Detecting Irregularities by Image Contour Based on Fuzzy Neural Network

Visual analysis of human motion in video sequences has attached more and more attention to computer visions in recent years. In order to indicate pedestrian movement in Intelligent Monitoring System, a Euclidean distance based on centroid method is proposed. And then according to the movement of body a set of standard images contour are made. All matrixes which represent human silhouette are normalized using affine transformation, which cuts computational cost. The difference between two matrixes is regard as fuzzy function. Fuzzy neural network is proposed to infer abnormal behavior of the walker. First of all, a four layer fuzzy neural network is presented. And then Fuzzy C-means clustering algorithm is used to calculate the number of hidden layer nodes. Finally the degree of the anomaly is resulted from the fuzzy membership of the two matrixes difference. Fuzzy discriminant can detect irregularities and implements initiative analysis to body behavior. The results show that the new algorithm has better performance.

[1]  Hironobu Fujiyoshi,et al.  Real-time human motion analysis by image skeletonization , 1998, Proceedings Fourth IEEE Workshop on Applications of Computer Vision. WACV'98 (Cat. No.98EX201).

[2]  Yunjie Zhang,et al.  The Global Fuzzy C-Means Clustering Algorithm , 2006, 2006 6th World Congress on Intelligent Control and Automation.

[3]  Tieniu Tan,et al.  A survey on visual surveillance of object motion and behaviors , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[4]  Michal Irani,et al.  Detecting Irregularities in Images and in Video , 2005, ICCV.

[5]  Michio Sugeno,et al.  Fuzzy identification of systems and its applications to modeling and control , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[6]  Yangsheng Xu,et al.  Abnormal Behavior Detection by Multi-SVM-Based Bayesian Network , 2007, 2007 International Conference on Information Acquisition.

[7]  Paulo Cortez,et al.  Prediction of Abnormal Behaviors for Intelligent Video Surveillance Systems , 2007, 2007 IEEE Symposium on Computational Intelligence and Data Mining.

[8]  R. Hagege,et al.  Parametric estimation of multi-dimensional affine transformations in the presence of noise: a linear solution , 2005, IEEE/SP 13th Workshop on Statistical Signal Processing, 2005.

[9]  A. Mecocci,et al.  Automatic detection of anomalous behavioural events for advanced real-time video surveillance , 2003, The 3rd International Workshop on Scientific Use of Submarine Cables and Related Technologies, 2003..