Open AccessExtracting Multiple Features for Dynamic Hand Gesture Recognition
Author(s) -
S.S. Suni,
K. Gopakumar
Publication year - 2021
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.d2343.0410421
Subject(s) - gesture , artificial intelligence , local binary patterns , computer science , computer vision , histogram , translation (biology) , optical flow , motion (physics) , gesture recognition , invariant (physics) , binary number , pattern recognition (psychology) , image (mathematics) , mathematics , mathematical physics , biochemistry , chemistry , arithmetic , messenger rna , gene
In this work a framework based on histogram oforientation of optical flow (HOOF) and local binary pattern fromthree orthogonal planes (LBP_TOP) is proposed for recognizingdynamic hand gestures. HOOF algorithm extracts local shapeand dynamic motion information of gestures from imagesequences and local descriptor LBP is extended to threeorthogonal planes to create an efficient motion descriptor. Thesefeatures are invariant to scale, translation, illumination anddirection of motion. The performance of the new framework istested in two different ways. The first one is by fusing the globaland local features as one descriptor and the other is using featuresseparately to train the multi class support vector machine.Performance analysis shows that the proposed approach producesbetter results for recognizing dynamic hand gestures whencompared with state of the art methods