• Citer

Outdoor group counting based on micro-Doppler signatures obtained with a 77GHz FMCW radar

par Cakoni, Dejvi ;Storrer, Laurent ;Cornelis, Bruno;De Doncker, Philippe ;Horlin, François
Référence (22-27 September, 2024: Paris, France), Proc. of the 21st European Radar Conference (EuRad)
Publication A Paraître, 2024-09

Publication dans des actes

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

In numerous mass gathering settings along withdaily commutes, maintaining an accurate count of individualsis imperative. Radar systems, known for their cost-effectivenessand low energy consumption, facilitate discreet monitoring acrossvarious applications. In this work, data was collected via a77GHz frequency-modulated continuous wave radar (FMCW)in an outdoor pedestrian street. We leverage the unique gaitmodel of each individual, which results in a distinct instantaneousvelocity pattern as a function of time to be able to countpeople. Therefore, we analyze and process our data in thetime-frequency domain to generate the so called micro-Dopplersignatures (MDS). Then, these MDS are fed to a ConvolutionalNeural Network (CNN) to classify groups of different sizes.Furthermore, due to the lack of significant amount of data,the CNN was firstly trained with synthetic data and later onwith the measurement data, to increase the system performance.The proposed system overcomes the limitations of existingcamera-based people counting techniques such as being affectedby lighting conditions and distinctly from other radar relatedwork, targets an outdoor scenario.