dc.rights.license | CC-BY-NC-ND | |
dc.contributor.advisor | Feelders, Ad | |
dc.contributor.advisor | Siebes, Arno | |
dc.contributor.advisor | Smit, Selmar | |
dc.contributor.author | Renckens, I.R. | |
dc.date.accessioned | 2014-09-16T17:01:09Z | |
dc.date.available | 2014-09-16T17:01:09Z | |
dc.date.issued | 2014 | |
dc.identifier.uri | https://studenttheses.uu.nl/handle/20.500.12932/18346 | |
dc.description.abstract | This thesis presents an implementation of an anomaly detection method that can be used for surveillance applications. The method is based on the perspective that a flying vehicle should be able to derive the objects in an environment that show anomalous behaviour. The objective is to make an method that is able to examine a very large urban environment in real time. This is done with a clustering algorithm that takes into account the paths, speed and lifetime of objects. The clusters are learned with only coordinates of the object in an environment. These clusters form the basis of the distance calculation in the detection phase. This method is based on a method proposed by Piciarelli et al. \cite{Piciarelli06} where a number of important improvements have been made. With our method we solved some essential problems of the method of Piciarelli, such as the time dependency of the model and the amount of data needed to create a good model. We also added environmental features to be taken into account. These improvements resulted in a faster and better working method for this particular purpose. | |
dc.description.sponsorship | Utrecht University | |
dc.format.extent | 2563991 | |
dc.format.mimetype | application/pdf | |
dc.language.iso | en | |
dc.title | Automatic Detection of Suspicious Behaviour | |
dc.type.content | Master Thesis | |
dc.rights.accessrights | Open Access | |
dc.subject.keywords | anomaly detection;suspicious behaviour;UaV;TNO;Clustering;Security | |
dc.subject.courseuu | Technical Artificial Intelligence | |