An in-air hand gesture signature verification system that doesn't require a handheld device is proposed. depth image sensor captures signature gestures and records each signature as a 3D volume. A structured projection is then applied to the directionally accumulated images for feature extraction. For trajectory features, the fingertip and palm-mass trajectories are extracted from a signature data sample. Subsequently, these features are fused for possible performance enhancement. The signature data acquired using the prototype system contains not only the region of the body but also noise such as imaging distortion and background clutter. The usage of palm-mass center features for identity verification yields better accuracies than that of using the fingertip features. This could be due to stability of the extracted features.
|Number of pages||4|
|Journal||IEEE Intelligent Systems|
|Publication status||Published - 2013|
All Science Journal Classification (ASJC) codes
- Computer Networks and Communications
- Artificial Intelligence