Can we do a morphological analysis on WIRCAM data%3F
Huertas-Company, M.; Rouan, D.; Tasca, L.; Kneib, J.P; Soucail, G.; Le F%E8vre, O.

Abstract: We present preliminary results on morphological classification up to z~2 of ~1500 objects in the COSMOS field, observed with WIRCAM (Ks band, 2.16um). Near-infrared data have the key advantage of probing old stellar populations in the rest-frame, enabling a determination of galaxy morphological types unaffected by recent star formation, which are more closely linked to the underlying mass than classical optical morphology studies. The classification, in two broad morphological types (early type and late type), has been performed using a new automated multi-parameter method based on a support vector learning machine. The technique, which is a generalization of the classical C-A classification to non-linear n-dimensional boundaries, is proved to be reliable (error < 15%25) up to the sample completeness limit (KAB~22) and represents thus a significant improvement. We obtain a morphological distribution as a function of redshift for the two main morphological types. We do not find any significant clue of evolution.