Extraction of ROI of Graphical Attributes on the Image of the Document of Cooperation

For the needs of data collection, reporting, and synchronization of the large number of documents of cooperation, an automatic content exploration system is needed for the collection of documents of cooperation. The first step is to detect and extract the graphical attributes on the image of the document of cooperation. Document image analysis methods must be applied to detect and extract region of interest (ROI) containing graphical attributes in documents, separating them from document areas containing textual content. This paper presents a proposed method to detect and extract the graphical attributes on the image of the document of cooperation, such as logo, stamp, signature, and seal. The proposed method consists of binarization method, connected component analysis, and image morphology processing. The results show that the proposed method is performed well in extracting the ROI for logo, stamp, signature, and seal in the image of document of cooperation.