Recently, many studies show that an indoor horse riding exercise has a positive effect on promoting health and diet. However, if a rider has an incorrect posture, it will be the cause of back pain. In spite of this problem, there is only few research on analyzing rider’s posture. Therefore, the purpose of this study is to estimate a rider pose from a depth image using the Asus’s Xtion sensor in real time. In the experiments, we show the performance of our pose estimation algorithm in order to comparing the results between our joint estimation algorithm and ground truth data.
We propose a novel face detection algorithm in order to improve higher detection rate of face-detector than conventional haar - adaboost face detector. Our purposed method not only improves detection rate of a face but decreases the number of false-positive component. In order to get improved detection ability, we merged two classifiers: adaboost and support vector machine. Because SVM and Adaboost use different feature, they are complementary each other. We can get 2~4% improved performance using proposed method than previous our detector that is not applied proposed method. This method makes improved detector that shows better performance without algorithm replacement.
In this paper, we will briefly outline our contributions for the physical realization of coded OTDR, along with its
principles and also highlight recent key results related with its applications. For the communication network application,
we report a multi-port / multi-wavelength, high-speed supervisory system for the in-service monitoring of a bidirectional
WDM-PON system transmission line up to 16 ports x 32 nodes (512 users) capacity. Monitoring of individual branch
traces up to 60 km was achieved with the application of a 127-bit simplex code, corresponding to a 7.5dB SNR coding
gain effectively reducing the measurement time about 30 times when compared to conventional average mode OTDR.
Transmission experiments showed negligible penalty from the monitoring system to the transmission signal quality, at a
2.5Gbps / 125Mbps (down / up stream) data rate. As an application to sensor network, a Raman scattering based coded-OTDR distributed temperature sensor system will be presented. Utilizing a 255-bit Simplex coded OTDR together with
optimized sensing link (composed of cascaded fibers with different Raman coefficients), significant enhancement in the
interrogation distance (19.5km from coding gain, and 9.6km from link-combination optimization) was achieved to result
a total sensing range of 37km (at 17m/3K spatial/temperature resolution), employing a conventional off-shelf low power
(80mW) laser diode.