Evaluating sensors for the measurement of public life: A future in image processing

Abstract

William Whyte, one of the most well-known urban planners, documented hundreds of hours of street life using videos, cameras, and interviews to develop social and physical policy recommendations for cities. Since then, studies of public life have primarily depended on human observation for data collection. Our research sets out to test whether Do-it-Yourself sensor technologies can automate this data collection process. To answer this question, our team embedded sensors in moveable benches and evaluated their performance according to the Gehl Method, a popular guideline that measures public life. During three field tests, we gathered information on public life via several sensors including image capture, location tracking, weight measurement, and other environmental sensing techniques. Ultimately, we determined that analysis derived from image processing was the most effective method for measuring public life. Our research demonstrates that it is possible to use sensors to automate the measurement of public life and highlights the value and precision of using video footage for collecting these data. Since image processing algorithms have become more accessible and can be applied to Do-it-Yourself projects, future work can build on this research to develop open access image processing tools to evaluate and advocate for urban design strategies.

Watching TikTok Live Streaming: A Data Collection for Public Life Study in Hutong

“…There are also two researchers, Williams et al [13] and Hoffimann et al [14] get breaks in the methodology of public life studies. Williams et al [13] embedded sensors inside a bench and a sandwich board to measure people's interaction with stationary and pedestrians in public spaces.…”

“…There are also two researchers, Williams et al [13] and Hoffimann et al [14] get breaks in the methodology of public life studies. Williams et al [13] embedded sensors inside a bench and a sandwich board to measure people's interaction with stationary and pedestrians in public spaces. In order to test whether their sensor-embedded furniture is practical or not, they compared the data collected by devices with those collected by the workforce.…”

“…Qualitative data collection Unusual tool Instrument used Aulia and Napitupulu [25] Documentations; observations; interviews No Aliyas and Jafari [26] Observations No Manta et al [8] Interviews No Telephone Zapata and Honey-Rosé s [27] Observations No Honey-Rosé s [28] Observations No Mattsson [29] Documentations; observations; audiovisual and digital materials No Williams et al [13] Audiovisual and digital materials Yes Bench; sandwich board Shen and Liu [9] Observations; audiovisual and digital materials No Drone photography Hoffimann et al [14] Interviews; Audiovisual and digital materials No Smartphone application Antonio [30] Interviews No Hooper et al [10] Document; audiovisual and digital materials No GIS software Hipp et al [7] Documentations, interviews No Telephone, E-mail Van Hecke et al [31] Interviews No Akers [24] Interviews; observation; case study No Wang [12] Documents; interviews; observations, audiovisual and digital materials No GIS software Li et al [32] Observations; interviews; audiovisual and digital materials No DEPTHMAP software Mateo-babiano [33] Observations No Akira et al [34] Observations No Moreno [35] Audiovisual and digital materials No AJAX software Simões Aelbrecht [36] Observations No…”

“…Figure 2 is the flow chart of this research. Similar to Williams et al [13] method to investigate the reliability of the new data collection method. This research compared the results from TikTok live streaming with those from the field research to investigate whether watching TikTok live streaming is reliable.…”

Quantifying the Impact of Urban Infill on the Urban Heat Island Effect – A Case Study for an Alternative Medium Density Model

“…Instead, urban planners often use scaled models, e.g. consisting of an array of concrete cubes (Park et al, 2012) or expansive sensor systems (Cohen et al, 2012), or low-cost sensors (Williams, et al, 2019), to quantify the urban heat impact and its different factors.…”