This project aims to perform feature extraction on time and frequency domain signals and subsequently classify them into three categories: Tensile, Shear, and Mixed. The features extracted from the signals include both time and frequency domain characteristics.
- NumPy
- SciPy
- Matplotlib
- Scikit-Learn
- Pandas
The training and testing datasets are loaded from MAT files (x_train.mat and y_train.mat for training, and Testing_dataset_1650AE_1000_samples.mat for testing) using SciPy's loadmat function.
The features_extraction function is defined to extract various statistical features from both time and frequency domains. These features include minimum, maximum, mean, RMS (Root Mean Square), variance, standard deviation, power, peak value, peak-to-peak value, crest factor, skewness, kurtosis, and various statistical measures from the frequency domain.
The extracted features are computed for both training and testing datasets. The results are stored in Pandas DataFrames for better visualization and organization.
The training data is standardized using Scikit-Learn's StandardScaler. Principal Component Analysis (PCA) is then applied to reduce the dimensionality of the feature space while retaining important information.
A Linear Discriminant Analysis (LDA) model is trained on the transformed feature space using the training data. The model is trained to classify signals into three classes: Tensile, Shear, and Mixed.
The testing data is standardized and transformed using the same scaler and PCA model. The trained LDA model is then used to predict the classes of the testing data. The results are visualized by plotting the first two principal components against each other, with different colors indicating different classes. Additionally, predicted classes for the testing data are marked with 'x' symbols on the plot.