Recognition of Ancient Tamil Handwritten Characters in Palm Manuscripts Using Genetic Algorithm

The objective of this research is to develop computer software that can recognize the Ancient Tamil handwritten characters by using the genetic algorithm technique (RATHCPM). The system consists of 5 main modules, which are: 1) image acquisition module, 2) image preprocessing module, 3) feature extraction module, 4) character recognition module, and 5) display result module. Each module has the following details. First, the image acquisition module collects an unknown input character from a user. Second, the input image is transformed into a suitable image for the feature extraction module. Third, the system extracts character features from the image. There are 3 main features of Tamil characters which are stroke, loop and location of loop and stroke connection. Fourth, the extracted character information is kept in the form of bits string chromosome in a genetic algorithm. Finally, the system displays the best fitness chromosome for the recognition result.