• 한국의학물리학회지:의학물리
• /
• 제29권4호
• /
• pp.137-142
• /
• 2018

This study aimed to evaluate and verify a process for correcting the extended source-to-imager distance (SID) in portal dosimetry (PD). In this study, eight treatment plans (four volumetric modulated arc therapy and four intensity-modulated radiation therapy plans) at different treatment sites and beam energies were selected for measurement. A Varian PD system with portal dose image prediction (PDIP) was used for the measurement and verification. To verify the integrity of the plan, independent measurements were performed with the MapCHECK device. The predicted and measured fluence were evaluated using the gamma passing rate. The output ratio was defined as the ratio of the absolute dose of the reference SID (100 cm) to that of each SID (120 cm or 140 cm). The measured fluence for each SID was absolutely and relatively compared. The average SID output ratios were 0.687 and 0.518 for 120 SID and 140 SID, respectively; the ratio showed less than 1% agreement with the calculation obtained by using the inverse square law. The resolution of the acquired EPIDs were 0.336, 0.280, and 0.240 for 100, 120, and 140 SID, respectively. The gamma passing rates with PD and MapCHECK exceeded 98% for all treatment plans and SIDs. When autoalignment was performed in PD, the X-offset showed no change, and the Y-offset decreased with increasing SID. The PD-generated PDIP can be used for extended SID without additional correction.

• 한국IT서비스학회지
• /
• 제18권2호
• /
• pp.143-159
• /
• 2019

Machine learning (ML) is a method of fitting given data to a mathematical model to derive insights or to predict. In the age of big data, where the amount of available data increases exponentially due to the development of information technology and smart devices, ML shows high prediction performance due to pattern detection without bias. The feature engineering that generates the features that can explain the problem to be solved in the ML process has a great influence on the performance and its importance is continuously emphasized. Despite this importance, however, it is still considered a difficult task as it requires a thorough understanding of the domain characteristics as well as an understanding of source data and the iterative procedure. Therefore, we propose methods to apply deep learning for solving the complexity and difficulty of feature extraction and improving the performance of ML model. Unlike other techniques, the most common reason for the superior performance of deep learning techniques in complex unstructured data processing is that it is possible to extract features from the source data itself. In order to apply these advantages to the business problems, we propose deep learning based methods that can automatically extract features from transaction data or directly predict and classify target variables. In particular, we applied techniques that show high performance in existing text processing based on the structural similarity between transaction data and text data. And we also verified the suitability of each method according to the characteristics of transaction data. Through our study, it is possible not only to search for the possibility of automated feature extraction but also to obtain a benchmark model that shows a certain level of performance before performing the feature extraction task by a human. In addition, it is expected that it will be able to provide guidelines for choosing a suitable deep learning model based on the business problem and the data characteristics.

• 농촌계획
• /
• 제25권3호
• /
• pp.59-66
• /
• 2019

This study saw developed to build a landscape monitoring methodology by simulation of landscape effect prediction. A Visual landscape planning and management system has been introduced and implemented by each ministry so as to solve the problems of visual landscape destruction due to recognition on the value of natural landscape of beautiful territory and various development projects. At present, this system emphasizes the importance of the visual and perceptual aspect of the landscape however, there is a lack of techniques required for comprehensively predicting, evaluating, and managing it. Furthermore, sustainable landscape management after the completion of development projects has been inadequately carried out, as the focus has been only on consultation in the planning process of the development project in institutional performance. The viewpoint for judging the change in the visual landscape of the development plan and development project should be selected as the effective point where the development project is expected to result in a remarkable landscape change. As for the method of selecting effective viewpoints, the main viewpoints are selected by analyzing the visible area of the target viewpoint. When selecting the viewpoint centered on the viewpoint target, it was judged that it is possible to reduce the procedure of selecting and checking the existing preliminary viewpoints and widening the effective visible range. The proposed visual landscape monitoring is expected to be able to solve the existing institutional problems, and to be used when the implementers and authors of the development projects review the effects on the landscape.

• 한국항공우주학회지
• /
• 제47권9호
• /
• pp.617-628
• /
• 2019

벤트홀로 연결된 두 개의 가변 체적 챔버를 가지고 있는 파이로테크닉 분리 너트는 복잡한 과정을 통해 분리를 수행하므로 많은 설계 인자를 가지고 있다. 접촉부의 각도, 구성품의 질량, 누름봉 돌출부 안쪽 지름, 환형 챔버의 초기 부피, 확장 챔버의 초기 부피, 화약질량, 벤트홀 지름과 같은 설계 인자들의 변화에 따라 분리 너트의 성능이 어떻게 달라지는지를 파악하면 설계 개선 방향을 결정하는데 도움이 될 수 있다. 이에, 선행 연구를 통해 개발한 파이로테크닉 분리 너트 거동 예측 모델을 이용하여 매개변수 연구를 수행하였다. 다른 설계 인자의 값은 고정한 상태에서 한 설계인자의 값을 독립적으로 변화시키며 분리 시간이 어떻게 달라지는지, 변화의 원인이 무엇인지를 분석하였다.

• Asian-Australasian Journal of Animal Sciences
• /
• 제32권10호
• /
• pp.1483-1490
• /
• 2019

Objective: To explore the molecular mechanisms of fat metabolism and deposition in pigs, an experiment was conducted to identify hepatic mRNAs and miRNAs expression and determine the potential interaction of them in two phenotypically extreme pig breeds. Methods: mRNA and miRNA profiling of liver from 70-day Jinhua (JH) and Landrace (LD) pigs were performed using RNA sequencing. Blood samples were taken to detect results of serum biochemistry. Bioinformatics analysis were applied to construct differentially expressed miRNA-mRNA network. Results: Serum total triiodothyronine and total thyroxine were significantly lower in Jinhua pigs, but the content of serum total cholesterol (TCH) and low-density lipoprotein cholesterol were strikingly higher. A total of 467 differentially expressed genes (DEGs) and 35 differentially expressed miRNAs (DE miRNAs) were identified between JH and LD groups. Gene ontology analysis suggested that DEGs were involved in oxidation-reduction, lipid biosynthetic and lipid metabolism process. Interaction network of DEGs and DE miRNAs were constructed, according to target prediction results. Conclusion: We generated transcriptome and miRNAome profiles of liver from JH and LD pig breeds which represent distinguishing phenotypes of growth and metabolism. The potential miRNA-mRNA interaction networks may provide a comprehensive understanding in the mechanism of lipid metabolism. These results serve as a basis for further investigation on biological functions of miRNAs in the porcine liver.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권10호
• /
• pp.3729-3749
• /
• 2021

At present, deep convolution network-based salient object detection (SOD) has achieved impressive performance. However, it is still a challenging problem to make full use of the multi-scale information of the extracted features and which appropriate feature fusion method is adopted to process feature mapping. In this paper, we propose a new adjacency auxiliary network (AANet) based on multi-scale feature fusion for SOD. Firstly, we design the parallel connection feature enhancement module (PFEM) for each layer of feature extraction, which improves the feature density by connecting different dilated convolution branches in parallel, and add channel attention flow to fully extract the context information of features. Then the adjacent layer features with close degree of abstraction but different characteristic properties are fused through the adjacent auxiliary module (AAM) to eliminate the ambiguity and noise of the features. Besides, in order to refine the features effectively to get more accurate object boundaries, we design adjacency decoder (AAM_D) based on adjacency auxiliary module (AAM), which concatenates the features of adjacent layers, extracts their spatial attention, and then combines them with the output of AAM. The outputs of AAM_D features with semantic information and spatial detail obtained from each feature are used as salient prediction maps for multi-level feature joint supervising. Experiment results on six benchmark SOD datasets demonstrate that the proposed method outperforms similar previous methods.

• 한국지반신소재학회논문집
• /
• 제18권3호
• /
• pp.55-67
• /
• 2019

낙동강 하구 일대 대심도의 연약지반 지역은 시간 의존적인 2차압밀 특성 등의 원인으로 예측보다 실제 과다한 침하가 발생하므로 설계 시점부터 2차압밀 특성을 고려한 예측기법의 필요성이 요구된다. 따라서 본 연구에서는 2,000일 이상의 장기 계측자료 분석과 1차원 이론, 탄소성, 점 탄소성의 수치해석을 수행하여 장기침하 특성을 분석하였다. 그 결과 1차원 이론, 탄소성 해석 모두 성토 높이와 연약층 깊이에 따른 뚜렷한 상관관계를 찾을 수 없었으나, 점 탄소성 해석 방법은 장기계측 결과와 아주 유사하게 침하를 예측하였다. 따라서 연구 대상 지역과 같은 대심도 지역에서 장기 침하 예측 시 점 탄소성 해석을 시공 전에 적용하면 2차압밀 침하 거동 특성을 미리 예측하여 시공 중 발생할 수 있는 공기 지연과 경제적 손실을 사전에 방지할 수 있을 것으로 판단된다.

• 한국융합학회논문지
• /
• 제12권1호
• /
• pp.105-110
• /
• 2021

의료기관에서 최근 많이 사용되고 있는 친환경 소재 방사선 차폐체는 시트 형태로 제작되어 Apron의 재료로 활용되고 있다. 친환경 Apron의 차폐성능은 납당량을 기준으로 제시되고 있으며, 납당량은 0.25~0.50mmPb로 제시되고 있다. 납이 주재료 사용되는 차폐체인 경우 납의 우수한 가공성으로 인해 두께로 차폐성능을 조절할 수 있다. 그러나 친환경 차폐시트는 차폐재료의 함량, 베이스 재료인 고분자 물질의 물성, 공정과정의 기술적 차이에 따라 차폐성능이 변화되어 두께 기준의 차폐성능을 제어하기가 어렵다. 본 연구에서는 이러한 문제점을 해결하고자 두께를 기준으로 차폐시트를 제작하여 차폐성능을 평가, 비교하였다. 동일한 시트 제작공정을 제시하여 두께를 제어할 수 있는 캘린더 공정의 압연 기술을 적용하였고 여러층의 적층 구조와 단일 구조로 제작된 두 시트의 두께별 차폐성능을 비교하여 5%대의 차이를 관찰하였다. 그 결과 여러 층으로 차폐한 적층 구조 차폐시트가 더 효과적임을 증명하는 동시에 두께 중심의 차폐성능의 가능성을 제시하였다.

• 산업경영시스템학회지
• /
• 제43권4호
• /
• pp.161-167
• /
• 2020

In the process of cutting large aircraft parts, the tool may be abnormally worn or damaged due to various factors such as mechanical vibration, disturbances such as chips, and physical properties of the workpiece, which may result in deterioration of the surface quality of the workpiece. Because workpieces used for large aircrafts parts are expensive and require strict processing quality, a maintenance plan is required to minimize the deterioration of the workpiece quality that can be caused by unexpected abnormalities of the tool and take maintenance measures at an earlier stage that does not adversely affect the machining. In this paper, we propose a method to indirectly monitor the tool condition that can affect the machining quality of large aircraft parts through real-time monitoring of the current signal applied to the spindle motor during machining by comparing whether the monitored current shows an abnormal pattern during actual machining by using this as a reference pattern. First, 30 types of tools are used for machining large aircraft parts, and three tools with relatively frequent breakages among these tools were selected as monitoring targets by reflecting the opinions of processing experts in the field. Second, when creating the CNC machining program, the M code, which is a CNC auxiliary function, is inserted at the starting and ending positions of the tool to be monitored using the editing tool, so that monitoring start and end times can be notified. Third, the monitoring program was run with the M code signal notified from the CNC controller by using the DAQ (Data Acquisition) device, and the machine learning algorithms for detecting abnormality of the current signal received in real time could be used to determine whether there was an abnormality. Fourth, through the implementation of the prototype system, the feasibility of the method proposed in this paper was shown and verified through an actual example.

• 한국추진공학회지
• /
• 제24권5호
• /
• pp.65-76
• /
• 2020

3차원 스크램제트 모델의 설계과정에 있어 3차원 전산해석은 2차원 해석에 비해 상대적으로 복잡한 격자 구성과 많은 해석 시간을 요구한다. 때문에 다양한 조건에서의 3차원 모델의 성능을 확인하는 것은 쉽지 않은 일이다. 따라서 본 연구에서는 2차원 스크램제트 비행체 모델의 전산해석 결과와 비점성 초음속 선형화 이론을 기반으로 2차원 모델의 비행 조건에 따른 공력계수 및 흡입구 질량 포획률 관계식을 도출하였다. 도출된 2차원 성능 관계식과 함께 최소한의 3차원 해석을 수행하여 3차원 스크램제트 모델의 공력 계수와 흡입구 질량 포획률 관계식을 이끌어내었다. 또한 추가적인 3차원 계산을 통해 확장된 3차원 관계식들의 공력 정확도를 검증하였다.