This paper presents an intelligent system to translate Arabic natural language to Arabic sign language based on knowledge base and image processing. This system was designed to help hearing impaired for improving their connection with hearing world and acquiring new concepts. The system plays a role for assisting teachers in sign language domain. Of the most important problems in this task, is that the number of words in the Arabic sign dictionary is very few compared with the words found in the Arabic language dictionaries. To solve this problem the proposed system includes a knowledge base to solve a number of Arabic language problems (e.g synonyms, inflectional, derivational, diacritical and plural).

This paper presents an intelligent translation system for the signs of some words and letters in the Arabic sign language. The proposed translation system does not depend on any visual markings or gloves used to complete the recognition process. The proposed translation system deals with images, which allows the user to interact with the system in a natural way. The proposed translation system consists of four main phases; Preprocessing images phase, feature extraction phase, matching strategy phase, and Display Text Translation phase. The extracted features used are combining intensity histogram features and Gray Level Co-occurrence Matrix (GLCM) features, Experiments revealed that the proposed system was able to recognize the 19 Arabic alphabets and word with an accuracy of 73%.

A spell checker is a basic requirement for any language to be digitized. It is a software that detects and corrects errors in a particular language. This paper proposes a model to spell error detection and auto-correction that is based on n-gram technique and it is applied in error detection and correction in English as a global language. The proposed model provides correction suggestions by selecting the most suitable suggestions from a list of corrective suggestions based on lexical resources and n-gram statistics. It depends on a lexicon of Microsoft words. The evaluation of the proposed model uses English standard datasets of misspelled words. Error detection, automatic error correction, and replacement are the main features of the proposed model. The results of the experiment reached approximately 93% of accuracy and acted similarly to Microsoft Word as well as outperformed both of Aspell and Google.