Human Gait Recognition Using Gait Flow Image and Extension Neural Network
This paper represents a new technique to recognize human gait using gait flow image (GFI) and extension neural network (ENN). GFI is a gait period-based technique, based on optical flow. ENN combines the extension theory and neural networks. So a novel ENN-based gait recognition method is proposed, which outperforms all existing methods. All the study is done on, CASIA-A database, which includes 20 persons. The results derived using ENN are compared with support vector machines (SVM) and nearest neighbor (NN) classifiers. ENN proved to have 98 % accuracy and lesser iterations as compared to other traditional methods.
KeywordsGait flow image Extension neural network Optical flow Support vector machine Nearest neighbor
- 6.Arora, P., Hanmandlu, M., Srivastava, S.: Gait based authentication using gait information image features. Pattern Recognition Letters (2015)Google Scholar
- 7.Arora, P., Srivastava, S.: Gait recognition using gait Gaussian image. In: IEEE Second International Conference on Signal Processing and Integrated Networks (SPIN), pp. 915–918. IEEE press (2015)Google Scholar
- 11.Horn, B.K., Schunck, B.G: Determining optical flow. In: Technical Symposium East, pp. 319–331. International Society for Optics and Photonics (1981)Google Scholar
- 12.Vapnik, V.N.: Estimation of dependences based on empirical data, vol. 41. Springer, New York (1982)Google Scholar
- 14.Chen, S., Gao, Y.: An invariant appearance model for gait recognition. In: IEEE International Conference on Multimedia and Expo, pp. 1375–1378. IEEE press (2007)Google Scholar