testbed/astropy__astropy/astropy/visualization/tests/test_lupton_rgb.py

# Licensed under a 3-clause BSD style license - see LICENSE.rst

"""
Tests for RGB Images
"""


import sys
import os
import tempfile

import pytest
import numpy as np
from numpy.testing import assert_equal

from astropy.convolution import convolve, Gaussian2DKernel
from astropy.visualization import lupton_rgb

try:
    import matplotlib  # noqa
    HAS_MATPLOTLIB = True
except ImportError:
    HAS_MATPLOTLIB = False

# Set display=True to get matplotlib imshow windows to help with debugging.
display = False


def display_rgb(rgb, title=None):
    """Display an rgb image using matplotlib (useful for debugging)"""
    import matplotlib.pyplot as plt
    plt.imshow(rgb, interpolation='nearest', origin='lower')
    if title:
        plt.title(title)
    plt.show()
    return plt


def saturate(image, satValue):
    """
    Return image with all points above satValue set to NaN.

    Simulates saturation on an image, so we can test 'replace_saturated_pixels'
    """
    result = image.copy()
    saturated = image > satValue
    result[saturated] = np.nan
    return result


def random_array(dtype, N=100):
    return np.array(np.random.random(10)*100, dtype=dtype)


def test_compute_intensity_1_float():
    image_r = random_array(np.float64)
    intensity = lupton_rgb.compute_intensity(image_r)
    assert image_r.dtype == intensity.dtype
    assert_equal(image_r, intensity)


def test_compute_intensity_1_uint():
    image_r = random_array(np.uint8)
    intensity = lupton_rgb.compute_intensity(image_r)
    assert image_r.dtype == intensity.dtype
    assert_equal(image_r, intensity)


def test_compute_intensity_3_float():
    image_r = random_array(np.float64)
    image_g = random_array(np.float64)
    image_b = random_array(np.float64)
    intensity = lupton_rgb.compute_intensity(image_r, image_g, image_b)
    assert image_r.dtype == intensity.dtype
    assert_equal(intensity, (image_r+image_g+image_b)/3.0)


def test_compute_intensity_3_uint():
    image_r = random_array(np.uint8)
    image_g = random_array(np.uint8)
    image_b = random_array(np.uint8)
    intensity = lupton_rgb.compute_intensity(image_r, image_g, image_b)
    assert image_r.dtype == intensity.dtype
    assert_equal(intensity, (image_r+image_g+image_b)//3)


class TestLuptonRgb:
    """A test case for Rgb"""

    def setup_method(self, method):
        np.random.seed(1000)  # so we always get the same images.

        self.min_, self.stretch_, self.Q = 0, 5, 20  # asinh

        width, height = 85, 75
        self.width = width
        self.height = height

        shape = (width, height)
        image_r = np.zeros(shape)
        image_g = np.zeros(shape)
        image_b = np.zeros(shape)

        # pixel locations, values and colors
        points = [[15, 15], [50, 45], [30, 30], [45, 15]]
        values = [1000, 5500, 600, 20000]
        g_r = [1.0, -1.0, 1.0, 1.0]
        r_i = [2.0, -0.5, 2.5, 1.0]

        # Put pixels in the images.
        for p, v, gr, ri in zip(points, values, g_r, r_i):
            image_r[p[0], p[1]] = v*pow(10, 0.4*ri)
            image_g[p[0], p[1]] = v*pow(10, 0.4*gr)
            image_b[p[0], p[1]] = v

        # convolve the image with a reasonable PSF, and add Gaussian background noise
        def convolve_with_noise(image, psf):
            convolvedImage = convolve(image, psf, boundary='extend', normalize_kernel=True)
            randomImage = np.random.normal(0, 2, image.shape)
            return randomImage + convolvedImage

        psf = Gaussian2DKernel(2.5)
        self.image_r = convolve_with_noise(image_r, psf)
        self.image_g = convolve_with_noise(image_g, psf)
        self.image_b = convolve_with_noise(image_b, psf)

    def test_Asinh(self):
        """Test creating an RGB image using an asinh stretch"""
        asinhMap = lupton_rgb.AsinhMapping(self.min_, self.stretch_, self.Q)
        rgbImage = asinhMap.make_rgb_image(self.image_r, self.image_g, self.image_b)
        if display:
            display_rgb(rgbImage, title=sys._getframe().f_code.co_name)

    def test_AsinhZscale(self):
        """Test creating an RGB image using an asinh stretch estimated using zscale"""

        map = lupton_rgb.AsinhZScaleMapping(self.image_r, self.image_g, self.image_b)
        rgbImage = map.make_rgb_image(self.image_r, self.image_g, self.image_b)
        if display:
            display_rgb(rgbImage, title=sys._getframe().f_code.co_name)

    def test_AsinhZscaleIntensity(self):
        """Test creating an RGB image using an asinh stretch estimated using zscale on the intensity"""

        map = lupton_rgb.AsinhZScaleMapping(self.image_r, self.image_g, self.image_b)
        rgbImage = map.make_rgb_image(self.image_r, self.image_g, self.image_b)
        if display:
            display_rgb(rgbImage, title=sys._getframe().f_code.co_name)

    def test_AsinhZscaleIntensityPedestal(self):
        """Test creating an RGB image using an asinh stretch estimated using zscale on the intensity
        where the images each have a pedestal added"""

        pedestal = [100, 400, -400]
        self.image_r += pedestal[0]
        self.image_g += pedestal[1]
        self.image_b += pedestal[2]

        map = lupton_rgb.AsinhZScaleMapping(self.image_r, self.image_g, self.image_b, pedestal=pedestal)
        rgbImage = map.make_rgb_image(self.image_r, self.image_g, self.image_b)
        if display:
            display_rgb(rgbImage, title=sys._getframe().f_code.co_name)

    def test_AsinhZscaleIntensityBW(self):
        """Test creating a black-and-white image using an asinh stretch estimated
        using zscale on the intensity"""

        map = lupton_rgb.AsinhZScaleMapping(self.image_r)
        rgbImage = map.make_rgb_image(self.image_r, self.image_r, self.image_r)
        if display:
            display_rgb(rgbImage, title=sys._getframe().f_code.co_name)

    @pytest.mark.skipif('not HAS_MATPLOTLIB')
    def test_make_rgb(self):
        """Test the function that does it all"""
        satValue = 1000.0
        with tempfile.NamedTemporaryFile(suffix=".png") as temp:
            red = saturate(self.image_r, satValue)
            green = saturate(self.image_g, satValue)
            blue = saturate(self.image_b, satValue)
            lupton_rgb.make_lupton_rgb(red, green, blue, self.min_, self.stretch_, self.Q, filename=temp)
            assert os.path.exists(temp.name)

    def test_make_rgb_saturated_fix(self):
        pytest.skip('saturation correction is not implemented')
        satValue = 1000.0
        # TODO: Cannot test with these options yet, as that part of the code is not implemented.
        with tempfile.NamedTemporaryFile(suffix=".png") as temp:
            red = saturate(self.image_r, satValue)
            green = saturate(self.image_g, satValue)
            blue = saturate(self.image_b, satValue)
            lupton_rgb.make_lupton_rgb(red, green, blue, self.min_, self.stretch_, self.Q,
                                       saturated_border_width=1, saturated_pixel_value=2000,
                                       filename=temp)

    def test_linear(self):
        """Test using a specified linear stretch"""

        map = lupton_rgb.LinearMapping(-8.45, 13.44)
        rgbImage = map.make_rgb_image(self.image_r, self.image_g, self.image_b)
        if display:
            display_rgb(rgbImage, title=sys._getframe().f_code.co_name)

    def test_linear_min_max(self):
        """Test using a min/max linear stretch determined from one image"""

        map = lupton_rgb.LinearMapping(image=self.image_b)
        rgbImage = map.make_rgb_image(self.image_r, self.image_g, self.image_b)
        if display:
            display_rgb(rgbImage, title=sys._getframe().f_code.co_name)

    def test_saturated(self):
        """Test interpolationolating saturated pixels"""
        pytest.skip('replaceSaturatedPixels is not implemented in astropy yet')

        satValue = 1000.0
        self.image_r = saturate(self.image_r, satValue)
        self.image_g = saturate(self.image_g, satValue)
        self.image_b = saturate(self.image_b, satValue)

        lupton_rgb.replaceSaturatedPixels(self.image_r, self.image_g, self.image_b, 1, 2000)
        # Check that we replaced those NaNs with some reasonable value
        assert np.isfinite(self.image_r.getImage().getArray()).all()
        assert np.isfinite(self.image_g.getImage().getArray()).all()
        assert np.isfinite(self.image_b.getImage().getArray()).all()

        # Prepare for generating an output file
        self.imagesR = self.imagesR.getImage()
        self.imagesR = self.imagesG.getImage()
        self.imagesR = self.imagesB.getImage()

        asinhMap = lupton_rgb.AsinhMapping(self.min_, self.stretch_, self.Q)
        rgbImage = asinhMap.make_rgb_image(self.image_r, self.image_g, self.image_b)
        if display:
            display_rgb(rgbImage, title=sys._getframe().f_code.co_name)

    def test_different_shapes_asserts(self):
        with pytest.raises(ValueError) as excinfo:
            # just swap the dimensions to get a differently-shaped 'r'
            image_r = self.image_r.reshape(self.height, self.width)
            lupton_rgb.make_lupton_rgb(image_r, self.image_g, self.image_b)
        assert "shapes must match" in str(excinfo.value)