Handwritten character recognition (HCR) is a complex process which involves identification of size, style and the pen used for writing. Therefore, it becomes crucial part of converting handwritten text into electric one in a precise way. This area comes under the purview of Image Processing, Pattern Recognition and Natural Language Processing. Although there has been tremendous research in this arena, the state of the art in the HCR has not witnessed a major progress. Even though a number of research scholars are working in this domain, a few among them are successful in achieving the desired accuracy in recognition and identification of kannada alphabets. This paper summarizes a review of existing work on structural features, global features and combined features of handwritten Kannada characters. Classifiers have been discussed in detail. Most of the authors have achieved good accuracy using structural features. While a few authors have preferred global features and achieved comparatively less accuracy.