Analysis of attribute domain for geometrical gesture performed by arm movements

Recently, the ability of the recognition systems to recognize gestures produced by human increases because of the advancement of the decision-making algorithm or the classifier. However sometimes the systems miss recognize the unknown gesture because the decision-making algorithm is in chaotic. The situation happens when the unknown gesture brings high similarity values compared with gestures stored in the database. Therefore, by adding a knowledge context of the gesture could reduce the chaotic and the systems output precise decision. The study proposes to employ the concept of ontology in the recognition of the arm gesture; however, this paper discusses the development of the attribute domain of the arm gesture. The attribute domain is defined as a stage where all the characteristic arm movement are presented. The characteristic arm movements are the features of the geometrical gesture. All the features are stored as the elements of the attribute domain. The relations among the elements must be formed and analyzed. Two analyses are conducted, which are the statistical and the precision measures. The statistical results have statistically significant to classify 9 registered gesture. Moreover, the precision measure is successfully removing the common data for all 9 gestures.