• 사상체질의학회지
• /
• 제19권1호
• /
• pp.78-89
• /
• 2007

1. Objectives This is the comparative study with hand-operated measurement method & Automatic measurement method, in order to convert the automatic measurement method. 2. Methods We measured the body of patients(hand-operated width 5 places and hand-operated circumference 8 places, Automatic width 5 places) and analyzed the anthropometric data dividing into sex, age, BMI. 1057 patients's data are used in the analysis. 3. Results and Conclusions 1) 5th Width variable which was measured during lying down was not a sasang constitutional difference. 2) Diagnostic accuracy rate of the body measurement by Width which was measured during lying down and Width which IBS standing was 40-60%. 3) Diagnostic accuracy rate of Automatic & hand-operated measuring was not a big difference.

• 한국정밀공학회지
• /
• 제19권10호
• /
• pp.99-106
• /
• 2002

This paper proposes an error compensation method that improves accuracy with geometry information of injection molding parts. Geometric information can give an improved accuracy in reverse engineering. Measuring data can not lead to get accurate geometric model, including errors of physical parts and measuring machines. Measuring data include errors which can be classified into two types. One is molding error in product, the other is measuring error. Measuring error includes optical error of laser scanner, deformation by probe forces of CMM and machine error. It is important to compensate these in reverse engineering. Least square method (LSM) provides the cloud data with a geometry compensation, improving accuracy of geometry. Also, the functional shape of a part and design concept can be reconstructed by error compensation using geometry information.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2000년도 춘계학술대회논문집A
• /
• pp.860-865
• /
• 2000

This Paper Proposes the test pieces of X, Y and Z axes to compensate the shape distortion of Selective Laser Sintering(SLS) parts resulting from the phase change during the sintering process. In no case of the proposed compensation test pieces of X, Y axes the accurate rates of shrinkage can be measured with the reduction of curling which is obtained from adjustment of build orientation and the formula used to get scale factors are proposed with the shrinkage rates of them. The scale factors of X, Y and Z axes are generated by building up proposed compensation test pieces. The generated scale iactors are required to satisfy the dimensional accuracy even if there are changes of the build position and the size of SLS parts in the build chamber. For this reason, it is proposed that the build positions and the size be considered to be noise factors against the compensation test pieces and a method is also proposed that scale factors be selected to robustly maintain the dimensional accuracy of SLS parts under the actual operating conditions with the application of the Taguchi Method.

• 대한인간공학회지
• /
• 제30권6호
• /
• pp.715-725
• /
• 2011

Objective: This study is aimed at developing Articulated Human Models(AHM) using superquadrics to improve the geometric accuracy of the body shape. Background: The previous work presents the AHM with geometrical simplification such as ellipsoids to improve analysis efficiency. However, because of the simplicity, their physical properties such as a center of mass and moment of inertia are computed with errors compared to their actual values. Method: This paper introduces a three steps method to present the AHM with superquadrics. First, a 3D whole body scan data are divided into 17 body segments according to body joints. Second, superquadric fitting is employed to minimize the Euclidean distance between body segments and superquadrics. Finally, Fee-Form Deformation is used to improve accuracy over superquadric fitting. Results: Our computational experiment shows that the superquadric models give better accuracy of dynamic analysis than that of ellipsoid ones. Conclusion: We generate the AHM composed of 17 superquadrics and 16 joints using superquadric fitting. Application: The AHM using superquadrics can be used as the base model for dynamics and ergonomics applications with better accuracy because it presents the human motion effectively.

• Design & Manufacturing
• /
• 제15권3호
• /
• pp.1-6
• /
• 2021

An artificial neural network model based on a deep learning algorithm is known to be more accurate than humans in image classification, but there is still a limit in the sense that there needs to be a lot of training data that can be called big data. Therefore, various techniques are being studied to build an artificial neural network model with high precision, even with small data. The transfer learning technique is assessed as an excellent alternative. As a result, the purpose of this study is to develop an artificial neural network system that can classify burr images of light guide plate products with 99% accuracy using transfer learning technique. Specifically, for the light guide plate product, 150 images of the normal product and the burr were taken at various angles, heights, positions, etc., respectively. Then, after the preprocessing of images such as thresholding and image augmentation, for a total of 3,300 images were generated. 2,970 images were separated for training, while the remaining 330 images were separated for model accuracy testing. For the transfer learning, a base model was developed using the NASNet-Large model that pre-trained 14 million ImageNet data. According to the final model accuracy test, the 99% accuracy in the image classification for training and test images was confirmed. Consequently, based on the results of this study, it is expected to help develop an integrated AI production management system by training not only the burr but also various defective images.

• 한국의류학회지
• /
• 제47권1호
• /
• pp.137-151
• /
• 2023

Stationary 3D whole-body scanners generally require 5 to 20 seconds of scanning time and cannot effectively detect armpit and crotch areas. Therefore, this study aimed to analyze the accuracy of a photogrammetric technique using a multi-camera system. First, dimensional accuracy was analyzed using a mannequin scan, comparing the differences between the scan-derived measurements and the direct measurements, with an allowable tolerance of ISO 20685-1:2018. Only 2 of 59 measurement items (ankle height and upper arm circumference, specifically) exceeded the ISO 20685-1:2018 criteria. When compared with the results of the eight stationary whole-body scanners assessed by the literature, the photogrammetric technique was found to have the advantage of scanning the top of the head, armpit, and crotch areas clearly. Second, this study found the photogrammetric technique is suitable for obtaining the body scans because it can minimize the perform scanning, resulting in a reduction of measurement errors due to breathing and uncontrolled movements. The error rate of the photogrammetry method was much lower than that of stationary 3D whole-body scanners.

• Nutrition Research and Practice
• /
• 제14권6호
• /
• pp.667-678
• /
• 2020

BACKGROUND/OBJECTIVES: This study aimed to improve portion size estimation aids (PSEAs) used in the nutrition survey of the Korea National Health and Nutrition Examination Survey (KNHANES) and validate the accuracy and precision of the newly developed aids. SUBJECTS/METHODS: We conducted intensive interviews with survey experts in KNHANES and consulted with experts to collect opinions about improvement of PSEAs. Based on the results of the interviews, 5 types of PSEAs (rice bowl, earthen pots, mounds, measuring spoons, and thickness sticks) were newly developed using 3-dimensional (3D) modeling or modification of color or shape. Validation tests were conducted with 96 adults 20 years old or older. For the rice bowl and earthen pots, the participants were asked to select the more similar PSEA in size after being shown the real dishes. For the mounds, measuring spoons, and thickness sticks, the participants were presented with actual plates of food and asked to estimate the given portion sizes using the given PSEAs. RESULTS: The improved 2-dimensional (2D) picture aid for the rice bowl reflecting the size distortion by angle of view using 3D modeling was perceived more closely to the actual size than the current 2D picture (P < 0.001). The change of the color of 2D pictures and 3D models, the change of shape of the measuring spoons, and the 3-dimensionalization of the 2D mounds had no significant improvement in the subjects' perception. CONCLUSIONS: The currently used 2D PSEAs need to be fully redesigned using 3D modeling to improve subjects' perception. However, change of color or shape will not be necessary. For amorphous foods, it is suggested that more evaluation be performed before reaching a final conclusion in the use of PSEAs, or alternative ways to improve accuracy of estimation need to be explored.

• 디지털융복합연구
• /
• 제16권3호
• /
• pp.303-309
• /
• 2018

컴퓨터 비전 기술은 인공지능의 한 분야로 인간의 눈의 기능을 기계가 수행 할 수 있도록 카메라로부터 영상 이미지를 취득하고, 알고리즘을 통하여 분석하고 판별한 후 생산부품의 양품과 불량을 판별하는 부품 검사에 많이 적용되고 있다. 볼트 생산 시스템에서 생산되는 볼트 탭의 유무를 컴퓨터 비전 기술을 적용하여 자동으로 불량품을 선별하는 장치를 개발하였다. 형상검사 방법으로는 라인 스캔방식으로 검사영역의 시작점과 끝점의 위치를 파악하여 높이를 측정하는 방법을 사용하였으며, 볼트 탭의 유무를 판별하기 위한 방식으로는 원형 스캔방식으로 검사영역의 평균 밝기 값의 차이를 측정하여 유사도를 구하는 방식을 사용하였다. 두 종류의 볼트 탭의 검사 성능실험에서 총 검사에 소요되는 시간은 분당 300개 검사가 가능하며, 완벽한 검사 정확도를 나타내어 생산라인에서 검사의 정확성과 효율성을 입증하였다.

• 전자공학회논문지S
• /
• 제36S권10호
• /
• pp.113-122
• /
• 1999

하나의 질의 영상에 대하여 영상 데이터 베이스로부터 유사도가 높은 영상을 찾고자 할 때에는 유사도의 척도로 컬러 히스토그램 인터섹션법과 형상 특징의 비교법, 질감 특징의 비교법 등이 사용된다. 그리고 앞의 개별 특징 중 2가지 특징을 조합한 방법은 유사도의 산출 결과의 정확도를 높이기 위하여 사용된다. 본 논문에서는 히스토그램 인터섹션법과 형상 특징을 반영하는 블록화된 영역별 히스토그램 인터섹션 법을 사용하여 유사도 높은 영상을 얻는 방법을 제안한다. 이 방법은 서로 다른 2가지 특징을 조합하여 사용한 방법과 비교할 때 비슷한 정확도와 상대적으로 빠른 처리 속도를 달성하는 고효율의 알고리즘이다. 시뮬레이션을 통하여 제안한 알고리즘은 600개로 구성된 다양한 영상의 데이터 베이스 내에서 영상 검색이 수행되고 기존의 다른 알고리즘에 비하여 상대적으로 빠르고 신뢰도가 높은 유사도 척도임을 보인다.

• 한국기계가공학회지
• /
• 제21권8호
• /
• pp.66-71
• /
• 2022

Metal additive manufacturing (AM) has revolutionized several manufacturing industries. AM can generate large-scale metal components and produce complex geometries close to net-shapes. WAAM is an AM technology that has garnered considerable interest among industries owing to its economics and relatively high deposition rates. However, the heat accumulation in the weld bead during deposition triggers distortion and residual stress. To address these problems, various methods of interpass pressure rolling systems have been suggested in recent research. In addition, combining the rolling and WAAM processes can mitigate residual stresses. The constant-pressure rolling of the interlayer also affect the microstructure. The coarse microstructure of the as-deposited sample was altered to finer equiaxed grains via these methods. However, the bead-shape accuracy of the interlayer constant-pressure method does not consider the heat accumulation in each layer. Therefore, this study develops an interpass variable pressure rolling system that considers the heat accumulation of each layer. The interpass variable pressure rolling system comprises deposition, detection, pressure, and transport units. Finally, verification tests are performed on the interpass variable-pressure rolling system (at 500 kg) with the WAAM process, and the obtained results are discussed.