iCore Smart Theme Park

This trial illustrates how the iCore functionalities can be utilized in entertainment attractions like theme parks. The motivation of this trial is to provide a personalized digital souvenir for theme park customers. This trial, on one hand, could capture the valuable experience of customers in the format of digital multimedia while they are enjoying the events. On the other hand, it creates a new product and theme parks could make profits from it. The main idea is to deploy various sensors including digital cameras, microphones, RFID sensors, infrared sensors around the thrilling rides to record expressions and screams of riders, starting and stopping time of rides and identification of riders. All these multimodal information are analyzed and fused to generate personalized video clips for each customer.
  • Tags:
  • Latest update: November 29, 2014
  • Developed by: Wuxi Smart Sensing Stars, Co. Ltd., China
  • Contact email: [email protected]
  • Source code repository: https://github.com/iot-icore/icore-smart-theme-park


Key components of the architecture are described below.

Digital Souvenir Generation (Service Level)

This component provides a simple GUI for the operator to offer service to customers and results in the dynamic creation/instantiation of the appropriate Composite Virtual Object (CVO) via interactions with other iCore components. It has been implemented using a number of technologies including Java Servlets, AJAX, PostgreSQL, as well as a RESTful interface to facilitate the connection between the Servlets and the iCore components.

CVO Level

  • CVO management unit
    At a service request, the CVO Management Unit tries to find the needed running CVO from CVO registry. If failed, it turns to the CVO factory to activate and instantiate a new one.
  • CVO registry
    Maintain the id and semantic description of the running CVO.
  • CVO factory
    The CVO factory is able to instantiate a CVO based on a CVO template. It answers to CVO Management Unit requests, which handles a description of the CVO functionalities.
  • Highlights retrieval CVO
    This CVO searches personal video and audio data from the Postgres database using customer information like name, phone number, RFID EPC as key words.
  • Highlights composition CVO
    This CVO edits and combines the personal video and audio data into a single digital souvenir in time order. It is implemented based on the open source Libav libraries from the FFmpeg project.
  • Souvenir preview CVO
    This CVO enables customers to preview their personal souvenirs. It is implemented as a streaming server using the free FFserver tool from FFmpeg project.

VO Level

  • VO management unit
    This component receive VO execution request from CVO level and search the VO registry for the required VO. If no VO is available, it interacts with the VO factory component to activate the VO and register it to VO registry.
  • VO registry
    The VO registry in Smart Theme Part Trial contains information about each sensor. The information describes the associations between VOs and sensors. Each VO is identified by a “Uniform Resource Identifier” (URI).
  • VO factory
    This component is responsible for the creation of VOs. Once a VO is created, installed and deployed it is registered in the VO registry.
  • Video capturing VO
    This VO maps to the real world digital cameras. Visual images are recorded and compressed by consumer digital cameras, which are installed on each seat of the ride and adjusted to be able to capture the detail expression of the rider. The recorded video streams from each seat are first saved to local storage and transmitted wirelessly to the backend servers when the ride vehicle returns to the station.
  • Audio capturing VO
    Similar to video capturing VO, audio capturing VO capture sounds including screams, laughs made by the riders during the ride using microphones installed on seats. The audio data are also transmitted through wireless links when the vehicle returns back.
  • Identity mapping VO
    This VO establishes the mapping between customer and the seat-mounted digital camera. This mapping information plays a key role for highlights retrieval CVO. The VO is implemented by attach a passive RFID tags to each seat-mounted camera. When the riders are seated, the security inspectors scan the wrist band of each rider and the tag on the camera installed on the same seat using a handheld smart terminal to set up the mapping, which is then transmitted wirelessly and stored in databases for further use.
  • Time sensing VO
    This VO records the time when ride vehicle departs from and returns to the station. Infrared sensors are deployed around the start and end of riding track to detect the passing vehicles.


  • RFID scanning terminal
    BayNexus SDA (Smart Digital Assistant), running Android 4.0 system, equipped with 2.4GHz RFID reader and 802.11b/g/n wireless interface card.
  • Data collection terminal
    A Cubietruck develop board, running Ubuntu 12.10, equipped with Allwinner A20 cpu, 64GB sdcard, 802.11g/n wireless interface, and Ethernet interface.
  • Digital camera
    Logitech Webcam C920, with embedded microphone. This camera uses H.264 encoded compression to provide high quality video at low bit rates and is connected to the Cubietruck develop board via USB 2.0.
  • Infrared sensor
    This sensor could detect objects within the range of 30-60 cm. It is connected to the Cubietruck board via the GPIO pins and deployed around the track of ride vehicles. Sensor data are transmitted via LAN to backend servers.


Service level

Reuse Readiness Levels

4 - Reuse is possible; the software might be reused by most users with some effort, cost, and risk.