data makes sense: if it's ramping up, then then I think at that point the acceleration should be linearly increasing, and then when it's ramping down, it should linearly decrease. If its moving constantly, the acceleration will be ~zero. Keep in mind though that sometimes the speed changes (is higher) from one "trial" to the other. In this case, there were ~120 "trials" or movements/sweeps, data sampled at 148 Hz.
For filtering, I've tried a low pass filter and then an exponentially decreasing moving average, and both plots weren't too hot. And although I'm not good at interpreting these: here is what I got when coding a power frequency plot:
What I was hoping to get help with here is, attain a really good method by which I can filter this data. The one thing that keeps coming up again time and time again (especially on this site) is the Kalman filter. While there's lots of code online that helps implementing these in MATLAB, I haven't been able to actually understand it that great, and therefore neglect to include my work on it here. So, is a kalman filter appropriate here, for rotational acceleration? If so, can someone help me implement one in matlab and interpret it? Is there something I'm not seeing that may be just as good/better that is relatively simple?
Here's the data I'm talking about. Looking at it more closely/zooming in gives a better appreciation for what's going on in the movement, I think:
http://cl.ly/433B1h3m1L0t?_ga=1.81885205.2093327149.1426657579
Edit: OK, here is the plot of both relavent dimensions collected from the accelerometer. I am neglecting to include the up and down dimension as the accelerometer shows a near constant ~1 G, so I think its safe to say its not capturing much rotational motion. Red is what I believe is the centripetal component, and blue is tangential. I have no idea how to combine them though, which is why I (maybe wrongfully?) ignored it in my post.
And here is the data for the other dimension:
http://cl.ly/1u133033182V?_ga=1.74069905.2093327149.1426657579
Forget the Kalman filter, see the note at the end of the answer for the reason why.
Using a simple moving average filter (like I showed you on an earlier reply if i recall) which is in essence a low-pass filter :
n = 30 ; %// length of the filter
kernel = ones(1,n)./n ;
ysm = filter( kernel , 1 , flipud(filter( kernel , 1 , flipud(y) )) ) ;
%// assuming your data "y" are in COLUMN (otherwise change 'flipud' to 'fliplr')
note: if you have access to the curvefit toolbox, you can simply use: ys = smooth(y,30) ; to get nearly the same result.
I get:
which once zoomed look like:
You can play with the parameter n to increase or decrease the smoothing.
The gray signal is your original signal. I strongly suspect that the noise spikes you are getting are just due to the vibrations of your rod. (depending on the ratio length/cross section of your rod, you can get significant vibrations at the end of your 38 cm rod. These vibrations will take the shape of oscillations around the main carrier signal, which definitely look like what I am seeing in your signal).
Note:
The Kalman filter is way overkill to do a simple filtering of noisy data. Kalman filter is used when you want to calculate a value (a position if I follow your example) based on some noisy measurement, but to refine the calculations, the Kalman filter will also use a prediction of the position based on the previous state (position) and the inertial data (how fast you were rotating for example). For that prediction you need a "model" of the behavior of your system, which you do not seem to have.
In your case, you would need to calculate the acceleration seen by the accelerometer based on the (known or theoretical) rotation speed of the shaft at any point of time, the distance of the accell to the center of rotation, and probably to make it more precise, a dynamic model of the main vibration modes of your rod. Then for each step, compare that to the actual measurement... seems a bit heavy for your case.
Look at the quick figure explaining the Kalman filter process in this wikipedia entry : Kalman filter, and read on if you want to understand it more.
I will propose for you low-pass filter, but ordinary first-order inertial model instead of Kalman. I designed filter with pass-band till 10 Hz (~~0,1 of your sample frequency). Discrete model has following equation:
y[k] = 0.9418*y[k-1] + 0.05824*u[k-1]
where u is your measured vector, and y is vector after filtering. This equation starts at sample number 1, so you can just assign 0 to the sample number 0.
I am trying to do some kind of image sorting.
I have 5 images and first one is my main image. I am trying to sort images according to their similarity.(Most similar image to less similar image).
Matlab had matchfeature method but I dont think I jave used it correctly because my results are wrong.I try to use:
[indexPairs,matchmetric] = matchFeatures(features1,features2,"MatchThreshold,10")
then i try to sort the matchmetric array.But it didnt work
Can anyone tell me some algorithm or any tips ?Thank you..
You could compute the correlation coefficient between every images and your main image and then sort them based on the coefficient.
doc corr2
For example, let's say you store all your images in a cell array (called ImageCellArray) in which the first image is your "main image":
for i = 2:size(ImageCellArray,2) % size(ImageCellArray,2) is the total # of images, i.e. the size of the cell array containing them.
CorrCoeff(i) = corr2(rgb2gray(ImageCellArray{1}),rgb2gray(ImageCellArray{i}));
end
[values indices] = sort(CorrCoeff); % sort the coefficients and get the number of the corresponging image.
Then you're good to go I guess.
You could compute the PSNR (peak signal-to-noise ratio) for each image compared to the main image. PSNR is a metric commonly used to measure the quality of a reconstructed compression against the original image.
It's implemented in Matlab in the Computer Vision System toolbox as a functional block, and there is also a psnr function in the Image Processing toolbox. The result will be a number in decibels you could use to rank the images. A higher PSNR value indicates greater similarity.
Take a look at this example of image retrieval. Instead of matching the features between pairs of images it uses the KDTreeSearcher from the Statistics Toolbox to find nearest neighbors of each feature from the query image across the whole set of database images.
I used the Database of Faces (formally the ORL Database) from the AT&T Laboratories Cambridge. The database consists of 400 images with 10 images per person, i.e, there is 10 images of each 40 person.
I separated 5 images of each person for training and the remaining 5 images of each person for testing.
So I have 2 folders:
1) Training (5 images/person = 200 images)
2) Testing (5 images/person = 200 images)
The photos in the training folder are different from those in the testing folder.
The percentage recognition rate I got is only 80%.
But if I pre-process the image before recognition I got:
pre-processing with imajust: 82%
pre-processing with sharpen: 83%
pre-processing with sharpen and imadjust: 84%
(If pre-processing is done, it is applied to bot training and testing images)
For the number of eigenfaces used, all eigenvalues of matrix L are sorted and those who are less than a specified threshold, are eliminated.
L_eig_vec = [];
for i = 1 : size(V,2)
if( D(i,i)>1 )
L_eig_vec = [L_eig_vec V(:,i)];
end
end
I use matlab to implement the face recognition system. Is it normal that the recognition rate is that low?
The accuracy would depend on the classifier you are using once you have the data in the PCA projected space. In the original Turk/Pentland eigenface paper
http://www.face-rec.org/algorithms/PCA/jcn.pdf
they just use kNN / Euclidean distance but a modern implementation might use SVMs e.g. with an rbf kernel as the classifier in the "face space", with C and gamma parameters optimized using a grid search. LibSVM would do that for you and there is a Matlab wrapper available.
Also you should register the faces first i.e. warping the images so they have facial landmarks e.g. eyes, nose, mouyth in a harmonised position across all the dataset? If the images aren't pre-registered then you will get a performance loss. I would expect a performance in the 90s for a dataset of 5 people using Eigenfaces with SVM and pre-registration. That figure is a gut feeling based on prior implementation / performance of past student projects. One thing to note however is your number of training examples is very low - 5 points in a high dimensional space is not much to train a classifier on.