The International Arab Journal of Information Technology (IAJIT)


A Framework for Saudi Uniform Gait Recognition Based on Kinect Skeletal Tracking

Due to its robustness in challenge variation in gait recognition domain, gait recognition is considered as one of the popular remote biometric identification technologies. Gait data may be reliably collected from a long distance and is difficult to conceal or copy. This article investigated the use of Kinect to identify gait in Saudis wearing loose-fitting apparel that conceals the majority of body shapes, such as thobes or abayas. Because these clothes cover the majority of the joints, it is difficult to determine gait. This research uses the Kinect sensor version 2 as a technique to choose the top three joints with the greatest identification results, which are then used for gait recognition. The Y coordinates of joints are used as features, which are then put into the K Nearest Neighbor classification algorithm. Several experiments were carried out, and the results demonstrate that the system has a promising identification rate and is capable of achieving a high recognition performance when identifying or recognizing a person while also dealing with obstacles associated with the types of loose clothing worn by the participants.

[1] AlKhateeb J., Ren J., Jiang J., and Al-Muhtaseb H., “Offline Handwritten Arabic Cursive Text Recognition Using Hidden Markov Models and Re-Ranking,” Pattern Recognition Letters, vol. 32, no. 8, pp. 1081-1088, 2011.

[2] Chen X., Luo X., Weng J., Lou W., and Li H., “Multi-View Gait Image Generation for Cross- View Gait Recognition,” IEEE Transactions on Image Processing, vol. 30, pp. 3041-3055, 2021. doi: 10.1109/TIP.2021.3055936.

[3] DiFilippo N. and Jouaneh M., “Characterization of Different Microsoft Kinect Sensor Models,” IEEE Sensors Journal, vol. 15, pp. 4554-4564, 2015. doi: 10.1109/JSEN.2015.2422611

[4] Ding M., Li W., Wang H., and Zhao Z., “Human Gait Recognition Based on Multi-Feature Fusion and Kinect Sensor,” in Proceedings of the 12th International Congress on Image and Signal Processing, Bio Medical Engineering and 278 The International Arab Journal of Information Technology, Vol. 21, No. 2, March 2024 Informatics, Suzhou, pp 1-6, 2019. doi:10.1109/CISP-BMEI48845.2019.8965833

[5] Ioannidis D., Tzovaras D., Damousis I., Argyropoulos S., and Moustakas K., “Gait Recognition Using Compact Feature Extraction Transforms and Depth Information,” IEEE Transactions on Information Forensics and Security, vol. 2, no. 3, pp. 623-630, 2007.

[6] Nixon M. and Carter J., “Gait Recognition Based on Kinect Skeletal Tracking,” IEEE Xplore, vol. 94, no. 11, pp. 2013-2024, 2006. doi:10.1109/JPROC.2006.886018

[7] Peterson L., “K-nearest Neighbor,” Scholarpedia, vol. 4, no. 2, pp. 1883, 2009. doi:10.4249/scholarpedia.1883

[8] Sinha A., Chakravarty K., and Bhowmick B., “Person Identification Using Skeleton Information from Kinect,” in Proceedings of the 6th International Conference on Advances in Computer-Human Interactions, pp. 101-108, 2013.

[9] Sharma H. and Grover J., “Human Identification Based on Gait Recognition for Multiple View Angles,” International Journal of Intelligent Robotics and Applications, vol. 2, no. 3, pp. 372– 380, 2018. 0061-y

[10] Tafazzoli F., Bebis G., Louis S., and Hussain M., “Genetic Feature Selection for Gait Recognition,” Journal of Electronic Imaging, vol. 24, no. 1, pp. 013036, 2015.

[11] Tao D., Li X., Wu X., and Maybank S., “General Tensor Discriminant Analysis and Gabor Features for Gait Recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 29, no. 10, pp. 1700-1715, 2007. doi:10.1109/TPAMI.2007.1096

[12] Tharwat A., “Classification Assessment Methods,” Applied Computing and Informatics, vol. 17, no. 1, pp. 186-192, 2018. DOI:10.1016/j.aci.2018.08.003

[13] Wang Y., Sun J., Li J., and Zhao D., “View- Invariant Gait Recognition Based on Kinect Skeleton Feature,” Journal of Multimedia Tools and Applications, vol. 77, pp 24909-24935, 2018.

[14] Yoo J. and Park K., “Skeleton Silhouette Based Disentangled Feature Extraction Network for Invariant Gait Recognition,” in Proceedings of the International Conference on Information Networking, Jeju Island, pp. 687-692, 2021. doi:10.1109/ICOIN50884.2021.9334007