8.1.4. hstools

class hs_process.hstools(spyfile)

Bases: object

Basic tools for manipulating Spyfiles and accessing their metadata.

Parameters

spyfile (SpyFile object) – The datacube being accessed and/or manipulated.

Methods Summary

clean_md_sets([metadata])	Converts metadata items that are expressed as a list to be expressed as a dictionary.
del_meta_item(metadata, key)	Deletes metadata item from SpyFile object.
get_band(target_wl[, spyfile])	Finds the band number of the closest target wavelength.
get_wavelength(target_band[, spyfile])	Returns actual wavelength of the closest target band.
get_wavelength_range(range_bands[, index, …])	Retrieves the wavelengths for all bands within a band range.
get_center_wl(wl_list[, spyfile, wls])	Gets band numbers and mean wavelength from all wavelengths (or bands) in wl_list.
get_band_index(band_name)	.
get_spectral_mean(band_list[, spyfile])	Gets the spectral mean of a datacube from a list of bands.
get_band_num(band_idx)	Adds 1 to band_idx and returns the band number(s).
get_band_range(range_wl[, index, spyfile])	Retrieves the band index or band name for all bands within a wavelength range.
get_meta_set(meta_set[, idx])	Reads metadata “set” (i.e., string representation of a Python set; common in .hdr files), taking care to remove leading and trailing spaces.
get_UTM(pix_e_ul, pix_n_ul, utm_x, utm_y, …)	Calculates the new UTM coordinate of cropped plot to modify the “map info” tag of the .hdr file.
load_spyfile(spyfile)	Loads a SpyFile (Spectral Python object) for data access and/or manipulation by the hstools class.
mask_array(array, metadata[, thresh, …])	Creates a masked numpy array based on a threshold value.
mean_datacube(spyfile[, mask, nodata])	Calculates the mean spectra for a datcube; if mask is passed (as a numpy.ndarray), then the mask is applied to spyfile prior to computing the mean spectra.
modify_meta_set(meta_set, idx, value)	Modifies metadata “set” (i.e., string representation of a Python set; common in .hdr files) by converting string to list, then adjusts the value of an item by its index.
plot_histogram(array[, fname_fig, title, …])	Plots a histogram with the percentile value labeled.

Methods Documentation

clean_md_sets(metadata=None)

Converts metadata items that are expressed as a list to be expressed as a dictionary.

Parameters

metadata (dict, optional) – Metadata dictionary to clean

Returns

metadata_out (dict) – Cleaned metadata dictionary.

Return type

dict

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Create sample metadata with “wavelength” expressed as a list of strings

>>> metadata = {'samples': 1300,
                'lines': 617,
                'bands': 4,
                'file type': 'ENVI Standard',
                'wavelength': ['394.6', '396.6528', '398.7056',
                '400.7584']}

Clean metadata using hstools.clean_md_sets. Notice how wavelength is now expressed as a str representation of a dict, which is required for properly writing the metadata to the .hdr file via save_image() in Spectral Python.

>>> io.tools.clean_md_sets(metadata=metadata)
{'samples': 1300,
 'lines': 617,
 'bands': 4,
 'file type': 'ENVI Standard',
 'wavelength': '{394.6, 396.6528, 398.7056, 400.7584}'}

del_meta_item(metadata, key)

Deletes metadata item from SpyFile object.

Parameters

• metadata (dict) – dictionary of the metadata

• key (str) – dictionary key to delete

Returns

metadata (dict) – Dictionary containing the modified metadata.

Return type

dict

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Create sample metadata

>>> metadata = {'samples': 1300,
                'lines': 617,
                'bands': 4,
                'file type': 'ENVI Standard',
                'map info': '{UTM, 1.0, 1.0, 421356.76707299997, 4844936.7317699995, 0.04, 0.04, 15, T, WGS-84, units  meters, rotation  0.000}',
                'wavelength': ['394.6', '396.6528', '398.7056',
                '400.7584']}

Delete “map info” from metadata using hstools.del_met_item

>>> io.tools.del_meta_item(metadata, 'map info')
{'samples': 1827,
 'lines': 617,
 'bands': 4,
 'file type': 'ENVI Standard',
 'wavelength': ['394.6', '396.6528', '398.7056', '400.7584']}

get_band(target_wl, spyfile=None)

Finds the band number of the closest target wavelength.

Parameters

• target_wl (int or float) – the target wavelength to retrieve the band number for (required).

• spyfile (SpyFile object, optional) – The datacube being accessed and/or manipulated; if None, uses hstools.spyfile (default: None).

Returns

key_band (int) – band number of the closest target wavelength (target_wl).

Return type

int

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Use hstools.get_band to find the band number corresponding to 703 nm

>>> io.tools.get_band(703, io.spyfile)
151

get_wavelength(target_band, spyfile=None)

Returns actual wavelength of the closest target band.

Parameters

• target_band (int or float) – the target band to retrieve wavelength number for (required).

• spyfile (SpyFile object, optional) – The datacube being accessed and/or manipulated; if None, uses hstools.spyfile (default: None).

Returns

key_wavelength (float) – wavelength of the closest target band (target_band).

Return type

float

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Use hstools.get_wavelength to find the wavelength value corresponding to the 151st band

>>> io.tools.get_wavelength(151, io.spyfile)
702.52

get_wavelength_range(range_bands, index=True, spyfile=None)

Retrieves the wavelengths for all bands within a band range.

Parameters

• range_bands (list) – the minimum and maximum band number to consider; values should be int.

• index (bool) – Indicates whether the bands in range_bands denote the band number (False; min=1) or the index number (True; min=0) (default: True).

Returns

wavelength_list (list): A list of all wavelengths between a range in band numbers or index values (depending how index is set).

Return type

list

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_hdr = r'F:\nigo0024\Documents\GitHub\hs_process\hs_process\data\Wells_rep2_20180628_16h56m_test_pika_gige_7-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_hdr)

Find the wavelengths from the 16th to 21st bands

>>> io.tools.get_wavelength_range([16, 21], index=False, spyfile=io.spyfile)
[425.392, 427.4448, 429.4976, 431.5504, 433.6032, 435.656]

Find the wavelengths from the 16th to the 21st index

>>> io.tools.get_wavelength_range([16, 21], index=True, spyfile=io.spyfile)
[427.4448, 429.4976, 431.5504, 433.6032, 435.656, 437.7088]

get_center_wl(wl_list, spyfile=None, wls=True)

Gets band numbers and mean wavelength from all wavelengths (or bands) in wl_list.

Parameters

• wl_list (list) – the list of bands to get information for (required).

• spyfile (SpyFile object) – The datacube being accessed and/or manipulated; if None, uses hstools.spyfile (default: None).

• wls (bool) – whether wavelengths are passed in wl_list or if bands are passed in wl_list (default: True - wavelenghts passed).

Returns

2-element tuple containing

• bands (list): the list of bands (band number) corresponding to wl_list.

• wls_mean (float): the mean wavelength from wl_list.

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Using hstools.get_center_wl, find the bands and actual mean wavelength of the bands closest to 700 and 710 nm.

>>> bands, wls_mean = io.tools.get_center_wl([700, 710], wls=True)
>>> bands
[150, 155]
>>> wls_mean
705.5992

get_band_index(band_name)

. Returns the index of band from “band names”.

Parameters

band_name (int or list) – the target band to retrieve the band index for (required).

Returns

band_idx (int) – band index of the passed band name (band_name).

Return type

int or list

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Using hstools.get_band_index, find the band index of bands 4, 43, and 111.

>>> io.tools.get_band_index([4, 43, 111])
[3, 42, 110]

get_spectral_mean(band_list, spyfile=None)

Gets the spectral mean of a datacube from a list of bands.

Parameters

• band_list (list) – the list of bands to calculate the spectral mean for on the datacube (required).

• spyfile (SpyFile object or numpy.ndarray) – The datacube being accessed and/or manipulated; if None, uses hstools.spyfile (default: None).

Returns

array_mean (numpy.array or pandas.DataFrame): The mean reflectance from spyfile for the bands in band_list.

Return type

numpy.array or pandas.DataFrame

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Calculate the spectral mean of the datacube via hstools.get_spectral_mean using all bands between 800 and 840 nm

>>> band_list = io.tools.get_band_range([800, 840], index=False)
>>> array_mean = io.tools.get_spectral_mean(band_list, spyfile=io.spyfile)
>>> io.show_img(array_mean)

get_band_num(band_idx)

Adds 1 to band_idx and returns the band number(s).

Parameters

band_idx (int or list) – the target band index(es) to retrive the band number for (required).

Returns

band_num (int or list): band number of the passed band index (band_idx).

Return type

int or list

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Using hstools.get_band_num, find the band number located at the 4th, 43rd, and 111th index values.

>>> io.tools.get_band_num([4, 43, 111])
[5, 44, 112]

get_band_range(range_wl, index=True, spyfile=None)

Retrieves the band index or band name for all bands within a wavelength range.

Parameters

• range_wl (list) – the minimum and maximum wavelength to consider; values should be int or float.

• index (bool) – Indicates whether to return the band number (False; min=1) or to return index number (True; min=0) (default: True).

Returns

band_list (list): A list of all bands (either index or number, depending on how index is set) between a range in wavelength values.

Return type

list

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> data_dir = r'F:\nigo0024\Documents\hs_process_demo'
>>> fname_in = os.path.join(data_dir, 'Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr')
>>> io = hsio(fname_in)

Find the band name of all bands between 700 and 710 nm

>>> io.tools.get_band_range([700, 710], index=False, spyfile=io.spyfile)
[150, 151, 152, 153, 154]

Find the band index values of all bands between 700 and 710 nm via hstools.get_band_range

>>> io.tools.get_band_range([700, 710], index=True, spyfile=io.spyfile)
[149, 150, 151, 152, 153]

Sometimes “band names” are not integers in sequential order. To demonstrate the utility of the index parameter, let’s take a look at Sentinel 2A mimicked imagery.

>>> data_dir = r'F:\nigo0024\Documents\hs_process_demo\spec_mod'
>>> fname_in = os.path.join(data_dir, 'Wells_rep2_20180628_16h56m_pika_gige_7-mimic-s2a.bip.hdr')
>>> io = hsio(fname_in)

Find the band name of all bands between 760 and 840 nm

>>> io.tools.get_band_range([760, 840], index=False, spyfile=io.spyfile)
['S2A_SR_AV_B7', 'S2A_SR_AV_B8']

Find the band index values of all bands between 760 and 840 nm via hstools.get_band_range

>>> io.tools.get_band_range([760, 840], index=True, spyfile=io.spyfile)
[6, 7]

get_meta_set(meta_set, idx=None)

Reads metadata “set” (i.e., string representation of a Python set; common in .hdr files), taking care to remove leading and trailing spaces.

Parameters

• meta_set (str) – the string representation of the metadata set

• idx (int) – index to be read; if None, the whole list is returned (default: None).

Returns

metadata_list (list or str): List of metadata set items (as str), or if idx is not None, the item in the position described by idx.

Return type

list or str

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Retrieve the “map info” set from the metadata via hstools.get_meta_set

>>> map_info_set = io.spyfile.metadata['map info']
['UTM',
 '1.0',
 '1.0',
 '441357.287073',
 '4855944.7717699995',
 '0.04',
 '0.04',
 '15',
 'T',
 'WGS-84',
 'units  meters',
 'rotation  0.000']

get_UTM(pix_e_ul, pix_n_ul, utm_x, utm_y, size_x, size_y)

Calculates the new UTM coordinate of cropped plot to modify the “map info” tag of the .hdr file.

Parameters

• pix_e_ul (int) – upper left column (easting) where image cropping begins.

• pix_n_ul (int) – upper left row (northing) where image cropping begins.

• utm_x (float) – UTM easting coordinates (meters) of the original image (from the upper left).

• utm_y (float) – UTM northing coordinates (meters) of the original image (from the upper left).

• size_x (float) – Ground resolved distance of the image pixels in the x (easting) direction (meters).

• size_y (float) – Ground resolved distance of the image pixels in the y (northing) direction (meters).

Returns

2-element tuple containing

• utm_x_new (float): The modified UTM x coordinate (easting) of cropped plot.

• utm_y_new (float): The modified UTM y coordinate (northing) of cropped plot.

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Retrieve UTM coordinates and pixel sizes from the metadata

>>> map_info_set = io.spyfile.metadata['map info']
>>> utm_x = io.tools.get_meta_set(map_info_set, 3)
>>> utm_y = io.tools.get_meta_set(map_info_set, 4)
>>> spy_ps_e = float(map_info_set[5])
>>> spy_ps_n = float(map_info_set[6])

Calculate the UTM coordinates at the 100th easting pixel and 50th northing pixel using hstools.get_UTM

>>> ul_x_utm, ul_y_utm = io.tools.get_UTM(100, 50,
                                          utm_x, utm_y,
                                          spy_ps_e,
                                          spy_ps_n)
>>> ul_x_utm
441360.80707299995
>>> ul_y_utm
4855934.691769999

load_spyfile(spyfile)

Loads a SpyFile (Spectral Python object) for data access and/or manipulation by the hstools class.

Parameters

spyfile (SpyFile object) – The datacube being accessed and/or manipulated.

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Load a new datacube using hstools.load_spyfile

>>> io.tools.load_spyfile(io.spyfile)
>>> io.tools.spyfile
Data Source:   'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip'
    # Rows:            617
    # Samples:        1300
    # Bands:           240
    Interleave:        BIP
    Quantization:  32 bits
    Data format:   float32

mask_array(array, metadata, thresh=None, percentile=None, side='lower')

Creates a masked numpy array based on a threshold value. If array is already a masked array, that mask is maintained and the new mask(s) is/ are added to the original mask.

Parameters

• array (numpy.ndarray) – The data array to mask.

• thresh (float or list) – The value for which to base the threshold; if thresh is list and side is None, then all values in thresh will be masked; if thresh is list and side is not None, then only the first value in the list will be considered for thresholding (default: None).

• percentile (float) – The percentile of pixels to mask; if percentile = 95 and side = ‘lower’, the lowest 95% of pixels will be masked prior to calculating the mean spectra across pixels (default: None; range: 0-100).

• side (str) – The side of the threshold for which to apply the mask. Must be either ‘lower’, ‘upper’, ‘outside’, or None; if ‘lower’, everything below the threshold will be masked; if ‘outside’, the thresh / percentile parameter must be list-like with two values indicating the lower and upper bounds - anything outside of these values will be masked out; if None, only the values that exactly match the threshold will be masked (default: ‘lower’).

Returns

2-element tuple containing

• array_mask (numpy.ndarray): The masked numpy.ndarray based on the passed threshold and/or percentile value.

• metadata (dict): The modified metadata.

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Retrieve the image band at 800 nm using hstools.get_band and hsio.spyfile.open_memmap

>>> band = io.tools.get_band(800)
>>> array = io.spyfile.open_memmap()[:, :, band]

Create a masked array of all values below the 75th percentile via hstools.mask_array

>>> array_mask, metadata = io.tools.mask_array(array, io.spyfile.metadata, percentile=75, side='lower')

See that the “history” tage in the metadata has been modified

>>> metadata['history'][-158:]
"hs_process.mask_array[<label: 'thresh?' value:None; label: 'percentile?' value:75; label: 'side?' value:lower; label: 'unmasked_pct?' value:24.9935170178282>]"

Visualize the unmasked array using hsio.show_img. Set vmin and vmax to ensure the same color scale is used in comparing the masked vs. unmasked arrays.

>>> vmin = array.min()
>>> vmax = array.max()
>>> io.show_img(array, vmin=vmin, vmax=vmax)

Visualize the unmasked array using hsio.show_img

>>> io.show_img(array_mask, vmin=vmin, vmax=vmax)

mean_datacube(spyfile, mask=None, nodata=0)

Calculates the mean spectra for a datcube; if mask is passed (as a numpy.ndarray), then the mask is applied to spyfile prior to computing the mean spectra.

Parameters

• spyfile (SpyFile object or numpy.ndarray) – The hyperspectral datacube to mask.

• mask (numpy.ndarray) – the mask to apply to spyfile; if mask does not have similar dimensions to spyfile, the first band (i.e., first two dimensions) of mask will be repeated n times to match the number of bands of spyfile (default: None).

• nodata (float or None) – If None, treats all pixels cells as they are repressented in the numpy.ndarray. Otherwise, replaces nodata with np.nan and these cells will not be considered when calculating the mean spectra.

Returns

3-element tuple containing

• spec_mean (SpyFile.SpyFile object): The mean spectra from the input datacube.

• spec_std (SpyFile.SpyFile object): The standard deviation of the spectra from the input datacube.

• datacube_masked (numpy.ndarray): The masked numpy.ndarray; if mask is None, datacube_masked is identical to the SpyFile data array.

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Retrieve the image band at 800 nm using hstools.get_band and hsio.spyfile.open_memmap, then mask out all pixels whose value falls below the 75th percentile.

>>> band = io.tools.get_band(800)
>>> array = io.spyfile.open_memmap()[:, :, band]
>>> array_mask, metadata = io.tools.mask_array(array, io.spyfile.metadata, percentile=75, side='lower')

Calculate the spectral mean from the remaining (i.e., unmasked) pixels using hstools.mean_datacube.

>>> spec_mean, spec_std, datacube_masked = io.tools.mean_datacube(io.spyfile, mask=array_mask)

Save using hsio.write_spec and hsio.write_cube, then load into Spectronon software for visualization.

>>> fname_hdr_spec = r'F:\nigo0024\Documents\hs_process_demo\hstools\Wells_rep2_20180628_16h56m_pika_gige_7-mean_800nm_75th.spec.hdr'
>>> fname_hdr_cube = r'F:\nigo0024\Documents\hs_process_demo\hstools\Wells_rep2_20180628_16h56m_pika_gige_7-mean_800nm_75th.bip.hdr'
>>> io.write_spec(fname_hdr_spec, spec_mean, spec_std, metadata=metadata, force=True)
Saving F:\nigo0024\Documents\hs_process_demo\hstools\Wells_rep2_20180628_16h56m_pika_gige_7-mean_800nm_75th.spec
>>> io.write_cube(fname_hdr_cube, datacube_masked, metadata=metadata, force=True)
Saving F:\nigo0024\Documents\hs_process_demo\hstools\Wells_rep2_20180628_16h56m_pika_gige_7-mean_800nm_75th.bip

modify_meta_set(meta_set, idx, value)

Modifies metadata “set” (i.e., string representation of a Python set; common in .hdr files) by converting string to list, then adjusts the value of an item by its index.

Parameters

• meta_set (str) – the string representation of the metadata set

• idx (int) – index to be modified; if None, the whole meta_set is returned (default: None).

• value (float, int, or str) – value to replace at idx

Returns

set_str (str): Modified metadata set string.

Return type

str

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Retrieve the “map info” set from the metadata via hstools.get_meta_set

>>> map_info_set = io.spyfile.metadata['map info']
>>> map_info_set
['UTM',
 '1.0',
 '1.0',
 '441357.287073',
 '4855944.7717699995',
 '0.04',
 '0.04',
 '15',
 'T',
 'WGS-84',
 'units  meters',
 'rotation  0.000']

Modify the value at index position 4 from 4855944.7717699995 to 441300.2 using hstools.modify_meta_set.

>>> io.tools.modify_meta_set(map_info_set, idx=4, value=441300.2)
'{UTM, 1.0, 1.0, 441357.287073, 441300.2, 0.04, 0.04, 15, T, WGS-84, units  meters, rotation  0.000}'

plot_histogram(array, fname_fig=None, title=None, xlabel=None, percentile=90, bins=50, fontsize=16, color='#444444')

Plots a histogram with the percentile value labeled.

Parameters

• array (numpy.ndarray) – The data array used to create the histogram for; if array is masked, the masked pixels are excluded from the histogram.

• fname_fig (str, optional) – The filename to save the figure to; if None, the figure will not be saved (default: None).

• title (str, optional) – The plot title (default: None).

• xlabel (str, optional) – The x-axis label of the histogram (default: None).

• percentile (scalar, optional) – The percentile to label and illustrate on the histogram; if percentile = 90, the band/spectral index value at the 90th percentile will be labeled on the plot (default: 90; range: 0-100).

• bins (int, optional) – Number of histogram bins (default: 50).

• fontsize (scalar) – Font size of the axes labels. The title and text annotations will be scaled relatively (default: 16).

• color (str, optional) – Color of the histogram columns (default: “#444444”)

Returns

Figure object showing the histogram.

Return type

fig (matplotlib.figure)

Example

Load and initialize hsio

>>> from hs_process import hsio
>>> fname_in = r'F:\nigo0024\Documents\hs_process_demo\Wells_rep2_20180628_16h56m_pika_gige_7-Radiance Conversion-Georectify Airborne Datacube-Convert Radiance Cube to Reflectance from Measured Reference Spectrum.bip.hdr'
>>> io = hsio(fname_in)

Retrieve the image band at 800 nm using hstools.get_band and hsio.spyfile.open_memmap

>>> band = io.tools.get_band(800)
>>> array = io.spyfile.open_memmap()[:, :, band]

Create a masked array of all values below the 5th percentile via hstools.mask_array

>>> array_mask, metadata = io.tools.mask_array(array, io.spyfile.metadata, percentile=5, side='lower')

Visualize the histogram of the unmasked pixels (i.e., those greater than the 5th percentile) using hstools.plot_histogram

>>> title = 'Reflectance at 800 nm'
>>> xlabel = 'Reflectance (%)'
>>> fig = io.tools.plot_histogram(array_mask, title=title, xlabel=xlabel)