The point containment predicate which specifies if a point is part of a mathematically well-defined object is a crucial problem in computergraphics. Point-driven algorithms can be executed for several points in paral...
详细信息
The point containment predicate which specifies if a point is part of a mathematically well-defined object is a crucial problem in computergraphics. Point-driven algorithms can be executed for several points in parallel as there is no interdependence among the computations for different points. The paper presents a variety of parallel configurations to counteract the main disadvantage of the point-driven algorithm: its slowness on a standard uniprocessor software implementation for interactive editing and manipulation.
We present a method to measure reflectance and texture of surfaces in a one step process. For later use in digital image synthesis, it is mandatory to separate the gathered intensity values into these two parts to eli...
详细信息
We present a method to measure reflectance and texture of surfaces in a one step process. For later use in digital image synthesis, it is mandatory to separate the gathered intensity values into these two parts to eliminate highlighting artifacts from textures. Our image based measurement system delivers bidirectional reflectance distribution function (BRDF) values distributed over the surface of the material under investigation. After fitting a reflectance model to the gathered data, we estimate the modulation of the diffuse reflectance coefficient which represents the texture. The last step analyzes the texture to get a parameterized and compact description of the measured surface properties. These results allow us to apply the gathered surface properties to objects with arbitrary shape and size. To keep the measurement system simple, a standard CCD camera and light source are used.
One of the difficulties of pattern recognition is developing a good evaluation of the classes presented in a scene. To suitably describe those classes it is necessary to find feature spaces which allow them to be dist...
详细信息
One of the difficulties of pattern recognition is developing a good evaluation of the classes presented in a scene. To suitably describe those classes it is necessary to find feature spaces which allow them to be distinguished from each other. We propose an unsupervised segmentation/classification technique associated with textural description and report the results obtained, which are quite encouraging.
The article presents a method for the automatic extraction of the contents of passive and/or active cells in forms. The approach is based on the analysis and recognition of the types of intersection of the lines that ...
详细信息
The article presents a method for the automatic extraction of the contents of passive and/or active cells in forms. The approach is based on the analysis and recognition of the types of intersection of the lines that make up such cells. Very little a priori knowledge of the form is required. The performance of this approach depends on the correction module mechanisms for detection and correction of errors generated during the intersection identification phase. The potentialities and advantages of this approach are described and illustrated with tests carried out on different form bases.
The work describes a new algorithm for segmentation of handwritten cursive words, based on the following two hypotheses: 1) natural segmentation points (SPs) exist corresponding to disconnected characters; and 2) the ...
详细信息
The work describes a new algorithm for segmentation of handwritten cursive words, based on the following two hypotheses: 1) natural segmentation points (SPs) exist corresponding to disconnected characters; and 2) the physical SPs between connected characters are located in ligatures. The morphological operations of opening and closing are utilized in the determination of word ligatures. A genetic algorithm was employed in the search for the optimum structuring element to determine ligatures.
Edge detection and image segmentation algorithms usually operate on an image to extract geometrical information based on pixel colors. For ray-traced images, the presence of geometrical information on the scene from w...
详细信息
Edge detection and image segmentation algorithms usually operate on an image to extract geometrical information based on pixel colors. For ray-traced images, the presence of geometrical information on the scene from which the image was rendered allows for a completely different approach. We present an algorithm that divides rays into equivalence classes, or categories. The category information is generated during the rendering process and used to determine edges in the resulting image. Detected edges can later be used to help determine areas subject to aliasing. Little effort is needed to implement the described algorithms over an existing ray tracer. Furthermore, the extra computational and memory requirements are modest.
The 3D reconstruction problem from a single endoscope image of a smooth object is studied in the context of the shape from shading methods and considering a single light source at the camera projection center. Based o...
详细信息
The 3D reconstruction problem from a single endoscope image of a smooth object is studied in the context of the shape from shading methods and considering a single light source at the camera projection center. Based on a curve expansion shape from shading algorithm, a spherical projection model for the endoscope camera and a dichromatic model for the surface reflectance, an approach to solve practical problems, namely the endoscope image distortion and the removal of the image specular reflection component, is presented. Results obtained from application of this approach to synthetic and real images are presented.
Morphological residues represent an image in a hierarchical way by means of a decomposition of its structures and according to a size parameter /spl lambda/. From this decomposition, we can obtain a relation between t...
详细信息
Morphological residues represent an image in a hierarchical way by means of a decomposition of its structures and according to a size parameter /spl lambda/. From this decomposition, we can obtain a relation between the different residual levels associated with the complexity of the image structures. In this work, we introduce a method for filtering of components in gray-scale images based on the morphological residue decomposition which takes into account a size parameter and a certain level of complexity of the different structures we want to be filtered.
This paper presents a system that is being developed for the recognition of the handwritten legal amount in brazilian bank checks. Our strategy used to approach the handwritten legal amount recognition problem puts on...
详细信息
This paper presents a system that is being developed for the recognition of the handwritten legal amount in brazilian bank checks. Our strategy used to approach the handwritten legal amount recognition problem puts on evidence the keywords: "mil", "reals/real", "centavos/centavo" which are almost always present in each amount. The recognizer, based on hidden markov models, does a global word analysis, therefore, it does not carry out an explicit segmentation of words into characters or pseudo-characters. In this context, each word image is transformed into a sequence of observations using pre-processing and feature extraction stages. Our system, when tested on our database simulating brazilian bank checks, shows the viability of our approach.
暂无评论