Day or Night Activity Recognition From Video Using Fuzzy Clustering Techniques

Abstract—We present an approach for activity state recognition implemented on data collected from various sensors—standard web cameras under normal illumination, web cameras using in-frared lighting, and the inexpensive Microsoft Kinect camera sys-tem. Sensors such as the Kinect ensure that activity segmentation is possible during the daytime as well as night. This is especially useful for activity monitoring of older adults since falls are more prevalent at night than during the day. This paper is an application of fuzzy set techniques to a new domain. The approach described herein is capable of accurately detecting several different activity states related to fall detection and fall risk assessment including sitting, being upright, and being on the floor to ensure that elderly residents get the help they need quickly in case of emergencies and ultimately to help prevent such emergencies. Index Terms—Activity labeling, depth images, fuzzy clustering, image moments, infrared images. I