Steven G. Blask

Jigsaw phase III: a miniaturized airborne 3-D imaging laser radar with photon-counting sensitivity for foliage penetration

Jigsaw three-dimensional (3D) imaging laser radar is a compact, light-weight system for imaging highly obscured targets through dense foliage semi-autonomously from an unmanned aircraft. The Jigsaw system uses a gimbaled sensor operating in a spot light mode to laser illuminate a cued target, and autonomously capture and produce the 3D image of hidden targets under trees at high 3D voxel resolution. With our MIT Lincoln Laboratory team members, the sensor system has been integrated into a geo-referenced 12-inch gimbal, and used in airborne data collections from a UH-1 manned helicopter, which served as a surrogate platform for the purpose of data collection and system validation. In this paper, we discuss the results from the ground integration and testing of the system, and the results from UH-1 flight data collections. We also discuss the performance results of the system obtained using ladar calibration targets.

Obscuration measurements of tree canopy structure using a 3D imaging ladar system

Recently-developed airborne imaging laser radar systems are capable of rapidly collecting accurate and precise spatial information for topographic characterization as well as surface imaging. However, the performance of airborne ladar (laser detection and ranging) collection systems often depends upon the density and distribution of tree canopy over the area of interest, which obscures the ground and objects close to the ground such as buildings or vehicles. Traditionally, estimates of canopy obscuration are made using ground-based methods, which are time-consuming, valid only for a small area and specific collection geometries when collecting data from an airborne platform. Since ladar systems are capable of collecting a spatially and temporally dense set of returns in 3D space, the return reflections can be used to differentiate and monitor the density of ground and tree canopy returns in order to measure, in near real-time, sensor performance for any arbitrary collection geometry or foliage density without relying on ground based measurements. Additionally, an agile airborne ladar collection system could utilize prior estimates of the degree and spatial distribution of the tree canopy for a given area in order to determine optimal geometries for future collections. In this paper, we report on methods to rapidly quantify the magnitude and distribution of the spatial structure of obscuring canopy for a series of airborne high-resolution imaging ladar collections in a mature, mixed deciduous forest.

Point Spread Function (PSF) noise filter strategy for geiger mode LiDAR

LiDAR is an efficient optical remote sensing technology that has application in geography, forestry, and defense. The effectiveness is often limited by signal-to-noise ratio (SNR). Geiger mode avalanche photodiode (APD) detectors are able to operate above critical voltage, and a single photoelectron can initiate the current surge, making the device very sensitive. These advantages come at the expense of requiring computationally intensive noise filtering techniques. Noise is a problem which affects the imaging system and reduces the capability. Common noise-reduction algorithms have drawbacks such as over aggressive filtering, or decimating in order to improve quality and performance. In recent years, there has been growing interest on GPUs (Graphics Processing Units) for their ability to perform powerful massive parallel processing. In this paper, we leverage this capability to reduce the processing latency. The Point Spread Function (PSF) filter algorithm is a local spatial measure that has been GPGPU accelerated. The idea is to use a kernel density estimation technique for point clustering. We associate a local likelihood measure with every point of the input data capturing the probability that a 3D point is true target-return photons or noise (background photons, dark-current). This process suppresses noise and allows for detection of outliers. We apply this approach to the LiDAR noise filtering problem for which we have recognized a speed-up factor of 30-50 times compared to traditional sequential CPU implementation.