Filtering

Python Functions

actag.median_filter.divide_array(arr_len: int, num_sections: int) list

Divides an array into a specified number of sections and returns a list containing the start and end indices for each section.

Parameters:
  • arr_len (int) – Length of the array.

  • num_sections (int) – Number of sections to divide the array into.

Returns:

List containing the start and end indices for each section.

Return type:

list

actag.median_filter.filter_sub_image(sub_img: ndarray, kernel_radius: int) ndarray

Filters a sub-image using a median filter.

Parameters:
  • sub_img (np.ndarray) – Sub-image to filter.

  • kernel_radius (int) – Radius of the kernel.

Returns:

Filtered sub-image.

Return type:

np.ndarray

actag.median_filter.median_filter(img: ndarray, kernel_radius: int) ndarray

Median filter for 2D images.

Parameters:
  • img (np.ndarray) – 2D image.

  • kernel_radius (int) – Radius of the kernel.

Returns:

Filtered image.

Return type:

np.ndarray

actag.median_filter.median_filter_multiprocessed(img: ndarray, kernel_radius: int, num_processes: int | None = None) ndarray

Same as median_filter(), but leverages multiprocessing to speed up computation time.

Parameters:
  • img (np.ndarray) – 2D image.

  • kernel_radius (int) – Radius of the kernel.

  • num_processes (int, default_value=None) – The number of processes to use. The default is None, which will attempt to use the maximum number of CPUs available, as reported by the OS.

Returns:

Filtered image.

Return type:

np.ndarray

Rust Functions

actag.rust_impl.median_filter()

Rust equivalent of actag.median_filter.median_filter().

Parameters:
  • data – 2D image.

  • radius (usize) – Radius of the kernel.

Returns:

Filtered image.

Return type:

Vec<Vec<i32>>

actag.rust_impl.median_filter_multithread()

Rust equivalent of actag.median_filter.median_filter_multiprocessed().

Parameters:
  • data – 2D image.

  • radius (usize) – Radius of the kernel.

  • cpu_option – The number of CPUS to use. If None, all available CPUs will be used.

Returns:

Filtered image.

Return type:

Vec<Vec<i32>>