Feature Selection Using Genetic Algorithms for Hand Posture Recognition

In this work, we propose a feature selection algorithm to perform hand posture recognition. The hand posture recognition is an important task to perform the human-computer interaction. The hand is a complex object to detect and recognize. That is because the hand morphology varies from human to human. The object recognition community has developed several approaches to recognize hand gestures, but still, there are not a perfect system to recognize hand gestures under diverse conditions and scenarios. We propose a method to perform the hand recognition based on feature selection. The feature selection is performed by a genetic algorithm that combines several features to build a descriptor. The evolved descriptor is used to train a perceptron, which is used as a weak classifier. Each weak learner is used in the AdaBoost algorithm to build a strong classifier. To test our approach, we use a standard image dataset and the full image evaluation methodology. The results were compared with a state of the art algorithm. Our approach demonstrated to be comparable with this algorithm and improve its performance in the some of the cases.

[1]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[2]  Raúl Enrique Sánchez-Yáñez,et al.  Integral split-and-merge methodology for real-time image segmentation , 2015, J. Electronic Imaging.

[3]  Byung Ro Moon,et al.  Hybrid Genetic Algorithms for Feature Selection , 2004, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  François Fleuret,et al.  Efficient Sample Mining for Object Detection , 2014, ACML.

[5]  Sanja Fidler,et al.  Selecting features for object detection using an AdaBoost-compatible evaluation function , 2008, Pattern Recognit. Lett..

[6]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[7]  Pietro Perona,et al.  Pedestrian Detection: An Evaluation of the State of the Art , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Alexei A. Efros,et al.  Ensemble of exemplar-SVMs for object detection and beyond , 2011, 2011 International Conference on Computer Vision.

[9]  Yang Yu,et al.  Subset Selection by Pareto Optimization , 2015, NIPS.

[10]  Takio Kurita,et al.  Fast training algorithm by Particle Swarm Optimization and random candidate selection for rectangular feature based boosted detector , 2008, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence).

[11]  Ai Poh Loh,et al.  Attention Based Detection and Recognition of Hand Postures Against Complex Backgrounds , 2012, International Journal of Computer Vision.

[12]  Qing Chen,et al.  Hand Gesture Recognition Using Haar-Like Features and a Stochastic Context-Free Grammar , 2008, IEEE Transactions on Instrumentation and Measurement.

[13]  Uriel Haile Hernández Belmonte,et al.  Real-Time Hand Posture Recognition for Human-Robot Interaction Tasks , 2016, Sensors.

[14]  Dah-Jye Lee,et al.  A feature construction method for general object recognition , 2013, Pattern Recognit..

[15]  Pascal Fua,et al.  A Real-Time Deformable Detector , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.