SNR

hyperquest.snr.hrdsdc(path_to_data, n_segments=200, compactness=0.1, n_pca=3, ncpus=1, snr_in_db=False, mask_waterbodies=True, no_data_value=-9999)

Homogeneous regions division and spectral de-correlation (Gao et al., 2008).

Neighbor pixel with band K is included in MLR, ( p_k-1 , p_k+1, p+1_k ).

Parameters:

  • path_to_data (str) – Path to the .hdr or .nc file that contains hyperspectral data.

  • n_segments (int) – The (approximate) number of labels in the segmented output image. see skimage.segmentation.slic for more.

  • compactness (float) – Balances color proximity and space proximity. Higher values give more weight to space proximity, making superpixel shapes more square/cubic.see skimage.segmentation.slic for more.

  • ncpus (int, optional) – Number of CPUs for parallel processing. Default is 1.

  • include_neighbor_pixel_in_mlr (bool, optional) – If True, neighbor pixel is used in MLR (for k`). Else, MLR only contains spectral data (k+1, k-1).

  • snr_in_db (bool, optional) – Whether the SNR is in dB. Default is False.

  • mask_waterbodies (bool, optional) – Whether to mask water bodies based on an NDWI threshold of 0.25. Default is True.

  • no_data_value (int or float) – Value used to describe no data regions.

Returns:

A tuple containing the Signal-to-Noise Ratio (SNR) and noise variance with respect to wavelength.

Return type:

tuple

hyperquest.snr.rlsd(path_to_data, block_size, nbins=150, ncpus=1, snr_in_db=False, mask_waterbodies=True, no_data_value=-9999)

Residual-scaled local standard deviation (Gao et al., 2007).

Neighbor pixel not included in MLR, ( p_k-1 , p_k+1).

Parameters:

  • path_to_data (str) – Path to the .hdr or .nc file that contains hyperspectral data.

  • block_size (int) – Block size for partitioning (e.g., 5 would be 5x5 pixels).

  • nbins (int, optional) – Number of bins for histogram analysis. Default is 150.

  • ncpus (int, optional) – Number of CPUs for parallel processing. Default is 1.

  • snr_in_db (bool, optional) – Whether the SNR is in dB. Default is False.

  • mask_waterbodies (bool, optional) – Whether to mask water bodies based on an NDWI threshold of 0.25. Default is True.

  • no_data_value (int or float) – Value used to describe no data regions.

Returns:

A tuple containing the Signal-to-Noise Ratio (SNR) and noise variance with respect to wavelength.

Return type:

tuple

hyperquest.snr.ssdc(path_to_data, block_size, nbins=150, ncpus=1, snr_in_db=False, mask_waterbodies=True, no_data_value=-9999)

Spectral and spatial de-correlation (Roger & Arnold, 1996).

Neighbor pixel with band K is included in MLR, ( p_k-1 , p_k+1, p+1_k ).

Parameters:

  • path_to_data (str) – Path to the .hdr or .nc file that contains hyperspectral data.

  • block_size (int) – Block size for partitioning (e.g., 5 would be 5x5 pixels).

  • nbins (int, optional) – Number of bins for histogram analysis. Default is 150.

  • ncpus (int, optional) – Number of CPUs for parallel processing. Default is 1.

  • snr_in_db (bool, optional) – Whether the SNR is in dB. Default is False.

  • mask_waterbodies (bool, optional) – Whether to mask water bodies based on an NDWI threshold of 0.25. Default is True.

  • no_data_value (int or float) – Value used to describe no data regions.

Returns:

A tuple containing the Signal-to-Noise Ratio (SNR) and noise variance with respect to wavelength.

Return type:

tuple