Traffic Monitoring Using Satellite and Ground Data: Preparation for FeasibilityTest and an Operational System

摘要

The authors address three issues in this report: demonstrate that vehicles can beidentified and classified accurately from satellite imagery, develop efficient image processing methods, and determine methods to integrate the imagery with ground-based data and assess the value of this integration. Field tests were replicated, where aerial photographs were used to simulate satellite imagery. The authors developed software to automate many of the calculations involved, and the empirical results show that the authors' approach and software work well. The