Questions tagged [machine-learning]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
0
votes
0answers
30 views

Is it possibel to use CNN to bandpass signal?

I have a time serie dataset and want to train a CNN-LSTM model to predict as well as detect outliers. How can I use CNN to filter the signal and extract features from specific frequency band?
0
votes
0answers
20 views

What image augmentations can help a neural network identify the lowest-value pixels within an image?

I am training a CNN to identify objects, and I believe the network will learn much faster if it can learn to focus on the pixels with the lowest value. One way to go about this would be to augment the ...
3
votes
1answer
90 views

Modern Method for 1D Signal Segmentation

I want to segment a signal in an unsupervised manner. The data is a 1D signal which has different segments which I want to be able to segment automatically for farther processing. I am looking for a ...
3
votes
1answer
39 views

The Meaning of $ \mathbb{E} $ Operator in the Pix2Pix Loss Formula of a Neural Network / Convolutional Neural Network

I've been observing the Pix2Pix Paper - Image to Image Translation with Conditional Adversarial Networks and wondered on formulas. For example, the objective of the CGAN is: , where x - observed image,...
0
votes
1answer
29 views

References of deep learning and ai for dsp researchers [closed]

Are there excellent references for machine learning, deep learning and ai for DSP researchers?
1
vote
0answers
19 views

Backpropagation in a Network [closed]

Hi does anyone know how to solve this. Im really struggling to figuring this one out. I am unclear of how the backpropagation part would work in this network and how to calculate it for each weight. ...
0
votes
2answers
38 views

Power Spectral Density as a single number confusion

I'm trying to recreate the results of a machine learning applied to the DSP classification in the article: link. I have a signal (activity measurements from a smartwatch) per patient, so about 30 ...
1
vote
1answer
55 views

Can a linear reconstruction in compressive sensing perform well?

I am trying to implement compressive sensing for grayscale 2D images, then reconstructing them using a multi-layer perceptron(MLP). It seems to perform well no matter how many layers I add or remove, ...
0
votes
0answers
16 views

Measure of Separability for Classification Problems

Hello are there any methods, for a given set of features (possibly high dimensional) and labels (class 1 and class 0), to measure how "separable" the data is? E.g., if the problem was 2 ...
2
votes
2answers
77 views

Recommendation for courses / studies on digital signal processing

I hold a master's degree in mechanical engineering. However at my job I am more and more diving into topics of signal processing and data science. I find it great to discover about new topics and to ...
0
votes
0answers
13 views

Is my interpretation of CAR filter correct?

I am a computer scientist working on a Brain-Computer Interface (BCI) project. So, i am using EEG data from a dataset and classifying them with the use of machine learning algorithms. Initial results ...
0
votes
1answer
73 views

Using STFT as an input to a Neural Net

I'm trying to use the STFT as an input to a neural network. After flattening, there are over 4,000 features for a few seconds of audio. Is there a recommended way of summarising these to be a more ...
0
votes
0answers
9 views

FPR for time series of 1 hour epilepsy recording

I am working on epilepsy seizure prediction classification using a convolution neural network. My dataset consisting of multiple recording of each record for 1 hour, first I segmented each record into ...
0
votes
0answers
15 views

Use of spectrograms in using deep learning on EEG related problems

I am currently working on a classification problem related to EEG signals. My professor asked me to convert those signals into their spectrograms and try various convolutional neural networks on them. ...
0
votes
0answers
15 views

Feature Based Methods for Quality Inspection Problem

I am currently working on an image classification problem to classify thermal images (can be interpreted as false-color images) of certain products as with or without defect. From an initial search in ...
1
vote
1answer
60 views

a neural network approach for FIR filter

I am trying to write a code for a neural network to do the digital filtering on some signals. Is there any neural network model for digital filtering?
0
votes
0answers
26 views

Source Localization in the context of EEG

I am having difficulty understanding how source localization is used in Brain-Computer Interface (BCI) as a feature extraction method. As a bit of introduction, the measured activity (Matrix M) is a ...
3
votes
2answers
63 views

Help with denoising signal and periodogram analysis resources

This is a cross posting from the crossvalidated stack exchange as I thought this may be a better forum to ask. I have a dataset consisting of respiratory time series signals of different lengths ...
2
votes
1answer
462 views

Upsampling vs downsampling. Which to use when?

Downsampling reduces dimensionality of the features while losing some information. It saves computation. Upsampling brings back the resolution to the resolution of previous layer. My question is which ...
0
votes
0answers
32 views

Finding arbitrary/random repetitive patterns in signals (both self and across two signals)

I am trying to figure out a direction for my research. I need to find random repetitive patterns between two signals and on each individual signal. I have read about FFT and time series-motif ...
0
votes
0answers
24 views

interpreting the post filter features in a sensor data

I am new to signal processing and I am working on Filtring signals(using basic analog filters, FIR filters, and IRR filters), for Human activity recognition, using sensor data, I Extract the features ...
0
votes
0answers
12 views

Applying ICA on 3rd order cumulant

Background: I was reading the article Application of Higher Order Statistics for Atrial Arrhythmia Classification, and they mentioned using the higher order cumulant $$C_3^x = m_3^x(i, j)=E[x(n)x(n+i)...
0
votes
0answers
37 views

Which audio features to use for content based music recommendation

I am attempting to write a content-based music recommendation system using machine learning. Using a python library, I am able to extract the features from raw audio files. For each audio file, I have ...
1
vote
1answer
63 views

how to handle different durations of audio data?

I am new to signal preprocessing, I read about mel_spectrograms, MFCC's. Now I want to apply it and use the CNN model, But the data which I have for practice is having audio of different durations, ...
1
vote
1answer
57 views

Understanding noise removal method using wavelets

I am trying to understand how wavelet transform can be used to denoise a time series or signal and how to plot the scalogram image. My signal has a lot of fluctuations and as such I am finding it ...
0
votes
2answers
128 views

What is the current “state of the art” in future audio sample prediction, given recent developments in “image inpainting” using deep learning?

In the last few years, there have been many breakthroughs in the image processing world regarding repairing "damaged" images, images with corrupt pixels, or even reversing artefacts from ...
0
votes
0answers
55 views

DWT and FFT on a very low frequency signals

I am trying to extract features from a very low frequency signal of $0.1$ HZ and less than $100$ samples. Is it advisable to use FFT and DWT on this signal?
0
votes
0answers
26 views

multi-frame image restoration

Suppose we have a sequence of still images each of which has been contaminated by some particles(ex, dust/sand/smoke) making the images very poor in certain areas. What approach would be best to teach ...
4
votes
2answers
227 views

Classic Signal Processing vs Deep Learning / Machine Learning (DNN / ML) Based Signal Processing

Are classic signal processing/statistics based approaches to optimum detection/estimation still relevant/important compared to ML based approaches using DL? There was a time when speech processing was ...
1
vote
1answer
25 views

Why not use synthetic datasets for training machine vision (deep learning) algorithms?

I am studying some examples of how to perform 3D instance segmentation in indoor scenes, and I have noticed many of the available datasets are from real environments. I was wondering with the ...
2
votes
1answer
63 views

Minimize the Cost Function of Values of Vectors Based on Their Amplitude

I have two vectors $X = [x_1,x_2,x_3,x_4]$; and $Y = [y_1,y_2,y_3,y_4]$; I know that $|x_1|$ = $|y_1|$, and $|x_2|$ = $|y_2|$,... so on. it means the difference is only in the sign. it might be ...
2
votes
2answers
35 views

Mathematics behind the creation of 3d face mask at runtime

I'm new in Image processing and Machine learning area. I want to create 3d mask of face at runtime. ...
0
votes
0answers
77 views

MWE of zero-sum mask in the context of machine learning

Federated Learning seems to have a lot of potential, I understand it can be done using different techniques, e.g. functional encryption, etc. The simplest seems to be zero-sum masks. For TensorFlow ...
1
vote
0answers
40 views

Why do people use STFT as a preprocessing step to using CNN?

Just briefly looking up some research papers on audio data and I have come across some papers that use STFT as a preprocessing step to using CNN. Why is this the case? What are the advantages and ...
0
votes
1answer
49 views

Automatic detection of noise-only segments in audio

This is my first question for this community. My connection to DSP comes via machine learning/deep learning, and I am working in the Python ecosystem. As a preprocessing step to audio classification, ...
0
votes
0answers
20 views

How can a CNN account for spectro-temporal constraints in neural data?

What are there the best ways to leverage the unique "geometrical" constraints of spectro-temporal signal representations (architecture, filter shapes, data augmentation, etc.)? For example, ...
0
votes
0answers
80 views

Hamilton-Jacobi-Bellman equation vs Riccati equation

Most of the literature on Reinforcement Learning discuss Hamilton Jacobi Bellman equations for optimality. In dynamics however Ricatti equations are used. I am curious if there are any parallels ...
0
votes
0answers
40 views

How to deal with different audio formats for audio classification?

I am working on an audio classification problem statement to classify between two audio classes. I have collected samples from jotform, they are providing audio widget to collect .wav audio but it ...
0
votes
1answer
221 views

Cochleagram vs STFT for CNN-based speech segregation

I’m working on a project (python-based) that would use ideal ratio masks (IRMs) as a basis for cleaning noisy speech in various environments. Specifically, this will be accomplished through the use of ...
1
vote
1answer
29 views

FFT question as it relates to Neural Networks / Supervised Learning Models

So I've been researching wavelet transforms and FFT. I want to feed wavelet transforms of a 1D signal in time into a neural net and train against a target variable at each time step. The idea being ...
0
votes
2answers
70 views

Understanding MFCCs

I am doing research about emotion recognition from speech, by applying machine learning. Most papers are recommending using MFCC features. Therefore, I am currently trying to understand the underlying ...
0
votes
0answers
38 views

Classifying motor imagery from EEG data: Feature selection

I am an undergraduate student who is trying to classify motor imagery from EEG data! I have no experience working with EEG or any neuroscience background, I only have a very basic knowledge of how ...
0
votes
0answers
66 views

Feature extraction for exponentially damped signals

I am looking into exponentially damped signals where it is a stationary signal (after implementing the Adfuller statistical test) and I would like to look into how can I extract meaningful features ...
0
votes
0answers
74 views

How do i calculate the accuracy of an algoruthm?

I'm doing research on an machine learning image processing algorithm (segmentation algorihtm)i.e. computer vision performs when an image takes as an input and objects in an image are segmented and ...
0
votes
1answer
29 views

Using MFCCs for acoustic machine failure prediction

MFCCs are ubiquitously extracted for speech processing tasks, but I would like to know how suitable they are for non-speech processing tasks. Intuitively, it is my understanding that MFCCs are ...
1
vote
0answers
33 views

Is it useful to apply edge detection filter as a preprocessing step before applying CNN?

Background: my future boss would like me to apply a number of low level filters to industrial images before applying machine learning methods on them. From my knowledge this doesn't make any sense as ...
0
votes
0answers
18 views

How to update Kalman Filter with n-th order state translation?

In Kalman Filter, the hidden state translation is defined by $X_t=F_tX_{t-1}+W_t$, where $X_t$ can be a vector or a single value. This is actually derived from Bayes filter, in which 1-th order markov ...
0
votes
1answer
70 views
0
votes
0answers
80 views

Frequency contour detection in spectrogram

I have an image of a spectrogram and I wish to detect the tracks/contours of prominent frequencies present in the spectrogram. In the end, I want to be able to get various prominent curves from the ...
0
votes
1answer
52 views

How to prepare different input size in CNN

I'm doing a research about personality identification based on their signature using CNN method, however the learning feature for the personality traits have a different input size. I understand that ...