Catálogo de publicaciones - libros

Texture analysis methods are widely used in various monitoring and measurement tasks in machine vision solutions. In this paper we present a novel method for the determination of grain size distributions in the manufacturing processes of crystalline products. Our method, (MDH), is based on statistical gray level differences in the texture images. Using this method, it is possible to estimate the grain size distributions in the images. It is also possible to monitor the average grain sizes in the image series acquired during the crystallization process. This is carried out by determining the center of gravity (CoG) of the distribution represented by MDH. Experimental results obtained from images acquired from a carbohydrate crystallization process reveal that the proposed method is useful in in-line grain size measurement tasks.

When an active contour is applied to a noisy image, the contour is sometimes attracted to a local energy minimum, since the noise gives rise to high rates of change of the image gray levels. In this paper we will describe a novel method of overcoming this problem by using a sparse set of points to represent the active contour and randomly varying the positions of these points.

Many image segmentation approaches rely upon or are enhanced by using spatial relationship information between image regions and their object correspondences. Spatial relationships are usually captured in terms of relative neighborhood graphs such as the Delaunay graph. Neighborhood graphs capture information about which objects are close to each other in the plane or in space but may not capture complete spatial relationships such as containment or holes. Additionally, the typical approach used to compute the Delaunay graph (or its dual, the Voronoi polytopes) is based on using only the point-based (i.e., centroid) representation of each object. This can lead to incorrect spatial neighborhood graphs for sized objects with complex topology, eventually resulting in poor segmentation. This paper proposes a new algorithm for efficiently, and accurately extracting accurate neighborhood graphs in linear time by computing the Hamilton-Jacobi generalized Voronoi diagram (GVD) using the exact Euclidean-distance transform with Laplacian-of-Gaussian, and morphological operators. The algorithm is validated using synthetic, and real biological imagery of epithelial cells.

A search engine for font recognition in very large font data-bases is presented and evaluated. The search engine analyzes an image of a text line, and responds with the name of the font used when writing the text. After segmenting the input image into single characters, the recognition is mainly based on eigenimages calculated from edge filtered character images. We evaluate the system with printed and scanned text lines and character images. The database used contains 2763 different fonts from the English alphabet. Our evaluation shows that for 99.8 % of the queries, the correct font name is one of the five best matches. Apart from finding fonts in large databases, the search engine can also be used as a pre-processor for Optical Character Recognition.

Non-photorealistic rendering (NPR) techniques aim to outline the shape of objects and reduce visual clutter such as shadows and inner texture edges. As the first phase result of our entire research, this work is concerned with a structured light based approach that efficiently detects depth edges in real world scenes. Depth edges directly represent object contours. We exploit distortion of the light pattern in the structured light image along depth discontinuities to reliably detect depth edges. However, in reality, distortion along depth discontinuities may not occur or be large enough to detect depending on the distance from the camera or projector. For practical application of the proposed approach, we have presented a novel method that guarantees the occurrence of the distortion along depth discontinuities for a continuous range of object location. Experimental results show a great promise that the technique can successfully provide object contours to be used for non-photorealistic rendering.

By combining weighted distances and distances based on neighbourhood sequences, a new family of distance functions with potentially low rotational dependency is obtained. The basic theory for these distance functions, including functional form of the distance between two points, is presented for the face-centered cubic grid and the body-centered cubic grid. By minimizing an error function, the optimal combination of weights and neighbourhood sequence is derived.

This paper proposes unsupervised perceptual segmentation of natural color images using a fuzzy-based hierarchical algorithm. color space is used to represent color features and statistical geometrical features are adopted as texture features. A fuzzy-based homogeneity measure makes a fusion of color features and texture features. Proposed hierarchical segmentation method is performed in four stages: simple splitting, local merging, global merging and boundary refinement. Experiments on segmentation of natural color images are presented to verify the effectiveness of the proposed method in obtaining perceptual segmentation.

This paper presents a new method for creating a thick shell tetrahedral mesh from a triangular surface mesh. Our main goal is to create the thickest possible shell mesh with the lowest possible number of tetrahedrons.

Low count tetrahedral meshes is desirable for animating deformable objects where accuracy is less important and to produce shell maps and signed distance fields. In this work we propose to improve convergence rate of past work.

In this paper we introduce affine invariants based on a multiscale framework combined with nonlinear comparison operations. The resulting descriptors are histograms, which are computed from a set of comparison results using binary coding. The new constructions are analogous to other multiscale affine invariants, but the use of highly nonlinear operations yields clear advantages in discriminability. This is also demonstrated by the experiments, where comparable recognition rates are achieved with only a fraction of the computational load. The new methods are straightforward to implement and fast to evaluate from given image patches.

This paper focuses on fuzzy image denoising techniques. In particular, we develop a new fuzzy impulse noise detection method. The main difference between the proposed method and other state-of-the-art methods is the usage of the colour components for the impulse noise detection method that are used in a more appropriate manner. The idea is to detect all noisy colour components by observing the similarity between (i) the neighbours in the same colour band and (ii) the colour components of the two other colour bands. Numerical and visual results illustrate that the proposed detection method can be used for an effective noise reduction method.