Skip to content

Deglinting

sensingpy.preprocessing.deglinting implements sun-glint correction algorithms for water remote sensing imagery.

hedley

hedley 

hedley(deep_area_mask: ndarray, to_correct: ndarray, nir: ndarray) -> np.ndarray

Hedley method for deglinting.

This method corrects sun glint in visible bands using a linear regression between NIR and visible bands over optically deep water areas. It assumes a linear relationship between sun glint in NIR and visible bands.

Parameters:

Name Type Description Default
deep_area_mask ndarray

Boolean mask identifying optically deep water areas to use for correction

 required
to_correct ndarray

Array of bands to correct for sun glint

 required
nir ndarray

Near-infrared band values used as the glint predictor

 required

Returns:

Type Description
ndarray

Array of bands with sun glint correction applied
Notes

The algorithm uses the slope of the linear regression between each visible band and the NIR band over deep water to determine the correction factor. Values < 0 after correction are set to NaN.

References

Hedley, J. D., Harborne, A. R., & Mumby, P. J. (2005). Simple and robust removal of sun glint for mapping shallow-water benthos. International Journal of Remote Sensing, 26(10), 2107-2112. https://doi.org/10.1080/01431160500034086

Source code in sensingpy/preprocessing/deglinting.py 
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
def hedley(
    deep_area_mask: np.ndarray, to_correct: np.ndarray, nir: np.ndarray
) -> np.ndarray:
    """
    Hedley method for deglinting.

    This method corrects sun glint in visible bands using a linear regression
    between NIR and visible bands over optically deep water areas. It assumes a
    linear relationship between sun glint in NIR and visible bands.

    Parameters
    ----------
    deep_area_mask : np.ndarray
        Boolean mask identifying optically deep water areas to use for correction
    to_correct : np.ndarray
        Array of bands to correct for sun glint
    nir : np.ndarray
        Near-infrared band values used as the glint predictor

    Returns
    -------
    np.ndarray
        Array of bands with sun glint correction applied

    Notes
    -----
    The algorithm uses the slope of the linear regression between each visible
    band and the NIR band over deep water to determine the correction factor.
    Values < 0 after correction are set to NaN.

    References
    ----------
    Hedley, J. D., Harborne, A. R., & Mumby, P. J. (2005). Simple and robust
    removal of sun glint for mapping shallow-water benthos. International Journal
    of Remote Sensing, 26(10), 2107-2112.
    https://doi.org/10.1080/01431160500034086
    """

    for idx in range(len(to_correct)):
        deep_value = to_correct[idx][deep_area_mask]
        deep_nir = nir[deep_area_mask]
        is_valid = ~np.isnan(deep_value) & ~np.isnan(deep_nir)

        m = np.polyfit(deep_nir[is_valid].ravel(), deep_value[is_valid].ravel(), 1)[0]
        to_correct[idx] = to_correct[idx] - m * (nir - np.nanmin(deep_nir[is_valid]))
        to_correct[idx][to_correct[idx] < 0] = np.nan

    return to_correct

lyzenga

lyzenga 

lyzenga(deep_area_mask: ndarray, to_correct: ndarray, nir: ndarray) -> np.ndarray

Lyzenga method for deglinting.

This method corrects sun glint in visible bands using covariance between NIR and visible bands over optically deep water areas. It is based on the statistical relationship between NIR brightness and glint intensity.

Parameters:

Name Type Description Default
deep_area_mask ndarray

Boolean mask identifying optically deep water areas to use for correction

 required
to_correct ndarray

Array of bands to correct for sun glint

 required
nir ndarray

Near-infrared band values used as the glint predictor

 required

Returns:

Type Description
ndarray

Array of bands with sun glint correction applied
Notes

This implementation uses the covariance between each visible band and the NIR band over deep water to determine the correction factor. The method uses mean NIR rather than minimum NIR as the reference point. Values < 0 after correction are set to NaN.

References

Lyzenga, D. R., Malinas, N. P., & Tanis, F. J. (2006). Multispectral bathymetry using a simple physically based algorithm. IEEE Transactions on Geoscience and Remote Sensing, 44(8), 2251-2259. https://doi.org/10.1109/TGRS.2006.872909

Source code in sensingpy/preprocessing/deglinting.py 
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
def lyzenga(
    deep_area_mask: np.ndarray, to_correct: np.ndarray, nir: np.ndarray
) -> np.ndarray:
    """
    Lyzenga method for deglinting.

    This method corrects sun glint in visible bands using covariance between
    NIR and visible bands over optically deep water areas. It is based on the
    statistical relationship between NIR brightness and glint intensity.

    Parameters
    ----------
    deep_area_mask : np.ndarray
        Boolean mask identifying optically deep water areas to use for correction
    to_correct : np.ndarray
        Array of bands to correct for sun glint
    nir : np.ndarray
        Near-infrared band values used as the glint predictor

    Returns
    -------
    np.ndarray
        Array of bands with sun glint correction applied

    Notes
    -----
    This implementation uses the covariance between each visible band and
    the NIR band over deep water to determine the correction factor. The method
    uses mean NIR rather than minimum NIR as the reference point. Values < 0
    after correction are set to NaN.

    References
    ----------
    Lyzenga, D. R., Malinas, N. P., & Tanis, F. J. (2006). Multispectral
    bathymetry using a simple physically based algorithm. IEEE Transactions on
    Geoscience and Remote Sensing, 44(8), 2251-2259.
    https://doi.org/10.1109/TGRS.2006.872909
    """

    for idx in range(len(to_correct)):
        deep_value = to_correct[idx][deep_area_mask]
        deep_nir = nir[deep_area_mask]
        is_valid = ~np.isnan(deep_value) & ~np.isnan(deep_nir)

        m = np.cov(deep_nir[is_valid].ravel(), deep_value[is_valid].ravel())[0, 1]
        to_correct[idx] = to_correct[idx] - m * (nir - np.nanmean(deep_nir[is_valid]))
        to_correct[idx][to_correct[idx] < 0] = np.nan

    return to_correct

joyce

joyce 

joyce(deep_area_mask: ndarray, to_correct: ndarray, nir: ndarray) -> np.ndarray

Joyce method for deglinting.

This method corrects sun glint by using the statistical mode of NIR values in deep water as the reference point. It combines aspects of both Hedley and Hochberg approaches to glint removal.

Parameters:

Name Type Description Default
deep_area_mask ndarray

Boolean mask identifying optically deep water areas to use for correction

 required
to_correct ndarray

Array of bands to correct for sun glint

 required
nir ndarray

Near-infrared band values used as the glint predictor

 required

Returns:

Type Description
ndarray

Array of bands with sun glint correction applied
Notes

The algorithm uses the slope of the linear regression between each visible band and the NIR band over deep water, similar to Hedley's method. However, it uses the mode of NIR values as the reference point rather than the minimum. Values < 0 after correction are set to NaN.

References

Joyce, K. E. (2004). A method for mapping live coral cover using remote sensing. PhD Thesis, University of Queensland, Brisbane, Australia.

Kay, S., Hedley, J. D., & Lavender, S. (2009). Sun glint correction of high and low spatial resolution images of aquatic scenes: a review of methods for visible and near-infrared wavelengths. Remote Sensing, 1(4), 697-730. https://doi.org/10.3390/rs1040697

Source code in sensingpy/preprocessing/deglinting.py 
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
def joyce(
    deep_area_mask: np.ndarray, to_correct: np.ndarray, nir: np.ndarray
) -> np.ndarray:
    """
    Joyce method for deglinting.

    This method corrects sun glint by using the statistical mode of NIR values
    in deep water as the reference point. It combines aspects of both Hedley and
    Hochberg approaches to glint removal.

    Parameters
    ----------
    deep_area_mask : np.ndarray
        Boolean mask identifying optically deep water areas to use for correction
    to_correct : np.ndarray
        Array of bands to correct for sun glint
    nir : np.ndarray
        Near-infrared band values used as the glint predictor

    Returns
    -------
    np.ndarray
        Array of bands with sun glint correction applied

    Notes
    -----
    The algorithm uses the slope of the linear regression between each visible
    band and the NIR band over deep water, similar to Hedley's method. However,
    it uses the mode of NIR values as the reference point rather than the minimum.
    Values < 0 after correction are set to NaN.

    References
    ----------
    Joyce, K. E. (2004). A method for mapping live coral cover using remote sensing.
    PhD Thesis, University of Queensland, Brisbane, Australia.

    Kay, S., Hedley, J. D., & Lavender, S. (2009). Sun glint correction of high
    and low spatial resolution images of aquatic scenes: a review of methods for
    visible and near-infrared wavelengths. Remote Sensing, 1(4), 697-730.
    https://doi.org/10.3390/rs1040697
    """

    for idx in range(len(to_correct)):
        deep_value = to_correct[idx][deep_area_mask]
        deep_nir = nir[deep_area_mask]
        is_valid = ~np.isnan(deep_value) & ~np.isnan(deep_nir)

        m = np.polyfit(deep_nir[is_valid].ravel(), deep_value[is_valid].ravel(), 1)[0]
        to_correct[idx] = to_correct[idx] - m * (
            nir - mode(deep_nir[is_valid].ravel()).mode
        )
        to_correct[idx][to_correct[idx] < 0] = np.nan

    return to_correct