Activities of daily living of the elderly is often monitored using passive sensor networks. With the reduction of camera prices, there is a growing interest of video-based approaches to provide a smart, safe and independent living environment for the elderly. In this paper, activity level in context of tracking the movement pattern of an individual as a metric to monitor the daily living of the elderly is explored. Activity levels can be an effective indicator that would denote the amount of busyness of an individual by modelling motion features over time. The novel framework uses two different variants of the motion features captured from two camera angles and classifies them into different activity levels using neural networks. A new dataset for assisted living research called the Sheffield Activities of Daily Living (SADL) dataset is used where each activity is simulated by 6 subjects and is captured under two different illumination conditions within a simulated assisted living environment. The experiments show that the overall detection rate using a single camera setup and a dual camera setup is above 80%.

No abstract

No abstract

Abstract-The mobile market has taken huge leap in the last two decades, re-defining the rules of communication, networking, socializing and transactions among individuals and organizations. Authentication based on verification of signature on mobile devices, is slowly gaining popularity. Most online signature verification algorithms focus on computing the global Equal Error Rate across all users for a dataset. In this work, contrary to such a representation, it is shown that there are user-specific differences on the combined features and user-specific differences on each feature of the Equal Error Rate(EER) values. The experiments to test the hypothesis is carried out on the two publicly available dataset using the dynamic time warping algorithm. From the experiments, it is observed that for the MCYT-100 dataset, which yields an overall EER of 0.08, the range of user-specific EER is between 0 and 0.27. I. INTRODUCTIONWith the rapid development of touch screen capabilities on mobile devices, virtual transactions and the need to have secure biometric authentication services have also increased. Along with facial recognition, voice recognition, one of the biometric modality that is gaining popularity recently is online signature verification. Signatures have been used to authenticate individuals for over centuries and is a legally accepted biometric trait for authenticating an individual. The fundamental advantage of the signature modality over other biometric modalities (like face, fingerprints etc.) is that it gives the user the control to change their signature in an event of a predicted impending security attack. Signatures are personalized gesture patterns within a finite space. There are two traditional approaches to signature verification: online and offline systems. In an offline system, the image attributes are matched for authenticating an individual [1] while in an online system, dynamic features like the x or y coordinates, pressure, azimuth etc. are used for verification purposes.In this work, it is shown through experiments that different dynamic attributes of a signature contributes differently for every user. The rest of paper is arranged as follows: Section II details the background of online signature verification and the dynamic time warping (DTW) algorithm, Section III methodology of the experiments, Section IV discusses the results of the experiments and Section V consist the conclusions.

No abstract