Source code and data sets

Core technical components of the hackAIR open technology platform

Open hardware sensors

  1. hackAIR home
    The hackAIR home sensor is an open hardware air quality sensor, based on Arduino (or Wemos for version 2). Connected through Ethernet or WiFi, it measures particulate matter concentrations and sends the results to the hackAIR platform.
    Documentation | Code
  2. hackAIR mobile
    The hackAIR mobile sensor is an open hardware air quality sensor, based on PSOC. It is powered with a mobile power bank and uploads its measurements through Bluetooth on your phone.
    Documentation | Code
  3. hackAIR cardboard
    The hackAIR cardboard sensor is low-cost and easy to use. You expose the cardboard for 24 hours, and then analyse a macro-photograph of the jelly to calculate pollution levels in the hackAIR app.

hackAIR open platform

  1. Web platform
    The hackAIR web platform integrates air quality data and provides user access to the hackAIR data via a web browser.
    Documentation | Code
  2. Mobile app
    The hackAIR mobile app provides access to hackAIR data and results via a mobile device, enables users to upload sky and cardboard photos and to add their perception. The app is available for both Android and iOS.
    Documentation | Code
  3. hackAIR API
    The hackAIR API provides free access to the hackAIR data for the mobile and web applications through a middleware, which applies all the business logic.
    Documentation | Code

Internal service modules

  1. Data retrieval
    The hackAIR data retrieval module collects air quality measurements from official web sites or web services and public images from media sharing platforms (i.e. Flickr) and public webcams. An image analysis process is applied on the collected images that involves detection of sky and localization of the sky part.
    Documentation | Code
  2. Data fusion
    The hackAIR data fusion module combines hackAIR observations with model output from the Copernicus Atmosphere Monitoring Service (CAMS) to provide estimates of the current air quality at any location within the study domains.
    Documentation included in Deliverable 4.1 | Code available on request from
  1. Knowledge base and reasoning framework
    The hackAIR knowledge base and reasoning framework handles both the semantic integration and reasoning of environmental and user-specific data, in order to provide recommendations to the hackAIR users, with respect to: (i) personal health and user preferences (activities, daily routine, etc.), and (ii) current air quality conditions of the location of interest.
    Documentation | Code 1 2 3
  2. Social media monitoring
    The hackAIR social media monitoring module monitors a number of keywords and accounts of interest and calculates a set of indicators that represent the “engagement” of social media users with the topic described by the keywords.
    Documentation | Code

Data sets

Open datasets included in the hackAIR data management plan

  • Geotagged Images Dataset (Github, Zenodo)
  • Environmental measurements Dataset (Github)
  • Twitter_AQ (Github)
  • PM measurements by Arduino hackAIR sensors (Zenodo)
  • PM measurements by PSOC hackAIR sensors (Zenodo)
  • Photos uploaded through the hackAIR mobile app (Zenodo)