Gait recognition based on spatial-temporal analysis of silhouette contours and joint angles

Ali Saadoon Abdulaali (P74183)

Please use this identifier to cite or link to this item: https://ptsldigital.ukm.my/jspui/handle/123456789/476273

Full metadata record

DC Field	Value	Language
dc.contributor.advisor	Md. Jan Nordin, Prof. Madya Dr.
dc.contributor.author	Ali Saadoon Abdulaali (P74183)
dc.date.accessioned	2023-10-06T09:15:38Z	-
dc.date.available	2023-10-06T09:15:38Z	-
dc.date.issued	2016-02-10
dc.identifier.other	ukmvital:80512
dc.identifier.uri	https://ptsldigital.ukm.my/jspui/handle/123456789/476273	-
dc.description	Biometrics means the identification of humans by their characteristics or traits. Every individual has features, therefore, biometrics is based on the unique feature of any person. Biometric attributes are of two types, physical and behavioral. Physiological characteristics are usually face, fingerprints, iris, palm print, DNA, etc. Behavioral characteristics are usually voice and gait. Gait recognition systems are usually applied in surveillance environments as well as for human-machine interaction purposes. Recently, gait recognition has become of interest to researchers. This study proposes an intelligent authentication of a human bt applying various feature extractions. The research implemented a gait recognition based on spatial-temporal analysis of silhouette contours and joint angles. In the present method, silhouettes are primary processed in order to eliminate noises and attaining the extraction of diverse angles. Based on the joints of human body, a silhouette image is segmented to 6 areas and the joints were extracted. In this study, the neck, waist, left and right knee, and left and right ankle joints are to be taken into consideration. In fact, six angle values start from the head to foot have been calculated through using the method of skeleton model. On the other hand, from the silhouette contours, height and width have been also measured. Subsequently, the features are collected and the feature vectors created. Since feature extraction is considered one of the critical steps to identify humans it is required to present an appropriate silhouette as a feature representation. The feature extraction requires low dimensionality and less memory usage as well. Finally the similarities between extracted features and dataset image features are identified and the recognition rate computed. Support Vector Machine classification technique is used for training and testing purpose. Gait representation technique that is used in this study is silhouette contours and Skeletonization model is which based on the appearance of the subject. The system accuracy could be improved by applying a gait representation technique that is based on the appearance and static features of the gait. Compared with previous work our experiments have been carried out on CASIA Gait Database. We achieved recognition rate of 98.41 % with CASIA Gait Database, which shows that this method is capable of doing a good recognition performance.,Master of Computer Science
dc.language.iso	eng
dc.publisher	UKM, Bangi
dc.relation	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat
dc.rights	UKM
dc.subject	Biometric identification
dc.subject	Pattern recognition systems
dc.subject	Computer vision
dc.subject	Universiti Kebangsaan Malaysia -- Dissertations
dc.subject	Dissertations, Academic -- Malaysia
dc.title	Gait recognition based on spatial-temporal analysis of silhouette contours and joint angles
dc.type	theses
dc.format.pages	99
dc.identifier.callno	TK7882.B56A235 2016 3 tesis
dc.identifier.barcode	002079 (2016)
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Files in This Item:

File	Description	Size	Format
ukmvital_80512+SOURCE1+SOURCE1.0.PDF Restricted Access		623.22 kB	Adobe PDF	View/Open

Show simple item record Recommend this item