Image processors and filters - inspired by Instagram, built on top of the PIL/Pillow, SciPy and scikit-image packages, accelerated with Cython, and ready to use with PILKit and the django-imagekit framework.
Included are filters for Atkinson and Floyd-Steinberg dithering, dot-pitch halftoning (with GCR and per-channel pipeline processors), classes exposing image-processing pipeline data as NumPy ND-arrays, Gaussian kernel functions, processors for applying channel-based LUT curves to images from Photoshop .acv files, imagekit-ready processors furnishing streamlined access to a wide schmorgasbord of Pillow's many image adjustment algorithms (e.g. noise, blur, and sharpen functions, histogram-based operations like Brightness/Contrast, among others), an implementation of the entropy-based smart-crop algorithm many will recognize from the easy-thumbnails Django app - and much more.