• Title/Summary/Keyword: form accuracy

Search Result 1,360, Processing Time 0.033 seconds

Method Discrimination for Product Traceability and Identification of Korean Native Chicken using Microsatellite DNA (초위성체를 이용한 한국 재래닭의 원산지 추적 및 개체 식별 방법에 관한 연구)

  • Park, Mi-Hyun;Oh, Jae-Don;Jeon, Gwang-Joo;Kong, Hong-Sik;Sang, Byong-Don;Choi, Chull-Hwan;Yeon, Sung-Hum;Cho, Byong-Wok;Lee, Hak-Kyu
    • Korean Journal of Organic Agriculture
    • /
    • v.12 no.4
    • /
    • pp.451-461
    • /
    • 2004
  • In an animals, identification system has been widely used by ear tag with dummy code and blood typing for parernity. Also, genotyping methods were using for useful mean of individual identification for live animals. In the case of genotyping estimation of gene in population of korean native chicken. In this study, we tested for development of genetic markers used it possible to determination of individual identification system. The candidate genetic markers were used already bow 10 of microstalite DNA sequence information in chromosome No. 1 and 14. Result of analysis for genotyping, the number of alleles of those microstatelites DNA was shown minimal 3 to 12 and the heterozygote expression frequency range was shown from 0.617 to 0.862. In our result, effective number of allele for each microsatellites DNA was shown 3~7, and the accuracy of individual identification was shown nearly 100%, when used with 6 genetic marker. This study was about genotyping method for identification used specific genetic marker form microsatellite DNA in the brand marketing of korean native chicken. Our results suggest that genotyping method used specific genetic marker from microsatellite DNA might be very useful for determination of individual identification.

  • PDF

3D Human Shape Deformation using Deep Learning (딥러닝을 이용한 3차원 사람모델형상 변형)

  • Kim, DaeHee;Hwang, Bon-Woo;Lee, SeungWook;Kwak, Sooyeong
    • Journal of Korea Society of Industrial Information Systems
    • /
    • v.25 no.2
    • /
    • pp.19-27
    • /
    • 2020
  • Recently, rapid and accurate 3D models creation is required in various applications using virtual reality and augmented reality technology. In this paper, we propose an on-site learning based shape deformation method which transforms the clothed 3D human model into the shape of an input point cloud. The proposed algorithm consists of two main parts: one is pre-learning and the other is on-site learning. Each learning consists of encoder, template transformation and decoder network. The proposed network is learned by unsupervised method, which uses the Chamfer distance between the input point cloud form and the template vertices as the loss function. By performing on-site learning on the input point clouds during the inference process, the high accuracy of the inference results can be obtained and presented through experiments.

Novel Auto White Balance Algorithm Using Adaptive Color Sampling Based on $CIEL^*a^*b^*$ color space for Mobile Phone Camera ($CIEL^*a^*b^*$ 색 공간에서 적응적 컬러 샘플링을 이용한 Mobile Phone 카메라용 자동화이트 밸런스 알고리즘)

  • Kim, Kyung-Rin;Son, Kyoung-Soo;Ha, Joo-Young;Kim, Sang-Choon;Kang, Bong-Soon
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.12 no.8
    • /
    • pp.1356-1362
    • /
    • 2008
  • In this paper. we propose a novel auto white balance algorithm which is one of the representative functions on cameras. White balance is the process of removing unrealistic color casts, which will make the captured white objects appear white. For white balance, we employ $CIEL^*a^*b^*$ color space which is the most complete color model available and is conventionally used to describe all the colors visible to the human eye and estimate the color difference on white objects with distribution of the image which is called the reference white estimation. For accuracy, we form groups or sets of pixels that are altered by the light sources and other elements. Moreover, Standard group is decided by judgment of specific-case images with the information of groups. Then, the reference white estimation is performed by the color sampling which is to choose all the accumulated pixels contained within the standard group. The color gain for image compensation by considering the color saturation is also computed. the proposed algorithm provides a significant performance.

Development of Cellulose Strip for Dry Eye Inspection (건성안검사용 셀룰로즈 스트립 개발)

  • Lee, Myeonggu;Jeong, Myeong-jin
    • The Journal of the Convergence on Culture Technology
    • /
    • v.5 no.4
    • /
    • pp.355-359
    • /
    • 2019
  • Schirmer test is one of the most used methods for the diagnosis of dry eye. We attempted to develop a tear level measurement tool to replace unreliable Schirmer test with inaccurate results. Absorbency tests for various absorbents were carried out. As a result, ${\alpha}$ cellulose pulp was selected as the absorbent. Prototypes were produced and evaluated twice. Finally a tear level measurement tool in the form of a strip of ${\alpha}$ cellulose pulp adhered to a polyurethane was prepared. Usability evaluation of prepared tear level measurement tool was performed. As a result, it was confirmed that it has a significant correlation with SM tube developed oversea recently. In addition, it was judged to be useful as an alternative to the Schirmer test in terms of measurement time and accuracy.

Integrated fire dynamics and thermomechanical modeling framework for steel-concrete composite structures

  • Choi, Joonho;Kim, Heesun;Haj-ali, Rami
    • Steel and Composite Structures
    • /
    • v.10 no.2
    • /
    • pp.129-149
    • /
    • 2010
  • The objective of this study is to formulate a general 3D material-structural analysis framework for the thermomechanical behavior of steel-concrete structures in a fire environment. The proposed analysis framework consists of three sequential modeling parts: fire dynamics simulation, heat transfer analysis, and a thermomechanical stress analysis of the structure. The first modeling part consists of applying the NIST (National Institute of Standards and Technology) Fire Dynamics Simulator (FDS) where coupled CFD (Computational Fluid Dynamics) with thermodynamics are combined to realistically model the fire progression within the steel-concrete structure. The goal is to generate the spatial-temporal (ST) solution variables (temperature, heat flux) on the surfaces of the structure. The FDS-ST solutions are generated in a discrete form. Continuous FDS-ST approximations are then developed to represent the temperature or heat-flux at any given time or point within the structure. An extensive numerical study is carried out to examine the best ST approximation functions that strike a balance between accuracy and simplicity. The second modeling part consists of a finite-element (FE) transient heat analysis of the structure using the continuous FDS-ST surface variables as prescribed thermal boundary conditions. The third modeling part is a thermomechanical FE structural analysis using both nonlinear material and geometry. The temperature history from the second modeling part is used at all nodal points. The ABAQUS (2003) FE code is used with external user subroutines for the second and third simulation parts in order to describe the specific heat temperature nonlinear dependency that drastically affects the transient thermal solution especially for concrete materials. User subroutines are also developed to apply the continuous FDS-ST surface nodal boundary conditions in the transient heat FE analysis. The proposed modeling framework is applied to predict the temperature and deflection of the well-documented third Cardington fire test.

Road Surface Damage Detection based on Object Recognition using Fast R-CNN (Fast R-CNN을 이용한 객체 인식 기반의 도로 노면 파손 탐지 기법)

  • Shim, Seungbo;Chun, Chanjun;Ryu, Seung-Ki
    • The Journal of The Korea Institute of Intelligent Transport Systems
    • /
    • v.18 no.2
    • /
    • pp.104-113
    • /
    • 2019
  • The road management institute needs lots of cost to repair road surface damage. These damages are inevitable due to natural factors and aging, but maintenance technologies for efficient repair of the broken road are needed. Various technologies have been developed and applied to cope with such a demand. Recently, maintenance technology for road surface damage repair is being developed using image information collected in the form of a black box installed in a vehicle. There are various methods to extract the damaged region, however, we will discuss the image recognition technology of the deep neural network structure that is actively studied recently. In this paper, we introduce a new neural network which can estimate the road damage and its location in the image by region-based convolution neural network algorithm. In order to develop the algorithm, about 600 images were collected through actual driving. Then, learning was carried out and compared with the existing model, we developed a neural network with 10.67% accuracy.

Produced Body Customized 3D Print Finger Brace using Dicom File (Dicom file을 이용하여 만든 신체 맞춤형 3D print 손가락 보조기 제작)

  • Choi, Hyeun-Woo;Park, Ji-Eun;Kim, Jung-Hun;Seo, An-Na;Lee, Jong-Min
    • Journal of the Korean Society of Radiology
    • /
    • v.13 no.4
    • /
    • pp.597-603
    • /
    • 2019
  • We obtained a Dicom file using a CT (Computed Tomography), a diagnostic test device used in clinical practice. Dicom files and 3D programs, and finger printers with 3D printers. Because the finger brace is intended for the human body, the accuracy of the shape is very important. 3D Print has the advantage of high precision, variety of materials, and short output time. In clinic, aluminum protector or medical device manufacturer's finger protector is limited. By creating a finger brace with a 3D printer, we expect to be able to apply a precise form of a custom finger brace to the patient that can be used to treat a patient's finger trauma, illness, or deformity.

Analysis of Accounts Receivable Aging Using Variable Order Markov Model (가변 마코프 모델을 활용한 매출 채권 연령 분석)

  • Kang, Yuncheol;Kang, Minji;Chung, Kwanghun
    • The Journal of Society for e-Business Studies
    • /
    • v.24 no.1
    • /
    • pp.91-103
    • /
    • 2019
  • An accurate prediction on near-future cash flows plays an important role for a company to attenuate the shortage risk of cash flow by preparing a plan for future investment in advance. Unfortunately, there exists a high level of uncertainty in the types of transactions that occur in the form of receivables in inter-company transactions, unlike other types of transactions, thereby making the prediction of cash flows difficult. In this study, we analyze the trend of cash flow related to account receivables that may arise between firms, by using a stochastic approach. In particular, we utilize Variable Order Markov (VOM) model to predict how future cash flows will change based on cash flow history. As a result of this study, we show that the average accuracy of the VOM model increases about 12.5% or more compared with that of other existing techniques.

A simple quasi-3D HSDT for the dynamics analysis of FG thick plate on elastic foundation

  • Boukhlif, Zoulikha;Bouremana, Mohammed;Bourada, Fouad;Bousahla, Abdelmoumen Anis;Bourada, Mohamed;Tounsi, Abdelouahed;Al-Osta, Mohammed A.
    • Steel and Composite Structures
    • /
    • v.31 no.5
    • /
    • pp.503-516
    • /
    • 2019
  • This work presents a dynamic investigation of functionally graded (FG) plates resting on elastic foundation using a simple quasi-3D higher shear deformation theory (quasi-3D HSDT) in which the stretching effect is considered. The culmination of this theory is that in addition to taking into account the effect of thickness extension (${\varepsilon}_z{\neq}0$), the kinematic is defined with only 4 unknowns, which is even lower than the first order shear deformation theory (FSDT). The elastic foundation is included in the formulation using the Pasternak mathematical model. The governing equations are deduced through the Hamilton's principle. These equations are then solved via closed-type solutions of the Navier type. The fundamental frequencies are predicted by solving the eigenvalue problem. The degree of accuracy of present solutions can be shown by comparing it to the 3D solution and other closed-form solutions available in the literature.

Development of Supervised Machine Learning based Catalog Entry Classification and Recommendation System (지도학습 머신러닝 기반 카테고리 목록 분류 및 추천 시스템 구현)

  • Lee, Hyung-Woo
    • Journal of Internet Computing and Services
    • /
    • v.20 no.1
    • /
    • pp.57-65
    • /
    • 2019
  • In the case of Domeggook B2B online shopping malls, it has a market share of over 70% with more than 2 million members and 800,000 items are sold per one day. However, since the same or similar items are stored and registered in different catalog entries, it is difficult for the buyer to search for items, and problems are also encountered in managing B2B large shopping malls. Therefore, in this study, we developed a catalog entry auto classification and recommendation system for products by using semi-supervised machine learning method based on previous huge shopping mall purchase information. Specifically, when the seller enters the item registration information in the form of natural language, KoNLPy morphological analysis process is performed, and the Naïve Bayes classification method is applied to implement a system that automatically recommends the most suitable catalog information for the article. As a result, it was possible to improve both the search speed and total sales of shopping mall by building accuracy in catalog entry efficiently.