This paper concerns detecting cell nuclei automatically from colour histopathology images. 
As a critical technique in a digital pathology laboratory, automatic nuclei detection has 
been investigated for more than one decade. Unlike previous watershed based algorithms relying 
on post processing of the watershed, we present a new method which incorporates the staining 
information of pathological slides in the analysis. This pre-processing step enhances the 
colour homogeneity within the nuclear areas as well as the background areas. We show that 
the proposed approach mitigates the over-segmentation problem of watershed based nuclei detection. 
Furthermore, this method is validated with 10 sets of histopathology images of lymphoma. 
The precision/recall performance obtained indicates that the proposed method outperforms 
several other alternatives. Proof of convergence for the proposed algorithm is also included.