Computer Graphics and Image Processing最新文献

This paper is concerned with the problem of approximating digitized pictures by polygons. The digitized picture is represented by a two-dimensional array of points, and its is desired to convert the given array into a set of polygons, such that each polygon has the least number of sides and the error between the initial points and the approximated lines is less than a given constant (E). There are many other solutions to this problem, but to evaluate the error, they use either the least-squares method or the cone intersection method. In this paper, it is shown that the minimax approximation that minimizes the maximum distance between the given points and the approximated line is the best approximation for the problem. A method is presented for obtaining the minimax approximated lines from the given N points in time proportional to N ∗ log N. From the obtained lines a polygon is then found using another algorithm. The polygon satisfies the condition that the number of sides is minimum and the maximum distance between the given points and the sides is less than the given E.

Automatic boundary detection in image signals obtained from an Anger scintillation camera requires special methods. On the one hand the mapped object is defocussed by the limited resolution of the camera and the radioactive process; on the other hand the true boundaries of the mapped organs for the comparison of different algorithms are not very easily obtainable. Therefore different filtering and boundary detection algorithms have been applied to phantom scintigrams from different objects, i.e., hollow spheres.

The complexity of computer-generated scenes is often greater than the display can handle. As a result, it is important to be able to select a subset of the scene which is appropriate for display. Without this subset filtering, small objects are aliased, producing Moiré patterns, flickering, and other disturbing display artifacts. This paper presents a scene representation and an associated display algorithm that together provide ease of subset filtering. The scene is hierarchically constructed and the filter selects an appropriate subtree of the hierarchy for display. The bottom nodes of the display subtree are visually faded with their parent nodes to produce a pleasing fade-out of objects that approach the limit of resolution. Combined with conventional antialiasing, this technique produces satisfying images, both still and animated, of scenes with a wide range of detail.

Standard cubics, i.e., nonrational cubics, are the simplest twisted curves—hence their considerable importance for CAGD. Rational cubics allow modification of their fullness even when the end tangents are kept fixed; this is the reason why they are occasionally preferred to standard cubics in CAGD. We shall point out connections between the various representations and their underlying geometric properties. This should serve as an easy and intuitive introduction and help the potential user choose a suitable representation. (This survey was initiated by Forrest's article “The twisted cubic curve” [14] and the subsequent correspondence with A. Ball [4] on some minor misunderstandings.) Some new algorithms can be obtained in a straightforward way.

One of the long-standing research problems in computer vision is the conversion of spatial information between the concrete domain of position in an image or in coordinate space and the more abstract domain of symbolic scene descriptions. This paper discusses a technique for solving the layout problem, one side of this general conversion problem. From a scene description composed of a set of symbolic spatial relations with fuzzy truth values, estimates of the objects' positions are derived. The position estimates consist of coordinate intervals with fuzzy truth values expressing the confidence associated with each interval.

Quasi-preservation of topological structures of binary pictures by a group of parallel local operations is considered. The topology is defined in terms of adjacency among binary components. Parallel local operations treated here are allowed to alter the topology only by deleting simply connected components. They also are required to annihilate all components except for the background. The window for these operations is 2 × 2, and is asymmetric with respect to the point whose value is to be calculated at the next step of operation. The group of operations are obtained by determining the necessary and sufficient conditions for a parallel operation to satisfy the quasi-preservation property thus defined. Some other considerations are also given.

A geometric modeling technique called Octree Encoding is presented. Arbitrary 3-D objects can be represented to any specified resolution in a hierarchical 8-ary tree structure or “octree” Objects may be concave or convex, have holes (including interior holes), consist of disjoint parts, and possess sculptured (i.e., “free-form”) surfaces. The memory required for representation and manipulation is on the order of the surface area of the object. A complexity metric is proposed based on the number of nodes in an object's tree representation. Efficient (linear time) algorithms have been developed for the Boolean operations (union, intersection and difference), geometric operations (translation, scaling and rotation), N-dimensional interference detection, and display from any point in space with hidden surfaces removed. The algorithms require neither floating-point operations, integer multiplications, nor integer divisions. In addition, many independent sets of very simple calculations are typically generated, allowing implementation over many inexpensive high-bandwidth processors operating in parallel. Real time analysis and manipulation of highly complex situations thus becomes possible.