Character Recognition Analysis

1297 Words6 Pages
In the fast moving world with the amazingly growing technology, character recognitions play a wide role by providing more scope to perform research in OCR techniques. Sanskrit Handwritten recognition has been one of the challenging research areas in the field of pattern recognition. Character recognition is the electronic translation of scanned images of handwritten or printed text into a machine encoded text. The character recognition is a standout amongst the most generally utilized biometric attributes for authentication of persons and document. In this paper proposed an off line handwritten character recognition framework utilizing feed forward neural network. A handwritten Sanskrit character is resized into 20x30 pixels and this character…show more content…
INTRODUCTION With emergence of the digital content the need for the development of an OCR engine with high performance has become essential. The idea of OCR is to analyze a document image by page, words and characters. To identify the exact characters, these characters are compared with image patterns. Character recognition can either be done from printed documents or from handwritten documents. Sanskrit handwritten is more complicated than other related works in offline mode, because Sanskrit letters have more consonants and modifiers Sanskrit is an ancient Indo Aryan language with a rich literary tradition. The traditional Sanskrit script is the well known script in India. It is an ancient language with written materials and no longer spoken. Most of the poetry, scientific, and technical texts are made with a rich tradition of Sanskrit literature. Sanskrit is a phonetic language which consists of 48 characters (15 vowels, 33 consonants).Sanskrit documents is written from left to right. 2. RELATED…show more content…
Thresholding concepts are usually used by researchers to extract the foreground image from background image. Histogram based thresholding approach can also be used to convert a gray-scale image into a two tone image. In contrast to this Adaptive Binarization method can be used to identify the local gray value contrast of the Image. This helps in extracting the text information from low quality documents. 3.1.2 Noise Removal Digital images are usually prone to so many types of noises. Noise can be termed as a document image which is due to poorly photocopied pages. Median Filtering [18], Wiener Filtering method [19] and morphological operations can be performed to remove noise [16]. To replace the intensity of the character image [24] Median filters are used. Gaussian filters can be used to smoothing the image [10]. 3.1.3 Normalization The process of converting a random sized image into a standard size is normalization. The Roi-Extraction [20], Bicubic interpolation [16], linear size normalization [14] and Java Image Class [12] normalization are used to get the single structural element from the
Open Document