شناسايي نوع فعاليت انسان در دنباله اي از تصاوير ويدئويي با استفاده از توصيفگر بافت پويا

چكيده انگليسي :

Human Action Recognition Using Dynamic Texture Descriptor Shima Ghasemi s ghasemi@cc iut ac ir Date of submission Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan Iran Degree M Sc Language PersianSupervisor Dr Saeed Sadri s sadri@ec iut ac ir Dr Niloofar Gheissari n gheissari@ec iut ac ir AbstractHuman action recognition is a dynamic and challenging field of study in computer vision It refers to the taskof classifying human action based on a video sequence taken from a subject performing that action This hasattracted a lot of attention because of its wide applications in surveillance systems sport video analysis intelligent environment and robot guidance Cluttered background camera motion xerography view point occlusion low frame rate and low resolution are major challenges to the analysis of human actions Dynamic textures are temporally continuous and infinitely varying sequences of images with certain spatialand temporal stationarity properties They include sea waves smoke foliage whirlwind etc Human actioncan be considered as a type of dynamic texture since it has statistical variations in spatio temporal domain The local interest features contain efficient information of these spatiotemporal variations The proposedmethod is based on dynamic texture description for analysis of human motion using visual dictionary In thisthesis we adopt the idea of spatio temporal analysis with dynamic textures on local features For this target spatio temporal interest points are extracted according to Laptev strategy These are points at which asignificant change occurs in both space and time domains This means that features selected by laptevoperator not only undergo an intensity change but also they undergo a change in the magnitude of motionvelocity or the direction Then these interest points are described by a dynamic texture descriptor localbinary pattern on three orthogonal plane LBP TOP LBP TOP is an extension of basic LBP operator that isapplied on three orthogonal planes XY XT YT We apply LBP HF a novel rotation invariant imagedescriptor computed from discrete Fourier transforms of local binary pattern histograms on each plane Next we cluster the features with k means clustering algorithm so that each center of cluster is a candidate ofother members of that cluster In order to compact information and simply compare features to classifyactions we construct a visual dictionary The concept of visual dictionary is often used for imagesegmentation and retrieval Centers of the clusters are the words of our visual dictionary Each action isdescribed by a histogram according to the number of occurrences of the words of visual dictionary Thenumber of bins of the histogram is equal to the number of words of visual dictionary and each bin shows theoccurrences of one word of the dictionary in the samples of the action Finally test sequences are given tothe system We should construct the histogram describing test sequences according to what we haveconducted for train sequences Then they will be classified by two classification algorithms K nearestneighbor KNN and support vector machine SVM that are among the most popular and powerfulclassification algorithms in computer vision In order to verify the proposed method we applied theexperiment to KTH dataset that contains six different actions boxing hand clapping hand waving jogging running and walking This method can recognize the actions of this dataset with the mean accuracy of This is a reasonable performance among other competing methods Keywords human action recognition motion analysis dynamic texture spatio temporal features visualdictionary bag of words

قاسمي، شيما

شناسايي نوع فعاليت انسان در دنباله اي از تصاوير ويدئويي با استفاده از توصيفگر بافت پويا

تشخيص نوع فعاليت انسان , تحليل حركت , بافت ديناميك , ويژگي هاي زماني-مكاني , فرهنگ لغت , سبدي از كلمات