Modified curvature scale space feature alignment approach for hand posture recognition

Abstract

In this paper, we present a modified feature alignment approach based on curvature scale space (CSS) for translation, scale, and rotation invariant recognition of hand postures. First, the CSS images are used to represent the shapes of contours of hand postures. Then, we extract and align the CSS features to overcome the problem of multiple deep concavities in contours of hand postures. Finally, nearest neighbor techniques are used to perform CSS matching between the input CSS features and the stored CSS features for hand posture identification. Results show the proposed approach performs well for recognition of hand postures.