The Swedish zoo “Lycksele Djurpark” is the northernmost zoo in the country. It is specialized in housing nordic animals and have animals like the bear, elk, musk-ox and wild boar to mention some. The zoo is home to more than 400 animals, divided into 30 species and covers an area of 47 hectares. It has been one of the biggest tourist attractions in northern Sweden. Due to the weather conditions and the sparsely populated location of the zoo, it is difficult to attract visitors to the zoo during winter. The opening season is for that reason from the end of May to the end of September with some additional openings around holidays like Christmas and Easter. Although the zoo is closed during the winter the animals’ normal life continues as usual and they are outdoors. Consequently, it is challenging for the zoo to make a profit and to develop further.
The goal of the project “Digital Djurpark” (Digital Zoo) funded by the EU regional development fund is to make the zoo digital and more attractive. The approach is to deploy a wireless sensor network covering the whole zoo. This network will consist of around 100 wireless sensors, which are able to configure themselves into a working network. The sensors will be used to collect data such as video, pictures, sound and temperature. The sensed data will be processed and semantic information will be used to support visitor interaction design, which is a key aspect of the new experience envisaged for the visitors. Figure 1 shows an overview of the working packages included in the project.
Work packages in the Digital Djurpark project. A wireless visual sensor network is used to capture information about the animals. This information is processed and interaction techniques are used to present the information for visitors.
In this chapter we will describe our work on creating the digital zoo, based on the proposed large scale testbed using wireless visual sensor networks. This will give us a unique opportunity to test long term wireless senor network deployments in an environment where there are a lot of events to detect. In the following sections we will describe the hardware platforms, sensors, computer vision system used in the network, and interaction design.
To collect information such as audio, video and position from individual animals a heterogeneous set of wireless sensors are used. At some locations we have access to the mains power supply and can use more powerful cameras that are able to perform advanced computer vision functions and deliver high-quality video streaming. At other locations we need to use low-power cameras that can operate using only harvested solar energy. These sensors will not be at continuous full operation. Instead they will be triggered on specific events and will process the collected media information only when some interesting events are detected.