HUMAN ACTIVITIES RECOGNITION USING SHAPE MOMENTS AND HISTOGRAM OF NORMALIZED DISTANCES “HND”

This paper presents an algorithm for human activities recognition in videos based on a combination of two different feature types the first feature type concerns the shape and is called the shape moments. the second feature type concerns the contour boundary coordinates and the feature is called histogram of normalized distances from center of gravity of the object shape “cog” and it’s contour points “hnd”. combining these features leads to the formation of a strong complementary feature vector that captures effective discriminate details of human action videos. we use two classifiers; the first is multi-class support vector machine and the second is naïve bayes classifier. the recognition rate by using multi-class svm classifier is up to 95.6 % but by using naive bayes classifier is 97.2%.