Anomaly detection aims to identify the data samples that do not conform to a known normal (regular) behavior. As the definition of an anomaly is often ambiguous, unsupervised and semi-supervised deep learning (DL) algorithms that primarily use unlabeled datasets to model normal (regular) behaviors, are popularly studied in this context. The unmanned aerial system (UAS) can use contextual anomaly detection algorithms to identify interesting objects of concern in applications like search and rescue, disaster management, public security etc. This thesis presents a novel multi-stage framework that supports detection of frames with unknown anomalies, localization of anomalies in the detected frames, and validation of detected frames for incremental semi-supervised learning, with the help of a human operator. The proposed architecture is tested on two new datasets collected for a UAV-based system. In order to detect and localize anomalies, it is important to both model the normal data distribution accurately as well as formulate powerful discriminant (anomaly scoring) techniques. We implement a generative adversarial network (GAN)-based anomaly detection architecture to study the effect of loss terms and regularization on the modeling of normal (regular) data and arrive at the most effective anomaly scoring method for the given application. Following this, we use incremental semi-supervised learning techniques that utilize a small set of labeled data (obtained through validation from a human operator), with large unlabeled datasets to improve the knowledge-base of the anomaly detection system. / Master of Science / Anomaly detection aims to identify the data samples that do not conform to a known normal (regular) behavior. As the definition of an anomaly is often ambiguous, most techniques use unlabeled datasets, to model normal (regular) behaviors. The availability of large unlabeled datasets combined with novel applications in various domains, has led to an increasing interest in the study of anomaly detection. In particular, the unmanned aerial system (UAS) can use contextual anomaly detection algorithms to identify interesting objects of concern in applications like search and rescue (SAR), disaster management, public security etc. This thesis presents a novel multi-stage framework that supports detection and localization of unknown anomalies, as well as the validation of detected anomalies, for incremental learning, with the help of a human operator. The proposed architecture is tested on two new datasets collected for a UAV-based system. In order to detect and localize anomalies, it is important to both model the normal data distribution accurately and formulate powerful discriminant (anomaly scoring) techniques. To this end, we study the state-of-the-art generative adversarial networks (GAN)-based anomaly detection algorithms for modeling of normal (regular) behavior and formulate effective anomaly detection scores. We also propose techniques to incrementally learn the new normal data as well as anomalies, using the validation provided by a human operator. This framework is introduced with the aim to support temporally critical applications that involve human search and rescue, particularly in disaster management.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

Säkerhetsläget i Sverige och Europa är osäkert och ett allmänt behov av övervakning har ökat. Övervakningskameror kan både förhindra brott och lösa redan begångna brott. Axis Communication är världsledande inom nätverkskameror. Syftet med arbetet är att integrera högtalare och mikrofon i övervakningskameror. Målet är att till en ny övervakningskamera kunna lyfta in ett färdigt koncept, en audiomodul, med högtalare och mikrofon som passar givna krav på ljudkvalitet och ljudtryck. Val av koncept, högtalare och mikrofon, har genomomförts genom en noggrann produktutvecklingsprocess med delar som intern och extern sökning, utvärdering av lösningsvarianter, detaljutveckling och produkttestning. Ett koncept med integrerad högtalare och mikrofon har tagits fram. En högtalare har valts och placerats i väggen i kamerans innerchassi. En Smart Amp används för att högtalaren ska kunna spela så högt som möjligt utan att ljudkvalitén försämras. Mikrofonen placeras på kamerans ytterchassi så långt ifrån högtalare som möjligt och med en ljudisolerande kåpa för att undvika eko från högtalaren till mikrofonen. Arbetet resulterar i ett koncept med högtalare och mikrofon som uppnår på förhand uppställda krav på ljudtryck och ljudkvalité. Konceptet bidrar till en tryggare miljö genom att en snabbare detektion uppnås och bättre ljud leder till att meddelandeutskick blir lättare att höra. / The security situation in Sweden and Europe is uncertain and a general need for surveillance has increased. Surveillance cameras can both prevent crimes and resolve already committed crimes. Axis Communication is the world leader in network cameras. The purpose of the work is to integrate loudspeakers and microphones into surveillance cameras. The goal is for a new surveillance camera to lift a ready-made concept, an audio module, with speakers and microphones that meet the requirements for sound quality and sound pressure. Selection of concepts, speaker and microphone has been implemented through a careful product development process with parts such as internal and external search, evaluation of solution variants, detail development and product testing. A concept with integrated speaker and microphone has been developed. A speaker has been selected and placed in the camera's inner chassis. A Smart Amp is used to allow the speaker to play as loud as possible without impairing the sound quality. The microphone is placed on the camera's outer chassis as far away from the speaker as possible and with an acoustic cover to avoid echo from the loudspeaker to the microphone. The work results in a loudspeaker and microphone concept that meets prerequisites for sound pressure and sound quality. The concept contributes to a safer environment by achieving a faster detection and better sound, making public address messaging easier to hear.