Abstract
Traffic operations field computational resources as well as the bandwidth of field communication links are often quite limited. Accordingly, for on-line implementation of Advanced Transportation Management and Information Systems (ATMIS) strategies, such as vehicle reidentification, there is strong interest in development of fieldâ??based techniques and models that can perform satisfactorily while minimizing field computational and communication requirements. A new vehicle reidentification algorithm (REID-2) developed previously by the authors (1) was oriented toward algorithm simplification, but also demonstrated the added benefits of improved performance and much broader potential applicability (to both round and square single inductive loops) compared with earlier methods. However, the basis of REID-2 is directly matching inductive vehicle signatures, which typically consist of 200~1,200 data points (stored as integers, and obtained from IST-222 detector cards) per signature. The purpose of this research was to investigate if a relatively simple data compression and transformation technique could be applied successfully to the raw inductive signatures for each vehicle, and then use the resulting transformed vehicle signatures as inputs to vehicle reidentification. A Piecewise Slope Rate (PSR) approach was used to compress and transform the raw vehicle signatures. The results of this investigation, including sensitivity analyses, vehicle reidentification performance, and the accuracy of section travel time measurement, are very promising and suggest that the reduction in both computational effort and computer memory needed to store individual signatures with this approach could potentially benefit both the field computational and communication requirements needed for real-time implementation of this modified vehicle reidentification technique.
