ASPRS

PE&RS March 2002

VOLUME 68, NUMBER 3
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
JOURNAL OF THE AMERICAN SOCIETY FOR PHOTOGRAMMETRY AND REMOTE SENSING

Peer-Reviewed Articles

225 Optimized Resource Allocation for GIS-Based Model Implementation
Michele Crosetto, Francesco Crosetto, and Stefano Tarantola

Abstract Download Full Article
This paper focuses on a new procedure to support the planning and implementation of GIS-based models. Critical decisions are often based on the outputs of such models. A major goal of the GIS planning stage is to implement a model whose output can reliably support the decision process. The procedure allows the above goal to be achieved with an optimized allocation of resources for GIS data acquisition. It is based on two important modeling tools: uncertainty analysis and sensitivity analysis. An application of the procedure to a GIS-based hydrologic model for flood forecasting is discussed.

233 Detection of Areas Associated with Flood and Erosion Caused by a Heavy Rainfall Using Multitemporal Landsat TM Data
Amod Sagar Dhakal, Takaaki Amada, Masamu Aniya, and Rishi Ram Sharma

Abstract Download Full Article
The potential of multitemporal Landsat Thematic Mapper (TM) data was examined for its use in detecting areas affected by flood and erosion from a heavy rainfall. The study area is the Kulekhani watershed (124 km2) located in the central region of Nepal. Four change-detection techniques were compared for their effectiveness including (1) Spectral Image Differencing (SID), (2) Tasseled Cap Brightness Image Differencing (TCBID), (3) Principal Component Analysis (PCA), and (4) Spectral Change Vector Analysis (SCVA). SID was performed on four raw bands (bands 1, 2, 3, and 7), and altogether seven new images (change images) were produced. Visible bands were effective in detecting affected areas. SCVA (using bands 1, 2, and 3) was found to be most accurate for detecting areas affected by flood and erosion followed by SID (band 2), PCA (using bands 1, 2, and 3), SID (band 1), and SID (band 3). The change image produced from SCVA showed overall and Khat accuracies of 88.3 percent and 75.4 percent, respectively. The analysis of spatial agreement conducted among the seven change images, produced from different techniques, varied from 89 percent to 98 percent. The change image produced from SCVA showed high spatial agreements with change images produced from PCA, SID (band 3), and SID (band 2). SCVA and SID (band 3) showed the spatial agreement of 88.1 percent and 98.7 percent in the change and no-change categories, respectively.

241 Applying GIS for Developing Regional Forest Agreements in Australia
Adrian L. Bugg, Ray D. Spencer, and Alex Lee

Abstract Download Full Article
Australia's Regional Forest Agreement (RFA) program was a national initiative to plan forest use in key areas for the next 20 years. It represents the largest undertaking of its kind in Australia, covering 12 regions in five states with a total area of 45 million hectares. Initiated in 1995 and completed in 2001, the program advanced understanding of the application of GIS in natural resources planning. This paper reviews developments in data collection, database construction, and data management for integrating large complex data sets for multiple objective planning. The data structures and methods presaged the use of object-oriented and relational database technology for spatial data and web-based metadata systems. Major advances covered linking GIS with a relational database management system (RDBMS) and process models with GIS, adopting sophisticated tools for analyzing complex problems, and development of standards and tools for data management and more user-friendly systems for data access and analysis.

251 Mapping Chinese Tallow with Color-Infrared Photography
Elijah W. Ramsey III, Gene A. Nelson, Sijan K. Sapkota, Eric B. Seeger, and Kristine D. Martella

Abstract Download Full Article
Airborne color-infrared photography (CIR) (1:12,000 scale) was used to map localized occurrences of the widespread and aggressive Chinese tallow (Sapium sebiferum), an invasive species. Photography was collected during senescence when Chinese tallow's bright red leaves presented a high spectral contrast within the native bottomland hardwood and upland forests and marsh land-cover types. Mapped occurrences were conservative because not all senescing tallow leaves are bright red simultaneously. To simulate low spectral but high spatial resolution satellite/airborne image and digital video data, the CIR photography was transformed into raster images at spatial resolutions approximating 0.5 m and 1.0 m. The image data were then spectrally classified for the occurrence of bright red leaves associated with senescing Chinese tallow. Classification accuracies were greater than 95 percent at both spatial resolutions. There was no significant difference in either forest in the detection of tallow or inclusion of non-tallow trees associated with the two spatial resolutions. In marshes, slightly more tallow occurrences were mapped with the lower spatial resolution, but there were also more misclassifications of native land covers as tallow. Combining all land covers, there was no difference at detecting tallow occurrences (equal omission errors) between the two resolutions, but the higher spatial resolution was associated with less inclusion of non-tallow land covers as tallow (lower commission error). Overall, these results confirm that high spatial ( < 1 m) but low spectral resolution remote sensing data can be used for mapping Chinese tallow trees in dominant environments found in coastal and adjacent upland landscapes.

257 Visualizing Topography by Openness: A New Application of Image Processing to Digital Elevation Models
Ryuzo Yokoyama, Michio Shirasawa, and Richard J. Pike

Abstract Download Full Article
A new parameter, here termed openness, expressing the degree of dominance or enclosure of a location on an irregular surface, is developed to visualize topographic character. Openness is an angular measure of the relation between surface relief and horizontal distance. For angles less than 90º, it is equivalent to the internal angle of a cone, its apex at a DEM location, constrained by neighboring elevations within a specified radial distance. Openness incorporates the terrain line-of-sight, or viewshed, concept and is calculated from multiple zenith and nadir angles-here along eight azimuths. Openness has two viewer perspectives. Positive values, expressing openness above the surface, are high for convex forms, whereas negative values describe this attribute below the surface and are high for concave forms. Openness values are mapped by gray-scale tones. The emphasis of terrain convexity and concavity in openness maps facilitates the interpretation of landforms on the Earth's surface and its seafloor, and on the planets, as well as features on any irregular surface-such as those generated by industrial procedures.

267 A Framework for Automatic Recognition of Spatial Features from Mobile Mapping Imagery
Zhuowen Tu and Rongxing Li

Abstract Download Full Article
Mobile mapping is a new technology for capturing georeferenced data. It is, however, still not practical to extract spatial and attribute information of objects such as infrastructure elements fully automatically. In this article, a new framework for 3D object recognition by hypothesis-and-test techniques is proposed and developed. An example of traffic-light recognition from mobile mapping images is given in detail. The hypothesis is generated according to the viewpoint dependent theory. We formulate the hypothesis test problem based on Bayesian inference and, in particular, the MAP (Maximize A Posteriori Probability). This approach functions in two major steps: (1) generation of hot-spot maps by vanishing point detection and template matching, and (2) estimation of the parameters of 3D objects (traffic lights) by Markov Chain Monte Carlo (MCMC). The developed hot-spot map generation method is, in general, faster than general color image segmentation algorithms. For example, it can handle the recognition problem with a color image of 720 by 400 pixels within a couple of minutes rather than tens of minutes to even hours when using the segmentation algorithms. The parameter estimation method uses MCMC to simulate an ergodic stochastic process so that a robust and global optimal solution can be found. The approach shows great potential for automatic object recognition in image sequences acquired by mobile mapping systems.
Top Home