Christian Menard

Stereo and Structured Light as Acquisition Methods in the Field of Archaeology

In this paper two acquisition methods for archaeological finds are proposed that could help the archaeologist in his work. First we present these very different acquisition methods, stereo and structured light acquisition to get the 3D-surface representation (a so-called 3D-object model) of a sherd. Further we discuss the accuracy of the acquisition methods for archaeological applications. The results are compared with each other and an outlook for a possible fusion of these two methods for an archaeological application is given.

Evaluation of the Geometric Accuracy of Automatically Recorded 3D – City Models Compared to GIS-Data

In this paper we propose methods for evaluating the geometric accuracy of three-dimensional city models. The approach based on the concept of an error matrix, statistical analysis of height differences and a buffer-overlay-statistic leads to accuracy parameters for an automated city modelling workflow from aerial images footnote The modelling workflow comes from Microsoft Photogrammetrywhich is a research and development unit of Microsoft. It emerged from the merger of Microsoft and Vexcel Imaging GmbH.. We study the theoretical properties of our approach and we show that in practice the concept of the paper behaves very well on real test data from the 3D-model of the inner city of Graz. The work concludes with a summary and an outlook to future work.

Adaptive Stereo Matching in Correlation Scale-Space

Stereo computes the distance of objects, “their depth”, from two images of two cameras using the triangulation principle. Points of imaged objects are mapped in different locations in the two stereo images. A central problem in stereo matching using correlation techniques lies in selecting the size of the search window. Small windows contain only a small number of data points, and thus are very sensitive to noise and therefore result in false matches. Whereas large search windows contain data from two or more different objects or surfaces, thus the estimated disparity is not accurate due to different projective distortions in the left and the right image. The new method introduces a continuous scale parameter for the matching process. It allows the adaption of the scale for every individual region and overcomes the drawbacks of fixed window sizes which is impressively demonstrated by the experimental results.

Anonymity and Application Privacy in Context of Mobile Computing in eHealth

In the area of health care and sports in recent years a variety of mobile applications have been established. Mobile devices are of emerging interest due to their high availability and increasing computing power in many different health scenarios. In this paper we present a scalable secure sensor monitoring platform (SSMP) which collects vital data of users. Vital parameters can be collected by just one single sensor or in a multi-sensor configuration. Nowadays a wide spectrum of sensors is available which provide wireless connectivity (e.g. Bluetooth). Vital data can then easily be transmitted to a mobile device which subsequently transmits these data to an eHealth portal. There are already solutions implementing these capabilities, however privacy aspects of users are very often neglected. Since health data may enable people to draw potentially compromising conclusions (e.g. insurance companies), it is absolutely necessary to design an enhanced security concept in this context. To complicate matters further, the trustworthiness of providers which are operating with users health data can not be determined by users a priori. This means that the security concept implemented by the provider may bear security flaws. Additionally there is no guarantee that the provider preserves the users privacy claims. In this work we propose a security concept incorporating privacy aspects using mobile devices for transferring and storing health data at a portal. In addition, the concept guarantees anonymity in the transfer process as well as for stored data at a service provider. Hence, insider attacks based on stored data can be prevented.

Robust stereo on multiple resolutions

Stereo computation is one of the vision problems where the presence of outliers cannot be neglected. Most standard algorithms make unrealistic assumptions about noise distributions, which leads to erroneous results that cannot be corrected in subsequent postprocessing stages. In this paper we present a modification of the standard area-based correlation approach so that it can tolerate a significant number of outliers. The approach exhibits a robust behavior not only in the presence of mismatches but also in the case of depth discontinuities. The confidence measure of the correlation and the number of outliers provide two complementary sources of information which, when implemented in a multiresolution framework, result in a robust and efficient method. We present the results of this approach on a number of synthetic and real images.

On estimating the position of fragments on rotational symmetric pottery

The activation of ancient vessel-fragments is a time consuming but important task for archaeologists. The basis for classification and reconstruction is the profile which is the cross-section of the fragment in the direction of the rotational axis of symmetry. Hence the position of a fragment (orientation) on a vessel is important. In this work the estimation of the axis of rotation out of range data by using a Hough inspired method is proposed. In order to avoid outliers a robust method for estimation of the axis is used. Classification and reconstruction are performed in a bottom-up manner using a description language, which holds all features of the fragment as primitives and all properties among features as relations. Classification of newly found fragments of unknown type is performed by comparing the description of the new fragment with the description of already classified fragments by completing graph similarity. The sub-graph with the highest similarity is then used to reconstruct the complete vessel out of the fragment.

Representing the object model for automatic visual inspection using a description language

The major drawbacks of automated visual inspection systems are the high set-up costs resulting from hard- and software development costs, labor, and maintenance costs. The key to solving the problem of flexibility is the development of visual inspection systems which are able to inspect a large variety of different objects without or only partly changing the analysis algorithm. One aspect in this design is the representation of the object to be inspection. A priori knowledge is used implicitly or explicitly by all visual inspection systems, since inspection can only be performed by matching the object under inspection with a set of predefined conditions of acceptability. These specifications are described by an explicit object model, that includes all relevant object features. The second representation of the object is the image containing the object. Within the image, object features are represented as image features, that have to be detected by feature detection algorithms. This paper shows an inspection model that allows a flexible object- specific description, defined in a so-called description language. This model has primitives as nodes and relations between the primitives as arcs. Furthermore tolerances and weights indicating the importance of detection are also part of the model. On the case study of analogue display instruments the representation and generation of the inspection model is shown.

Stereo Correspondences in Scale Space

A central problem in stereo matching using correlation techniques lies in selecting the size of the search window. Small windows contain only a small number of data points, and thus are very sensitive to noise and therefore result in false matches. Whereas large search windows contain data from two or more different objects or surfaces, thus the estimated disparity is not accurate due to different projective distortions in the left and the right image.

Adaptive area-based stereo matching

The stereo analysis method is similar to the human visual system. Due to the way our eyes are positioned and controlled, our brains usually receive similar images of a scene taken from nearby points of the same horizontal level. Stereo tries to imitate this principle by computing the distance of objects, their depth, from two images of two cameras using the triangulation principle. Points of imaged objects are mapped in different locations in the two stereo images. These points need to be identified in both images. The correspondences are established by correlating windows of the left and right image and finding a maximum. A central problem in stereo matching using correlation techniques lies in selecting the size of the search window. Small windows contain only a small number of data points, and thus are very sensitive to noise and therefore result in false matches. Whereas large search windows contain data from two or more different objects or surfaces, thus the estimated disparity is not accurate due to different projective distortions in the left and the right image. The new method introduces a continuous scale parameter for the matching process. It allows the adaption of the scale for every individual region and overcomes the drawbacks of fixed window sizes which is impressively demonstrated by the experimental results.

Stereo Matching Using M-Estimators

Stereo computation is just one of the vision problems where the presence of outliers cannot be neglected. Most standard algorithms make unrealistic assumptions about noise distributions, which leads to erroneous results that cannot be corrected in subsequent processing stages. In this work the standard area-based correlation approach is modified so that it can tolerate a significant number of outliers. The approach exhibits a robust behaviour not only in the presence of mismatches but also in the case of depth discontinuities. Experimental results are given on synthetic and real images.