“…Luo, Wu [ 11 ] extracted connected sidewalk network from aerial images and the common occlusion problem resulting from trees and their shadows can bring large uncertainty to the detected sidewalks under the bird’s eye view [ 12 ]. Hosseini et al [ 10 ] developed a scalable satellite imagery-based sidewalk inventory method. Street-view images are another data source used for assessing sidewalk accessibilities.…”
Section: Literature Reviewmentioning
confidence: 99%
“…A significant portion of the work in this domain relied on field observations [ 5 , 6 ] and street-view photo-based analysis [ 7 ]. Remote sensing technologies enabled the utilization of satellite images and point-cloud data for digitizing pedestrian infrastructure [ 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 ]. Among these technologies, vehicular mapping systems such as mobile LiDAR have recently gained increased attention from the transportation community due to their capability to collect accurate and dense urban point-cloud sets at street level.…”
In this paper, a Segment Anything Model (SAM)-based pedestrian infrastructure segmentation workflow is designed and optimized, which is capable of efficiently processing multi-sourced geospatial data, including LiDAR data and satellite imagery data. We used an expanded definition of pedestrian infrastructure inventory, which goes beyond the traditional transportation elements to include street furniture objects that are important for accessibility but are often omitted from the traditional definition. Our contributions lie in producing the necessary knowledge to answer the following three questions. First, how can mobile LiDAR technology be leveraged to produce comprehensive pedestrian-accessible infrastructure inventory? Second, which data representation can facilitate zero-shot segmentation of infrastructure objects with SAM? Third, how well does the SAM-based method perform on segmenting pedestrian infrastructure objects? Our proposed method is designed to efficiently create pedestrian-accessible infrastructure inventory through the zero-shot segmentation of multi-sourced geospatial datasets. Through addressing three research questions, we show how the multi-mode data should be prepared, what data representation works best for what asset features, and how SAM performs on these data presentations. Our findings indicate that street-view images generated from mobile LiDAR point-cloud data, when paired with satellite imagery data, can work efficiently with SAM to create a scalable pedestrian infrastructure inventory approach with immediate benefits to GIS professionals, city managers, transportation owners, and walkers, especially those with travel-limiting disabilities, such as individuals who are blind, have low vision, or experience mobility disabilities.
“…Luo, Wu [ 11 ] extracted connected sidewalk network from aerial images and the common occlusion problem resulting from trees and their shadows can bring large uncertainty to the detected sidewalks under the bird’s eye view [ 12 ]. Hosseini et al [ 10 ] developed a scalable satellite imagery-based sidewalk inventory method. Street-view images are another data source used for assessing sidewalk accessibilities.…”
Section: Literature Reviewmentioning
confidence: 99%
“…A significant portion of the work in this domain relied on field observations [ 5 , 6 ] and street-view photo-based analysis [ 7 ]. Remote sensing technologies enabled the utilization of satellite images and point-cloud data for digitizing pedestrian infrastructure [ 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 ]. Among these technologies, vehicular mapping systems such as mobile LiDAR have recently gained increased attention from the transportation community due to their capability to collect accurate and dense urban point-cloud sets at street level.…”
In this paper, a Segment Anything Model (SAM)-based pedestrian infrastructure segmentation workflow is designed and optimized, which is capable of efficiently processing multi-sourced geospatial data, including LiDAR data and satellite imagery data. We used an expanded definition of pedestrian infrastructure inventory, which goes beyond the traditional transportation elements to include street furniture objects that are important for accessibility but are often omitted from the traditional definition. Our contributions lie in producing the necessary knowledge to answer the following three questions. First, how can mobile LiDAR technology be leveraged to produce comprehensive pedestrian-accessible infrastructure inventory? Second, which data representation can facilitate zero-shot segmentation of infrastructure objects with SAM? Third, how well does the SAM-based method perform on segmenting pedestrian infrastructure objects? Our proposed method is designed to efficiently create pedestrian-accessible infrastructure inventory through the zero-shot segmentation of multi-sourced geospatial datasets. Through addressing three research questions, we show how the multi-mode data should be prepared, what data representation works best for what asset features, and how SAM performs on these data presentations. Our findings indicate that street-view images generated from mobile LiDAR point-cloud data, when paired with satellite imagery data, can work efficiently with SAM to create a scalable pedestrian infrastructure inventory approach with immediate benefits to GIS professionals, city managers, transportation owners, and walkers, especially those with travel-limiting disabilities, such as individuals who are blind, have low vision, or experience mobility disabilities.
“…Particularly in developing countries, the spatial resolution of open source satellite images is often insufficient to accurately detect small-scale features like sidewalks (Roy and Saha, 2016; Yu et al, 2020). Recently, Hosseini et al (2023) achieved high performance metrics in identifying sidewalks and crosswalks from satellite images of three cities in the United States which were captured at 20 zoom level representing 0.19 m on the earth’s surface. However, the publicly available satellite images for developing countries like Bangladesh are mostly captured at 18 zoom level representing greater than 0.3 m on the surface of earth (Google Maps Static API, 2023).…”
Section: Literature Reviewmentioning
confidence: 99%
“…Also, integrating the labeled dataset generated from this study will add geographic and socio-economic diversity to popular open-source datasets (e.g., Cityscapes and Mapillary) for semantic segmentation model benchmarking. Besides, heterogeneity in the physical features, materials, and dimensions of roads and sidewalks exist among different cities, and the differences are more prominent between international cities (Hosseini et al, 2023). So, in order to implement the computer vision models in developing countries which were initially developed using images collected from developed settings, these models need to be retrained (i.e., training the model from scratch) using local training images for achieving higher model performance.…”
Section: Introductionmentioning
confidence: 99%
“…By using this tool, transportation authorities, practitioners, and other stakeholders will become well informed about the existing state of pedestrian infrastructure, and plan better for the future developments. Although GSV images are already being used to create sidewalk inventories with the help of image segmentation models, such approaches have mostly been adopted in developed countries (Hosseini et al, 2023; Kang et al, 2021). As a result, adequate well-labeled images belonging to developing countries are scarce.…”
Although reliable and accurate inventorying of sidewalks is time consuming, it can aid urban planners in decision making for infrastructure development. Recent advancements in computer vision and machine learning algorithms have improved the reliability and accuracy of automated inventorying. This research uses a deep learning architecture-based semantic segmentation model (i.e., HRNet + OCR) to automate sidewalk identification using Google Street View (GSV) images. The results show that retraining the model using local training images yields 114.16% and 178.11% higher performance in terms of intersection over union (IoU) metric compared to pretrained model using Cityscapes and Mapillary datasets, respectively. The developed model showed excellent performance in predicting the presence of sidewalks in an image by achieving high accuracy (0.9557), precision (0.9447), recall (0.9900), and F1- score (0.9668). Further, in comparison with EfficientNet, a computationally efficient image classification model, the present model showed superior performance in predicting sidewalk presence at the image level. Therefore, integrating local training images containing minimum required labels (in this study, roads, sidewalks, buildings, and walls) with publicly available training datasets can help increase the performance of the semantic segmentation model for extracting the required features (in this study, roads and sidewalks) from GSV images, especially in developing countries like Bangladesh. This study generates sidewalk maps on a neighborhood scale, which can be useful visualization tools for researchers and practitioners to understand the existing pedestrian infrastructure and plan for future improvements.
Encouraging sustainable mobility through sidewalk condition improvement is a critical concern for urban transportation. Sidewalk condition affects pedestrian safety, satisfaction, and mobility inclusiveness. Early sidewalk defect detection and repair ensure transport justice by addressing pedestrian inequality caused by walkability issues. This study presents novel Sidewalk Defect Detection Models (SDDMs) using computer vision to identify and delineate sidewalk defect boundaries accurately. The SDDMs provide a cost‐effective and efficient sidewalk inspection method, achieving high accuracy in recognizing defects for concrete and brick materials (mIoU of 0.91 and mAP of 0.99 for concrete, mIoU of 0.90, and mAP of 0.97 for brick). Integrated with Google Street View for data acquisition, it offers a rapid solution for monitoring sidewalk conditions remotely, promoting sustainability through timely repairs. This research provides significant advancements in urban planning and transport research, ultimately improving pedestrian safety and satisfaction. Thus, it makes human settlements more inclusive, safe, and sustainable.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.