Compressive sensing based velocity estimation in video data

TL;DR

This paper introduces a compressive sensing approach combined with time-frequency analysis for accurate vehicle velocity estimation from limited video data, demonstrating robustness across various vehicle types.

Contribution

It presents a novel algorithm that effectively estimates velocity using sparse reconstruction and time-frequency analysis, even with few video frames, advancing video-based velocity measurement methods.

Findings

Accurate velocity estimation with limited video frames.

Robust performance across different vehicle types.

Analysis of parameter influence on estimation accuracy.

Abstract

This paper considers the use of compressive sensing based algorithms for velocity estimation of moving vehicles. The procedure is based on sparse reconstruction algorithms combined with time-frequency analysis applied to video data. This algorithm provides an accurate estimation of object's velocity even in the case of a very reduced number of available video frames. The influence of crucial parameters is analysed for different types of moving vehicles.