Computer vision supported pedestrian tracking: A demonstration on trail bridges in rural Rwanda

Abstract

Trail bridges can improve access to critical services such as health care, schools, and markets. In order to evaluate the impact of trail bridges in rural Rwanda, it is helpful to objectively know how and when they are being used. In this study, we deployed motion-activated digital cameras across several trail bridges installed by the non-profit Bridges to Prosperity. We conducted and validated manual counting of bridge use to establish a ground truth. We adapted an open source computer vision algorithm to identify and count bridge use reflected in the digital images. We found a reliable correlation with less than 3% error bias of bridge crossings per hour between manual counting and those sites at which the cameras logged short video clips. We applied this algorithm across 186 total days of observation at four sites in fall 2019, and observed a total of 33,800 daily bridge crossings ranging from about 20 to over 1,100 individual uses per day, with no apparent correlation between daily or total weekly rainfall and bridge use, potentially indicating that transportation behaviors, after a bridge is installed, are no longer impacted by rainfall conditions. Higher bridge use was observed in the late afternoons, on market and church days, and roughly equal use of the bridge crossings in each direction. These trends are consistent with the design-intent of these bridges.

Factors Affecting Cost and Execution Period of Trail Bridge Construction Project in Nyamagabe District, Rwanda

“…Email: info@stratfordjournals.org ISSN: 2616-8464 to Prosperity (B2P), a non-profit organization, builds trail bridges to connect rural communities to critical destinations, and studies have established economic and livelihood benefits associated with these bridges (Thomas, 2020;Brooks & Donovan, 2019). Objective measurement technologies, such as motion-activated cameras, are used to assess bridge use (Thomas, 2020). Various methods, including in-person counting and digital video analysis, have been deployed to track bridge usage (Kothuri, 2017;Ghadiyaram, 2017).…”

Rural Transportation Infrastructure in Low- and Middle-Income Countries: A Review of Impacts, Implications, and Interventions

“…To this end, B2P works actively within a research agenda related to rural transportation including usage impacts and evaluations of rural bridges, and plans to continue this work with construction of an additional 150 bridges in Rwanda and an associated randomized controlled trial study by 2023 [232,233,239]. Researchers should also consider further investigation of seasonal bridge trends, expanding on the conclusion that no apparent change in bridge usage patterns during rain events in Rwanda may suggest that bridges minimize seasonal and extreme weather-related transportation interruptions [239,242]. Additional future efforts should look toward the formalized and institutionalized maintenance of bridges [243] and toward the adaptation of bridges to climate-resilient standards [241].…”

“…In this way, innovations and advanced technologies that are designed to provide faster, easier, less resource-intensive, more accurate, and more cost-effective data collection are playing an increasingly vital role throughout the sector [60,63,272]. Innovative technologies and strategies are becoming increasingly useful and widely implemented in the rural transportation sector worldwide, even despite a history of technological innovations being used mainly in higher income and urban settings in LMICs [15,242,272]. In accordance, several studies have proposed and investigated the impacts of new data collection methodologies, techniques, platforms, and tools, with the objectives of reducing cost, requiring less time, effort, and professional skill, and providing policymakers with critical information.…”

“…Namely, a Mozambican case study developed a methodology to prioritize rural transportation investments amidst uncertainty from climate change related disasters by utilizing geospatial data coupled with multidimensional cost-benefit analyses, resulting in low-budget and resilient improvements to the network [63]. Data collection tools used to examine road or bridge conditions and usage levels, details of particular networks, and socioeconomic impacts include motion-activated digital cameras, accelerometers, smartphone applications, satellite imaging, and may further expand into the realms of drones or other unmanned aerial vehicles [UAVs] moving forward [60,63,242,272]. Technological innovations in the rural transportation sector have begun to expand beyond data collection, and into the storage, analysis, and presentation of data, through the usage of global road databases and open-source platforms [15,50].…”

“…However, even crowdsourced tools such as OSM are subject to biases toward urban and more populous areas, which reiterates a need for further innovation within the collection of data in rural areas [15]. Other steps within future research may involve expansion upon the data processing capacities of computer vision, machine learning, and artificial intelligence, which could be used to automate the collection, translation, and export of rural transportation related data more accurately and efficiently [242,256,272].…”