Computer vision is a field of study that covers a huge area starting from object detection, tracking, and recognition to human-computer interaction. The demand for computer vision is increasing day by day with applications to surveillance, video retrieval, and man-machine interaction. Human activity recognition has become an important topic in this field of research since it is the key to automatic video surveillance. In recent years, a large number of datasets, related to both sensors and videos, are created for human activity analysis. Most of these datasets are for single person activity recognition. In this research, we undertook even more challenging two-person interaction video datasets. We worked on two benchmark datasets of such research, the University of Texas at Austin (UTA) interaction dataset and the dataset of Stony Brook University. We have proposed a gradient-based technique for interaction recognition. The gradients at points within the region of interest of video frames are taken as features. In this process, we introduced a technique to find the region of interest based on moving object tracking. Variations in motion performance, inter-personal difference, and recording settings make the task extremely challenging. The proposed method yields a recognition rate of 68.33% for University of Texas Austin two-person interaction dataset and proves to be more efficient for indoor videos of the dataset of Stony Brook University with an accuracy of 74.40%.
With the increase in the sizes of the farms in Japan’s livestock industry, managing individual animals has become quite difficult. Under such circumstances, disease detection becomes complicated, and the delay of detection increases the spread and severity of the disease. The purpose of this study is to establish an early detection system to identify respiratory diseases in pigs using body-conducted sound (BCS). Biological information such as respiratory sounds and heartbeat is necessary to properly determine the occurrence of disease. Therefore, a bio-information monitoring method for the disease-detection systems, which extracts periodic components in body-conducted sounds using independent component analysis (ICA) and adaptive signal processing (ALE), is proposed. Furthermore, significant differences were found through the analysis of BCS acoustics before and after inoculations in zero crossing and mel-frequency cepstral coefficient acoustic features, which are features of BCS when a pig has a respiratory disease. Therefore, it is suggested that early detection of respiratory diseases can be achieved by assessing these acoustic features.
The active drainage method involves mobilizing the affected limb after total knee arthroplasty (TKA) to promote effective drainage of the blood within the knee. This study aimed to prospectively compare the active drainage method with the conventional drainage method, and to clarify its effects on drain management after TKA. Patients who underwent TKA were randomly assigned to the conventional drainage control group (conventional group) or the active drainage method group (active group). The volume of postoperative blood drainage was measured in each group. The swelling of the affected limb was determined by measuring the circumferential diameter of the limb preoperatively and on the second, third and fourth postoperative days. A significant increase in blood drainage was observed after TKA in the active group (555 ± 42 mL) compared with the conventional group (248 ± 25 mL) (p = 0.001). The limb circumference was significantly smaller in the active group than in the conventional group above the knee (45.2 ± 0.74 cm vs. 42.3 ± 0.6cm, p = 0.002), at the knee center (43.1 ± 0.62 cm vs. 40.3 ± 0.44 cm, p = 0.004), and below the knee (36.5 ± 0.5 cm vs. 35.0 ± 0.44 cm, p = 0.03), while the circumference decreased in the early postoperative days. The active drainage method reduced knee swelling in the early postoperative period via active mobilization of the affected limb, and did not increase the incidence of complications.
In order to diagnose osteoarthritis of the shoulder joint, the 3-D shape of the humerus provides the essential information. Also, the segmented region can be utilized for analyzing individual variety of the 3-D shape between normal and anomaly, etc. Since now, there is no study which automatically segments the humerus region from computed tomography (CT) images. U-Net is a fully-convolutional network architecture, and has been applied to some image segmentation problems. This research introduces U-Net architecture to automatically segment the humerus region in shoulder CT images. To validate the proposed method, it has been applied to 19 male subjects. The method achieved 0.946 Dice coefficient, which demonstrates that it successfully segmented the humerus region with a high level of accuracy and precision.