Improving range estimation of a 3-dimensional FLASH LADAR via blind deconvolution

The purpose of this research effort is to improve and characterize range estimation in a three-dimensional FLASH LAser Detection And Ranging (3D FLASH LADAR) by investigating spatial dimension blurring effects. The myriad of emerging applications for 3D FLASH LADAR both as primary and supplemental sensor necessitate superior performance including accurate range estimates. Along with range information, this sensor also provides an imaging or laser vision capability. Consequently, accurate range estimates would also greatly aid in image quality of a target or remote scene under interrogation.;Unlike previous efforts, this research accounts for pixel coupling by defining the range image mathematical model as a 2D convolution between the system spatial impulse response and the object (target or remote scene) at a particular range slice. Using this model, improved range estimation is possible by object restoration from the data observations. Object estimation is principally performed by deriving a blind deconvolution Generalized Expectation Maximization (GEM) algorithm with the range determined from the estimated object by a normalized correlation method. Theoretical derivations and simulation results are verified with experimental data of a bar target taken from a 3D FLASH LADAR system in a laboratory environment. Simulation examples show that the GEM object restoration improves range estimation over the unprocessed data and a Wiener filter method by 75% and 26% respectively. In the laboratory experiment, the GEM object restoration improves range estimation by 34% and 18% over the unprocessed data and Wiener filter method respectively.;This research also derives the Cramer-Rao bound (CRB) on range separation estimation of two point sources interrogated by a 3D FLASH LADAR system. Using an unbiased estimator, range separation estimation variance was attained through simulation and compared favorably to the range separation CRB theory. The results show that the CRB does indeed provide a lower bound on the range separation estimation variance and that the estimator is nearly efficient. With respect to the estimator, traditional pixel-based estimators like peak detection and matched filtering are biased because they assume there is only one target in the pixel. Therefore, an unbiased estimator was derived accounting for the possibility of two targets within a single pixel.;Additionally, among other factors, the range separation CRB is a function of two LADAR design parameters (range sampling interval and transmitted pulse-width), which can be optimized using the expected range resolution between two point sources. Typically, a shorter transmitted pulse-width corresponds to better range resolution (the ability to resolve two targets in range). Given a particular range sampling capability determined by the receiver electronics, the CRB theory shows there is an optimal pulse-width where a shorter pulse-width would increase estimation variance due to the under-sampling of the pulse and a longer pulse-width would degrade the resolving capability. Using both CRB theory and simulation results, an investigation is accomplished that finds the optimal pulse-width for several range sampling scenarios. For example, given a Gaussian received pulse model sampled every 1.876 ns, both range separation CRB theory and simulation predict an optimal pulse-width standard deviation equal to 0.88 ns. As the speed of the optical receiver is increased, the range resolution is improved with a corresponding narrower optimal pulse-width attained by the ability to sufficiently sample a shorter pulse-width. Conversely, the optimal pulse-width is wider with slower electronics with an associated negative impact on range resolution.