CLASSIFICATION OF MOTIONS IN A PLANE
Applying a technique from coding theory we develop a classification of motions of a point in a plane. Not only the common shifts but also both vibemtions and turns in the movements are consided in our classification. Using vectors with four components to express the code words for each movement. Then we also standardize each componets into 0 or 1 which leads to 16 standard states in order to simplify the analysis.Based on the standard code words we can describe the whole movement in the form of the sequence of the standard states. Furthermore we give one method to measure the similarity of two movements based on our tecknique of classification, which can be used to determine the type of any given movement and it is proved to distinquish the different movements quite well. Then these encodings are tested on data in the form of measurements of body motions obtained from video recordings. First the encodings from manual and automatic measurements are compared, and the result shows us the reliability of the automatic measuremetns. The recordings of certain types of movements during different activities are investigated to see whether they can be used for classifying the activities. Because of the difficulty to decide when an activity begins and stops during a movement, finally a measure called agitation, which can be seen as an indicator for a change in activity, is introduced.
Classification Cding Curves Movement Agitation
XIN WANG CORNELIS HOEDE XIAODONG LIU
Department of Mathematics, Dalian Maritime University, Dalian 116026, PR China Department of Applied Mathematics, University of Twente, P.O.Box 217, 7500 AE, Enschede, The Netherl Research Center of Information and Control, Dalian University of Technology, Dalian 116024, PR China
国际会议
The Second International Conference on Information & Systems Sciences(ICISS2008)(第二届信息与系统科学国际会议)
大连
英文
1092-1108
2008-12-18（万方平台首次上网日期，不代表论文的发表时间）