• 한국컴퓨터정보학회논문지
• /
• 제20권11호
• /
• pp.9-16
• /
• 2015

We propose a hybrid pattern recognition method that effectively combines two different features for improving data classification. We first extract the PCA (Principal Component Analysis) and LDA (Linear Discriminant Analysis) features, both of which are widely used in pattern recognition, to construct a set of basic features, and then evaluate the separability of each basic feature. According to the results of evaluation, we select only the basic features that contain a large amount of discriminative information for construction of the combined features. The experimental results for the various data sets in the UCI machine learning repository show that using the proposed combined features give better recognition rates than when solely using the PCA or LDA features.

• 대한기계학회논문집
• /
• 제16권2호
• /
• pp.259-266
• /
• 1992

For CAD/CAPP integration, part information on not only geometry but also machining characteristics should be delivered and commonly used between designers and process planners. In this study, the machining features, as linking factors of the integration, are represented as the combination of functional features and atomic features and grouped into a hierarchical database. And the feature based modelling approach is used by generating information on the machining features in design stage. These features are drawn by analyzing real decision rules of process planners. The database using the machining features is built and used for application modules of process planning, operation planning and standard time estimation.

• 대한의용생체공학회:의공학회지
• /
• 제15권2호
• /
• pp.201-208
• /
• 1994

Many three-dimensional object modeling and display methods for computer graphics and computer vision have been developed. Recently, with the help of medical imaging devices such as computerized tomography, magnetic resonance image, etc., some of those object modeling and display methods have been widely used for capturing the shape, structure and other properties of real objects in many medical applications. In this paper, we propose the reconstruction and display method of the three-dimensional object from a series of the cross sectonal image. It is implemented by using the automatic threshold selection method and the contour following algorithm. The combination of curvature and distance, we select feature points. Those feature points are the candidates for the tiling method. As a results, it is proven that this proposed method is very effective and useful in the comprehension of the object's structure. Without the technician's responce, it can be automated.

• Journal of Information Processing Systems
• /
• 제10권4호
• /
• pp.555-567
• /
• 2014

Biometric performance improvement is a challenging task. In this paper, a hierarchical strategy fusion based on multimodal biometric system is presented. This strategy relies on a combination of several biometric traits using a multi-level biometric fusion hierarchy. The multi-level biometric fusion includes a pre-classification fusion with optimal feature selection and a post-classification fusion that is based on the similarity of the maximum of matching scores. The proposed solution enhances biometric recognition performances based on suitable feature selection and reduction, such as principal component analysis (PCA) and linear discriminant analysis (LDA), as much as not all of the feature vectors components support the performance improvement degree.

• Journal of Information Processing Systems
• /
• 제13권6호
• /
• pp.1628-1639
• /
• 2017

The combination texture feature extraction approach for texture image retrieval is proposed in this paper. Two kinds of low level texture features were combined in the approach. One of them was extracted from singular value decomposition (SVD) based dual-tree complex wavelet transform (DTCWT) coefficients, and the other one was extracted from multi-scale local binary patterns (LBPs). The fusion features of SVD based multi-directional wavelet features and multi-scale LBP features have short dimensions of feature vector. The comparing experiments are conducted on Brodatz and Vistex datasets. According to the experimental results, the proposed method has a relatively better performance in aspect of retrieval accuracy and time complexity upon the existing methods.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 1996년도 10월 학술대회지
• /
• pp.349-354
• /
• 1996

The phonetic mastery of English has been considered next to impossible to many non-native speakers of English, including even some teachers of English. This paper takes issue with this phonetic problem of second language acquisition and proposes that combination of cognitive and physical approaches can help master English faster and more easily.

• 동굴
• /
• 제78호
• /
• pp.1-8
• /
• 2007

The combination of mechano-chemical activation (MCA)and Self-propagating High-temperature Synthesis (SHS) has widened the technical possibilities for both methods. For YBCO systems the investigation showed that a short-term MCA of initial powders before SHS leads to single-phase and ultra-fine products. A new technique for preparation ultra-fine high-temperature superconductors (HTS) of YBCO composition with a grain size d <1m is developed using combination of MCA and SHS. The specific feature of the technique is formation of the $YBa_2Cu_3O_7-$ crystalline lattice directly from an X-ray amorphous state arising as a result of mechanical activation of the original oxide mixture. The technique allows the stage of formation of any intermediate reaction products to be ruled out. X-ray and magnetic studies of ultra-fine high temperature superconductors are carried out. Dimension effects associated with the microstructure peculiarities are revealed. A considerable enhancement of inter-grain critical currents is found to take place in the ultra-fine samples.

• 한국컴퓨터정보학회논문지
• /
• 제10권6호
• /
• pp.1-8
• /
• 2005

얼굴 분석은 얼굴 인식 머리 움직임과 얼굴 표정을 이용한 인간과 컴퓨터사이의 인터페이스, 모델 기반 코딩, 가상현실 등 많은 응용 분야에서 유용하게 활용된다. 이러한 응용 분야에서는 얼굴의 특징점들을 정확하게 추출해야 한다. 본 논문에서는 눈, 눈썹, 입술의 코너와 같은 얼굴 특징을 자동으로 추출하는 방법을 제안한다. 먼저, 입력 영상으로부터 AdaBoost 기반의 객체 검출 기법을 이용하여 얼굴 영역을 추출한다. 그 다음에는 계곡 에너지. 명도 에너지, 경계선 에너지의 세 가지 특징 에너지를 계산하여 결합한다. 구해진 특징 에너지 영상에 대하여 에너지 값이 큰 수평 방향향의 사각형을 탐색함으로써 특징 영역을 검출한다. 마지막으로 특징 영역의 가장자리 부분에서 코너 검출 알고리즘을 적용함으로써 눈, 눈썹, 입술의 코너를 검출한다. 본 논문에서 제안된 얼굴 특징 추출 방법은 세 가지의 특징 에너지를 결합하여 사용하고 계곡 에너지와 명도 에너지의 계산이 조명 변화에 적응적인 특성을 갖도록 함으로써, 다양한 환경 조건하에서 견고하게 얼굴 특징을 추출할 수 있다.

• 한국컴퓨터정보학회논문지
• /
• 제28권4호
• /
• pp.31-39
• /
• 2023

본 논문에서는 이종 모델의 특징맵 간 상관관계인 외부적 상관관계와 동종 모델 내부 특징맵 간 상관관계인 내부적 상관관계를 활용하여 교사 모델로부터 학생 모델로 지식을 전이하는 Internal/External Knowledge Distillation (IEKD)를 제안한다. 두 상관관계를 모두 활용하기 위하여 특징맵을 시퀀스 형태로 변환하고, 트랜스포머를 통해 내부적/외부적 상관관계를 고려하여 지식 증류에 적합한 새로운 특징맵을 추출한다. 추출된 특징맵을 증류함으로써 내부적 상관관계와 외부적 상관관계를 함께 학습할 수 있다. 또한 추출된 특징맵을 활용하여 feature matching을 수행함으로써 학생 모델의 정확도 향상을 도모한다. 제안한 지식 증류 방법의 효과를 증명하기 위해, CIFAR-100 데이터 셋에서 "ResNet-32×4/VGG-8" 교사/학생 모델 조합으로 최신 지식 증류 방법보다 향상된 76.23% Top-1 이미지 분류 정확도를 달성하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권11호
• /
• pp.2720-2736
• /
• 2013

Multi-view face detection has become an active area for research in the last few years. In this paper, a novel multi-view human face detection algorithm based on improved real Adaboost is presented. Real Adaboost algorithm is improved by weighted combination of weak classifiers and the approximately best combination coefficients are obtained. After that, we proved that the function of sample weight adjusting method and weak classifier training method is to guarantee the independence of weak classifiers. A coarse-to-fine hierarchical face detector combining the high efficiency of Haar feature with pose estimation phase based on our real Adaboost algorithm is proposed. This algorithm reduces training time cost greatly compared with classical real Adaboost algorithm. In addition, it speeds up strong classifier converging and reduces the number of weak classifiers. For frontal face detection, the experiments on MIT+CMU frontal face test set result a 96.4% correct rate with 528 false alarms; for multi-view face in real time test set result a 94.7 % correct rate. The experimental results verified the effectiveness of the proposed approach.