Within the image processing arena, the field of pattern recognition or matching has been the focus of extensive research. Many researchers have attempted to solve the problem of pattern recognition in general, while many others have been interested in the specific problem of character recognition because of its potential application. In addition, the automated recognition and processing of (Hand-) printed characters is also the basic function that should be given attention in order to improve man-machine communication. A large number of research papers and reports have already been published on Latin, Chinese and Japanese characters. However, although almost a third of a billion people worldwide, in several different languages, use Arabic characters for writing, little research progress, in both on-line and off-line, has been achieved towards the recognition of Arabic characters. This is a result of the lack of adequate support in terms of funding, and other utilities such as Arabic database, dictionaries, etc.. The main objectives of this paper are: to identify the major problems related to handwritten and printed Arabic characters; to present a general panorama on the research technique in the domain of Arabic character recognition and, in particular, to present some different systems that we have developed during the past two decades in both on-line and off-line recognition