Pattern recognition has become a very interesting topic for researchers during last few decades. Handwriting recognition is very challenging area of pattern recognition with various practical applications. There are many applications of this form of recognition. Like postal code verification, vehicle number plate recognition, bank cheque processing, Assigning ZIP Codes to letter mail, automatic reading of area code and address from the letter, various data form processing etc. MEETEILON is an Indo-Aryan language. It is the main language of the state Manipur in India. Around 25 lakhs people speak MEETEILON language in Manipur. Many researchers are doing efforts to convert human written things into the machine readable format which saves time and money. Handwritten digit recognition is very difficult because it depends on various persons and their writing styles. It’s easy to recognize English digits compare to MEETEI-MAYEK digits because in English most of the digits have straight lines and less shapes compare to other script digits like Gujarati ,Devanagiri. MEETEI-MAYEK digits are having different characteristics. They are having various shapes and it’s really difficult to recognize those shapes. Due to varieties in shapes there are some characters that are confusing and possibilities for misclassification are very high. Neural Networks are widely applied to pattern recognition areas. Neural Networks can be trained and then tested on various handwritten digits. This paper describes feed forward neural network with back propagation learning approach for the handwritten digit recognition. Optical Character Recognition (OCR) is a very well-studied problem in the vast area of pattern recognition. Its origins can be found as early as 1870 when an image transmission system was invented which used an array of photocells to recognize patterns. Until the middle of the 20th century OCR was primarily developed as an aid to the visually handicapped. With the advent of digital computers in the 1940s, OCR was realized as a data processing approach for the ﬁrst time. The ﬁrst commercial OCR systems began to appear in the early 1950s and soon they were being used by the US postal service to sort mail. According to Wikipedia “The accurate recognition of Latin-script, typewritten text is now considered largely a solved problem on applications where clear imaging is available such as scanning of printed documents. Typical accuracy rates on these exceed 99%; total accuracy can only be achieved by human review. Other areas including recognition of hand printing, cursive handwriting, and printed text in other scripts (especially those with a very large number of characters) are still the subject of active research.”
Offline Character recognition of handwritten Meitei mayek digits
The major schemes in OCR starting from the relatively easier to the most diﬃcult are as follows : (i) Fixed-font character recognition is the recognition of speciﬁc fonts (Ariel, Courier, etc.) of typewritten characters. (ii) On-line character recognition is the recognition of single hand-drawn characters where not only the character image is provided but also the timing information of each stroke. (iii) Handwritten character recognition is the recognition of single hand-drawn characters of an alphabet which are unconnected and not written in calligraphy. (iv) Script recognition is the recognition of unrestricted handwritten characters which may be connected and cursive. Standard pattern recognition methodologies that have been successfully applied to OCR include point by point global comparisons, global transformations, extraction of local properties, template matching, analysis by means of curvatures and structural methods. However, the application of these methods for handwritten character recognition is discussable because of the inﬁnite variations of character shapes resulting from writing habits, style, education, region of...
Please join StudyMode to read the full document