OCR
Project Name: Optical Character Recognizer.
Duration: 5 Months.
Client: Department of Computer Science and Information Technology. Islamic University of Technology, Dhaka, Bangladesh.
Status: Successfully Completed.
Project Group: This project is accomplished by a group of 3 members. I worked as a developer and designer.
Language: C.
Project Description:
This is a Optical Character recognizer to recognize printed Bengali characters. For implementing the software, BAM (Bidirectional Associative Memories) neural network is uesd. For recognizing the charaters, it is assumed that the texts are scanned and each charater are separated by some feature extraction mechanizm. Then this software will represent the chracters into matrix format and recognize it through previously trained network. The recognition accuracy of this software is shown in the following chart. It is worth to mention that this product is only has the accademic version to study the recognition accuracy of complex Bengali characters (as Bengali characters has more variations than English characters).
