International Journal of Computer Applications (0975 – 8887) Volume 160 – No 2, February 2017 31 A Survey of Contextual Handwritten Recognition Systems based HMMs for Cursive Arabic and Latin Script Mouhcine Rabi Laboratory IRF-SIC, faculty of sciences, Ibn Zohr University Agadir, Morocco Mustapha Amrouch Laboratory IRF-SIC, faculty of sciences Ibn Zohr University, Agadir, Morocco Zouhir Mahani High school of technology. Ibn Zohr University, Agadir, Morocco ABSTRACT Offline handwriting recognition has become lately a very popular research area and the number of its possible application is very large. Most recognition system are based on modeling characters to recognize, then the concatenation of these models to recognize a word, while modeling character allows deformations related to its context. This paper provides a survey of handwritten recognition systems based on context-dependent character modeling to account possible deformations related to its context. It examines the literature on the most significant work in contextual handwritten text recognition for two different alphabets, Latin and Arabic. Finally discussing the comparative results to achieve a comprehensive summary of the various approaches and systems taking account the character’s context which could help open up some interesting new prospects. Keywords Offline handwriting Recognition, Latin, Arabic, Context, Cursive 1. INTRODUCTION Handwritten character recognition is a process of transforming handwritten text into machine executable format. There are mainly three steps in pattern recognition: observation, pattern segmentation and pattern classification. Character’s recognition has become very interesting topic in pattern recognition for the researchers during last few decades. In general, handwritten recognition is classified into two types as on-line and off-line recognition methods [01]. Offline handwriting recognition involves the text’s automatic conversion into an image into letter codes which are usable within computer and text-processing applications. The data obtained by this form is regarded as a static representation of handwriting. But, in the on-line system, the two dimensional coordinates of successive points are represented as a function of time and the order of strokes made by the writer are also available. Offline character recognition is comparatively more challenging due to shape of characters, great variation of character symbol, different handwriting style and document quality. Several applications including mail sorting, bank processing, document reading and postal address recognition require off- line handwriting recognition systems. As a result, the off-line handwriting recognition continues to be an active area of research towards exploring the newer techniques that would improve recognition accuracy. The study defines the five major stages in any handwritten character recognition which is shown in Figure1. Figure 1 : stages of character recognition In image preprocessing we apply a series of operations like Binarization, Complement, size normalisation, Morphological operation, noise removal using filters, thresholding, skeletonization, thinning, cleaning techniques and filtering mechanisms on scanned image which are taken as input. It makes the input image easier to process in order to increase the overall efficiency of recognition system [02]. Segmentation is process of extracting the basic constituent symbols of the script. Image is subdivided into many parts so that each part of the image is readable. To accomplish this task the image is subdivided considering three aspects, i.e. line wise segmentation, word wise segmentation and character wise segmentation.[03] Feature extraction is a special form of dimension reduction. This approach is useful when image sizes are large and a reduced feature representation is required to quickly complete tasks such as image matching and retrieval [04][05]. The classification stage is the decision making stage of the recognition system, is the process of assigning the sensed data to their corresponding class with respect to groups with homogeneous characteristics, with the aim of discriminating multiple objects from each other within the image. There are many existing techniques available for handwriting classification.[06] The last stage of the character recognition system is Post- processing. It prints the corresponding recognized characters in the structured text form. The accuracy of character recognition stages can be improved if the semantic information is available up to great extent. Researches have tried various approaches employing various techniques for pre-processing, features extraction and classification [07][08]. The cursive nature of handwriting text, the word’s variability, letter shapes are context sensitive, inter and intra word spaces, the skew and slant of characters and words makes the construction of offline recognition system a challenging task, opening the way to a new approaches using