Abstract

Video and audio monitoring systems are used across the globe to observe certain events and processes. Usually, such systems are connected to a communications network and electricity grid and are not equipped with any kind of processing logic ("intelligence") other than data compression. How can such systems be used if there is no electricity nor a broadband communications network available? The aim of this thesis is to develop an autonomous monitoring system built entirely from commercially-available standard components, which can operate independently. The first step to achieve this is to conduct a pre-evaluation of existing hardware and software components according to various criteria. The next step involves the development of a hardware prototype as well as the implementation of an efficient and modular software framework that can be adapted for different applications. The system should support policy-based recording modes such as modes controlled by input data from defined sensors, like motion and light sensors, that make autonomous decisions about the use of available resources. The software framework should be designed to support the integration of image processing algorithms to allow pre-processing of recorded (video) data on board in order to save bandwidth. In the final prototype, both the recording of audio and video (among other) data was realized to the desired extent. Using an application, the recording mode can be set easily using a XML configuration file. The power consumption of the build system is, depending on used sensors and executed tasks, between 0.68 to 1.75 watts and lies within the range of comparable commercial solutions, such as "leanXcam".

Reference

Buturovic, A. (2016). Entwicklung eines autonomen Überwachungssystems [Diploma Thesis, Technische Universität Wien]. reposiTUm. https://doi.org/10.34726/hss.2016.39671