In this master thesis, the problem of tracking groups using an image sequence dataset is examined. Target tracking can be defined as the problem of estimating a target's state given prior knowledge about its motion and some sensor measurements related to the target's state. A popular method for target tracking is e.g. the Kalman filter. However, the Kalman filter is insufficient when there are multiple targets in the scene. Consequently, alternative multitarget tracking methods must be applied along with methods for estimating the number of targets in the scene. Multitarget tracking can however be difficult when there are many unresolved targets, e.g. associating observations with targets in dense crowds. A viable simplification is group target tracking, keeping track of groups rather than individual targets. Furthermore, group target tracking is preferred when the user wants to know the motion and extension of a group in e.g. evacuation scenarios. To solve the problem of group target tracking in video surveillance, a combination of GM-PHD filtering and mean shift clustering is proposed. The GM-PHD filter is an approximation of Bayes multitarget filter. Pedestrian detections converted into flat world coordinates from the image dataset are used as input to the filter. The output of the GM-PHD filter consists of Gaussian mixture components with corresponding mean state vectors. The components are divided into groups by using mean shift clustering. An estimate of the number of members and group shape is presented for each group. The method is evaluated using both single camera measurements and two cameras partly surveilling the same area. The results are promising and present a nice visual representation of the groups' characteristics. However, using two cameras gives no improvement in performance, probably due to differences in detections between the two cameras, e.g. a single pedestrian can be observed being at two positions several meters apart making it difficult to determine if it is a single pedestrian or multiple pedestrians.

We propose a framework for detecting and tracking multiple interacting objects from a single, static, uncalibrated camera. The number of objects is variable and unknown, and object-class-specific models are not available. We use background subtraction results as measurements for object detection and tracking. Given these constraints, the main challenge is to associate pixel measurements with (possibly interacting) object targets. We first track clusters of pixels, and note when they merge or split. We then build an inference graph, representing relations between the tracked clusters. Using this graph and a generic object model based on spatial connectedness and coherent motion, we label the tracked clusters as whole objects, fragments of objects or groups of interacting objects. The outputs of our algorithm are entire tracks of objects, which may include corresponding tracks from groups of objects during interactions. Experimental results on multiple video sequences are shown.

An algorithm for tracking single visual targets is developed in this study. Feature detection is the necessary and appropriate image processing technique for this algorithm. The main point of this approach is to use the data supplied by the feature detection as the observation from a group of targets having similar motion dynamics. Therefore a single visual target is regarded as a group of multiple targets. Accurate data association and state estimation under clutter are desired for this application similar to other multi-target tracking applications. The group tracking approach is used with the well-known probabilistic data association technique to cope with data association and estimation problems. The applicability of this method particularly for visual tracking and for other cases is also discussed.

This thesis is concerned with development of a navigation algorithm primarily for the aircraft fighter SAAB JAS 39 Gripen, in swarms of other units. The algorithm uses information from conventional navigation systems and additional information from a radio data link as aiding information, relative range measurements. As the GPS can get jammed, this group tracking solution can provide an increased navigation performance in these conditions. For simplicity, simplified characteristics are used in the simulations where simple generated trajectories and measurements are used. This measurement information can then be fused by using filter theory applied from the sensor fusionarea with statistical approaches. By using the radio data link and the external information sources, i.e. other aircraft and different types of landmarks with often good performance, navigation is aided when the GPS is not usable, at e.g. hostile GPS conditions. A number of scenarios with operative sense of reality were simulated for verifying and studying these conditions, to give results with conclusions. / Det här examensarbetet syftar till utveckling av en algoritm för navigering, primärt för stridsflygplanet SAAB JAS 39 Gripen, i svärmar av andra enheter. Algoritmen använder information från konventionella navigeringssystem och ytterligare information från en radiodatalänk som ger understödjande information, relativa avståndsmätningar. Då den förlitade GPS:en kan störas ut, kan denna gruppspårande lösning öka navigeringsprestandan i dessa förhållanden. För enkelhetens skull, används förenklade karaktäristiker i simuleringarna där enkla genererade trajektorier och mätningar används. Denna mätinformation kan sedan ihopviktas genom att använda filterteori från statistisk sensorfusion. Genom att använda radiodatalänkar och den tillförda informationen från externa informationskällor, således andra flygplan och olika typer av landmärken som väldigt ofta har god prestanda, är navigeringen understödd när GPS inte är användbar, t.ex. i GPS-fientliga miljöer. Ett antal scenarion med operativ verklighetsanknytning simulerades för att verifiera och studera dessa förhållanden, för att ge resultat med slutsatser. / © Daniel Magnusson.