After upgrading to skimage 0.19.0 in #362 we have several failing tests:

• test_kmeans_segmentation_filter_on_rgb_image
• test_rag_threshold_filter_on_rgb_image
• test_rag_threshold_filter_return_labels
• test_rag_threshold_filter_on_gs_image
• test_canny_edges_filter_on_gs_image

Moreover, to be adapted:

• "Additional colorspace illuminants and observers parameter options were added to skimage.color.lab2rgb, skimage.color.rgb2lab, skimage.color.xyz2lab, skimage.color.lab2xyz, skimage.color.xyz2luv and skimage.color.luv2xyz."
• "The level parameter of measure.find_contours is now a keyword argument"
• "Ensure skimage.color.separate_stains does not return negative values."
• deprecation of the selem argument in favor of footprint throughout the library
• deprecation of indices argument in peak_local_max

Notes:

• kmeans_segmentation and rag_threshold tests are probably failing due to accelerate the masked slic superpixels algorithm scikit-image/scikit-image#4903 or "SLIC superpixel segmentation outputs may differ from previous versions for data that was not already scaled to [0, 1] range. There is now an automatic internal rescaling of the input to [0, 1] so that the compactness parameter has an effect that is independent of the input image's scaling." -- for this we just need to update the expectations
• test_canny_edges_filter_on_gs_image is failing because "Seeding random number generators will not give the same results as the underlying generator was updated to use numpy.random.Generator." MAINT/ENH: refactor np.random.x to use np.random.Generator scikit-image/scikit-image#5357