A Comprehensive Framework for OCR Web Services System for Arabic Calligraphy Documents
Hassanin M. Al-Barhamtoshy, Abdullah Saad Al-Malaise ALGhamdi · International Journal of Engineering & Technology · 2019
This paper describes document layout analysis web services approach for OCR systems, in case of integrate with web-based applications using SOAP and REST interfaces. The proposed solution provides accessing way to use different OCR systems. Therefore, these web services are implemented using SOAP and REST interfaces through HTTP or HTTPS requests. Consequently, different developers can communicate with each other’s without time consuming to customize code implementation, operating system barriers, and programming language conditions.The scientific scope of this paper focuses on three objectives:(1) The document categories on which they are included in the dataset, (2) The related algorithms that are used in the level of document analysis, and (3) The Arabic document image segmentation algorithms they are used. Consequently, the connected components method is used to remove page frame in the old and calligraphy documents. Also, shadow noises in the old and historical documents are removed using the adapted sparse algorithm.This paper discusses a number of the major areas where OCR web services have been working comprehensively: in supporting document analysis and OCR service-oriented architecture computing. Using the OCR web services approaches, we are dealing with heterogeneous large scale documents with wide varying structured category. Furthermore, there could be multipage document with different languages.Accordingly, the language domain will be identified within the language script specification module.