• Journal of the Korea Society of Computer and Information
• /
• v.28 no.3
• /
• pp.25-33
• /
• 2023

Noise generated during the acquisition and transmission of CT images acts as a factor that degrades image quality. Therefore, noise removal to solve this problem is an important preprocessing process in image processing. In this paper, we remove noise by using a deformable convolutional autoencoder (DeCAE) model in which deformable convolution operation is applied instead of the existing convolution operation in the convolutional autoencoder (CAE) model of deep learning. Here, the deformable convolution operation can extract features of an image in a more flexible area than the conventional convolution operation. The proposed DeCAE model has the same encoder-decoder structure as the existing CAE model, but the encoder is composed of deformable convolutional layers and the decoder is composed of conventional convolutional layers for efficient noise removal. To evaluate the performance of the DeCAE model proposed in this paper, experiments were conducted on CT images corrupted by various noises, that is, Gaussian noise, impulse noise, and Poisson noise. As a result of the performance experiment, the DeCAE model has more qualitative and quantitative measures than the traditional filters, that is, the Mean filter, Median filter, Bilateral filter and NL-means method, as well as the existing CAE models, that is, MAE (Mean Absolute Error), PSNR (Peak Signal-to-Noise Ratio) and SSIM. (Structural Similarity Index Measure) showed excellent results.

• Korean Journal of Computational Design and Engineering
• /
• v.16 no.5
• /
• pp.344-350
• /
• 2011

In injection molding industries, CAE analyses are generally used to find out problems predicted during the process of manufacturing. The results of CAE analyses consist of much in formation such as meshes and stress, so that the size of data is pretty large. To reduce the size of the data and to make it easy to share, the CAE result to JT translator is proposed in this paper. The translator consists of three modules to translate CAE result to JT format; Extracting module gets ASCII data of product shape and the result values of CAE analysis. Sorting module and mapping module make an element data set and JT file with the data extracted from Extracting module respectively. To the JT files, engineers are able to append product properties and their comments, so that they can share the whole history of the analysis process. In addition, our case study shows that the size of JT format is reduced by almost 90% of its original data format.

• Journal of the KSME
• /
• v.33 no.2
• /
• pp.145-151
• /
• 1993

전자산업이 고도화되어감에 따라 전자기술은 소재 . 기계분야와 복합된 기술의 발전이 가속화 되어 가고 있다. 특히 이 글에서는 응력해석 CAE 기술이 전자부품산업에 적용되는 사례를 소 개하고자 한다. 그 중 하나는 하드디스크의 정밀부품 중의 하나인 스핀들 모터용 Hub의 절삭 가공시 고려하여야 하는 탄성 변형량을 해석하고 이를 적용하는 사례이고 또 하나는 Audio Deck의 Capatan Shaft를 지지하는 플리스틱 Base가 신뢰성 시험도중 또는 고온 부하시 발생하는 크리프현상을 해석하여 적용하는 사례이다. 이 글에서 소개하는 두 가지 사례는 전자부품산업에 응력해석 CAE를 적용할 수 있는 예들 중 빙산의 일각에 불과할 뿐이고, 이와 유사한 전자부품에 응력해석 CAE를 적용할 수 있다.

• Journal of Korea Foundry Society
• /
• v.31 no.4
• /
• pp.218-223
• /
• 2011

Computer Aided Engineering (CAE) is very helpful field for every manufacturing industry including foundry. It covers CAD, CAM, and simulation technology also, and becomes as common sense in developing new products and processes. In South Korea, more than 600 foundries exist, and their average employee number is less than 40. Moreover, average age of them becomes higher. To break out these situations of foundry, software tools can be effective, and many commercial software tools had already been introduced. But their high costs and risks of investment act as difficulties in introducing the software tools to SMEs (Small and Medium size Enterprise). So we had developed cloud computing platform to propagate the CAE technologies to foundries. It includes HPC (High Performance Computing), platforms and software. So that users can try, enjoy, and utilize CAE software at cyber space without any investment. In addition, we also developed platform APIs (Application Programming Interface) to import not only our own CAE codes but also 3rd-party's packages to our cloud-computing platforms. As a result, CAE developers can upload their products on cloud platforms and distribute them through internet.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2007.05a
• /
• pp.318-321
• /
• 2007

This paper is concerned with improvement of surface quality of a sheet metal member in the stamping process. The CAE procedure of the stamping process is utilized in order to investigate cause of surface troubles and to improve surface quality. A complicated shape of the sheet metal member can induce surface troubles such as wrinkles because of insufficient tensions force and non-uniform contact of a blank sheet with tools. This paper proposes two guidelines such as a change of tool shape and added draw-beads on the tool surface in order to increase tension forces and to induce uniform contact. The proposed guidelines are verified with the CAE of the stamping process. The CAE results show that the changed shape of tools and added draw-beads can reduce the amount of wrinkles and improve surface quality.

• Proceedings of the KSME Conference
• /
• 2001.06c
• /
• pp.269-274
• /
• 2001

In order to reduce trial-and-errors in the die design and production, CAE systems for analysis of stamping tools have been introduced at the initial design stage recently. For optimal design, the CAE engineers may need to correct the meshes generated by automatic mesh generation programs. However, they may need help of CAD engineers as they are usually not skilled in manipulation of CAD systems. In order to get around these problems, automatic shell mesh modification method is proposed, which utilizes existing CAD/CAE package (in this study, I-DEAS) without user interaction. The developed method and optimization techniques are applied to a stamping die rounding optimization problem. The optimization results show that the manpower and the time required at virtual tryout can be reduced by using the developed systems.

• Proceedings of the KSME Conference
• /
• 2007.05a
• /
• pp.614-619
• /
• 2007

The importance of simulation of plastic flow in the injection mold is increasing as the parts are more complex, but the small-scale enterprises can't afford to invest for the infra individually. CAE (Computer Aided Engineering) service was naturally born based on these needs. This paper presents the engineering collaboration model between the large and the small/medium enterprises in the field of injection molds. Based on the engineering collaboration hub and CAE service, small-scale enterprises could research the necessary technology and develop the proper products. The analysis results of CAE are provided by the integrated visualization system on the web. This paper also deals with the delivery shorting process of CAE service for electric/electronic parts which should meet the needs of customers as soon as possible.

• 정보저장시스템학회:학술대회논문집
• /
• 2005.10a
• /
• pp.246-247
• /
• 2005

This study suggested fanless thermal design using CAE technique for the information storage system under the serious thermal problem. At first, main heat flow was controlled by CAE based fanless heat sink design not to influence sensitive optical pick-up sensor. Then, vent parametric studies found a thermal solution about highly concentrated case top heat due to fanless. These CAE results were verified by experimental methods. As a consequence of newly designed thermal path, thermal specification of optical pick-up sensor was satisfied and fanless thermal design for the information storage system was achieved.

• Korean Journal of Computational Design and Engineering
• /
• v.15 no.6
• /
• pp.418-424
• /
• 2010

For an engineering analysis to predict a bending of PCB by heat, geometry information from CAM data should be transferred to CAE system. But because this work is done by people, not computer, in PCB manufacturing company, it takes too much time, makes many errors and a result is unreliable. To solve these problems, we analyze a working process of company, then develop a CAM to CAE interface to translate a CAM data into a data to be input into a CAE system automatically.

• Journal of the korean Society of Automotive Engineers
• /
• v.12 no.5
• /
• pp.14-18
• /
• 1990

본 논문은 당사에서 지금까지 개발한 성형 난이성 평가 CAE system에 관한 것으로 전체적인 system 구성, 기능 및 적용예등을 소개하기로 한다.