A Robust Method Based on Static Hand Gesture Recognition for Human–Computer Interaction Under Complex Background
An appearance-based approach needs usually a perfect segmentation. However, it is a difficult task especially under complex background. As a result, it limits the robustness for application. In this chapter, we design a new method for static hand gesture recognition in complex background for human–computer interface (HCI). In this method, we do not need perfect segmentation or hand tracking. The Hu invariant moment features are extracted from a binary image after simple segmentation and served as the input of our classifier, which is constructed beforehand based on support vector machines (SVM) algorithm. Furthermore, a Euclidean distance is calculated to combine with SVM model for avoiding the non-hand gestures. Tests on the testing dataset show the proposed method exhibits a recognition rate near 100%. Experimental results on a simple HCI system on real-time demonstrated the effectiveness, speediness and robustness of the system under cluttered background.
KeywordsHand gesture recognition Support vector machine (SVM) Hu invariant moments Human–computer interaction (HCI)
This work was supported by National Natural Science Foundation of China under the Grants 60978006.
- 8.Otsu N (1979) A threshold selection method from gray level histogram, IEEE SMC-9 (1):62–66Google Scholar
- 9.Tin H, Maung H (2009) Real-time hand tracking and gesture recognition system using neural networks. In: Proceedings of World Academy of Science, Engineering and Technology, p 50Google Scholar
- 11.Bradski G, Kaehler A (2008) Machine learning. In: learning OpenCV: computer vision with the OpenCV library, O’Reilly Media USA pp 495−517Google Scholar
- 12.Jianjun Y, Hongxun Y, Feng J (2004) Based on HMM and SVM multilayer architecture classifier for Chinese sign language recognition with large vocabulary. In: Proceedings of the third international conference on image and graphicGoogle Scholar
- 13.Platt JC, Cristianini N, Shawe-Taylor J (2000) Large margin DAGs for multiclass classification. Adv Neur Inform Process Syst 12:547–553Google Scholar