Recently, research in the field of content-based image retrieval has attracted a lot of attention. Nevertheless, most existing methods cannot be easily applied to medical image databases, as global image descriptions based on color, texture, or shape do not supply sufficient semantics for medical applications. The concept for content-based image retrieval in medical applications (IRMA) is therefore based on the separation of the following processing steps: categorization of the entire image; registration with respect to prototypes; extraction and query-dependent selection of local features; hierarchical blob representation including object identification; and finally, image retrieval. Within the first step of processing, images are classified according to image modality, body orientation, anatomic region, and biological system. The statistical classifier for the anatomic region is based on Gaussian kernel densities within a probabilistic framework for multiobject recognition. Special emphasis is placed on invariance, employing a probabilistic model of variability based on tangent distance and an image distortion model. The performance of the classifier is evaluated using a set of 1617 radiographs from daily routine, where the error rate of 8.0% in this six-class problem is an excellent result, taking into account the difficulty of the task. The computed posterior probabilities are furthermore used in the subsequent steps of the retrieval process.

D object recognition has been tackled by passive approaches in the past. This means that based on one image a decision for a certain class and pose must be made or the image must be rejected. This neglects the fact that some other views might exist, which allow for a more reliable classification. This situation especially arises if certain views of or between objects are ambiguous.

We present a modular architecture for image understanding and active computer vision which consists of the following major components: Sensor and actor interfaces required for datatriven active vision are encapsulated to hide machine-dependent parts; image segmentation is implemented in object-oriented programming as a hierarchy of image operator classes, guaranteeing simple and uniform interfaces. We apply this architecture to appearance-based object recognition. This is used for an autonomous mobile service robot which has to locate objects using visual sensors.

. This special research area combines visualization and interpretation of sensor data by exploring the central subjects #models", #optimization ", #hierarchies", and #data fusion". This article describes the structural and thematic organization of the research project and illustrates #rst results of one sub#project as an example. 1 Introduction The Sonderforschungsbereich #SFB, special research area# 603 #model based analysis and visualization of complex scenes and sensor data" has been established by the German Research Foundation in the beginning of 1998, and for the #rst research period it is granted until the end of 2000. At the moment, it is divided into twelve sub#projects in which seven institutes of the Technical Department of the University of Erlangen#Nuremberg and two clinics are involved. This article describes the thematic questions and goals for which section 2 gives an overview. The whole SFB is divided into sub#projects, which are explained in section 3. In secti...

This paper deals with an aspect of active object recognition for improving the classification and localization results by choosing optimal next views at an object. The knowledge of "good" next views at an object is learned automatically and unsupervised from the results of the used classifier. For that purpose methods of reinforcement learning are used in combination with numerical optimization. The major advantages of the presented approach are its classifier independence and that the approach does not require a priori assumptions about the objects. The presented results for synthetically generated images show that our approach is well suited for choosing optimal views at objects.