Novel Representations of Elevation Data

Dynamic use of increasing, high fidelity GIS data remains a significant challenge in processing GIS data. As the demand by users to exploit detailed DEM data in real-time increases, novel data compression methodologies resulting in compact datasets are in high demand. Especially those that can be decompressed in real-time while rigorously maintaining error constraints. The aim of this effort to develop, test, demonstrate and deploy a working encoding/decoding system that facilitates the generation and subsequent real time consumption of highly-compressed and accurate terrain data. A prime focus of the development is to generate DEM representations that can be used within a wide array of industry standard GIS software. Our algorithms naturally lead to representations which provide richer information than can be embraced by standard DEM files. For example, processing of LiDAR data to produce DEM files is inherently lossy. The prototype developed will comprise an end-to-end solution by which input data (such as LiDAR) may be directly processed and compressed, leading to more compact and accurate representations. The final output may be represented as DEM data for downstream processing, but we also contend that having downstream software access the more accurate representations provides significant advantages.