• 한국컴퓨터정보학회논문지
• /
• 제28권3호
• /
• pp.25-33
• /
• 2023

CT 영상의 획득 및 전송 등의 과정에서 발생하는 잡음은 영상의 질을 저하시키는 요소로 작용한다. 따라서 이를 해결하기 위한 잡음제거는 영상처리에서 중요한 전처리 과정이다. 본 논문에서는 딥러닝의 convolutional autoencoder (CAE) 모형에서 기존 컨볼루션 연산 대신 deformable 컨볼루션 연산을 적용한 deformable convolutional autoencoder (DeCAE) 모형을 이용하여 잡음을 제거하고자 한다. 여기서 deformable 컨볼루션 연산은 기존 컨볼루션 연산보다 유연한 영역에서 영상의 특징들을 추출할 수 있다. 제안된 DeCAE 모형은 기존 CAE 모형과 같은 인코더-디코더 구조로 되어있으나 효율적인 잡음제거를 위해 인코더는 deformable 컨볼루션 층으로 구성하고, 디코더는 기존 컨볼루션 층으로 구성하였다. 본 논문에서 제안된 DeCAE 모형의 성능 평가를 위해 다양한 잡음, 즉, 가우시안 잡음, 임펄스 잡음 그리고 포아송 잡음에 의해 훼손된 CT 영상을 대상으로 실험하였다. 성능 실험 결과, DeCAE 모형은 전통적인 필터 즉, Mean 필터, Median 필터와 이를 개선한 Bilateral 필터, NL-means 방법 뿐만 아니라 기존의 CAE 모형보다 정성적이고, 정량적인 척도 즉, MAE (Mean Absolute Error), PSNR (Peak Signal-to-Noise Ratio) 그리고 SSIM (Structural Similarity Index Measure) 면에서 우수한 결과를 보였다.

• 한국CDE학회지
• /
• 제4권3호
• /
• pp.26-31
• /
• 1998

Recent trend of popularity of CAE(Computer Aided Engineering) tools and its availability, partly contributed by significant price reduction of H/W and SW, makes us believe CAE has already established terra firma as de jure tools, enablint the design improvement in the manufacturing am. However, if your jobs are required to be working closely with engineers located in the front line of manufacturing site, CAE is observed far from popularity, still being regarded as exclusive tools for engineers holding advanced degrees. Conventional methodologies depending on knowledge that was accumulated thorough trial and errors, depending primarily on engineering tables and formulas or proprietary know-how, are preferred as the de factor standard under the roof of contemporary development shops. Samsung camera, having stared its business since early 、80, has accumulated enough technological strength to compete in the world market. As today's consumers demand more sophisticated featured-lighter weight, built-in multi zoom and miniature size fitting in the palm-from camera manufacturer, so should Samsung camera respond to ever-delicate consumer's needs with great flexibility. Consequently conventional designers, without sophisticated analytical tools, with encounter solving the critical design factors that have never been treated as seriousness-marginal safety factors induced by reduced size of parts. In the study, CAE results of boss and gears were demonstrated as examples, which confirms the facts that the simple analysis done by front line designers, can bring distinguishable effects on the potential improvements of design and on the consequential influences on the future design process-simulation before actual tooling and productions.

• Nuclear Engineering and Technology
• /
• 제53권3호
• /
• pp.888-893
• /
• 2021

The International Atomic Energy Agency has developed a tomographic imaging system for accomplishing the total fuel rod-by-rod verification time of fuel assemblies within the order of 1-2 h, however, there are still limitations for some fuel types. The aim of this study is to develop a deep learning-based denoising process resulting in increasing the tomographic image acquisition speed of fuel assembly compared to the conventional techniques. Convolutional AutoEncoder (CAE) was employed for denoising the low-quality images reconstructed by filtered back-projection (FBP) algorithm. The image data set was constructed by the Monte Carlo method with the FBP and ground truth (GT) images for 511 patterns of missing fuel rods. The de-noising performance of the CAE model was evaluated by comparing the pixel-by-pixel subtracted images between the GT and FBP images and the GT and CAE images; the average differences of the pixel values for the sample image 1, 2, and 3 were 7.7%, 28.0% and 44.7% for the FBP images, and 0.5%, 1.4% and 1.9% for the predicted image, respectively. Even for the FBP images not discriminable the source patterns, the CAE model could successfully estimate the patterns similarly with the GT image.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2012년도 춘계학술대회 논문집
• /
• pp.132-137
• /
• 2012

To enhance the CAE accuracy, the definition of material behavior is one of key influence on the result. In case of plastic material with fiber reinforcement, the anisotropic material behavior should be taken into account to increase of CAE accuracy. BASF has developed an innovative CAE tool, ULTRASIM$^R$, which is capable of generating material models of thermoplastic materials for structural simulation. ULTRASIM$^R$, not only the glass fiber orientation effect, but also the weld line effect, tensile-compression anisotropy, strain rate effect are combined in a non-linear material law, which will be evaluated in a unique failure criterion, thus resulting in an highly accurate CAE approach.

• 천문학논총
• /
• 제30권2호
• /
• pp.289-292
• /
• 2015

Most of the stars in the Galaxy are in binary systems. Binaries should be possible as the hosting stars of planets. Searching for planetary companions to binaries, especially evolved close binary stars, can provide insight into the formation and the ultimate fate of circumbinary planets and shed light on the late evolution of binary stars. In order to do this, we have chosen some post common envelope binaries including sdB-type eclipsing binaries and detached WD+dM eclipsing binaries as our targets and monitored them for several years. In this paper, we will present some of our new observations and results for three targets, NSVS 07826147, NSVS14256825 and RR Cae.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2011년도 추계학술대회 논문집
• /
• pp.821-832
• /
• 2011

Recently, the noise of vehicle is the one of the key factors for customers to purchase a vehicle and the most important part which is related to the noise is the exhaust system. Thus, car makers have their own ways to assess this exhaust noise not only to decrease the level of noise but to enhance the feeling of it. Typically, to do these things in the early stage of development, the tuning code of the exhaust system has to be made by CAE tool, which is very reliable but expensive, and the prototype parts of this code would be made for the validation test. Then this process can be iterated to meet the target of the performance. In this study, a new algorithm which adapts the '3 dimensional convective sound wave theory 'and 'Doppler effect' has been developed. With this new algorithm, a brand new system for the calculation of tail pipe noise has been developed and validated by acoustic wind tunnel test. As a result of this study, various comparisons and have been carried out, for example, the comparison with other CAE tool has been performed for the validity and the improvement of the new calculation code could be achieved.

• Interaction and multiscale mechanics
• /
• 제2권2호
• /
• pp.129-151
• /
• 2009

This paper presents a meshfree procedure using a convex generalized meshfree (GMF) approximation for the large deformation analysis of particle-reinforced rubber compounds on microscopic level. The convex GMF approximation possesses the weak-Kronecker-delta property that guarantees the continuity of displacement across the material interface in the rubber compounds. The convex approximation also ensures the positive mass in the discrete system and is less sensitive to the meshfree nodal support size and integration order effects. In this study, the convex approximation is generated in the GMF method by choosing the positive and monotonic increasing basis function. In order to impose the periodic boundary condition in the unit cell method for the microscopic analysis, a singular kernel is introduced on the periodic boundary nodes in the construction of GMF approximation. The periodic boundary condition is solved by the transformation method in both explicit and implicit analyses. To simulate the interface de-bonding phenomena in the rubber compound, the cohesive interface element method is employed in corporation with meshfree method in this study. Several numerical examples are presented to demonstrate the effectiveness of the proposed numerical procedure in the large deformation analysis.