• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.6
• /
• pp.748-753
• /
• 2012

In this paper, Face recognition algorithm is designed based on optimized pRBFNNs pattern classifier using three-dimensional scanner. Generally two-dimensional image-based face recognition system enables us to extract the facial features using gray-level of images. The environmental variation parameters such as natural sunlight, artificial light and face pose lead to the deterioration of the performance of the system. In this paper, the proposed face recognition algorithm is designed by using three-dimensional scanner to overcome the drawback of two-dimensional face recognition system. First face shape is scanned using three-dimensional scanner and then the pose of scanned face is converted to front image through pose compensation process. Secondly, data with face depth is extracted using point signature method. Finally, the recognition performance is confirmed by using the optimized pRBFNNs for solving high-dimensional pattern recognition problems.

• Journal of Korea Society of Industrial Information Systems
• /
• v.24 no.5
• /
• pp.17-25
• /
• 2019

In this paper, we introduce the technology to convert a single human image into a fashion show animation video clip. The technology can help the customers confirm the dynamic fitting result when combined with the virtual try on technique as well as the interesting experience to a normal person of being a fashion model. We developed an extended technique of full human 2D to 3D inverse modeling based on SMPLify human body inverse modeling technique, and a rigged model animation method. The 3D shape deformation of the full human from the body model was performed by 2 part deformation in the image domain and reconstruction using the estimated depth information. The quality of resultant animation videos are made to be publically available for evaluation. We consider it is a promising approach for commercial application when supplemented with the post - processing technology such as image segmentation technique, mapping technique and restoration technique of obscured area.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.370-370
• /
• 2017

최근 기후변화와 이상기후의 영향으로 인하여 홍수재해의 시 공간적 패턴은 보다 복잡해지고, 예측이 어려워지고 있다. 이러한 기상이변에 따른 홍수피해를 예방하기 위한 비구조적 대책으로 홍수위험등급 및 범람범위 등의 정보를 포함하고 있는 홍수위험지도의 작성이 필요하다. 실제로 고정밀도 홍수위험지도를 작성하기 위해서는 지형, 지질, 기상 등의 디지털 정보 및 사회 경제와 관련된 다양한 DB를 필요로 하며, 강우-유출-범람해석 모델링을 통해 범람면적 및 침수깊이 등의 정보를 획득하게 된다. 하지만 일부지역, 특히 개발도상국에서는 이러한 계측 홍수 데이터가 부족하거나 획득할 수가 없어 홍수위험지도 제작이 불가능하거나 그 정확도가 매우 낮은 실정이다. 따라서 본 연구에서는 ASTER 또는 SRTM과 같은 범용 DEM 등 지형자료만을 기반으로 한 선형이진분류법(Liner binary classifiers)과 ROC분석(Receiver Operation Characteristics)을 이용하여 미계측 유역 (DB부재 또는 부족으로 강우-유출-범람해석 모델링이 불가능한 북한지역)의 홍수위험지역을 탐지하고, 적용성을 평가하고자 한다. 5개의 단일 지형학적 지수와 6개의 복합 지형학적 지수를 이용하여 Area Under the Curve (AUC)를 계산하고, Sensitivity (민감도)와 Specificity (특이도)가 가장 높은 지수를 선별하여 홍수위험지도를 작성하고, 실제 홍수범람 영상(2007년 북한 함경남도지역 용흥강 홍수)과 비교 분석하였다. 본 연구에서 제시하는 선형이진분류법과 ROC분석 방법은 홍수범람해석을 위한 다양한 기초정보를 필요로 하지 않고, 지형정보만을 사용하기 때문에 관측 데이터가 없거나 부족한 지역에 대해서 우선적으로 홍수위험지역을 탐지하고, 선별하는데 유용할 것으로 판단된다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.31-31
• /
• 2016

국내 해안선을 따라 발생되는 침식 및 퇴적현상으로 인하여 해안구조물의 안정성과 같은 안전 문제가 발생되고 있다. 이러한 자연현상을 대비하여 연안지역 특성에 맞는 침식 저감 대책 마련에 대한 연구가 되고 있다. 침식현상은 다양한 외력조건과 지리적 특성을 함께 고려해야 한다. 고립파로 인한 침식 발생 현상에 대한 국외 연구들이 다수 수행되어 왔으며, 국내에는 이와 관련된 연구가 상대적으로 부족하다. 고립파는 파고의 특성에 따라 파형이 결정되는 특징을 가지고 있어, 침식현상과 지진해일의 실험적 연구에 있어 입사파로 주로 활용되고 있다. 기존의 방법과 동일하게 sluice gate를 순간적으로 개방하여 고립파를 재현하는 방식으로 파랑을 재현하였다. 본 연구에서는 폭 0.80 m, 높이 0.75 m, 수로연장 12.00 m의 강화유리로 설계된 직사각형 단면 수로에서 실험을 수행하였다. 측면에 두 대의 카메라(Model No. : Sony, HDR-XR550)를 고정설치하여 경사면 하상변화를 촬영하였고, 영상분석을 통하여 하상변동을 측정하였다. 유사특성에 따른 하상변동을 연구하기 위하여 경사면에 포설된 유사로는 주문진 표준사($d_{50}=0.6268mm$, $c_u=1.68$)와 안트라사이트($d_{50}=1.547mm$, $c_u=1.387$)를 동일한 실험조건에서 함께 사용하였다. 경사면에서 고립파의 처오름, 처내림을 추적하여 그 경향을 분석하였다. 하상의 침식 및 퇴적구간 길이, 침식, 퇴적의 높이 및 깊이와 같은 형상분석을 통한 영향인자를 함께 무차원화 하여 실험결과를 제시하였다. 본 연구의 결과를 바탕으로 Sluice gate를 이용한 고립파 수리실험으로 경사면 하상의 침식 및 퇴적 현상에 관한 기초적인 실험자료로 활용될 것이다.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.5
• /
• pp.221-228
• /
• 2022

With the application of artificial intelligence technology, robots can provide efficient services in real life. Unlike industrial manipulators that do simple repetitive work, this study presented design methods of 6 degree of freedom robot arm and intelligent object search and movement methods for use alone or in collaboration with no place restrictions in the service robot field and verified performance. Using a depth camera and deep learning in the ROS environment of the embedded board included in the robot arm, the robot arm detects objects and moves to the object area through inverse kinematics analysis. In addition, when contacting an object, it was possible to accurately hold and move the object through the analysis of the force sensor value. To verify the performance of the manufactured robot arm, experiments were conducted on accurate positioning of objects through deep learning and image processing, motor control, and object separation, and finally robot arm was tested to separate various cups commonly used in cafes to check whether they actually operate.

• The Journal of the Korea Contents Association
• /
• v.22 no.3
• /
• pp.430-441
• /
• 2022

With the development of ICT technology, various mobile applications have appeared one after another, and they have penetrated deeply into the lives of individuals. In order to explore the key factors that movie content-related applications should have through empirical studies, this study used the information system success model to investigate the influence of each quality factor on movie consumers' attitudes toward movies and their viewing intentions. As a result of the study, it was derived that the most important factor among the quality factors of movie content-related applications was the information quality factor, followed by the service quality being the next most important factor. In this study, based on the results of this study, the direction of application related to movie contents as a new distribution channel was proposed by suggesting a way to improve applications related to movie contents for the development of the domestic movie industry.

• The Journal of Korean Society for Radiation Therapy
• /
• v.25 no.1
• /
• pp.15-24
• /
• 2013

Purpose: In Asan Medical Center, Two parallel opposite beams are employed for total body irradiation. Patients are required to be in supine position where two arms are attached to mid axillary line. Normally, physical compensators are required to compensate the large dose difference for different parts of body due to the different thicknesses compared to the umbilicus separation. There was the maximum dose difference up to 30% in lung and chest wall compared to the prescription dose. In order to resolve the dose discrepancy occurring on different body regions, the feasibility of using Fieid-in-Field Technique is investigated in this study. Materials and Methods: CT scan was performed to The RANDO Phantom with fabricated two arms and sent to Eclipse treatment planning system (version 10.0, Varian, USA). Conventional plan with physical lead compensator and new plan using Field-in-Field Technique were established on TPS. AAA (Anisotropic Analytical Algorithm) dose calculation algorithm was employed for two parallel opposite beams attenuation. Results: The dose difference between two methods was compared with the prescription dose. The dose distribution of chest and anterior chest wall uncovered by patient arms was 114~124% for physical lead compensator while Field-in-Field Technique gave 106~107% of the dose distribution. In-vivo dosimetry result using TLD showed that the dose distribution to the same region was 110~117% for conventional physical compensator and 104~107% for Field-in-Field Technique. Conclusion: In this study, the feasibility of using FIF technique has been investigated with fabricated arms attached Rando phantom. The dose difference was up to 17% due to the attached arms. It is shown that the dose homogeneity is within ${\pm}10%$ with the CT based 3-dimensional 4 step FIF technique. The in-vivo dosimetry result using TLD was showed that 95~107% dose distribution compared to prescription dose. It is considered that CT based 3-dimensional Field-in-Field Technique for the total body irradiation gives much homogeneous dose distribution for different body parts than the conventional physical compensator method and might be useful to evaluate the dose on each part of patient body.

• Journal of radiological science and technology
• /
• v.34 no.1
• /
• pp.51-58
• /
• 2011

The aim of the study is to analyze the dose delivery error depending on the depth variation according to target positions and arc trajectories by comparing the simulated treatment planning with the actual dose delivery in conformal arc therapy. We simulated the conformal arc treatment planning with the three target positions (center, 2.5 cm, and 5 cm in the phantom). For the experiments, IMRT body phantom (I’mRT Phantom, Wellhofer Dosimetry, Germany) was used for treatment planning with CT (Computed Tomography, Light speed 16, GE, USA). The simulated treatment plans were established by three different target positions using treatment planning system (Eclipse, ver. 6.5, VMS, Palo Alto, USA). The radiochromic film (Gafchromic EBT2, ISP, Wayne, USA) and dose analysis software (OmniPro-IMRT, ver. 1.4, Wellhofer Dosimetry, Germany) were used for the measurement of the planned arc delivery using 6 MV photon beam from linear accelerator (CL21EX, VMS, Palo Alto, USA). Gamma index (DD: 3%, DTA: 2 mm) histogram and dose profile were evaluated for a quantitative analysis. The dose distributions surrounded by targets were also compared with each plans and measurements by conformity index (CI), and homogeneity index (HI). The area covered by 100% isodose line was compared to the whole target area. The results for the 5 cm-shifted target plan show that 23.8%, 35.6%, and 37% for multiple conformal arc therapy (MCAT), single conformal arc therapy (SCAT), and multiple static beam therapy, respectively. In the 2.5 cm-shifted target plan, it was shown that 61%, 21.5%, and 14.2%, while in case of center-located target, 70.5%, 14.1%, and 36.3% for MCAT, SCAT, and multiple static beam therapy, respectively. The values were resulted by most superior in the MCAT, except the case of the 5 cm-shifted target. In the analysis of gamma index histogram, it was resulted of 37.1, 27.3, 29.2 in the SCAT, while 9.2, 8.4, 10.3 in the MCAT, for the target positions of center, shifted 2.5 cm and 5 cm, respectively. The fail proportions of the SCAT were 2.8 to 4 times as compared to those of the MCAT. In conclusion, dose delivery error could be occurred depending on the target positions and arc trajectories. Hence, if the target were located in the biased position, the accurate dose delivery could be performed through the optimization of depth according to arc trajectory.

• Journal of the Korea Computer Graphics Society
• /
• v.14 no.3
• /
• pp.31-46
• /
• 2008

The depth of field is the range that the objects inside of this range treated to be focused. Objects that are placed out of this range are out of focus and become blurred. In computer graphics, generating depth of field effects gives a great reality to rendered images. The previous researches on the depth of field in computer graphics can be divided into two major categories. One of them is the distributed ray tracing that samples the lens area against each pixel. It is possible to obtain precise results without noise if enough number of samples are taken. However, to make a good result, a great number of samples are needed, resulting in an enormous timing requirement. The other approach is the method that approximates depth of field effect by post-processing an image and its depth values computed using a pin-hole camera. Though the second technique is not that physically correct like distributed ray tracing, many approaches which using this idea have been introduced because it is much faster than the first approach. But the post-processing have some limitations because of the lack of ray information. In this paper, we first present an improvement technique that corrects the previous post-processing methods and then propose another one that accelerates the distributed ray tracing by using a radiance caching method.

• Journal of KIISE:Software and Applications
• /
• v.35 no.10
• /
• pp.600-607
• /
• 2008

This paper proposes a robust human detection method irrespective of their pose variation using the multiple oriented 2D elliptical filters (MO2DEFs). The MO2DEFs can detect the humans regardless of their poses unlike existing object oriented scale adaptive filter (OOSAF). To overcome OOSAF's limitation, we introduce the MO2DEFs whose shapes look like the oriented ellipses. We perform human detection by applying four different 2D elliptical filters with specific orientations to the 2D spatial-depth histogram and then by taking the thresholds over the filtered histograms. In addition, we determine the human pose by using convolution results which are computed by using the MO2DEFs. We verify the human candidates by either detecting the face or matching head-shoulder shapes over the estimated rotation. The experimental results showed that the accuracy of pose angle estimation was about 88%, the human detection using the MO2DEFs outperformed that of using the OOSAF by $15{\sim}20%$ especially in case of the posed human.