The Application of Inverse Filters to 3D Microscanning of LADAR Imagery

Abstract

Microscanning is an effective technique for re-of the scene. One particular system that collects 3-D images ducing aliasing and increasing resolution in 2D images pro-is a FLASH LADAR imager [1], [2], [3], [4]. One drawback duced by non-coherent imaging systems. Both the aliasing of this readout technology is that it has not been miniaturized reduction and resolution enhancement are accomplished by to the same degree other image acquisition devices have, such increasing the effective spatial sampling interval through subas traditional silicon CCD (Charge Coupled Device) arrays. pixel movements of the field of view on the detector array.As a consequence ofthis the pixel pitches in FLASH LADAR In this paper the authors examine (using both simulated and imaging arrays are much larger than other imaging systems. real data) the application of microscanning on 3D coherent Large pixel pitches make it difficult to sample the scene at the imagery (with application to 2D coherent imagery). The ap-diffraction limit of the imaging system. This undersampling plication of Inverse filtering techniques are also examined to of the image produces a loss of spatial resolution. resolve difficulties associated with blurring by the detector and optical systems. The effects of blurring by the detec-Microscanning has been proposed as a method for attaintor and optical system decrease image resolution when miing diffraction-limited spatial resolution from imaging syscroscanning is attempted at sub-pixel movements of less than tems whose spatial sampling is inadequate to achieve the half the detector width. A discrete filter is designed using diffraction-limit [5], [6]. Application ofmicro-scanning techthe MTF of the imaging system. This filter is then applied niques to 3-D LADAR imaging systems imparts technical to images collected with a 128x128x20 FLASH 3D LADAR challenges not encountered in the 2-D micro-scanning probsystem. The authors also investigate the difficulties that arise lem. Chief among these problems is the fact that each 3-D when the MTF of the system is not well characterized and image cube will phase with the target differently than others when image registration errors and a low signal to noise ratio due to the fact that the time the FLASH LADAR imager beare present in the system. gins recording data depending on when the returning pulse exceeds some predetermined threshold. This means that the

Imaging flash LIDAR for safe landing on solar system bodies and spacecraft rendezvous and docking

“…The SR algorithm, as illustrated in Fig. 8, is based on three major blocks: 3-D modification of the back-projection method 25 , a 6-D version of the Lucas-Kanade registration algorithm 28 , and the modified inverse filtering algorithm 29 . A backprojection method has recently been applied to solve the super-resolution problem in the 2-D environment 25 .…”

Multiframe fusion of undersampled 3D imagery

“…This technique has been successfully applied to the reconstruction of 3D images taken with range-gated LADAR systems, like the ASC tigereye system [4]. These types of 3D imagers collect 2D images corresponding to a slice of a 3D scene by filtering out the light that comes from distances that are outside the set range gate.…”

A super-resolution algorithm for enhancement of flash lidar data: flight test results

“…SR is a wellestablished technique for enhancing two-dimensional (2-D) images and over the years, a large number of algorithms have been developed for processing intensity images produced by different types of imaging systems [1][2][3][4][5]. The emergence of Flash Lidar technology that is capable of generating three-dimensional (3-D) images at video rates, with a relatively large number of pixels has created a need to extend 2-D SR techniques to 3-D images [7][8][9][10][11][12][13].…”

“…To overcome this limitation, we developed a new SR method to enhance the resolution of individual Flash Lidar 3D image frames and generate a sufficiently large DEM [14]. Over approach is can be divided into three major blocks: 3D modification of back projection method [11], 6D version of the Lucas-Kanade registration algorithm [15], and the modified inverse filtering algorithm [13,16]. This paper first describes the static test data gathered at the NASA Langley Research Center Lidar Test Range facility.…”