| Feature | Description |
|---|---|
| Max | Calculates the maximum value in the window. |
| Min | Calculates the minimum value in the window. |
| Mean | Calculates the arithmetic mean of the window. |
| Range | Calculates the difference between the maximum and minimum values of a signal within a given time window or measurement period. It is a simple yet informative feature that provides insight into the signal's amplitude variation. |
| Absolute Mean | Calculates the average of the absolute values of the samples, measuring the average signal magnitude while ignoring polarity (positive or negative values). It is a simple way to estimate signal energy and intensity, similar to RMS but less compute-intensive. |
| Standard Deviation | Quantifies the amount of variation or dispersion in a set of data points from their mean value. In feature extraction, standard deviation is often used alongside other statistical measures such as mean, variance, skewness, and kurtosis to provide a comprehensive representation of signal characteristics. It is particularly useful in: • Activity recognition: helping distinguish between high-variance and low-variance movements in wearable devices. • Anomaly detection: identifying unusual patterns or outliers in data. • Audio processing: measuring the dynamic range of audio signals. • Image processing: extracting texture features from images. |
| Mean Absolute Deviation | Quantifies the variability or dispersion of a signal. It is calculated as the average of the absolute differences between each data point and the mean of the dataset. |
| Root Mean Square | Calculates the root of the arithmetic mean of the squares of a set of numbers. |
| Kurtosis | Measures the combined weight of a distribution's tails relative to the center of the distribution. |
| Skewness | Measures the asymmetry of the distribution of a variable. |