Where scikit-image shines
scikit-image is a peer-reviewed collection of image processing algorithms from the scientific Python community. It works on NumPy arrays and fits naturally with SciPy, matplotlib, and the rest of the scientific stack, which makes it the go-to library for segmentation, measurement, and analysis.
Do it with BrushCue
BrushCue is for when the goal is a better-looking image rather than a measurement. Some operations overlap, like edge detection:
scikit-image
import skimage as ski
image = ski.io.imread("photo.jpg")
edges = ski.filters.sobel(ski.color.rgb2gray(image))
ski.io.imsave("photo-edges.png", ski.util.img_as_ubyte(edges))BrushCue (Python)
import brushcue as bc
ctx = bc.Context()
image = bc.Composition.load("photo.jpg")
# Sobel edge detection
edited = image.sobel_edge_detection()
result = edited.execute(ctx)
with open("photo-edited.png", "wb") as f:
f.write(result.to_image_bytes(ctx))scikit-image gives you the raw edge magnitudes as an array, which is what you want for analysis. BrushCue's version renders a ready-to-use image, and sits alongside creative tools like the Inked Filter and Pencil Sketch.
BrushCue Script (bcue)
image = Composition::from_path("photo.jpg")
edited = image.sobel_edge_detection()
Bytes::save_to_path(Image::to_bytes(Composition::render_to_image(edited)), "photo-edited.png")



