During the last ten years, the availability of images acquired from unmanned aerial vehicles (UAVs) has been continuously increasing due to the improvements and economic success of flight and sensor systems. From our point of view, reliable and automatic image-based change detection may contribute to overcoming several challenging problems in military reconnaissance, civil security, and disaster management. Changes within a scene can be caused by functional activities, i.e., footprints or skid marks, excavations, or humidity penetration; these might be recognizable in aerial images, but are almost overlooked when change detection is executed manually. With respect to the circumstances, these kinds of changes may be an indication of sabotage, terroristic activity, or threatening natural disasters. Although image-based change detection is possible from both ground and aerial perspectives, in this paper we primarily address the latter. We have applied an extended approach to change detection as described by Saur and Kruger,<sup>1</sup> and Saur et al.<sup>2</sup> and have built upon the ideas of Saur and Bartelsen.<sup>3</sup> The commercial simulation environment Virtual Battle Space 3 (VBS3) is used to simulate aerial "before" and "after" image acquisition concerning flight path, weather conditions and objects within the scene and to obtain synthetic videos. Video frames, which depict the same part of the scene, including "before" and "after" changes and not necessarily from the same perspective, are registered pixel-wise against each other by a photogrammetric concept, which is based on a homography. The pixel-wise registration is used to apply an automatic difference analysis, which, to a limited extent, is able to suppress typical errors caused by imprecise frame registration, sensor noise, vegetation and especially parallax effects. The primary concern of this paper is to seriously evaluate the possibilities and limitations of our current approach for image-based change detection with respect to the flight path, viewpoint change and parametrization. Hence, based on synthetic "before" and "after" videos of a simulated scene, we estimated the precision and recall of automatically detected changes. In addition and based on our approach, we illustrate the results showing the change detection in short, but real video sequences. Future work will improve the photogrammetric approach for frame registration, and extensive real video material, capable of change detection, will be acquired.
We present a fast point cloud clustering technique which is suitable for outlier detection, object segmentation and region
labeling for large multi-dimensional data sets. The basis is a minimal data structure similar to a kd-tree which enables us
to detect connected subsets very fast. The proposed algorithms utilizing this tree structure are parallelizable which
further increases the computation speed for very large data sets. The procedures given are a vital part of the data preprocessing.
They improve the input data properties for a more reliable computation of surface measures, polygonal
meshes and other visualization techniques. In order to show the effectiveness of our techniques we evaluate sets of point
clouds from different 3D scanning devices.
Modern manufacturing requires continuous quality control to assure a constant quality. For highly automated processes, industrial 2D and 3D image processing methods are often used. In the recent years the need for 3D geometry measuring has increased significantly. Therefore, we present a photogrammetric stereo vision system, including hardware as well as software components, which is designed to measure the 3D shape of complex extruded plastic profiles online. Thus, the main challenge is to capture the profiles while they are moving and deforming at the same time. The profiles are measured at several defined points in time and the geometry of the front surfaces are extracted respectively. This enables us to reconstruct and evaluate the deformation and shaping process. Based on these results we are able to simulate the viscous flow of plastics in the extrusion process. The 3D geometry capturing and the simulation are repeated iteratively to generate reliable simulation input close to reality on the one hand and to optimize the extrusion tool and the involved mechanical parts on the other. Our work is primarily concerned with the 3D data acquisition. Thus, we particularly focus on the stereo matching procedures used and discuss the optimal, system specific configuration.
One of the economically most important branches is the automotive industry with their component suppliers. The high degree of automation in manufacturing processes, requires automated control and quality assurance equally. In this scope, we present a complex 3D measuring device, consisting of multiple optical 3D sensors, which is designed to capture the geometry of wheel rims. The principal challenge for automated measurements is the variety of rims with respect to design, dimensions and the production flow. Together with connected conveyers, the system automatically sorts good rims without interrupting the manufacturing process. In this work we consider three major steps. At first we discuss the application of the used 3D sensors and the underlying measuring principles for the 3D geometry acquisition. Therefore, we examine the hardware architecture, which is needed to fulfill the requirements concerning to the variety of shapes and to the measuring conditions in industrial environments. In the second part we focus on the automated calibration procedure to integrate and combine the data from the set of sensors. Finally, we introduce the algorithms for the 3D geometry extraction and the mathematical methods which are used for the data preprocessing and interpretation.
The detection of varying 2D shapes is a recurrent task for Computer Vision applications, and camera based object recognition has become a standard procedure. Due to the discrete nature of digital images and aliasing effects, shape recognition can be complicated. There are many existing algorithms that discuss the identification of circles and ellipses, but they are very often limited in flexibility or speed or require high quality input data. Our work considers the application of shape recognition for processes in industrial environments and, especially the automatization requires reliable and fast algorithms at the same time. We take a very practical look at the automated shape recognition for common industrial tasks and present a very fast novel approach for the detection of deformed shapes which are in the broadest sense elliptic. Furthermore, we consider the automated recognition of bacteria colonies and coded markers for both 3D object tracking and an automated camera calibration procedure.
The digitalization of real-world objects is of great importance in various application domains. E.g. in industrial processes quality assurance is very important. Geometric properties of workpieces have to be measured. Traditionally, this is done with gauges which is somewhat subjective and time-consuming. We developed a robust optical laser scanner for the digitalization of arbitrary objects, primary, industrial workpieces. As measuring principle we use triangulation with structured lighting and a multi-axis locomotor system. Measurements on the generated data leads to incorrect results if the contained error is too high. Therefore, processes for geometric inspection under non-laboratory conditions are needed that are robust in permanent use and provide high accuracy as well as high operation speed. The many existing methods for polygonal mesh optimization produce very esthetic 3D models but often require user interaction and are limited in processing speed and/or accuracy. Furthermore, operations on optimized meshes consider the entire model and pay only little attention to individual measurements. However, many measurements contribute to parts or single scans with possibly strong differences between neighboring scans being lost during mesh construction. Also, most algorithms consider unsorted point clouds although the scanned data is structured through device properties and measuring principles. We use this underlying structure to achieve high
processing speeds and extract intrinsic system parameters and use them for fast pre-processing.