Supoj Chinveeraphan

Structural Features by MCR Expression for Printed Arabic Character Recognition

This paper discribes how stroke features in document images are extracted and used for the recognition of printed Arabic characters. It is of importance to provide a good base representation that facilitate analysis and processing of document images. The strokes are extracted by a method called Minimum Covering Runs (MCR)[1]. This method of representing binary images by a minimum number of horizontal and vertical runs is used as a preprocessing step. The strokes are labeled and ordered, a feature space for the 100 shapes of the 28 Arabic characters is build. The system is under developement but the recognition rate obtained at this stage, 95.5% is encouraging.

A hierarchical description of digital grayscale images based on image dipoles

Deals with a hierarchical description of digital grayscale images based on non-local structural features, called image dipoles. An image dipole is defined to be a region whose lines of slope emanate from the same maximum, and reach the same minimum. Incorporating an image dipole based representation of an image into a scale-space filtering, we propose in this paper a hierarchical image dipole description of the image. The proposed description consists of image dipole networks of some scales, and linking of image dipoles across the scales such that the description contains all local transformations of the image dipole network in the scale-space.

Understanding of Ridge-Valley Lines on Image -Intensity Surfaces in Scale-Space

When a grey-scale image is regarded as a surface, the considerable topographic primal structures are stationary points, ridge and valley lines. The ridge and valley lines are rich in information because they pass through all stationary points, and are invariant with respect to monotonic image transformations. As images can be interpreted at multiresolutions by using scale-space filtering, this paper presents an analysis of the local topology of ridge-valley lines and its change in the scale-space. The result shows that the topology of ridge-valley lines along a stationary path will change when the path reaches a singular stationary point or an umbilicus; whereas the change at any non-stationary point occurs when a ridge-valley line gets contact with another or vanishes.

MCR expression of document images based on maximum matching of bipartite graph

Proposes a new method for expressing binary images, named minimum covering run (MCR) expression. The proposed technique adapted from horizontal-vertical run expression, is derived by selecting suitable runs from horizontal and vertical run types which minimize the total number of runs that still represent the original images perfectly. It is shown that horizontal and vertical runs of binary image can be thought as partite sets of a bipartite graph. Consequently, the MCR expression that corresponds to maximum matching in the graph, can be found by using an algorithm solving this graph-theoretic problem. An efficient algorithm for the expression method is presented, and results show the validity of stroke extraction of information in document images.<<ETX>>

Arabic Characters Recognition Based on MCR

In this paper we propose an off-line recognition system for machine printed Arabic characters based on the Minimum Covering Run expression method (MCR). The recognition is achieved by matching to reference prototypes. We report a recognition rate of 97.5% and possibility of generalising the system to other symbols and characters.

Multiscale image dipole description by fine-to-coarse aggregation algorithm

This paper deals with a hierarchical description of grayscale images based on image dipoles. An image dipole is a region whose lines of slope emanate from the maximum, and reach the same minimum. Incorporating an image dipole based representation of an image into a scale-space filtering, we proposed here a hierarchical image dipole description of the image. The proposed description consists of image dipole networks of selected scales, and linking of image dipoles across the scales such that it contains all transformations of the image dipole network in the scale-space.