Ground Target Recognition Using Rectangle Estimation

Abstract

We propose a ground target recognition method based on 3D laser radar data. The method handles general 3D scattered data. It is based on the fact that manmade objects of complex shape can be decomposed to a set of rectangles. The ground target recognition method consists of four steps; 3D size and orientation estimation, target segmentation into parts of approximately rectangular shape, identification of segments that represent the targets functional/main parts and target matching with CAD models. The core in this approach is rectangle estimation. The performance of the rectangle estimation method is evaluated statistically using Monte Carlo simulations. A case study on tank recognition is shown, where 3D data from four fundamentally different types of laser radar systems are used. Although the approach is tested on rather few examples, we believe that the approach is promising.Keywords: Rectangle estimation, laser radar, automatic target recognition 1 Ground target recognition using rectangle estimation Christina Grönwall, Fredrik Gustafsson, Mille Millnert Abstract-We propose a ground target recognition method based on 3D laser radar data. The method handles general 3D scattered data. It is based on the fact that man-made objects of complex shape can be decomposed to a set of rectangles. The ground target recognition method consists of four steps; 3D size and orientation estimation, target segmentation into parts of approximately rectangular shape, identi cation of segments that represent the target's functional/main parts and target matching with CAD models.The core in this approach is rectangle estimation. The performance of the rectangle estimation method is evaluated statistically using Monte Carlo simulations. A case study on tank recognition is shown, where 3D data from four fundamentally different types of laser radar systems are used. Although the approach is tested on rather few examples, we believe that the approach is promising.

Simulating the performance of laser imaging and range profiling of small surface vessels

“…This can be done by a rectangle estimator developed by Gronwall et al [22]. We tested this rectangle estimator to derive width and height of three boats (No.…”

“…A somewhat simpler, but maybe less robust, 3D target recognition technique have been developed at FOI [22]. It was developed for ground vehicle recognition but has been used for land mine recognition as well.…”

A GPS-based system design for the recognition and tracking of moving targets

“…The formula which gives the middle point of the object is as follows [15]: (3) N indicates the number of comer pixels. The Euclidean distance between pixels P and Q is calculated as follows for P=(xl,yl) and Q=(x2,y2) points [16]: In Figure 6, the distances to the point taken as reference for the determination of the middle point of the object are shown. When calculating the middle point, the entire image received from the camera is considered as an object, and the locations outside of the object are treated as empty space.…”

Registration and change detection techniques using 3D laser radar data from natural environments

“…Tests have shown that when there are at least 50 samples distributed over the target, its dimensions can be determined with a 10% uncertainty [4]. Usually this is enough to determine its class.…”

“…Usually this is enough to determine its class. To determine the target type, usually at least 200 well-distributed target samples are needed [3,4]. It is thus desirable to be able to increase the number of such samples by incorporating data obtained from several measurements, either with one sensor moved to different positions or with several sensors in a sensor network.…”