Describes the interpolated number of pedestrians per hour moving in either direction along the sidewalk in the visible video field. The count refers to a single sidewalk on only one side of the street, even if there are two sidewalks on both sides of a road. The original video clips typically capture a 15-minute observation period on a sidewalk, which are played back in time lapse videos at 12 times the original speed. The counted number of pedestrians observed in the said 15-minute period is generalized to an hour by multiplying by four (e.g. 40 pedestrians per 15 minutes equals 40*4=160 pedestrians per hour). A handy tool for performing the counts from time lapse videos is Google Chrome browser's Video Speed Controller, which allows you to slow the time lapse down for easier counting.
The betweenness from network analysis describes how in-between given origins and destinations on a network a particular location is, under the assumptions that travelers take either shortest, or near shortest paths from the origins to the destinations. The metric thus depends on what origins and destinations are used, at what walking radius and whether only shortest routes or also slightly longer routes are considered. The Street Catalog betweenness metric uses all business establishments, obtained from Google Maps in a 400 meter (1,300 ft) radius around the camera as origin points and all public transit stops found in the same radius as destinations. The metric effectively models predicted pedestrian flow from nearby businesses to nearby transit stations. The maximum allowable travel distance is set at 400 meters and walks can run on not only shortest paths, but paths that are up to 15% longer than shortest paths. Each of the paths from an origin to a destination that is up to 15% longer than the shortest path is given an equal probability. Betweenness was measured at the location of the camera (i.e. camera location is set as an Observer point in the Rhino UNA betweenness tool).
Measures how many individual business establishments are found within a 400 meter network walk from the camera location. The data describing the business locations as points or coordinates is obtained from Google Maps or Open Street Map. The metric provides a basic indicator of how many potential trip destinations are found in the vicinity of the camera location, which are likely to affect the observed pedestrian traffic on the street.
Measures how many public transit stops are found within a 400 meter network walk from the camera location. Since stops are not equal and MRT stops tend to serve more riders than bus stops, the following convention was adopted for weighing the stops: An MRT (or subway) stop has a weight of "1000"; a light rail or tram stop a weight of "200"; and a bus stop a wight of "100". These weights roughly correspond to the differences in carrying capacity of the respective vehicles. An MRT train can carry up to 5 times more people than a tram and 10 times more people than a bus. If there is one MRT, one LRT and one bus stop within a 400 meter buffer around the camera, then the comined transit access result would show 1000+200+100=1300. Ideally, stop weights should also distinguish how many lines pass trhough each stop. Since such data is difficult to obtain in most cities, we have omitted this factor.
Measures how much gross floor area (GFA) in square meters is found in the buildings around the camera location within a 400 meter
(1,300 ft) network radius. All buildings, regardless of type or function, are included as destinations for simplicity. The GFA numbers at the individual building level can be approximated from a site survey, obtained from axonometric views in
Google Maps, precisely calculated from available data by multiplying the building footprint area with the number of floors, or obtained from city assessors records. Buildings of different size thus contribute differently to the result, with larger and nearer buildings creating a bigger impact on floor area accessibility.
Handy, S., & Niemeier, A. D. (1997). Measuring Accessibility: an exploration of issues and alternatives. Environment and Planning A, 29, 1175–1194.
Sevtsuk, A., Mekonnen, M., 2012, “Urban Network Analysis Toolbox,” International Journal of Geomatics and Spatial Analysis, vol. 22, no. 2, pp. pp. 287–305. PDF