IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308 _______________________________________________________________________________________ Volume: 03 Issue: 11 | Nov-2014, Available @ http://www.ijret.org 510` TO IDENTIFY THE PERSON USING GAIT: KNN BASED APPROACH P.B.Shelke 1 , P.R.Deshmukh 2 1 Assistant Professor, Department of E&TC, Pankaj Ladhad Institute of Technology, Maharastra, India 2 Professor, Department of CSE, Sipna College of Engg &Technology, Maharashtra, India Abstract In human identification, the process of identifying the person by their gait is an emerging research trend in the field of visual surveillance. Gait is a new biometrics, has been recently used to recognize a person via style of his walking. While person walking variation becomes take place in different parts of body. On the basis of these variations, proposed method is evaluated on CASIA gait database by using K-nearest Neighbor classifier. Experimental results demonstrate that the proposed method has an encouraging recognition performance also the results indicate that the classification ability of KNN with correlation measure perform better than with other type of distance measure functions. Keywords: Gait biometrics, KNN, visual surveillance, CASIA, silhouette. --------------------------------------------------------------------***---------------------------------------------------------------------- 1. INTRODUCTION To identify the people, Gait biometrics play very important role. The fusion of human gait and biometrics [1] has become a popular research direction over the past few years. This concept is used for automated person identification systems for visual surveillance and monitoring applications in security-sensitive environments such as banks, railway stations, and shopping malls. Main aim of person identification system is to discriminate individuals by the way they walk. Gait has some unique features such as it is unobtrusive in natures. It can be captured at a distance without prior consent of the observed object and difficult to hide and steal [2] compare to face, iris, palm biometrics applications. This paper is organized into five sections as follows: Section1Introduction.Section2 Literature review Section3 Proposed Methodology Sections 4.Experiments and Results finally Conclusion and Future Scope are presented in section 5. 2. LITERATURE REVIEW Gait identification approaches are classified into two categories namely model based methods and model free methods [2]. In the model-based methods, the human body silhouette structure or motion is model and then the image features are extracted by the measure of structural components of models or by the motion trajectories of body parts [3, 4, 5, 6]. Most existing model free approaches can be further divided into two main classes, state-space methods and spatiotemporal methods [7, 8, 9, 10]. In the state-space methods consider gait motion to be composed of a sequence of static body poses, and recognize it by considering temporal variations observations with respect to those static pose. The spatiotemporal method characterizes the spatiotemporal distribution generated during their gait motion. Model based approach it has high computational complexity and more difficult in low resolution images so found difficulty in real time system due to feature extraction process. In model free approach as its computational complexity remain low. This approach is well suitable for real time system as it is easy to extract the features comparatively. Philips et.al [11] demonstrated human ID gait analysis and presented the result on Baseline algorithm. Raul MartinFeiez, Ramon A.Mollineda, J.Salvador Sanchez [13] proposed a realistic appearance based representation of gait sequences for automatic gender recognition. It is based on the method where set of appearance based feature of gait sample is used for gender recognition where silhouette appearance of gait sequence was recognized and resulting part of silhouette were fitted by a collection of ellipses. As a result of more realistic ellipses and more meaningful feature space are obtained. Sun Xiaoying, Zhang Qinhong,Xu Yangun[14] proposed a use of new robust gait recognition algorithm based on kinematics characteristics optical and establishes a multi area ellipse model of human body structure by dividing the human body area into several Sub areas according to characteristics of human body and fitting each sub-area with an ellipse model into which Optical flow feature are integrated. For this experiment they have used CASIA gait database and have achieved 84.25% correct classification rate. Haitao Liu, Yang Cao, Zengfu Wang [15] proposed a simple and effective approach for gait recognition based on stereo vision .In this methods, 3D silhouette Silhouette vector then stereo gait features are extracted for analysis and recognition.PCA is used for dimensionality reduction of gait features. Finally by using NN and ENN classifiers, achieved 59.27% and 70.18% correct recognition rate. 3. PROPOSED METHODOLOGY The proposed system consist of four phases namely, preprocessing, segmentation, feature extraction and classification. Initially gait video sequence is captured by using static camera. By using approximate background subtraction method binary silhouette of the moving objects