Modeling Human Activity From Voxel Person Using Fuzzy Logic

As part of an interdisciplinary collaboration on elder-care monitoring, a sensor suite for the home has been augmented with video cameras. Multiple cameras are used to view the same environment and the world is quantized into nonoverlapping volume elements (voxels). Through the use of silhouettes, a privacy protected image representation of the human acquired from multiple cameras, a 3-D representation of the human is built in real time, called voxel person. Features are extracted from voxel person and fuzzy logic is used to reason about the membership degree of a predetermined number of states at each frame. Fuzzy logic enables human activity, which is inherently fuzzy and case-based, to be reliably modeled. Membership values provide the foundation for rejecting unknown activities, something that nearly all current approaches are insufficient in doing. We discuss temporal fuzzy confidence curves for the common elderly abnormal activity of falling. The automated system is also compared to a ground truth acquired by a human. The proposed soft computing activity analysis framework is extremely flexible. Rules can be modified, added, or removed, allowing per-resident customization based on knowledge about their cognitive and functionality ability. To the best of our knowledge, this is a new application of fuzzy logic in a novel approach to modeling and monitoring human activity, in particular, the well-being of an elderly resident, from video.

I. Introduction

Human activity analysis is an open problem that has been studied intensely within the areas of video surveillance [1], [2], homeland security [3], [4], and more recently, eldercare [5], [6], to name a few. Our goal is the continuous monitoring of human activity for the assessment of the “well-being” of a resident and the detection of abnormal or dangerous events, such as falls. It is important that both theories and realistic technologies be developed for recognizing elderly activity, and that they do so in a noninvasive fashion. Video sensors are a rich source of information that can be used to monitor a scene, but privacy is always a concern. To preserve privacy, segmentation of the human from an image results in a silhouette, a binary map that distinguishes the individual from the background. The raw video is not stored and only silhouettes are used to track the individual’s activity.