Digital Djurpark
Digital Djurpark funded by EU mål2 is a research project aiming at developing new experiences around the traditional visit to Lycksele djurpark, and making the zoo digital and more attractive. The approach is to build a wireless sensor network covering the whole zoo. The network will consist of a set of wireless sensors, which are able to configure themselves into a working network. The sensors could be collecting data such as video, pictures, sound and temperature. The sensed data will be processed and semantic information will be used to support interaction design, which is a key component to design a new experience for the visitor.
To capture information from the animals we have also developed a camera platform based on the Gumstix OveroTM Fire computer-on-module and a custom image sensor board The Gumstix OveroTM Fire consists of an Texas Instrument Omap 3530 application processor with ARM Cortex-A8 CPU and a C64x digital signal processor (DSP) core both running at 600MHz. The board has 256MB of RAM and 256MB of flash memory. For communication IEEE 802.11g wireless LAN is used in a wireless mesh network. The board measures only 17mm x 58mm x 4.2mm which enable us to build a really small and powerful camera platform.
The camera board contains an OmniVision OV9710 image sensor supporting 1280x800 at 30 fps. The image sensor also has a good low light sensitivity of 3.3V per lux-second. For this platform we use a multi-hop wireless ad-hoc network on 802.11 wireless links. Since this platform has high computational capacity and also high communication capacity it will enable the possibility to send high-quality real-time video. This platform has however rather high power consumption, typically a few watts, and will in general be connected to the mains power supply.
The current deployment in the zoo consists of 25 video transmitting nodes. The nodes used in this network are based on our high-performance wireless LAN platform. They are distributed in a way to generate a large network covering the entire zoo, At a rather central location in the zoo, close to the bear den, we have a fiber connection to the Internet and this is the location of the sink node in the wireless network. The nodes configure themselves into a working ad-hoc network and send all collected multimedia information back to the sink node. The network is rather large and to reach some nodes four hops are needed from the sink node. This deployment has been in place and running for over one year in the zoo.
The Computer vision added to the system, concerned with the theory behind artificial systems that extract information from videos and images, make the camera as an artificial intelligence machine, not only can “see”, but also can “understand”. In the Digital Djurpark project, the computer vision system is involved to analysis the videos captured from the ad-hoc camera system, Examples of applications of computer vision include systems for:
• Detect the motion of the animals.
• Separate the foreground and background. Get the location information of the animals.
• Track the moving animals
• Even more, understand the behavior of the animals, and control the animals.
The computer vision system in the digital djurpark projectis divided into several parts. First the background andforeground estimation is used to extract the moving objects,that is the animals in the zoo. Then the motion detection,scoring, and tracking is performed based on the results ofthe estimation.
We saw that with the wireless sensor network in the zoo we would be able to give the visitors of the zoo a new experience. Our idea was to give people an experience that would give them a chance get to know the animal on a more personal level.
With the applications we made, based on the wireless sensor network information, the concept “zoo visitor” become a wider meaning. The visitor would not only be someone in the zoo at Lycksele but also become a user of our applications, applications that can be located anywhere where we have Internet connection. We wanted this visitor to be an active user of the information. Interactions of the collected information become something important for us. The visitors/users became “interactors”.
The information collected with the wireless sensor network is in form of “moving pictures”. These moving pictures were both film clips and live stream videos from the animals in the zoo.
So with the to following in mind, letting the visitor get to know the animals and to make them interactors, we made some applications based on the information from the wireless sensor network.
