Human body contour data based activity recognition

Abstract

This research work is aimed to develop autonomous bio-monitoring mobile robots, which are capable of tracking and measuring patients' motions, recognizing the patients' behavior based on observation data, and providing calling for medical personnel in emergency situations in home environment. The robots to be developed will bring about cost-effective, safe and easier at-home rehabilitation to most motor-function impaired patients (MIPs). In our previous research, a full framework was established towards this research goal. In this research, we aimed at improving the human activity recognition by using contour data of the tracked human subject extracted from the depth images as the signal source, instead of the lower limb joint angle data used in the previous research, which are more likely to be affected by the motion of the robot and human subjects. Several geometric parameters, such as, the ratio of height to weight of the tracked human subject, and distance (pixels) between centroid points of upper and lower parts of human body, were calculated from the contour data, and used as the features for the activity recognition. A Hidden Markov Model (HMM) is employed to classify different human activities from the features. Experimental results showed that the human activity recognition could be achieved with a high correct rate.

Development of Robust Behaviour Recognition for an at-Home Biomonitoring Robot with Assistance of Subject Localization and Enhanced Visual Tracking

“…At-home biomonitoring systems have become an important solution in the medical field as an assistive technology for the people who have difficulties leaving their houses, such as elderly people or motor-function impaired persons (MIPs) [ 1 – 8 ]. Benefits of these systems include more convenient and comfortable ways of healthcare for patients and reduction in the workload of the therapists [ 5 – 8 ].…”

“…As proposed in our previous works [ 5 , 8 , 19 , 20 ], our main idea is to use assistive robotics to support elderly or motor-function impaired persons biomonitoring. Although a mobile system brings more complexity to handle daily observation tasks, a mobile robot solution includes several advantages: (i) it is a cheaper smart system than smart houses, (ii) it could make full use of the capability of sensors, since the sensors could be brought to a position and an angle optimal or suboptimal to observation, (iii) it avoids the multiple vision modules by using a mobile monitoring system, and (iv) it prevents the need of the subjects for wearing any sensors [ 5 , 8 , 19 , 20 ] compared to wearable sensor applications. These advantages make the mobile robot a good candidate for assistive technologies.…”

“…Therefore, in our previous research [ 5 , 8 , 19 , 20 ], we presented an at-home biomonitoring mobile robot project to improve the quality of life of motor-function impaired persons (MIPs). The robot observes the subject and recognizes the activity he is performing.…”

“…The robot observes the subject and recognizes the activity he is performing. It is also able to provide analysis of walking pattern if the lower limb joint trajectories were traceable [ 5 , 8 ]. To achieve these tasks, there are several modules that have to be developed.…”

Vision-based Recognition of Activities by a Humanoid Robot

“…One of our main contributions is to show how optimizing vision algorithms and robot settings can help ensure a good recognition performance. It is worth noting that most current robot-human activity or gesture interaction systems are based on a Kinect camera [5][6][7][8]. As a result, these robots need the Kinect sensor for data acquisition, as well as a computer to process the skeleton and RGBD raw data on one hand, and to serve as an interface between the Kinect and the robot on the other.…”