• 소성∙가공
• /
• 제33권1호
• /
• pp.43-49
• /
• 2024

A high strength steel sheet (HSSS) has widely used to improve the specific rigidity of parts and the safety of the passenger in automotive industries. However, the HSSS is difficult to manufacture precise parts through a forming process due to the reduced elongation and the increased elastic recovery. The goal of the paper is to investigate the improved design of bending dies for DP590 HSSS. The over forming type bending dies with cam systems added to the side of the formed part is proposed to improve the quality of the part. The effects of the die design and the forming parameter on formability and elastic recovery characteristics is examined using finite element analyses (FEAs). From the results of FEAs, proper die design and forming parameters are predicted.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2024년도 춘계학술발표대회
• /
• pp.794-797
• /
• 2024

Despite advancements in deep learning, existing semantic segmentation models exhibit suboptimal performance under adverse weather conditions, such as fog or rain, whereas they perform well in clear weather conditions. To address this issue, much of the research has focused on making image or feature-level representations weather-independent. However, disentangling the style and content of images remains a challenge. In this work, we propose a novel fine-tuning method, 'freeze-n-update.' We identify a subset of model parameters that are weather-independent and demonstrate that by freezing these parameters and fine-tuning others, segmentation performance can be significantly improved. Experiments on a test dataset confirm both the effectiveness and practicality of our approach.

• 대한토목학회논문집
• /
• 제42권2호
• /
• pp.291-298
• /
• 2022

최근 개발되는 드론은 저가이면서 운용의 편의성 또한 높아 드론을 이용한 공간정보 제작 및 활용이 나날이 증대되고 있다. 그러나, 대부분 저가형의 드론은 저정밀도 GNSS (Global Navigation Satellite System), IMU (Inertial Measurement Unit)를 탑재하여 영상을 취득하기 때문에 영상이 초기에 가지고 있는 위치정보 및 회전각 요소의 정확도가 낮다. 또한, 기체가 작고 가벼워 바람의 영향을 많이 받기 때문에 일정한 중복도를 유지하기에 어려움이 있으며 이러한 문제가 위치결정 정확도에 영향을 미친다. 이에 본 연구에서는 외부표정요소 변화에 따른 위치 정확도 변화를 분석하기 위하여 서로 다른 시기에 촬영한 영상을 Pix4D Mapper로 영상처리하고 성과물의 정확도를 분석하였다. 외부표정요소에 따른 정확도 변화를 세밀하게 분석하기 위하여 1차 처리 결과의 외부표정요소를 2차 처리 시 메타데이터로 활용하였다. 이후 외부표정요소를 스트립별로 나누어 변화량을 분석하였다. 분석 결과 회전각 요소 중 Omega, Phi값의 변화는 표고위치 정확도, Kappa값은 평면위치 정확도의 저하에 더욱 관련되어 있음을 입증하였다.

• 한국생산제조학회지
• /
• 제19권1호
• /
• pp.114-120
• /
• 2010

Optimization of the turning process has been concentrated on the selection of the optimal cutting parameters, such as cutting speed, feed rate and depth of cut. However, optimization of the cutting parameters does not necessarily guarantee the maximum profit. For the maximization of the profit, parameters other than cutting parameters have to be taken care of. In this study, 8 price-related parameters were considered to maximize the profit of the product. Regression equations obtained from RSM technique to relate the cutting parameters and maximum cutting volume with a given insert were used. The experiments with four combinations of cutting inserts and material were executed to compare the results that made the profit and cutting volume maximized. The results showed that the cutting parameters for volume and profit maximization were totally different. Contrary to our intuition, global optimization was achieved when the number of inserts change was larger than those for volume maximization. It is attributed to the faster cutting velocity, which decreases processing time and increasing the number of tool used and the total tool changing time.

• Fibers and Polymers
• /
• 제7권4호
• /
• pp.344-351
• /
• 2006

This paper is intended to determine the optimal processing parameters applied to the dyeing procedure so that the desired color strength of a raw fabric can be achieved. Moreover, the processing parameters are also used for constructing a system to predict the fabric quality. The fabric selected is the nylon and Lycra blend. The dyestuff used for dyeing is acid dyestuff and the dyeing method is one-bath-two-section. The Taguchi quality method is applied for parameter design. The analysis of variance (ANOVA) is applied to arrange the optimal condition, significant factors and the percentage contributions. In the experiment, according to the target value, a confirmation experiment is conducted to evaluate the reliability. Furthermore, the genetic algorithm (GA) is combined with the back propagation neural network (BPNN) in order to establish the forecasting system for searching the best connecting weights of BPNN. It can be shown that this combination not only enhances the efficiency of the learning algorithm, but also decreases the dependency of the initial condition during the network training. Most of all, the robustness of the learning algorithm will be increased and the quality characteristic of fabric will be precisely predicted.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권10호
• /
• pp.2679-2691
• /
• 2012

In this paper, a high accuracy stereo reconstruction method for surgery instruments positioning is proposed. Usually, the problem of surgical instruments reconstruction is considered as a basic task in computer vision to estimate the 3-D position of each marker on a surgery instrument from three pairs of image points. However, the existing methods considered the 3-D reconstruction of the points separately thus ignore the structure information. Meanwhile, the errors from light variation, imaging noise and quantization still affect the reconstruction accuracy. This paper proposes a method which takes the structure information of surgical instruments as constraints, and reconstructs the whole markers on one surgical instrument together. Firstly, we calibrate the instruments before navigation to get the structure parameters. The structure parameters consist of markers' number, distances between each markers and a linearity sign of each instrument. Then, the structure constraints are added to stereo reconstruction. Finally, weighted filter is used to reduce the jitter. Experiments conducted on surgery navigation system showed that our method not only improve accuracy effectively but also reduce the jitter of surgical instrument greatly.

• 열처리공학회지
• /
• 제7권4호
• /
• pp.288-297
• /
• 1994

In this study, lattice parameter, binding energy and microstructures of TiC layer according to the addition of Fe, Cr were investigated in the reactive deposition coating. From the results, the lattice parameters of the TiC layers by using ferro-titanium as a precursor were 4.329~4.339A but the lattice parameters of the TiC layers formed by ferro-titanium and ferro-chromium decreased to 4.316~4.330A. The hardness of the former's was HV(100g) 3,000~3,400kg/mm and the hardness of the latter's was HV (100g) 3,800~3,900. But, regardless of Cr and Fe, the binding energy of TiC layers were 454.75 eV for $Ti2p_{3/2}$ and were 281.85 eV for Cls. Meanwhile, the TiC layers were densified by addition of Fe, Cr and internal defects were reduced Therefore. it can be concluded that the remarkable hardness increment was obtained by the improvement of microstructures of TiC rather than the increase of bond strength or Peierls stress.

• 한국재료학회지
• /
• 제13권11호
• /
• pp.725-731
• /
• 2003

The effect of organic processing parameters on the dispersion stability of the BaTiO$_3$-based dielectric particles and borosilicate glass particulate suspensions was investigated in a system where organic solvents, dispersant, binder and modifier were used as processing additives in a low temperature cofired ceramic fabrication processes. Two- and three-component organic solvents were used to disperse ceramic particles and it was found the better stability in the particulate suspension prepared in a bi-solvent, which was consists of toluene and ethanol in a non-azeotropic composition. The addition amount of organic additives had a great impact on dispersion in the present investigation. The flow curves of the suspensions prepared with binder and modifier were fitted according to the power-law equation, which indicates that the internal structure of the suspension could be disturbed under the applied shear stress. Finally, the LTCC green tapes were successfully tape-cast based upon the optimum formulation of LTCC suspension and its microstructure was compared with that of the hard-agglomerates.

• 한국정밀공학회지
• /
• 제11권6호
• /
• pp.185-199
• /
• 1994

This study has been performed to present a new automatic tool wear measurement by digital image processing. The purpose of this paper is to develop an automatic tool wear measuring system based on the image processing which can be applied to the quasi-real time measurement of the characteristics of insert tip in turning. Tool wear monitoring is one of the key-problems, for the development of control systems of modern unmanned factory which are not completely solved now. In oredr words at present complete qualitative and quantitative information on tool wear morphology is required, at least on the following aspects : flank wear, its dimensions and distribution on the maximum and mean values on VB pqrqmeter in the various zones of the wearland. crater wear, its main dimensions and values of KT parameters. This research has been performed to this technique made possible by designing a proper lighting system to the worn tool with following features : The flank wear is measured by observing the active cutting part from a proper direction and by lighting the wearland by a diffuser optic system. The crater wear is visualized by lighting the tool by a He-Ne gas laser system developed in this study. By means of this system it is research to evaluate classical parameters of tool wear and to have complete information about tool wear morphology.

• 한국정보처리학회논문지
• /
• 제4권12호
• /
• pp.3200-3210
• /
• 1997

훼손된 타원 및 복잡한 형태의 영상에서 타원 검출에 최소자승법(LSM : Least Square Method)을 적용할 수 있는데 이는 데이터가 비정규 오류 분포를 따르거나 특이한점들이 있는 상태에서는 신뢰할 수 있는 결과를 얻을 수 없다. 특히 최소자승법은 훼손된 부분을 데이터가 없는 것으로 가정 하고 모든 데이터를 동일한 비중으로 연산하므로 훼손된 부분은 더욱 훼손된 모양으로 검출되는 문제점 있다. 본 논문에서 변형된 최소자승 법(MLSM: Modified Least Square Method)이란 훼손부분의 가까운점에 큰 비중을 둠으로 원래의 모양에 접근하는 형상(feature)의 타원을 검출하려는 것으로 훼손점 부근의 2점과 그외 중요한 l점을 강제로 만족하는 방법이다. 3점을 만족시키는 제한 조건을 주고 2개의 파라미터는 최소자승법으로 구하고, 나머지 3개는 제한 조건으로 구하여 타원 검출에 적용한 결과 실제 영상에서 타원의 검출 및 판별에 좋은 효과가 있었으며, 특히 인간의 치열의 곡선 모양을 결정하는테 좋은 효과가 있음을 보였다.