Devices, people and vehicles counting - Liverpool

Liverpool City Council has device counters deployed throughout the city centre to understand how the city is used. Data in this dashboard is derived from Meshed nCounters and CCTV retrofitted with an algorithm developed by University of Wollongong. The algorithm has used machine learning to identify objects in the camera viewfinder and categorise them into pedestrian, vehicle or bicycle. This transforms the CCTV into visual sensors. The nCounters generate data by counting the number of Wi-Fi signals emitted by non-identifiable mobile devices within a specified proximity and performing certain filtering and processing. No individuals are identified by either method. Both methods have advantages and limitations.Advantages of the nCounter method:Can provide insightful data on crowd sizes and individualsAn individual with one device will be counted onceThe nCounter can report the average ‘dwell time’ of the deviceLimitations of the nCounter method:individuals without devices will not be counted (for example young children or people without smart phones),if someone is carrying a smart phone which is in aeroplane mode or switched off then it will not be counted, andindividuals with multiple devices will be counted by the number of devices they have. For example, one person may have two smart phones, therefore the individual will be counted more than once.The visual sensors (CCTV) count the number of bicycles, people and vehicles in the location.Advantages of the visual sensor method:Can provide insightful data on pedestrian, vehicle and bicycle numbers,Re-uses existing common technology on city streets without further visual clutter,Does not rely on individuals carrying their own devices, so useful in areas with lower technology uptakeLimitations of visual sensor method:Individuals can be counted multiple times as they exit and re-enter the camera viewfinder,The machine learning cannot differentiate between bicycles and motorbikesData collected in this dataset can be visualised in the Liverpool City Centre activity dashboard