2026-04-30T23:29:54Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/3422 2021-12-13T06:06:23Z 1582963302567:1597824273898 user-ucsy

Geometric Kinect Joints Computing for Human Fall Recognition Htoo, Chit Kyin Sein, Myint Myint This paper proposes an computing analysis onhuman geometric shape features to detect a fallbehavior. The system mainly computes the changeson human orientation (torso angle) and centroidheight via the human skeleton joints extracted byKinect sensor. The system computes and tracks thespatial changes of these human orientation andcentroid height and distinguishs a fall behavioramong other daily activities by using a thresholdingalgorithm. The main objective of this computation isto minize the computational time and to increase thetrue alarms in developing a fall detection. The systemworks the feature extraction on our collected falldetection dataset containing the fall data along withdaily activities such as sitting down, lying, combing.Standing, etc., are collected by Microsoft Kinectsensor. 2019-02-27 http://hdl.handle.net/20.500.12678/0000003422 https://meral.edu.mm/records/3422