Journal of Transportation Engineering · 2006 · 29 citations · 13 references
ComputingComputer ScienceData ScienceComputer VisionMachine VisionTransportData CleaningData QualityIntelligent Transportation SystemsMatching MethodsOptical Character RecognitionPattern RecognitionAutomated License Plate RecognitionData Quality CleaningPlate CleaningTemporal Structure Analysis
In recent years automatic number plate recognition systems have been increasingly used to monitor urban link travel time. However, the set of travel times produced will contain erroneous records for a variety of reasons including the misreading of the licence plates by the camera, vehicles stopping en-route, and vehicles taking nondirect routes between the two camera sites. This paper proposes a novel technique called the overtaking rule, which uses the serial temporal structure of the data to filter out these erroneous travel time records. The paper proposes a methodology to measure the performance of a licence plate cleaning method, and it is subsequently shown that the overtaking rule outperforms existing cleaning methods.
13
<i>Outliers in Statistical Data</i>
V. Barnett, T. Lewis, Francine F. Abeles · Physics Today · 1979 · 2.8K citations
N. Balakrishnan, V. Barnett, T. Lewis · Biometrics · 1995 · 1.2K citations
Computing, Data Science, Mathematics +12