This paper describes the basis for techniques such as stochastic subdivision in the theory of random processes and estimation theory. The popular stochastic subdivision construction is then generalized to provide control of the autocorrelation and spectral properties of the synthesized random functions. The generalized construction is suitable for generating a variety of perceptually distinct high-quality random functions, including those with non-fractal spectra and directional or oscillatory characteristics. It is argued that a spectral modeling approach provides a more powerful and somewhat more intuitive perceptual characterization of random processes than does the fractal model. Synthetic textures and terrains are presented as a means of visually evaluating the generalized subdivision technique. Categories and Subject Descriptors: I.3.3 [Computer Graphics]: Picture/Image Generation; I.3.7 [Computer Graphics]: Three Dimensional Graphics and Realism -<F11.

Abstract not found

Dans cet article nous allons présenter une méthode de détection d’objets basée sur l’analyse multifractale. La détection d’objets est particulièrement difficile lorsque l’on ne connaît pas les caractéristiques des objets pouvant être présents dans l’image. Contrairement aux méthodes le plus souvent proposées, cette méthode est destinée à détecter des objets artificiels dans des images photographiques prises depuis le sol. Cette méthode peut trouver des applications dans le domaine de la robotique en evironnement inconnu ou hostile. Parmi de nombreux exemples d’applications potentielles, on peut imaginer d’utiliser cette méthode pour trouver des détritus sur une plage. Cette méthode est basée sur un calcul de spectre multifractal en utilisant la transformée de Legendre. Elle tire profit de la capacité du spectre multifractal de Legendre d’opérer une discrimination entre les textures. La méthode d’analyse d’image consiste à diviser l’image en imagettes et à calculer le spectre multifractal de Legendre de chaque imagette en utilisant une approche par comptage de rectangles. Des critères de détection sont ensuite évalués pour déterminer la position des objets, en se basant sur le fait que le spectre multifractal de Legendre a une forme différente selon la texture. La méthode utilisée pour le calcul du spectre multifractal de Legendre sera présentée dans cet article, ainsi que son application pour la détection d’objets, et enfin on interprétera les résultats expérimentaux. Mots-clé: Reconnaissance d’objet, spectre multifractal, transformée de Legendre, détection d’objet, groupement et segmentation

One of the overall aims of this study has been is to develop a set of tools that describe the general geomorphometry of a surface. On the whole, this is quite distinct from the process of identifying specific geomorphometric features such as cirques or floodplains. There are however, a number of surface features that may be used both in the specific and general geomorphometric identification process. These features can be thought of as morphometric features rather than geomorphometric in that they are characteristic of any surface. The most widely used set of morphometric characteristics, is the subdivision of all points on a surface into one of pits, peaks, channels, ridges, passes and planes (see Figure 5.1). The names of these features suggest a geomorphological interpretation, but they may be unambiguously described in terms of rates of change of three orthogonal components (see Table 5.1). Note that the components x and y are not necessarily parallel to the axes of the DEM, but are in the direction of maximum and minimum profile convexity.

this document preserves this structure and enhances it. The first 42 months of the NCGIA have demonstrated that a multi-institutional, multi-disciplinary center for the advancement of fundamental geographic and GIS research can function effectively. The new research plan builds on the previous one, but also reflects the advancement in our thinking. The five research areas listed in the original solicitation from the National Science Foundation, often referred to as "the five bullets", are herein replaced by three general research areas: spatial representation; spatial analysis; and spatial informatics. This new framework serves as the common base from which individual research efforts can be selected and planned, and into which they are integrated. The Research Initiative continues to be the NCGIA's principal vehicle for research, but is augmented by additional vehicles designed to provide greater flexibility and responsiveness to the Center's ability to conduct its mission and achieve its goals. If NCGIA is to play a truly national role, it must coordinate its research activities with those of other individuals and groups, and act as a focal point and catalyst. The Specialist Meetings held at the start of each NCGIA Research Initiative have performed a useful function in bringing together researchers interested in each topic, and in establishing a common research agenda for the community as a whole. Our published research plan (NCGIA, 1989) has strong similarities to those GIS and GIA research agendas developed and published by other groups (Craig, 1989; Maguire, 1990; Masser, 1990). At its June 1991 meeting the NCGIA Board of Directors gave strong encouragement to undertake the development of a national GIS and GIA research agenda as an NCGIA activity. Thus we see the...

this document preserves this structure and enhances it. The first 42 months of the NCGIA have demonstrated that a multi-institutional, multi-disciplinary center for the advancement of fundamental geographic and GIS research can function effectively. The new research plan builds on the previous one, but also reflects the advancement in our thinking. The five research areas listed in the original solicitation from the National Science Foundation, often referred to as "the five bullets", are herein replaced by three general research areas: spatial representation; spatial analysis; and spatial informatics. This new framework serves as the common base from which individual research efforts can be selected and planned, and into which they are integrated. The Research Initiative continues to be the NCGIA's principal vehicle for research, but is augmented by additional vehicles designed to provide greater flexibility and responsiveness to the Center's ability to conduct its mission and achieve its goals.

Abstract not found

Digital elevation model data were used to partition a mountainous landscape (northwestern Montana, USA) into terrain units at several different spatial scales. Fractal analysis of the perimeter to area relationships of the resulting partition polygons identified statistical self-similarityacross a range of spa- tial scales (approximately four orders of magnitude in partition area). The fractal dimension was higher for a relatively complex fluvially-dominated terrain than for a structurally simpler glacially-dominated terrain (1.23 vs. 1.02, respectively). The structural self-similarityexhibited by this landscape has direct implications in scaling up ecosystem process models for landscape to regional simulations.

The major problem with high spatial resolution satellite images from Landsat 7 is that imagery is not available very often (i.e. every 16 days or longer) and the coverage area is relatively small (swath width 185km), while images of lower spatial resolution from MODIS are available daily and one image covers a relatively large area (swath width 2,330km). This paper considers the feasibility of applying various down-scaling methods to combine MODIS and Landsat imagery in order to obtain both high temporal and high spatial resolution. The Surface Energy Balance Algorithm for Land (SEBAL) was used to derive daily evapotranspiration (ET) distributions from Landsat 7 and MODIS images. Two down-scaling procedures were evaluated: input down-scaling and output down-scaling. In each down-scaling scheme, disaggregated imagery was obtained by two different processes: subtraction and regression. The primary objective of this study was to investigate the effect of the different down-scaling schemes on the spatial distribution of SEBAL derived ET. We found that all of the four proposed down-scaling methodologies can generate reasonable spatial patterns of the disaggregated ET map. The results 1 25 26