Efficient and fast distorted character recognition algorithm

Abstract EN

An algorithm to recognize distorted Machine and handwritten characters is proposed.

It uses a feature point extraction-based recognition approach.

A new verification scheme, which deals with this problem, is presented.

A new feature extractor set is introduced.

This research explores best sets of feature extraction techniques and studies the accuracy and speed of the suggested procedures.

Finally, experimental results on one database are presented, their results showing the robustness of the algorithm using small database.