Catálogo de publicaciones - libros

Graphics recognition for raster-based input discovers primitives such as lines, arrowheads, and circles. This paper focuses on graphics recognition of figures in vector-based PDF documents. The first stage consists of extracting the graphic and text primitives corresponding to figures. An interpreter was constructed to translate PDF content into a set of self-contained graphics and text objects (in Java), freed from the intricacies of the PDF file. The second stage consists of discovering simple graphics entities which we call , e.g., a pair of primitive graphic objects satisfying certain geometric constraints. The third stage uses machine learning to classify figures using grapheme statistics as attributes. A boosting-based learner (LogitBoost in the Weka toolkit) was able to achieve 100% classification accuracy in hold-out-one training/testing using 16 grapheme types extracted from 36 figures from BioMed Central journal research papers. The approach can readily be adapted to raster graphics recognition.

Syntactic approaches on structural symbol recognition are characterized by defining symbols using a grammar. Following the grammar productions a parser is constructed to recognize symbols: given an input, the parser detects whether it belongs to the language generated by the grammar, recognizing the symbol, or not. In this paper, we describe a parsing methodology to recognize a set of symbols represented by an adjacency grammar. An adjacency grammar is a grammar that describes a symbol in terms of the primitives that form it and the relations among these primitives. These relations are called , which are validated using a defined cost function. The cost function approximates the distortion degree associated to the constraint. When a symbol has been recognized the cost associated to the symbol is like a similarity value. The evaluation of the method has been realized from a qualitative point of view, asking some users to draw some sketches. From a quantitative point of view a benchmarking database of sketched symbols has been used.

This paper presents a solution for online composite sketchy shape recognition. The kernel of the strategy treats both stroke segmentation and sketch recognition as an optimization problem of “fitting to a template”. A nested recursive optimization process is then designed by means of dynamic programming to do stroke segmentation and symbol recognition cooperatively by minimizing the fitting errors between inputting patterns and templates. Experimental results prove the effectiveness of the proposed method.

To develop a method for structured document recognition, it is necessary to know the relative position of the graphical elements in a document. In order to deal with this notion, we build a neighbourhood graph based on Voronoï tessellation. We propose to combine the use of this interesting notion of neighbourhood with an existing generic document recognition method, DMOS, which has been used to describe various kinds of documents. This association allows exploiting different aspects of the neighbourhood graph, separating the graph analysis from the knowledge linked to a kind of document, and establishing a bi-directional context-based relation between the analyser and the graph. We apply this method on the analysis of various documents.

Optical Music Recognition consists in the identification of music information from images of scores. In this paper, we propose a method for the early stages of the recognition: segmentation of staff lines and graphical primitives in handwritten scores. After introducing our work with modern musical scores (where projections and Hough Transform are effectively used), an approach to deal with ancient handwritten scores is exposed. The recognition of such these old scores is more difficult due to paper degradation and the lack of a standard in musical notation. Our method has been tested with several scores of 19th century with high performance rates.

We present a shape classification technique for structural content–based retrieval of two-dimensional vector drawings. Our method has two distinguishing features. For one, it relies on explicit hierarchical descriptions of drawing structure by means of spatial relationships and shape characterization. However, unlike other approaches which attempt rigid shape classification, our method relies on estimating the likeness of a given shape to a restricted set of simple forms. It yields for a given shape, a feature vector describing its geometric properties, which is invariant to scale, rotation and translation. This provides the advantage of being able to characterize arbitrary two–dimensional shapes with few restrictions. Moreover, our technique seemingly works well when compared to established methods for two dimensional shapes.

In this paper, a polygonal approximation approach based on a multi-objective genetic algorithm is proposed. In this method, the optimization/exploration algorithm locates breakpoints on the digital curve by minimizing simultaneously the number of breakpoints and the approximation error. Using such an approach, the algorithm proposes a set of solutions at its end. This set which is called the Pareto Front in the multi objective optimization field contains solutions that represent trade-offs between the two classical quality criteria of polygonal approximation : the Integral Square Error (ISE) and the number of vertices. The user may choose his own solution according to its objective. The proposed approach is evaluated on curves issued from the literature and compared with many classical approaches.

Rotation and scale variations complicate the matters of shape description and recognition because these variations change the location of points composing the shape. However, some geometric invariant points and the relations among them are not changed by these variations. Therefore, if points in image space depicted with the coordinates system can be transformed into a new coordinates system that are invariant to rotation and scale, the problem of shape description and recognition becomes easier. This paper presents a shape description method via transformation from the image space into the invariant feature space having - and -axes: representing relative distance from a centroid and contour segment curvature (CSC) respectively. The relative distance describes how far a point departs from the centroid, and the CSC represents the degree of fluctuation in a contour segment. After transformation, mesh features were used to describe the shape mapped onto the plane. Traditional mesh features extracted from the plane are sensitive to rotation, whereas the mesh features from the plane are robust to it. Experimental results show that the proposed method is robust to rotation and scale variations.

We propose a method for extracting the geometric feature and the comprehensive fluctuation from time-series data and also a method for detecting a reference sequence effectively on the basis of the distance graph. The prevalent methods such as one based on the frequency characteristics do not deal with time-series data in the time dimention. Therefore, our method for extracting the features is temporally sensitive to fluctuations of time-series data. We experimented using the time-series data whose frequency bands were changed variously in order to make clear the availability of the proposal procedures such as smoothing and encoding.

This paper presents a method of parameterization for online freehand drawing objects based on a piecewise cubic Bezier curve approximation. The target is to represent sketches in a compact format within a certain error tolerance with lower computation to be practically adaptable for the online graphics input. A set of user’s intended breakpoints in digital ink is firstly produced in terms of pen speed and local curvatures. Each of strokes of a skechy shape is then parameterized by the optimization of piecewise Bezier curve approximation to minimize the fitting error between stroke path and the curve. The experimental results show both effective and efficient for a wide range of drawing graphic objects.