Morlet wavelets in time and in frequency

Поділитися
Вставка
  • Опубліковано 21 гру 2024

КОМЕНТАРІ • 68

  • @thomasalderson368
    @thomasalderson368 3 роки тому +5

    don't quit your day job...
    these videos are insanely helpful.

    • @mikexcohen1
      @mikexcohen1  3 роки тому +3

      lol, making these videos is my night job ;)

  • @exoticcoder5365
    @exoticcoder5365 Рік тому +1

    This is a very good explanation series that other channels didn’t provide, even with many views one, much appreciation sir !

    • @mikexcohen1
      @mikexcohen1  Рік тому

      Thank you kindly, exoticcoder5365 :)

  • @BruinChang
    @BruinChang 3 роки тому +5

    The time frequency representation is an art.

  • @donjon9333
    @donjon9333 4 роки тому +17

    U killed me with that joke-let

    • @mikexcohen1
      @mikexcohen1  4 роки тому +8

      In these times, even bad jokes can be useful!

  • @Tapuzi
    @Tapuzi Рік тому

    High quality! Thank you. I am trying to base my intuition in this subject for a while, this is very beneficial.

  • @videofountain
    @videofountain Рік тому +2

    I think you may have inspired students to learn morelet about this topic.

  • @vtrandal
    @vtrandal 9 місяців тому

    Excellent video. Thank you. I'm looking for a transcipt to this video, but I cannot find one.

    • @mikexcohen1
      @mikexcohen1  9 місяців тому

      Hmm, that does seem to be the case. They're auto-generated. From some googling, this seems to happen sometimes with long videos, although this isn't my longest YT video. It doesn't seem possible to re-start the autogen process. I'll look into it, thanks for letting me know.

  • @dnranjit
    @dnranjit 4 роки тому +1

    Really interesting video and really appreciate the amount of passion and hardwork has gone into this.One point though: @11:04 ..the amplitude of the sine wave is fluctuating. Only if it had been a complex signal rather than a sinusoid would its amplitude be constant.

    • @mikexcohen1
      @mikexcohen1  4 роки тому

      Great point, thanks Ranjit. Indeed, we always use complex-valued wavelets for the analyses, but I find it useful to start by introducing real-valued wavelets.

    • @dnranjit
      @dnranjit 4 роки тому

      @@mikexcohen1 I must say, your videos are quite lucid and easy to comprehend and most importantly ..a good sense of humor.Good content and really love the color palette for the waveforms. Please keep up the good work.

  • @tobi3497
    @tobi3497 4 роки тому

    This sliding of the Morley wavelet over the series seems like the exact same thing as the STFT. Where instead of applying this window function on the input signal, you're applying the window function on the convolution wave

    • @mikexcohen1
      @mikexcohen1  4 роки тому

      Yes, that's a good summary. Most "different" methods for time-frequency analysis are actually very similar to each other. Wavelet convolution has some advantages over the STFT (e.g., easier and faster to implement, fewer parameters), but the end result will be basically the same.

    • @tobi3497
      @tobi3497 4 роки тому

      @@mikexcohen1 it seems like it would be slower to run, as youd need to run it for every frequency and time window: O(t x f) . Where was with STFT, you're only running for every time window (getting all the frequencies for that window at once): O(t)

    • @tobi3497
      @tobi3497 4 роки тому

      @@mikexcohen1 I asked a question about this here: dsp.stackexchange.com/questions/68015/wavelets-vs-fourier-transforms

    • @mikexcohen1
      @mikexcohen1  4 роки тому

      It depends on how you set it up and on how many time bins you have. Let's say you have 20 frequencies to extract. You need 1 FFT for the data, 20 FFTs for each wavelet, and 20 IFFTs. So that's 41 FFTs in total for the entire TF map. In neuroscience data analysis, we might have 2000 time points. If you extract 41 time bins for the STFFT, then the number of FFTs is the same, but 41 bins for 2000 time points is rather sparse. Furthermore, people often want to change the width of the time bins over different ranges of frequencies. Let's say you have three bins of frequency ranges with different time windows for the same center time point. So then the number of FFTs for the STFFT is actually 3N where N is the number of time time bins. To match the temporal resolution between STFFT and wavelet convolution, you'd need 2000 FFTs X 3 frequency bins = 6000 FFTs. So the way I've described it here, it's 41 FFTs for the wavelet analysis and 6000 for the STFFT analysis. But again, this is not trivially always the case; it all depends on how you set things up.

    • @tobi3497
      @tobi3497 4 роки тому

      @@mikexcohen1 from what I've read, you've missed out the convolution step for the wavelet transforms. U use 41 ffts to the relevant frequencies, but you still have to do this for every time step - unless I'm confused.

  • @sukursukur3617
    @sukursukur3617 Рік тому

    Is morlet wavelet useful in image processing? I googled and there are a few study about that.

    • @mikexcohen1
      @mikexcohen1  Рік тому

      Yes, 2D Morlet wavelets are used for filtering images and feature-detection. Although image processing is increasingly done through deep learning models like CNNs.

    • @sukursukur3617
      @sukursukur3617 Рік тому

      @@mikexcohen1 Thank you very much for answer. I actually want to ask whether morlet wavelet can be used as kernel in convolution. Why is it not popular in image processing?

    • @mikexcohen1
      @mikexcohen1  Рік тому

      Yes, Morlet wavelets are commonly used as convolution kernels. They work great for signals and for images. A lot of image processing is moving towards ML/DL/AI techniques.

  • @周建-s2x
    @周建-s2x Рік тому

    May I ask one question¿ why the icwt in matlab of one wavelet coefficient is just a point ¿

  • @mimArmand
    @mimArmand 3 роки тому

    You are a great teacher!
    I wonder what are some practical use-cases of these transformations? Is it to eliminate some of the noise?

    • @mikexcohen1
      @mikexcohen1  3 роки тому +2

      Thanks! Morlet wavelets are often used in time-frequency analysis, i.e., creating a spectrogram, which shows how spectral energy changes over time.

  • @elevate4181
    @elevate4181 2 роки тому

    Can you multiply elementwise the sine waves by non Gaussian distribution?

    • @mikexcohen1
      @mikexcohen1  2 роки тому

      Sure, you can use any tapering function. But a Gaussian is a great to use for several reasons, including the smoothness and spectral properties.

  • @SampleroftheMultiverse
    @SampleroftheMultiverse 2 місяці тому +1

    i love waves

  • @yaarobmohammad1020
    @yaarobmohammad1020 4 роки тому

    Can you please apply wavelet on geophysical data ?
    Thank you so much
    Hope you will answer me about this

    • @mikexcohen1
      @mikexcohen1  4 роки тому

      Sure, you can apply wavelet convolution to any regularly sampled time series data.

  • @BeSaj22
    @BeSaj22 2 роки тому

    A life saver! Thanks for these helpful tutorials!
    I love the dad-tier jokes too btw.

    • @mikexcohen1
      @mikexcohen1  2 роки тому

      Glad you like them! (The videos and the jokes.)

  • @prankinjp
    @prankinjp 5 місяців тому

    Is there python version for this course

    • @mikexcohen1
      @mikexcohen1  5 місяців тому

      Nope, but the book that this course is based on has been translated into Python. See my github repo for the ANTS book for links.
      That said, I do have a signal-processing course (non-neuroscience-related) that is in both MATLAB and Python.

  • @MatthewKelley-mq4ce
    @MatthewKelley-mq4ce 4 місяці тому

    Hehe. 😅 Let's assume I have the background I don't. It'll be fine with time. Topically relevant to what I'm looking into. You're videos seem fun and informative from what little I've seen so far.

    • @mikexcohen1
      @mikexcohen1  4 місяці тому

      You're awesome, Matthew :)

  • @edwardody4838
    @edwardody4838 4 роки тому

    Why are the edges of the Morlet Wavelet only close to 0 and not exactly 0?
    Thanks for the great course!

    • @mikexcohen1
      @mikexcohen1  4 роки тому +2

      Good question. One reason is theoretical: A Morlet wavelet is created by multiplying by a Gaussian [ exp(-x^2) ], and a Gaussian can never be zero; it can only asymptote to zero. The second reason is that due to underflow, digital computers have a hard time computing exact zero, so you end up with tiny numbers like 1e-16.
      On the other hand, there is a "practical zero" here where the values at the edges get so tiny that they have no real effect on the data.

    • @edwardody4838
      @edwardody4838 4 роки тому

      @@mikexcohen1 Hi Mike, thanks for replying to that so quickly! It didn't occur to me that the Gaussian would also never be 0 but it makes sense from the way that it's calculated.

  • @isbestlizard
    @isbestlizard 4 роки тому +1

    Third times the charm. Will this video actually give an example of a wavelet function or continue to handwave

    • @isbestlizard
      @isbestlizard 4 роки тому +1

      YES thank you best wavelet intro on youtube!

  • @ITriedThis
    @ITriedThis 2 роки тому

    Thanks for the video. I'm doing analysis on Pilot Induced Oscillations using Morlet wavelets and this was a great introduction.
    By the way, is a Morlet wavelet a small version of a Mor wavelet?

    • @mikexcohen1
      @mikexcohen1  2 роки тому

      lol, yeah I think so. They're named after Jean Morlet, a French geophysicist. He was the son of Jean Mor, who invented the sine wave. (OK ok, that last sentence is just a joke.)

  • @kevalan1042
    @kevalan1042 3 роки тому +1

    I see that your course uses Matlab, would you consider doing a PyTorch version?

    • @mikexcohen1
      @mikexcohen1  3 роки тому +1

      This whole series is in MATLAB. I have a bunch of other videos and courses in Python. I'm currently working on a course about deep learning in PyTorch, which will probably be finished in the summer...

    • @kevalan1042
      @kevalan1042 3 роки тому

      @@mikexcohen1 thanks for the reply, I like your style

  • @silviamaradacostacampos8971
    @silviamaradacostacampos8971 3 роки тому

    Congratulation for the explanation!!

  • @tienernwoon8637
    @tienernwoon8637 4 роки тому +1

    that joke with let okayyyy srsly AHAHA but before that much thanks and appreciation for your vids! much thorough explanation and Im getting interested along the way!

  • @gr_it_solutions
    @gr_it_solutions 9 місяців тому

    Superb sir

  • @exoticcoder5365
    @exoticcoder5365 Рік тому +1

    7:09 I caught that joke lol

  • @salihaamoura232
    @salihaamoura232 5 років тому +1

    Thank you very much 👏👏👏

  • @ninepuchar1
    @ninepuchar1 19 днів тому

    7:04😂 Good one.

  • @xchen3132
    @xchen3132 4 роки тому +1

    I like the jokelet :3

  • @jyanguas3251
    @jyanguas3251 2 роки тому

    so useful!

  • @aayushbajaj2260
    @aayushbajaj2260 Рік тому +1

    I love you

    • @mikexcohen1
      @mikexcohen1  Рік тому

      Aww... I love you too, aayushbajaj2260.

  • @DataMount1
    @DataMount1 4 роки тому +1

    Best joke ever

  • @mrx42
    @mrx42 4 роки тому

    Great joke buddy ;) Keep on going =)