Both Cellular Morphometric Feature and features from unsupervised feature learning lead to superior performance when compared to SIFT and [Color, Texture];

Cellular saliency incorporation impairs the performance for systems built upon pixel-/patch-level features; and,

The effect of model architecture is correlated with the robustness of features, and the performance can be consistently improved by the deep-extension of systems built upon both Cellular Morphometric Feature and features from unsupervised feature learning.