Gerhard Paar

Window Detection in Facades

This work is about a novel methodology for window detection in urban environments and its multiple use in vision system applications. The presented method for window detection includes appropriate early image processing, provides a multi-scale Haar wavelet representation for the determination of image tiles which is then fed into a cascaded classifier for the task of window detection. The classifier is learned from a Gentle Adaboost driven cascaded decision tree on masked information from training imagery and is tested towards window based ground truth information which is together with the original building image databases publicly available. The experimental results demonstrate that single window detection is to a sufficient degree successful, e.g., for the purpose of building recognition, and, furthermore, that the classifier is in general capable to provide a region of interest operator for the interpretation of urban environments. The extraction of this categorical information is beneficial to index into search spaces for urban object recognition as well as aiming towards providing a semantic focus for accurate post-processing in 3D information processing systems. Targeted applications are (i) mobile services on uncalibrated imagery, e.g. , for tourist guidance, (ii) sparse 3D city modeling, and (iii) deformation analysis from high resolution imagery.

Robust disparity estimation in terrain modeling for spacecraft navigation

Navigation and imagery in the orbit, descent, and landing phases during an interplanetary mission require methods that are able to derive the elevation map of a planetary body using remote sensing tools. The authors propose stereovision techniques for this task. An algorithm for correspondence matching, which is one of the crucial steps in automatic terrain modeling, is introduced. It uses well known pyramid-based data structures, but is novel in its direct application of methods from statistical pattern recognition. Feature vectors for correspondence matching and feature selection techniques are used to find optimal features. These include grey-level statistics (mean variance) as well as more sophisticated features derived from operators like local frequency edge gradient or, as an extension, Moravec-, Gabor- or Fourier-features. The applicability of the algorithm in the remote sensing scenario of interplanetary missions is verified using a mockup simulation of the Martian surface.<<ETX>>

Mass Movement Monitoring Using Terrestrial Laser Scanner for Rock Fall Management

The danger of a rock fall or rockslide event is omnipresent, mainly due to dense settlement, excessive land usage even in alpine regions, and the global warming. In the case of a rock fall event the rapid operational availability of a measurement system is important for disaster management to assess the risk and to take appropriate measures.

Robust Window Detection from 3D Laser Scanner Data

In this paper we propose a robust system for window detection using popular descriptive statistics and image based methods, making use of 3D information from a laser scanner. The scanner generates 3D point clouds containing intensity and distance information in a spherical coordinate system, with optional additional RGB texture information. The applied descriptive statistical method exploits basic local features such as mean, variance and standard deviation of the distance measurement data. The laser distance information shows high variability in windows region, due to specular reflections on window screens on one hand, and screen penetration on the other hand. Therefore we determine an adaptive threshold on the basis of local absolute differences of adjacent laser-measured distances in the image formed by the angular coordinate system of the scanner. For window segmentation the image is binarized using the derived threshold, and morphological operations such as closing using adaptive (i.e. distance - dependent) structural elements are performed. After contour analysis the resulting bounding rectangles are used to retrieve the positions and global shapes of windows in the image. The system provides a sufficient windows detection rate for direct application in a deformation measurement system.

A 3D optical deformation measurement system supported by knowledge-based and learning techniques

Abstract High accuracy 3D representation and monitoring of objects is receiving increasing interest both in science and industrial applications. Up to now tasks like monitoring of building displacements or deformations were solved by means of artificial targets on the objects of interest, although mature optical 3D measurement and laser scanning techniques are available. Such systems can perform their measurements even without targeting. This paper presents a new optical 3D measurement system, based on the fusion between a geodetic image sensor and a laser scanner. The main goal of its development was the automation of the whole measurement process, including the tasks of point identification and measurement, deformation analysis, and interpretation. This was only possible by means of new methods and techniques originally developed in the area of Artificial Intelligence; both point detection and deformation analysis are supported by decision systems that use such techniques. The resulting complex multi-sensor system is able to measure and analyse the deformation of objects, as shown in experiments. In this article we focus on specific key components and novel techniques that have been developed, and briefly report on the current stage of the whole system.

Algorithmic solution for autonomous vision-based off-road navigation

A vision based navigation system is a basic tool to provide autonomous operations of unmanned vehicles. For offroad navigation that means that the vehicle equipped with a stereo vision system and perhaps a laser ranging device shall be able to maintain a high level of autonomy under various illumination conditions and with little a priori information about the underlying scene. The task becomes particularly important for unmanned planetary exploration with the help of autonomous rovers. For example in the LEDA Moon exploration project currently under focus by the European Space Agency (ESA), during the autonomous mode the vehicle (rover) should perform the following operations: on-board absolute localization, elevation model (DEM) generation, obstacle detection and relative localization, global path planning and execution. Focus of this article is a computational solution for fully autonomous path planning and path execution. An operational DEM generation method based on stereoscopy is introduced. Self-localization on the DEM and robust natural feature tracking are used as basic navigation steps, supported by inertial sensor systems. The following operations are performed on the basis of stereo image sequences: 3D scene reconstruction, risk map generation, local path planning, camera position update during the motion on the basis of landmarks tracking, obstacle avoidance. Experimental verification is done with the help of a laboratory terrain mockup and a high precision camera mounting device. It is shown that standalone tracking using automatically identified landmarks is robust enough to give navigation data for further stereoscopic reconstruction of the surrounding terrain. Iterative tracking and reconstruction leads to a complete description of the vehicle path and its surrounding with an accuracy high enough to meet the specifications for autonomous outdoor navigation.

Optical crack following on tunnel surfaces

One of the most important monitoring tasks of tunnel inspection is the observation of cracks. This paper describes an approach for crack following using mid-resolution (2-5mm per pixel) images of the tunnel surface. A mosaic on the basis of the tunnel design surface is built from images taken with a mobile platform. On this image representing the unwrapped tunnel surface texture the starting points of each crack are found semiautomatically using a modified Hough transform. Crack following takes place on the basis of local line fitting and exhaustive search in both directions of the crack, taking into account several restrictions, rules and optimization criteria to find the correct crack trajectory. A practical implementation polygonizes the extracted cracks and feeds them into a tunnel inspection data base. The method is applicable to various types of background texture as expected in the tunnel environment.

The PanCam instrument for the ExoMars Rover

Abstract The scientific objectives of the ExoMars rover are designed to answer several key questions in the search for life on Mars. In particular, the unique subsurface drill will address some of these, such as the possible existence and stability of subsurface organics. PanCam will establish the surface geological and morphological context for the mission, working in collaboration with other context instruments. Here, we describe the PanCam scientific objectives in geology, atmospheric science, and 3-D vision. We discuss the design of PanCam, which includes a stereo pair of Wide Angle Cameras (WACs), each of which has an 11-position filter wheel and a High Resolution Camera (HRC) for high-resolution investigations of rock texture at a distance. The cameras and electronics are housed in an optical bench that provides the mechanical interface to the rover mast and a planetary protection barrier. The electronic interface is via the PanCam Interface Unit (PIU), and power conditioning is via a DC-DC converter. PanCam also includes a calibration target mounted on the rover deck for radiometric calibration, fiducial markers for geometric calibration, and a rover inspection mirror. Key Words: Mars—ExoMars—Instrumentation—Geology—Atmosphere—Exobiology—Context. Astrobiology 17, 511–541.

Photogrammetric fingerprint unwrapping

Abstract Fingerprints are important biometric cues. Compared to conventional fingerprint sensors the use of contact-free stereoscopic image acquisition of the front-most finger segment has a set of advantages: Finger deformation is avoided, the entire relevant area for biometric use is covered, some technical aspects like sensor maintenance and cleaning are facilitated, and access to a three-dimensional reconstruction of the covered area is possible. We describe a photogrammetric workflow for nail-to-nail fingerprint reconstruction: A calibrated sensor setup with typically 5 cameras and dedicated illumination acquires adjacent stereo pairs. Using the silhouettes of the segmented finger a raw cylindrical model is generated. After preprocessing (shading correction, dust removal, lens distortion correction), each individual camera texture is projected onto the model. Image-to-image matching on these pseudo ortho images and dense 3D reconstruction obtains a textured cylindrical digital surface model with radial distances around the major axis and a grid size in the range of 25–50 µm. The model allows for objective fingerprint unwrapping and novel fingerprint matching algorithms since 3D relations between fingerprint features are available as additional cues. Moreover, covering the entire region with relevant fingerprint texture is particularly important for establishing a comprehensive forensic database. The workflow has been implemented in portable C and is ready for industrial exploitation. Further improvement issues are code optimization, unwrapping method, illumination strategy to avoid highlights and to improve the initial segmentation, and the comparison of the unwrapping result to conventional fingerprint acquisition technology.

Stereo reconstruction from dense disparity maps using the locus method

The reconstruction of a surface having already matched corresponding points from stereo images (disparities) is a nontrivial task. This paper presents a new technique, the so-called Locus method, that exploits sensor geometry to efficiently build a terrain representation from stereo disparities. The power of this approach is the efficient and direct computation of a dense elevation map in arbitrary resolution. Additionally it proposes to solve problems like occlusions, ambiguities, and uncertainties caused by stereo matching errors. We extended the Locus method for active range finder data to the stereo disparity mapping case. For this reason, a newly developed fast matching method is utilized that provides dense disparity maps, hence a disparity for each input pixel. Once this data set is given, the Locus method can be applied in a straightforward and efficient way to gain a robust 3D reconstruction of the observed surface. It operates directly in image space, using dense and uniform measurements instead of first converting to object space. Experiments on synthetic and natural environment data show that the Locus method is less sensitive to disparity noise than traditional reconstruction.