Abstract. Text/graphics separation aims at segmenting the document into two layers: a layer assumed to contain text and a layer containing graphical objects. In this paper, we present a consolidation of a method proposed by Fletcher and Kasturi, with a number of improvements to make it more suitable for graphics-rich documents. We discuss the right choice of thresholds for this method, and their stability. We also propose a post-processing step for retrieving text components touching the graphics, through local segmentation of the distance skeleton.

In this paper, we present a method for symbol recognition based on the spatio-structural description of a 'vocabulary' of extracted visual elementary parts. It is applied to symbols in electrical wiring diagrams. The method consists of first identifying vocabulary elements into different groups based on their types (e.g., circle, corner ). We then compute spatial relations between the possible pairs of labelled vocabulary types which are further used as a basis for building an Attributed Relational Graph that fully describes the symbol. These spatial relations integrate both topology and directional information.The experiments reported in this paper show that this approach, used for recognition, significantly outperforms both structural and signal-based state-of-the-art methods.

In this paper, we present a pattern recognition method that uses dynamic programming (DP) for the alignment of Radon features. The key characteristic of the method is to use dynamic time warping (DTW) to match corresponding pairs of the Radon features for all possible projections. Thanks to DTW, we avoid compressing the feature matrix into a single vector which would otherwise miss information. To reduce the possible number of matchings, we rely on a initial normalisation based on the pattern orientation. A comprehensive study is made using major state-of-the-art shape descriptors over several public datasets of shapes such as graphical symbols (both printed and hand-drawn), handwritten characters and footwear prints. In all tests, the method proves its generic behaviour by providing better recognition performance. Overall, we validate that our method is robust to deformed shape due to distortion, degradation and occlusion. KeywordsThe Radon Transform; Dynamic Programming; Shape Descriptor; Pattern Recognition. ! 1. Fixed size vector, here refers to the set-up feature vector space. For example, the feature vector size in R-signature Tabbone et al. (2006) is 180 i.e., a single compressed projection value at each projecting angle in the range [0, π[, and in generic Fourier descriptor (GFD) Zhang and Lu (2002), it has a size of m × n, where m and n respectively refers to radial and angular frequencies.