The elderly population is expected to more than triple by 2050 in the United States alone. This indicates the growing need of medical innovations that are expected to deliver convenient, user-friendly, and intelligent health care in the home. In particular, the use of computer vision and artificial intelligence offers a new promising solution to analyze human behavior and detect unusual events. We propose a novel method to detect unintentional falls, which are one of the greatest risks for seniors living alone. Our approach is based on a machine learning technique, in which we use a human skeletal joint data from a Red Green Blue Depth (RGBD) sensor, Kinect. To the best of our knowledge, this is the first fall detection research utilizing machine learning mechanism with joint information. After building a solid definition of various types of falls, the system is trained using Support Vector Machine (SVM) in two separate layers, which we call Double Layer SVM (DLSVM). Our evaluation results show that our proposed system can efficiently detect different types of falls facing various directions from the camera and is capable of accurately distinguishing an actual fall versus a fall-like behavior. ⓒ 2013 DGIST
Table Of Contents
Ⅰ. INTRODUCTION 10 -- Ⅱ. BACKGROUND – CURRENT FALL DETECTION APPROACHES 11 -- 1. Accelerometer 11 -- 2. Floor Vibration 12 -- 3. Video Based Fall Detection 12 -- III. SYSTEM OVERVIEW – INTELLIGENT FALL DETECTION SYSTEM 13 -- 1. Kinect 14 -- 2. Architecture 14 -- IV. FALL BEHAVIOR 16 -- 1. Types of Falls 16 -- 2. Skeletal Joints 16 -- 3. Body Shape Features and Motion Sequence 17 -- 3.1 Body Shape Features 17 -- 3.2 Motion Sequence 18 -- V. BEHAVIOR CLASSIFCATION BY MACHINE LEARNING ALGORITHM 18 -- 1. Support Vector Machine 18 -- 2. Double Layer Support Vector Machine for Fall Detection 20 -- VI. EVALUATION 24 -- 1. Visualization of Joint Data 24 -- 2. Accuracy of First Layer of DLSVM 24 -- 3. Performance Evaluation 25 -- VII. RELATED WORK 28 -- VIII. CONCLUSIONS AND FUTURE WORK 31 -- REFERENCES 32 -- 요약문 35 -- Acknowledgement 36 -- Curriculum Vitae 37