Vision‐based vehicle speed estimation: A survey

The need to accurately estimate the speed of road vehicles is becoming increasingly important for at least two main reasons. First, number cameras installed worldwide has been growing in recent years, as introduction and enforcement appropriate limits are considered one most effective means increase safety. Second, traffic monitoring forecasting networks plays a fundamental role enhance traffic, emissions energy consumption smart cities, being relevant parameters state. Among technologies available accurate detection vehicle speed, use vision-based systems brings great challenges be solved, but also potential advantages, such drastic reduction costs due absence expensive range sensors, possibility identifying accurately. This paper provides review estimation. terminology application domains described complete taxonomy large selection works that categorizes all stages involved proposed. An overview performance evaluation metrics datasets provided. Finally, current limitations future directions discussed.