We developed the system that doctors can prevent forgotten medication of patients at any time to check the medication status. This system consists of smartphone application on Android and cloud application on Google App Engine. Smartphone application notifies the time of taking medicine by alarms. The patients can record the medication information by pressing the button. Medication information is sent to the cloud application automatically. The doctors are able to check the medication information at any time using cloud apps in the browser.
This paper proposes a direct communication-based information sharing system that is supposed to be used for disaster evacuation. Although it is important to quickly grasp the situation when disaster occurred, there are several problems that block efficient sharing and aggregation of information found by evacuees, such as difficulty of operating mobile devices during evacuation, and failure of global network. In order to tackle those problems, the proposed method employs direct communication based on Bluetooth device search of smart-phones, which can be used without global network. Furthermore, the proposed method automatically selects information to be sent to other smart-phones for realizing information sharing without manual operation. This paper shows the prototype system and simulation result regarding characteristics of automatic information selection strategies.
When using video chat, you may send private information to other party with background information (e.g, inside office) of the video that is not processing. To cope with this problem, it is effective to replace background information excluding an object with any image according to the purpose. In addition, the background replacement that replaces video background with a landscape image can do video chat with “Feeling like you are at the location shown in the image”. However, if it applies the background replacement to the video chat, uncomfortable feeling will be caused by the chroma difference between a human region of camera image and a new background image (e.g, a landscape image). This chroma difference can be caused by the difference of color temperature of the light source when capturing images. In this paper, we proposed a method of the uncomfortable feeling reduction of background replacement on the basis of features of color temperature. The algorithm comprises three steps. First, it initialized parameters of the adaptive background subtraction, which is a method to extract human region from camera image as a pre-process phase. It also generated a matrix for the chromatic adaptation transform (CAT) with the color temperature estimated from the camera image and a new background image. Next, the human region was extracted from the camera image by the background subtraction, and it applied the CAT to this region. Finally, it extracted the background region from the new background image using the background subtraction result. Moreover, it obtained the background replacement results as a logical sum of the result by the CAT and the background region. The results of experiment conducted with 14 evaluators for analyzing images suggest that the proposed method can reduce uncomfortable feeling of background replacement.
We developed an application that uses sensors in a smartphone to estimate a user's chest compression motion in performing cardiopulmonary resuscitation (CPR). In our system, the smartphone measures the motion of chest compression with accelerometer sensors and provides a visual, auditory, and tactile feedback to conduct proper chest compression CPR. We evaluated our estimation algorithms while conducting chest compressions on a mannequin and compared the error between the estimated position and ground truth position. Furthermore, we compared the efficacy of sensory feedback (auditory only, tactile only, and audiotactile) to instruct the proper chest compression rate using the smartphone application.