• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1295-1301
• /
• 2019

In this paper, we propose a novel method of denoising human motion using a bidirectional recurrent neural network (BRNN) with an attention mechanism. The corrupted motion captured from a single 3D depth sensor camera is automatically fixed in the well-established smooth motion manifold. Incorporating an attention mechanism into BRNN achieves better optimization results and higher accuracy than other deep learning frameworks because a higher weight value is selectively given to a more important input pose at a specific frame for encoding the input motion. Experimental results show that our approach effectively handles various types of motion and noise, and we believe that our method can sufficiently be used in motion capture applications as a post-processing step after capturing human motion.

• Journal of Broadcast Engineering
• /
• v.13 no.6
• /
• pp.859-874
• /
• 2008

In this paper, we develop a fast view synthesis method that generates multiple intermediate views in real-time for the 3D display system when the camera geometry and depth map of reference views are given in advance. The proposed method achieves faster view synthesis than previous approaches in GPU by processing in parallel the entire computations required for the view synthesis. Specifically, we use $CUDA^{TM}$ (by NVIDIA) to control GPU device. For increasing the processing speed, we adapted all the processes for the view synthesis to single instruction multiple data (SIMD) structure that is a main feature of CUDA, maximized the use of the high-speed memories on GPU device, and optimized the implementation. As a result, we could synthesize 9 intermediate view images with the size of 720 by 480 pixels within 0.128 second.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.764-767
• /
• 2003

The characteristics of diesel spray have much effect on the engine performances such as power. fuel consumption rate and emissions. Therefore, the measurement of fuel spray characteristics is very important for the improvement of heat engine. The factors which control diesel spray characteristics are injection pressure, ambient temperature and density etc. Spray behaviors are visualized by using the high speed video camera and spray angle, spray penetration are measured. Experimental equations of spray penetration and spray angle were derived by using the experimental results. 1) Ambient temperature and density influence on the characteristics of diesel spray. 2) Experimental equation of spray penetration is expressed as follows 0＜t＜ $t_{b}$ ; $S_1$=11.628$\Delta$ $P^{0.485}$ $\rho$$_{a}$ $^{-0.478}$ $t^{1.337}$, $t_{b}$ ＜t; $S_2$=7.457$\Delta$ $P^{0.523}$ $\rho$$_{a}$ $^{-0.382}$ $t^{0.548}$ 3) Experimental equation of spray Angie is expressed as follows $T_{a}$ =293K; Tan($\theta$/2)=059($\rho$$_{a}$ / $\rho$$_{f}$ )$^{0.437}$, $T_{a}$ =473K; Tan($\theta$/2)=0588($\rho$$_{a}$ / $\rho$$_{f}$ )$^{0.404}$_{f}$ )$^{0.404}$

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.29 no.5 s.236
• /
• pp.643-652
• /
• 2005

Saturated nucleate pool boiling experiments for binary mixtures, which are consisted of refrigerant R11 and R113, were performed with constant wall temperature condition. Results for binary mixtures were also compared with pure fluids. A microscale heater array and Wheatstone bridge circuits were used to maintain the constant temperature of the heating surface and to obtain heat flow rate measurements with high temporal and spatial resolutions. Bubble growth images were captured using a high speed CCD camera synchronized with the heat flow rate measurements. The departure time for binary mixtures was longer than that for pure fluids, and binary mixtures had a higher onset of nucleate boiling (ONB) temperature than pure fluids. In the asymptotic growth region, the bubble growth rate was proportional to a value between $t^{\frac{1}{6}}$ and $t^{\frac{1}{4}}$. The bubble growth behavior was analyzed to permit comparisons with binary mixtures and pure fluids at the same scale using dimensionless parameters. There was no discernable difference in the bubble growth behavior between binary mixtures and pure fluids for a given ONB temperature. And the departure radius and time were well predicted within a ${\pm}30{\%}$ error. The minimum heat transfer coefficient of binary mixtures occurred near the maximum ${\mid}y-x{\mid}$ value, and the average required heat flux during bubble growth did not depend on the mass fraction of R11 as more volatile component in binary mixtures. Finally, the results showed that for binary mixtures, a higher ONB temperature had the greatest effect on reducing the heat transfer coefficient.

• Safety and Health at Work
• /
• v.8 no.2
• /
• pp.151-155
• /
• 2017

Background: At the entrance of a tunnel, reflection of sunlight from the surrounding environment and a lack of adequate lighting usually cause some vision problems. The purpose of this study was to perform a safety evaluation of lighting on a very long road in Ilam, Iran. Methods: The average luminance was measured using a luminance meter (model S3; Hagner, Solna, Sweden). A camera (model 108, 35-mm single-lens reflex; Yashica, Nagano, Japan) was used to take photographs of the safe stopping distance from the tunnel entrance. Equivalent luminance was determined according to the Holliday polar diagram. Results: Considering the average luminance at the tunnel entrance ($116.7cd/m^2$) and using Adrian's equation, the safe level of lighting at the entrance of the tunnel was determined to be 0.7. Conclusion: A comparison between the results of the safe levels of lighting at the entrance of the tunnel and the De Boer scale showed that the phenomenon of black holes is created at the tunnel entrance. This may lead to a misadaptation of the drivers' eyes to the change in luminance level at the entrance of the tunnel, thereby increasing the risk of road accidents in this zone.

• Journal of Intelligence and Information Systems
• /
• v.14 no.4
• /
• pp.89-101
• /
• 2008

The most used tool for quality control is control chart in manufacturing industry. But it has limitations at current situation where most of manufacturing facilities are automated and several manufacturing processes have interdependent relationship such as CCM assembly line. To Solve problems, we propose quality management system based on data mining that are consisted of monitoring system where it monitors flows of processes at single window and feature extraction system where it predicts the yield of final product and identifies which processes have impact on the quality of final product. The quality management system uses decision tree, neural network, self-organizing map for data mining. We hope that the proposed system can help manufacturing process to produce stable quality of products and provides engineers useful information such as the predicted yield for current status, identification of causal processes for lots of abnormality.

• The Journal of the Korea Contents Association
• /
• v.13 no.11
• /
• pp.28-36
• /
• 2013

This study proposed a robust detection algorithm. It detects hands more stably with respect to changes in light and rotation for the identification of a hand shape. Also it satisfies both efficiency of calculation and the function of detection. The algorithm proposed segmented the hand area through pre-processing using a hand shape as input information in an environment with a single camera and then identified the shape using a Self Organized Feature Map(SOFM). However, as it is not easy to exactly recognize a hand area which is sensitive to light, it has a large degree of freedom, and there is a large error bound, to enhance the identification rate, rotation information on the hand shape was made into a database and then a principal component analysis was conducted. Also, as there were fewer calculations due to the fewer dimensions, the time for real-time identification could be decreased.

• Journal of the Institute of Convergence Signal Processing
• /
• v.15 no.2
• /
• pp.42-47
• /
• 2014

A finger-gesture game controller using the single camera is implemented in this paper, which is based on the recognition of the number of fingers and the index finger moving direction. Proposed method uses the CAMShift algorithm to trace the end-point of index finger effectively. The number of finger is recognized by using a double circle tracing method. Then, HSI color mode transformation is performed for the CAMShift algorithm, and YCbCr color model is used in the double circle tracing method. Also, all processing tasks are implemented by using the Intel OpenCV library and C++ language. In order to evaluate the performance of the proposed method, we developed a shooting game simulator and validated the proposed method. The proposed method showed the average recognition ratio of more than 90% for each of the game command-mode.

• The Journal of the Korea Contents Association
• /
• v.8 no.11
• /
• pp.85-93
• /
• 2008

For music score recognition, several approaches have been proposed including shape matching, statistical methods, neural network based methods and structural methods. In this paper, we deal with recognition for low resolution images which are captured by the digital camera of a mobile phone. Considerable distortions are included in these low resolution images, so when existing technology is used, many problems appear. First, captured images are not stable in the sense that they contain lots of distortions or non-uniform illumination changes. Therefore, notes or symbols in the music score are damaged and recognition process gets difficult. This paper presents recognition technology to overcome these problems. First, musical note to head, stick, tail part are separated. Then template matching on head part of musical note, and remainder part is applied. Experimental results show nearly 100% recognition rate for music scores with single musical notes.

• International Journal of Highway Engineering
• /
• v.10 no.3
• /
• pp.19-31
• /
• 2008

Video image processing systems(VIPS) offer numerous benefits to transportation models and applications, due to their ability to monitor traffic in real time. VIPS based on wide-area detection, i.e., multi-lane surveillance algorithm provide traffic parameters with single camera such as flow and velocity, as well as occupancy and density. However, most current commercial VIPS utilize a tripwire detection algorithm that examines image intensity changes in the detection regions to indicate vehicle presence and passage, i.e., they do not identify individual vehicles as unique targets. If VIPS are developed to track individual vehicles and thus trace vehicle trajectories, many existing transportation models will benefit from more detailed information of individual vehicles. Furthermore, additional information obtained from the vehicle trajectories will improve incident detection by identifying lane change maneuvers and acceleration/deceleration patterns. The objective of this research was to relate traffic safety to VIPS tracking and this paper has developed a computer vision system of monitoring individual vehicle trajectories based on image processing, and offer the detailed information, for example, volumes, speed, and occupancy rate as well as traffic information via tripwire image detectors. Also the developed system has been verified by comparing with commercial VIP detectors.