• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.7
• /
• pp.327-332
• /
• 2016

This study is a leading research project to develop an automatic grade decision making algorithm of a 6-years-old fresh ginseng. For this work, we developed a Ginseng image acquiring instrument which can take 4-direction's images of a Ginseng at the same time and obtained 245 jingen images using the instrument. The 12 parameters were extracted for each image by a manual way. Lastly, 4 parameters were selected depending on a Ginseng grade classification criteria of KGC Ginseng research institute and a survey result which a distribution of averaging 12 parameters. A pattern recognition classifier was used as a support vector machine, designed to "k-class classifier" using the OpenCV library which is a open-source platform. We had been surveyed the algorithm performance(Correct Matching Ratio, False Acceptance Ratio, False Reject Ratio) when the training data number was controlled 10 to 20. The result of the correct matching ratio is 94% of the $1^{st}$ ginseng grade, 98% of the $2^{nd}$ ginseng grade, 90% of the $3^{rd}$ ginseng grade, overall, showed high recognition performance with all grades when the number of training data are 10.

• Explosives and Blasting
• /
• v.28 no.2
• /
• pp.133-147
• /
• 2010

A Suggestion of In-situ Rock Mass Evaluation and Correlation between Rock Mass Classfication MethodsThe purpose of this study is to find out rock mass classification method which is practically applicable to a field and to consider a correlation between the new method and the old method. Rock mass is an aggregate of separated blocks. To express the aggregate, the properties of both intact rock and rock mass should be considered. In this study, therefore, parameters for rock mass description are classified into rock strength and rock structure. Indices for parameters evaluation are obtained from old method and the strength and structure property of rock is described by using those indices. Value of 25 is allocated to each parameter obtained. $RMR_{basic}$ =0.86(X=Method)+14.47 is derived between $RMR_{basic}$ and this study and $RMR^*$ = 0.87(X-Method)+9.20 is derived between revised RMR and this study. Coefficient of determination is $R^2$=0.841 and $R^2$=0.846 each.

• Journal of Korea Water Resources Association
• /
• v.32 no.1
• /
• pp.3-13
• /
• 1999

In this study, an alternative spatial analysis method against conventional methods such as Thiessen method, Inverse Distance method, and Kriging method, named Spatial-Analysis Neural-Network (SANN) is presented. It is based on neural network modeling and provides a nonparametric mean estimator and also estimators of high order statistics such as standard deviation and skewness. In addition, it provides a decision-making tool including an estimator of posterior probability that a spatial variable at a given point will belong to various classes representing the severity of the problem of interest and a Bayesian classifier to define the boundaries of subregions belonging to the classes. In this paper, the SANN is implemented to be used for analyzing a mean annual precipitation filed and classifying the field into dry, normal, and wet subregions. For an example, the whole area of South Korea with 39 precipitation sites is applied. Then, several useful results related with the spatial variability of mean annual precipitation on South Korea were obtained such as interpolated field, standard deviation field, and probability maps. In addition, the whole South Korea was classified with dry, normal, and wet regions.

• The KIPS Transactions:PartB
• /
• v.14B no.3 s.113
• /
• pp.153-162
• /
• 2007

Sasang medicine is peculiar medicine that constitution of a human classify four types and differ treatment method by physical constitution. In this way the most important thing is very difficult problem that classification of Sasang constitution and discriminate correctly. Therefore, in this paper targets diagnosis medical appliances development of hybrid form that can behave constitution classification and sees among for this paper to propose about method to grasp characteristic that is morphology about eye, nose, ear and mouth be based on appearance and manner of speaking. In this paper, classified and verified this for Sasang constitution through the QSCC II program at 1 step and present method that more exactly and conveniently analyzing measure each physical constitution feature by survey about eye, nose, ear and mouth at 2 steps. Also, extraction and analyze and verified main area of physical constitution classification based on front face and side face at 3 steps. Such propose method to extraction the principal face region based on face color from front face and side face for correct physical constitution classification diagnosis appliance development through experiment consideration and verification process.

• Journal of KIISE:Software and Applications
• /
• v.31 no.4
• /
• pp.498-504
• /
• 2004

This paper presents a new text classifier based on a feature projection technique. In feature projections, training documents are represented as the projections on each feature. A classification process is based on individual feature projections. The final classification is determined by the sum from the individual classification of each feature. In our experiments, the proposed classifier showed high performance. Especially, it have fast execution speed and robustness with noisy data in comparison with k-NN and SVM, which are among the state-of-art text classifiers. Since the algorithm of the proposed classifier is very simple, its implementation and training process can be done very simply. Therefore, it can be a useful classifier in text classification tasks which need fast execution speed, robustness, and high performance.

• Korean Journal of Remote Sensing
• /
• v.19 no.4
• /
• pp.329-339
• /
• 2003

In this study, an approach of image fusion in decision level has been proposed for unsupervised image classification using the images acquired from multiple sensors with different characteristics. The proposed method applies separately for each sensor the unsupervised image classification scheme based on spatial region growing segmentation, which makes use of hierarchical clustering, and computes iteratively the maximum likelihood estimates of fuzzy class vectors for the segmented regions by EM(expected maximization) algorithm. The fuzzy class vector is considered as an indicator vector whose elements represent the probabilities that the region belongs to the classes existed. Then, it combines the classification results of each sensor using the fuzzy class vectors. This approach does not require such a high precision in spatial coregistration between the images of different sensors as the image fusion scheme of pixel level does. In this study, the proposed method has been applied to multispectral SPOT and AIRSAR data observed over north-eastern area of Jeollabuk-do, and the experimental results show that it provides more correct information for the classification than the scheme using an augmented vector technique, which is the most conventional approach of image fusion in pixel level.

• Journal of Korean Library and Information Science Society
• /
• v.50 no.1
• /
• pp.105-124
• /
• 2019

This study was to suggest the improvement and the application of KDC 6 for classifying children's books by the literature review and survey. First, it was suggested to shorten the classification numbers by the divisions and subdivisions and to expand the classification numbers by sub-subdivisions according to the library-specific classification policy, using the subject statistics of the children's books held by 20 public libraries affiliated in Seoul Metropolitan Office of Education and representative C libraries. Second, the knowledge picture books and the fairy tales were suggested to be classified according to its subject, and the fairy tales in each country were suggested to be classified by adding sub-subdivisions and genre subdivisions. Third, it was suggested to shelve by collection and location codes that were distinguished by the ages and the reading level for user, to prepare a standard guideline for shelving, and to implement the regular user education about the classification system. This study could contribute to the development of the KDC abridged version for children's books in the future.

• Annual Conference on Human and Language Technology
• /
• 2012.10a
• /
• pp.201-205
• /
• 2012

지식검색은 방대한 지식정보 데이터를 바탕으로 사용자의 질문에 대한 답변을 검색하는 시스템이다. 이러한 사용자 참여로 구축된 지식정보는 잘못된 답변으로 인한 신뢰성 부족과 중복 답변 등의 문제점이 있어, 원하는 답변을 찾기 위해서는 지식검색에서 다수의 답변을 읽고 그 답변의 진위여부를 판단해야만 한다. 만일 정답에 포함되는 단어나 어구가 답변들에서 나타내는 통계적 특성을 활용하여 사용자가 원하는 답변을 제시할 수 있다면, 지식검색의 효용성과 신뢰성이 크게 향상될 수 있다. 본 논문에서는 지식정보 데이터 분석을 통해 사용자의 질문의 유형을 단어, 목록, 도표, 글의 4가지 유형으로 분류하고, 각 분류에 대한 사용자 질의어의 답변을 요약하는 방식을 제안한다. 단어, 목록, 글 유형은 TF와 IDF, 어휘 간의 거리 정보를 통해서 중요 단어를 추출하여 각 유형에 적합한 형식의 답변을 사용자에게 제시한다. 도표형은 답변들에서 사용자의 의견 정보를 추출하여 의견 통계를 도표로서 제시한다.

• Proceedings of the KIEE Conference
• /
• 2004.07d
• /
• pp.2540-2542
• /
• 2004

일반적으로 support vector machine (SVM)은 입력 데이터를 두개의 다른 클래스로 구별하는 결정면을 학습을 통하여 구한다. 특히 비분류 문제, 비선형 분류 문제들과 같은 두-클래스 문제를 해결하기 위해 데이터를 고차원의 특정 공간에서 다룬다. 많은 응용분야에서, 각 입력 데이터들은 이 두개의 클래스 중의 하나로 완전히 정의되지 않을 수도 있다. 이러한 문제를 해결하기 위해 우리는 본 논문에서 FSVM(fuzzy support vector machine)을 적용한다. 각 입력 데이터에 퍼지 멤버십(fuzzy membership)을 적용하여 결정면의 학습과정에 입력 데이터들이 다른 기여 (contribution)를 할 수 있게 한다. 본 논문에서는 기준 데이터 집합에 대해 제안된 방법을 실험하고, FSVM이 기존의 SVM보다 더 나음을 보인다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.450-452
• /
• 2012

본 논문에서는 CCD 카메라에 입력된 동영상에서 Random Forest를 이용하여 휴먼 행동을 인식하는 알고리즘을 제안한다. 행동 인식을 위한 특징 벡터 추출을 위해 가장 최근의 N개의 비디오 프레임들을 하나의 액션 볼륨으로 생성하고, 액션 볼륨 내에서 객체 트랙킹 된 영역을 서브 볼륨으로 생성한다. 이후 서브불륨을 $N{\times}N$개의 블록으로 나누고 각 블록에서 HOOF (Histogram of oriented optical flow)를 특징 벡터로 추출한다. 각 휴먼의 행동인식을 위해 사용된 Random Forest 분류기는 걷기, 뛰기, 발차기, 주먹질, 앉기, 쓰러지기, 넘어지기 7개의 행동을 나타내는 클래스로 분류하도록 학습되었으며 Random Forest에 의한 분류결과에 따라 어떤 행동을 취하는지 최종 판단한다.