Automatic and privacy-preserving systems to monitor elderly people in their home environment are one of the basic targets addressed by the wide research area of Ambient Assisted Living. Thanks to the low-cost Microsoft Kinect® device, high-resolution depth and visual sensing is now not limited to experimental and prototype implementations and is ready to address marketable solutions. This chapter emphasizes the advantages provided by Kinect in the field of automatic monitoring, discussing its performance in human subject detection and tracking. Two sample use cases are discussed in detail: the former deals with generating a numerical representation of the Get Up and Go Test outcome, the latter implements an automatic fall detection algorithm based on depth frames analysis, with the sensor in a top configuration. The chapter ends suggesting issues that need to be addressed to further extend the range of applications for the Kinect device and enhance the obtainable performance.
Depth Cameras in AAL Environments: Technology and Real-world Applications / Gasparrini, Samuele; Cippitelli, Enea; Spinsante, Susanna; Gambi, Ennio. - ELETTRONICO. - (2015), pp. 22-41. [10.4018/978-1-4666-7373-1.ch002]
Depth Cameras in AAL Environments: Technology and Real-world Applications
GASPARRINI, SAMUELE;CIPPITELLI, Enea;SPINSANTE, Susanna;GAMBI, Ennio
2015-01-01
Abstract
Automatic and privacy-preserving systems to monitor elderly people in their home environment are one of the basic targets addressed by the wide research area of Ambient Assisted Living. Thanks to the low-cost Microsoft Kinect® device, high-resolution depth and visual sensing is now not limited to experimental and prototype implementations and is ready to address marketable solutions. This chapter emphasizes the advantages provided by Kinect in the field of automatic monitoring, discussing its performance in human subject detection and tracking. Two sample use cases are discussed in detail: the former deals with generating a numerical representation of the Get Up and Go Test outcome, the latter implements an automatic fall detection algorithm based on depth frames analysis, with the sensor in a top configuration. The chapter ends suggesting issues that need to be addressed to further extend the range of applications for the Kinect device and enhance the obtainable performance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
