Vehicle detection from GSV imagery: Predicting travel behaviour for cycling and motorcycling using Computer Vision

2025-08-18

AI Summary

To address the scarcity of global cycling and motorcycle travel data, this study proposes an automated, survey-free method for estimating mode share using Google Street View (GSV) imagery. We employ YOLOv4 to detect bicycles and motorcycles in street-level images and integrate these detections with covariates—including population density—into a Beta regression model with log-transformed inputs to predict city-level cycling and motorcycle mode shares. This approach enables the first large-scale, cross-city comparable estimation across 185 cities without ground-truth travel surveys. The model achieves R² values of 0.614 (cycling) and 0.612 (motorcycle), with median absolute errors of only 1.3% and 1.4%, respectively. It delivers reliable modal split estimates for 60 cities lacking conventional transportation data, thereby filling a critical methodological gap in macro-level mobility behavior monitoring.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: The sharing economy
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Transportation influence health by shaping exposure to physical activity, air pollution and injury risk.Comparative data on cycling and motorcycling behaviours is scarce, particularly at a global scale.Street view imagery, such as Google Street View (GSV), combined with computer vision, is a valuable resource for efficiently capturing travel behaviour data.This study demonstrates a novel approach using deep learning on street view images to estimate cycling and motorcycling levels across diverse cities worldwide.We utilized data from 185 global cities.The data on mode shares of cycling and motorcycling estimated using travel surveys or censuses.We used GSV images to detect cycles and motorcycles in sampled locations, using 8000 images per city.The YOLOv4 model, fine-tuned using images from six cities, achieved a mean average precision of 89% for detecting cycles and motorcycles in GSV images.A global prediction model was developed using beta regression with city-level mode shares as outcome, with log transformed explanatory variables of counts of GSV-detected images with cycles and motorcycles, while controlling for population density.We found strong correlations between GSV motorcycle counts and motorcycle mode share (0.78) and moderate correlations between GSV cycle counts and cycling mode share (0.51).Beta regression models predicted mode shares with $R^2$ values of 0.614 for cycling and 0.612 for motorcycling, achieving median absolute errors (MDAE) of 1.3% and 1.4%, respectively.Scatterplots demonstrated consistent prediction accuracy, though cities like Utrecht and Cali were outliers.The model was applied to 60 cities globally for which we didn't have recent mode share data.We provided estimates for some cities in the Middle East, Latin America and East Asia.With computer vision, GSV images capture travel modes and activity, providing insights alongside traditional data sources.

Problem

• Estimating global cycling and motorcycling levels using street view imagery
• Developing a deep learning model for vehicle detection in GSV images
• Predicting travel behavior to address scarce comparative data

Innovation

• Uses YOLOv4 for vehicle detection
• Applies beta regression for global predictions
• Leverages GSV images for travel behavior analysis