Evaluating methods for implementing photogrammetric sensor platforms with various LIDAR components for use with close and mid-range particle detection systems.

摘要

Over the past decade, there has been an increase in the use of photogrammetry and LIDAR for various engineering applications. The measurement of a variety of volumes for both gaseous and solid particle clouds have been studied in this case using photogrammetric approaches. By implementing precisely calibrated high-resolution camera networks in a variety of placements, accurate volume measurements of distorted masses were found to be obtainable in closed environments. In addition, various LIDAR techniques have been reviewed to further improve the analytical capabilities of such a system in field applications. This can provide a means to analyze environmental changes in real-time.;Autodesk Maya was used to create a variety of theoretical test scenarios, in order to assist in predicting the accurate placement of each camera for a wide variety of theoretical test situations. The program allows for the full adjustment of the particle density, volume, refractive index, and luminance of gaseous volumes in addition to normal terrain and water body generation. In addition, the program is able to render physical data from any viewpoint within the generated area using the exact same settings of the cameras used in the experiment. An additional goal of this portion of the experiment was to create a means to record incidents for future evaluation.;Accurate volume measurements from distorted plume shapes were first demonstrated during the test that took place at the University of Nevada, Reno Fire Science Academy in September 2007. Over the course of three days, several smoke plumes resulting from a variety of firefighting training sessions were measured using eight cameras placed in a large-ranged circle. Using images obtained from the camera network with the PhotoModeler program, fairly accurate measurements for a variety of plume volumes were obtained during the initial experiment; unfortunately, results from this project are unavailable at this time. There were several synchronized images that proved to be unusable as well however, indicating that further improvements to the system should be made. Most of these seem to be the result of localized lighting variation changes affecting the intensity of each image. Therefore, a real-time aperture and shutter speed adjustment program should be implemented in future experiments, to increase the yield of usable images.;By implementing a variety of LIDAR systems with a sensor network, the density, velocity, spread, respective surface points, and chemical consistency of a plume can be determined. To prevent interference from the other sensors in such a network, a rapid-pulse LIDAR system implementing multiple wavelength pairing techniques would most likely be used in conjunction with the visual sensors to obtain a measurement with near-complete accuracy. In addition, techniques involving spread LIDAR applications have been overviewed, and can be implemented to improve the coverage area as well as the maximum resolution of the received data.