Segmentation and classification of breast cancer cells in histological images

In this paper, we propose a new treatment for the detection of tumoral cells of breast cancer. First segmentation is processed on breast cells with the watershed mathematical morphology process. The cells are then described by a set of complete and stable invariant based on Fourier descriptors. Finally the PCA is performed to classify the cells to normal and tumoral - marked and unmarked - ones. The rate of tumoral cells will allow a diagnostic help for pathologists