Background:

Sparsity Constrained Convolutional Regression is a machine learning algorithm that was initially developed for the segmentation of nuclear regions in tissue histology.

Description:

This project contains the source code (matlab) for Sparsity Constrained Convolutional Regression that is described in the publication as follows,

• Y. Zhou, H. Chang†, K. E. Barner and B. Parvin†. "Nuclei Segmentation via Sparsity Constrained Convoluational Regression." IEEE International Symposium on Biomedical Imaging (ISBI 2015), pp. 1284-1287, Brooklyn, NY, U.S., April 2015. († Co-Corresponding Author) (Oral Presentation)

It provides API functions, examplar data and a demo function, with which the learnt fiter banks and example of segmentation are illustrated as follows,

And the decision function of SCCR is illustrated as follows,