Modified Hough Spiker Algorithm

modified_hough_spiker(signal: ndarray, window_length: int, threshold: float) → ndarray[source]

Detect spikes in a signal using the Modified Hough Spiker Algorithm.

This function detects spikes in an input signal by incorporating a threshold-based error accumulation mechanism. The signal is compared with a convolution result using a boxcar filter, and the error is accumulated over time. If the error remains within a specified threshold, a spike is detected, and the signal is modified.

Refer to the Modified Hough Spiker Encoding for a detailed explanation of the Modified Hough Spiker Algorithm.

Code Example:

import numpy as np
from spikify.encoding.temporal.deconvolution import modified_hough_spiker
signal = np.array([0.1, 0.2, 0.3, 1.0, 0.5, 0.3, 0.1])
window_length = 3
threshold = 0.5
spikes = modified_hough_spiker(signal, window_length, threshold)
Parameters:

  • signal (numpy.ndarray) – The input signal to be analyzed. This should be a numpy ndarray.

  • window_length (int) – The length of the boxcar filter window.

  • threshold (float) – The threshold value for error accumulation.

Returns:

A 1D numpy array representing the detected spikes.

Return type:

numpy.ndarray

Raises:

  • ValueError – If the input signal is empty or if the window length is greater than the signal length.

  • TypeError – If the signal is not a numpy ndarray.