We use the Expectation-Maximization (EM) algorithm to classify 3D aerial lidar scattered height data into four
categories: road, grass, buildings, and trees. To do so we use five features: height, height variation, normal
variation, lidar return intensity, and image intensity. We also use only lidar-derived features to organize the data
into three classes (the road and grass classes are merged). We apply and test our results using ten regions taken
from lidar data collected over an area of approximately eight square miles, obtaining higher than 94% accuracy.
We also apply our classifier to our entire dataset, and present visual classification results both with and without
uncertainty. We use several approaches to evaluate the parameter and model choices possible when applying EM
to our data. We observe that our classification results are stable and robust over the various subregions of our
data which we tested. We also compare our results here with previous classification efforts using this data.
In this work, we introduce GSIFT (Geometric Scale Invariant Terrain Feature Transform), geometric descriptors that are invariant to translation, rotation, and scaling. SIFT (Scale Invariant Feature Transform) descriptors have been found to be very successful in a variety of computer vision tasks. GSIFT expands SIFT by using the novel technique of binding features at different scales to associate priorities/importance to features based on its scale and persistence among different scales. As a first step to obtain scale invariance, we create a multi-scale pyramid for detecting important features such as maxima, minima, and saddle points for 1D and 2D height fields. We use relative height histograms as GSIFT with support regions determined by the priority of the feature. We use symmetric chi-square as the similarity measure to compare geometric features. Experiments with GSIFT and SIFT (on images constructed from height fields) on both synthetic and real height field data sets show that GSIFT provide comparable and in some cases, better results for data registration. GSIFT has the added advantage that it can be used by scientists in compressing and registering (or non-registering) height data, where it is important for them to understand which features to keep/discard or why the registration is good/poor, when the data is obtained at different scales from independent sources.
Weather is one of the major causes of general aviation accidents. One possible cause is that the pilot may not absorb and retain all the weather information she is required to review prior to flight. A second cause is the inadequacy of in-flight weather updates: pilots
are limited to verbal updates via aircraft radio contact with a
ground-based weather specialist. We propose weather visualization and
interaction methods tailored for general aviation pilots to improve understanding of pre-flight weather data and improve in-flight weather updates. Our system, Aviation Weather Environment (AWE), utilizes information visualization techniques, a direct manipulation
graphical interface, and a speech-based interface to improve a pilot's situational awareness of relevant weather data. The system design is based on a user study and feedback from pilots.
The importance of accurate registration of GPS (Global Positioning System) tracked objects embedded within a GIS (Geographic Information Systems) context has emerged as a critical need in several land, marine, and air navigational systems both for civilian and defense applications. The object of this work is to measure, model and geo-spatially register the positional accuracy of objects carrying GPS receivers against a GIS background. Although positional accuracy is affected by a number of factors, in this work we have focused on GPS modes (standalone or differential), type of environment (urban or foliage), and type of expected movement of objects. The Ashtech Z-12 sensor is used to collect the data. Linear models are used to estimate the errors associated with the horizontal position information. This error is then visualized upon a `1/2 foot resolution aerial imagery of the UCSC (University of California, Santa Cruz) campus. Estimates of speed and direction errors are used to create visualizations of spatio-temporal uncertainty associated with an object walking through the campus.
We describe three systems that use natural or event-based sounds as means of data delivery. In these systems we have mapped data to natural sounds using metaphors. In the first system we evaluate the use of sounds of air, horn, and train to convey ordered numeric values between 1 to 6. An example of the metaphor used here is the association of speed values to the sound of a moving train at different speeds. In the second system, we use sounds of ocean waves to convey whether the exposure in a protein structural alignment is buried, partially exposed or fully exposed. The metaphor used here is the association of sound with how exposed the user is with respect to the ocean. In the third system, we map animal sounds such as the sound of a roaring lion or a chirping bird to certain stocks based on user preferences. The behavior of the stocks are then sounded by the use of whistles and car crash to signify the movement in process of the stocks. An up whistling sound can be clearly associated with an uptrend. We present and discuss the results of user evaluation studies for all the three systems.
WebVis, the Hierarchical Web Home Page Visualizer, is a tool for managing home web pages. The user can access this tool via the WWW and obtain a hierarchical visualization of one's home web pages. WebVis is a real time interactive tool that supports many different queries on the statistics of internal files such as sizes, age, and type. In addition, statistics on embedded information such as VRML files, Java applets, images and sound files can be extracted and queried. Results of these queries are visualized using color, shape and size of different nodes of the hierarchy. The visualization assists the user in a variety of task, such as quickly finding outdated information or locate large files. WebVIs is one solution to the growing web space maintenance problem. Implementation of WebVis is realized with Perl and Java. Perl pattern matching and file handling routines are used to collect and process web space linkage information and web document information. Java utilizes the collected information to produce visualization of the web space. Java also provides WebVis with real time interactivity, while running off the WWW. Some WebVis examples of home web page visualization are presented.
Many direct volume rendering algorithms are routinely used to render volumetric data in scientific applications. Different algorithms, however, produce different results and may lead to different interpretations of the scientific data. There are many factors that contribute to different results including the rendering algorithm such as the ray tracing or projection method. Within the algorithm itself such as ray tracing, there are many factors such as the number of samples, desired opacity, sample location etc., that lead to different images. In some of these cases, the differences between the images are significant enough to demand further investigation. In this work we investigate the sensitivity of differences between images to the viewing angle. In other words, we employ different visualization methods and obtain different images for the same viewing angle. The dependence of these differences on the viewing angle is then investigated. These difference images are visualized by pasting them on six sides of a cube corresponding to six different viewing angles. These differences are also visualized by using glyphs on a sphere, where each point on a sphere corresponds to a viewing angle. For most viewing angles, these differences are not significant and therefore, in such cases, inexpensive visualization algorithms can be employed. In some cases, where the differences are large, our technique compels the user to incorporate uncertainty while drawing conclusions from those images. We also discuss extensions of this work to incorporate uncertainty in volumetric visualization corresponding to different choices of color mapping or opacity.
Uncertainty or errors are introduced in scientific visualization as the data is acquired, transformed and rendered. In this work, we focus on the uncertainty introduced by the interpolation process. Appropriate tools for investigating the geometry of these interpolants are crucial in determining the accuracy of these interpolants or comparing and contrasting these interpolants. A visualization tool for investigating this geometric uncertainty of surface interpolants was recently reported by Lodha et al. This tool combines various traditional techniques such as side-by-side viewing, differencing, pseudo-coloring with more modern techniques such as glyphs, transparency and texture. One of the limitations of this software was that it could be used to compare only tow surfaces at a time. This was rather inconvenient if one were to compare two surface interpolants at the same time with a given 'true' surface. MUSURF fills this gap by allowing users to visually compare two surface interpolants together with the 'true' surface. This simplifies the task of comparative evaluation of two interpolants considerably. MUSURF also extends the software in several directions. First, it adds a few more capabilities to the geometric uncertainty system. The two most important capabilities are cross-sectional contours and correlating the errors with glyphs. Finally, MUSURF incorporates sound capabilities to distinguish between several surface interpolants at the same time. We also present applications of MUSURF to analytic and digital elevation data sets.
Hierarchical decomposition of data using Haar and Legendre scaling functions as well as multiresolution compression and decomposition of data using hyperbolic 3D Haar wavelets, Battle-Lemarie wavelets, and biorthogonal wavelets have been used in the past to visually explore large volumetric data sets. In this work we explore the use of Legendre wavelets for efficient volumetric compression and rendering of data. There are several advantages of using Legendre wavelets. First, by using wavelets rather than scaling functions, we gain the advantages associated with multiresolution decomposition of data. This includes efficient exploration of data at different levels of detail and advantages of incremental rendering of data and progressive transmission. Second, the main advantage of these wavelets over other wavelet models arises from the fact that they do not overlap and therefore require filters of only unit length. In contrast, Battle-Lemarie wavelets require filters of infinite length. Similarly, if B-spline wavelets are used, they will require filters of infinite length as well. Biorthogonal wavelets require filters of finite length; however Legendre wavelets use filters of only unit length. This results in relatively simple and efficient computation. We use coherent projection method and L<SUB>2</SUB> error criterion to compress and render the volumetric data, although the model is flexible to accommodate other volumetric rendering techniques and other error criteria. The Legendre wavelet model for volumetric data compression and rendering has been implemented. The system has been used for visual data exploration of several large volumetric data sets. Detailed statistical measures of compression ratios, rendering time, and associated errors have been derived for different threshold values of many volumetric data sets. Although for lossless compression Legendre wavelet model requires much more time and space, it clearly outperforms Haar wavelet model in compression and image quality for lossy compression with very small L<SUB>2</SUB> errors. This characteristic is very helpful in visual exploration of data.
Environmental data have inherent uncertainty which is often ignored in visualization. For example, meteorological stations measure wind with good accuracy, but winds are often averaged over minutes or hours. As another example, Doppler radars (wind profilers and ocean current radars) take thousands of samples and average the possibly spurious returns. Others, including time series data, have a wealth of uncertainty information that the traditional vector visualization methods such as using wind barbs and arrow glyphs simply ignore. We have developed new vector glyphs to visualize uncertain winds and ocean currents. Our approach is to include uncertainty in direction and magnitude, as well as the mean direction and length, in vector glyph plots. Our glyphs show the variation in uncertainty, and provide fair comparisons of data from instruments, models, and time averages of varying certainty. We use both qualitative and quantitative methods to compare our glyphs to traditional ones. Subjective comparison tests with experts (meteorologists and oceanographers) are provided, as well as objective tests (data ink manximization), where the information density of our new glyphs and traditional glyphs are compared. We have shown that visualizing data together with their uncertainty information enhances the understanding of the continuous range of data quality in environmental vector fields.
This work describes a degree reduction method for Bezier simplexes with the following properties: (1) Symmetry: The degree reduction method is symmetric with respect to the corner points of the Bezier simplex. (2) Restriction: The degree reduction method restricted to the boundary of a Bezier simplex yields the same result as the boundary of the degree-reduced Bezier simplex. (3) Interpolation: The degree-reduced simplex of degree e interpolates the value and the first [e-1/2] derivatives at the corner points of the original Bezier simplex. (4) Optimal order of approximation: The order of approximation of the given simplex by the degree-reduced simplex is O(he+1), (where h is the diameter of the domain simplex), which is optimal for functional approximation. The method, restricted to Bezier surface, yields a new technique for degree reduction, which is easy to implement.
We describe a new method for creating rectangular Bezier surface patches on an implicit cubic surface. Traditional techniques for representing surfaces have relied on parametric representations of surfaces, that, in general, generate surfaces of implicit degree eight in case of rectangular Bezier surfaces with rational biquadratic parametrization. Thus we have achieved low-degree algebraic surface patch construction by reducing the implicit degree from eight to three. The construction uses a rectangular biquadratic Bezier control polyhedron, embedded within a tetrahedron and satisfying a projective constraint. The control polyhedron and the resulting cubic surface patch satisfy all of the standard properties of parametric Bezier surfaces, including
Q uadric surfaces such as cylinders and spheres play a fundamental role in CAGD.
This paper describes a new method for creating triangular surface patches on a quadric
surface. The surface patches are defined using a restricted type of quadratic Bezier
control polyhedron. The control polyhedron and the resulting quadric surface patch
satisfy all of the standard properties of parametric Bezier surfaces, including interpolation
of the corners of the control polyhedron and the convex hull property. A new
technique for creating a C1 mesh of these quadric surface patches is also introduced.