Abstract: This paper proposes a new position detector using frosted glass and four photo detectors. Four photo detectors are arranged under the four corners of the rough surface of frosted glass. When this sensor is irradiated by laser pointer, the light repeats reflection in the frosted glass and arrives at the photo detectors. The light intensity decreases by the increasing of distance between the photo detector and the irradiating point. Therefore, the distance between a photo detector and the irradiation point can be worked out from the output of the photo detectors. Thus, the light irradiation point can be derived from the distances with the four photo detectors. This sensor is low cost and has a large sensing area. In addition, this sensor also can be used as a pointing device because the movement of the light on the frosted glass can be judged.
Abstract: To prevent people to forget to take medicines, we propose a new monitoring system to prevent forgetting to take medicines on time. In this system, we develop a funnel-shaped sensor consisting of a holder with an LED below the lid of the holder and a photo detector at the bottom to detect medicines. When medicines are dropped into the sensor, they fall down to the bottom of the sensor and the photo detector receives only part light of the LED. According to the value of the output of the photo detector, it can be easily to judge medicines with a proper threshold value. Highly accurate monitoring system to prevent forgetting to take medicines can be produced by using this sensor. If medicines are not taken at time to take medicines, the system announces to the user.
Abstract: This paper proposes a new non-contact type chewing sensor using photo-reflector. The modulated infrared rays from the infrared LED of photo-reflector irradiate the mandible. The reflected light is received by the photo detector of photo-reflector. The chewing sensor recognizes chewing from the big movement of the mandible by chewing. By wearing the sensor in the ear, this sensor can be used in any place. The unpleasantness when using the chewing sensor is a little because the chewing sensor is non-contact type. In addition, the chewing sensor can accurately judge whether the user is chewing or not.
Abstract: A measurement method of heart rate is suggested, that is able to calculate a heart rate from a heartbeat sound which is measured by a body conduction sound sensor. The heart rate is obtained to calculate the autocorrelation of the sound. The effectiveness of this method is supported by comparison of values measured by a blood pressure monitor and a body conduction sound sensor. When we make a measurement system using a microcomputer, it could not measure heartbeat sounds at a high frequency because of performance of the hardware. In case of lower sampling frequency is checked and, it is shown that the correct heart rates are measured in the case of a sampling frequency is hundreds of hertz in the method.
Abstract: A privacy-preserving sensor for person localization has been developed. In theory, the sensor can be constructed with a line sensor and cylindrical lens because only a one-dimensional brightness distribution is needed. However, a line sensor is expensive. In contrast, CMOS area sensors are low cost and are increasing in sensitivity according to recent rapid advancement in the technology. Therefore, we covered the CMOS area sensor physically so that it behaved as a line sensor, we substituted CMOS sensors for the line sensors in practice. The proposed sensor obtains a one-dimensional horizontal brightness distribution that is approximately equal to the integration value of each vertical pixel line of the two-dimensional image. It is impossible to restore the two-dimensional detail texture image from one-dimensional brightness distribution, although it obtains enough information to detect a person’s position and movement status. Thus, the privacy is protected. Moreover, the appearance of the proposed sensor is very different from the conventional video camera, so the psychological resistance of having a picture taken is reduced. In this work, we made the privacy preserving sensor practically, and verified whether a person’s state was able to be detected. The simulation results show that the proposed sensor can detect a present person’s state responsively without violating privacy.
Abstract: We propose a new method for a detection of a 3D sphere. This method uses polytope method, a kind of minimization algorithm. One-dimensional histogram is used to detect a 3D sphere. The histogram has two characteristics. (1) The distribution of the histogram changes if the parameters of representing the 3D sphere changes. (2) The value of highest frequency of histogram becomes maximum if the best parameters are obtained. Therefore, the maximum value of highest frequency of histogram is searched to obtain the best parameters of a 3D sphere by using polytope method. By using polytope method, proposed method can detect a 3D sphere from 3D vertex data including other shapes without a large memory space and a lot of processing time.
Abstract: The objects have potentially dangerousness for pedestrians, such as illegally-parked bicy-cles, power poles, stopping vehicles, etc., could be found in a street. Those objects are threats to thesafety of elderly and disabled people with impaired vision, because it is difficult for them to find thoseobjects. In this paper, to enhance the elderly and disabled people's mobility, we propose the methodto estimate the region which is feasible for people's mobility in life space. In this method, the imagesare obtained from a head-mounted wearable camera, and the suited region for walking is estimatedusing 3-dimensional analysis and genetic algorithm. The performance and the validity are shown byapplying the proposed method to a number of scenes in life space.
Abstract: In this paper, a method to extract building wall textures from an on-vehicle camera imagesis proposed as an aid to construct the 3D maps. The building wall textures are required to attach to 3Dpolygons which are obtained through 3D measurements in urban space.We assume that the on-vehiclecamera is under linear uniform motion and building walls are planar regions perpendicular to the opti-cal axis. Under the assumption, the same building wall region has the same depth, or disparities, overthe region among successive images. Since disparities derived from foreground objects are differentfrom the disparities derived from the building wall, we can use the disparity differences as a clue toeffectively distinguish the building walls from the foreground objects. We formulate the extractionof building wall textures incorporating these consideration as an optimization problem which can besolved by graph-cut algorithm. To show the effectiveness of the proposed method, it is applied to somesequential scenes.
Abstract: Character recognition is a classical issue which has been devoted by a lot of researchers.Making character recognition system more widely available in natural scene images might open upinteresting possibility to use as an input interface of characters and an annotation method for images.Nevertheless, it is still difficult to recognize all sorts of fonts including decorated characters such ascharacters depicted on signboards. The decorated characters are constructed by using some specialtechniques for attracting viewers' attentions. Therefore, it is hard to obtain good recognition results bythe existingOCRs. In this paper,we propose a newcharacter recognition systemusing SOM. The SOMis employed to extract an essential structure concerning the topology from a character. The extractedtopological structure from each character is used to matching and the recognition is performed on thebasis of the topological matching. Experimental results show the effectiveness of the proposed methodin most forms of characters.
Abstract: In this paper, a method which specifies the signboard region and extracts the charactersinside the signboard is proposed.We usually take notes not to forget what we should leave to memory.But it is often that the task is too troublesome. Our aim is the development of a new input-interface soas to input texts froma picture.Most of signboards are composed of almostmonochromatic region. Onthe basis of this observation, image segmentation using color information is applied, and then we getsome binary images by applying threshold for each segmented region. Each binary image is enclosedby the smallest circumscribed square. The signboard region is specified according to distribution andarea of the white pixels inside the square. As a result of experiment, we confirmed the effectivenessof the proposed method.