Supporting Data and Code: Fast clustering in linear independent 1D subspaces: segmentation of multi-channel images with high spatial resolution

摘要: Algorithms for subspace clustering (SC) such as sparse and low- rank representation SC are effective in terms of the accuracy but suffer from high computational complexity. We propose algorithm (highly) similar data points drawn union linear independent one-dimensional subspaces with complexity that is number points. The finds a dictionary represents reproducible kernel Hilbert space (RKHS). Afterwards, projected into RKHS by using empirical map (EKM). Segmentation realized applying max operator on data. provide rigorous proof noise free proposed approach yields exact subspaces. also prove EKM-based projection less correlated Due to nonlinear projection, method can adopt linearly nonseparable demonstrate efficiency synthetic dataset well segmentation tissue components image unstained specimen histopathology.