OFFLINE HANDWRITTEN WRITER INDEPENDENT TAMIL CHARACTER RECOGNITION

Today handwritten character recognition is one of the challenging computational processes. To the best of our knowledge, little work has been done in the area of tamil handwritten character recognition and they experimented with their own database. The main objective of this project is to get better the recognition ratewhile compared with the previous approaches for offline tamil handwritten character recognition. In the character recognition system feature extraction is one of the important phases. In region-based invariants, all the pixels within a shape are taken into account to obtain the mathematical representation. The most popular region-based methods include various moment-based invariants such as Hu's seven moment invariants, Zernike moments, Complex moments, etc. In region-based invariants, all of the pixels of the image are taken into account to represent the shape. Because region-based invariants combine information of an entire image region rather than exploiting information just along the boundary pixels, they can capture more information from the image. The region-based invariants can also be employed to describe disjoint shapes. In Hu's moment invariants, the whole concept is based on the central moments which have integrated the translation and scale normalization in the definitions. The Zernike moments, are only invariant to image rotation for them. To compute the Zernike moments of a digital image, the range of the image should be mapped to the unit circle first with its origin at the image's center. The pixels falling outside the unit circle are discarded in the computation process. A Hidden Markov Model (HMM) classifier is used for recognition purpose