Effective Computational Tool for the Segmentation of Images from a Printed Document

671 Words Feb 1st, 2018 3 Pages
The main purpose of this work is to analyze how a better and the efficient system can be designed to segments two or more pictures from the same printed document page at the same time. Most of the previous methods for achieving this particular problem is useful only when single image has to segment from the printed document. This particular system will also help us to extract title images and the images in between two pages. For this either digital cameras or flatbed scanners can be used but there is various advantages of digital camera over flatbed scanners such as portability, acquisition without contacting and fast response, whereas the flat scanners requires high computing power.
So clearly there is need to develop such effective computational tool for the segmentation of the images from the printed document.

3.2 OBJECTIVES
By keeping following objectives in mind we are developing such a computational tool:
I. To extract more than 1 Image from a Single Document Page so that the limitation of single image extraction can be removed.
II. To extract the small and large images by using the method of detecting edges of some predefined range of orientation in order to remove the problem related to detecting such boundaries.
III. To extract Segmented Images which are lie between 2 pages which can be possible only using digital camera because flatbed scanners can never meet…
Open Document