• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권1호
• /
• pp.368-391
• /
• 2018

Face recognition (FR) with a single sample per person (SSPP) is common in real-world face recognition applications. In this scenario, it is hard to predict intra-class variations of query samples by gallery samples due to the lack of sufficient training samples. Inspired by the fact that similar faces have similar intra-class variations, we propose a virtual sample generating algorithm called k nearest neighbors based virtual sample generating (kNNVSG) to enrich intra-class variation information for training samples. Furthermore, in order to use the intra-class variation information of the virtual samples generated by kNNVSG algorithm, we propose image set based multimanifold discriminant learning (ISMMDL) algorithm. For ISMMDL algorithm, it learns a projection matrix for each manifold modeled by the local patches of the images of each class, which aims to minimize the margins of intra-manifold and maximize the margins of inter-manifold simultaneously in low-dimensional feature space. Finally, by comprehensively using kNNVSG and ISMMDL algorithms, we propose k nearest neighbor virtual image set based multimanifold discriminant learning (kNNMMDL) approach for single sample face recognition (SSFR) tasks. Experimental results on AR, Multi-PIE and LFW face datasets demonstrate that our approach has promising abilities for SSFR with expression, illumination and disguise variations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권12호
• /
• pp.5464-5484
• /
• 2016

Few samples face recognition has become a highly challenging task due to the limitation of available labeled samples. As two popular paradigms in face image representation, sparse component analysis is highly robust while parts-based paradigm is particularly flexible. In this paper, we propose a probabilistic generative model to incorporate the strengths of the two paradigms for face representation. This model finds a common spatial partition for given images and simultaneously learns a sparse component analysis model for each part of the partition. The two procedures are built into a probabilistic generative model. Then we derive the score function (i.e. feature mapping) from the generative score space. A similarity measure is defined over the derived score function for few samples face recognition. This model is driven by data and specifically good at representing face images. The derived generative score function and similarity measure encode information hidden in the data distribution. To validate the effectiveness of the proposed method, we perform few samples face recognition on two face datasets. The results show its advantages.

• 한국정보처리학회논문지
• /
• 제1권1호
• /
• pp.73-80
• /
• 1994

푸리에 서술자는 물체의 경계를 표현하는 일반적인 방법이다. 본 논문에서는 이 푸리에 서술자를 이용하여 물체를 인식할 수 있는 알고리즘을 구축하고 실제로 8가지 의 도구들을 구분 인식할 수 있는지를 직접 적용하여 알고리즘의 타당성 검토와 여러 분야에 응용될 수 있음을 추론하였다. 이 알고리즘 구축은 이미지 형성과정과 물체 인식 과정으로 분리하여 생각한다. 이미지 형성 과정은 가능한 인식을 용이하도록 적 절한 조도, 투영시각, 배경과 물체간에 선명한 대비를 이루는 것이 중요하다. 물체 인식과정은 푸리에 서술자와 경계선 매칭을 사용하여 여러 물체 가운데 최소 거리를 산출한 것을 인식한다. 또한 물체가 회전, 위치 변위, 확대 축소 변형에도 인식이 가 가능한지를 시험한다. 인식 과정을 신속히 수행하기 위해 푸리에 서술자 1024개중에 1/4 만을 사용하여 이를 달성한다.

• Journal of Audiology & Otology
• /
• 제23권1호
• /
• pp.1-7
• /
• 2019

Self-reported difficulties in speech-in-noise (SiN) recognition are common among tinnitus patients. Whereas hearing impairment that usually co-occurs with tinnitus can explain such difficulties, recent studies suggest that tinnitus patients with normal hearing sensitivity still show decreased SiN understanding, indicating that SiN difficulties cannot be solely attributed to changes in hearing sensitivity. In fact, cognitive control, which refers to a variety of top-down processes that human beings use to complete their daily tasks, has been shown to be critical for SiN recognition, as well as the key to understand cognitive inefficiencies caused by tinnitus. In this article, we review studies investigating the association between tinnitus and cognitive control using behavioral and brain imaging assessments, as well as those examining the effect of tinnitus on SiN recognition. In addition, three factors that can affect cognitive control in tinnitus patients, including hearing sensitivity, age, and severity of tinnitus, are discussed to elucidate the association among tinnitus, cognitive control, and SiN recognition. Although a possible central or cognitive involvement has always been postulated in the observed SiN impairments in tinnitus patients, there is as yet no direct evidence to underpin this assumption, as few studies have addressed both SiN performance and cognitive control in one tinnitus cohort. Future studies should aim at incorporating SiN tests with various subjective and objective methods that evaluate cognitive performance to better understand the relationship between SiN difficulties and cognitive control in tinnitus patients.

• 한국지역지리학회지
• /
• 제16권5호
• /
• pp.457-473
• /
• 2010

본 연구에서는 재해 및 방재 관련 지명의 유형과 분포, 사례지역 조사를 토대로 지명을 통한 재해 및 방재 가능성을 탐색하였다. "한국지명총람"을 대상으로 106개의 검색 지명어에서 37,901개의 지명을 추출하였으며, 연구결과는 다음과 같다. 지명의 유형별 빈도는 지형재해 및 방재 관련 지명이 월등히 많았고, 특히 호우, 범람, 침수 재해와 관련된 지명이 탁월하였다. 지역적 분포는 영 호남 지역의 점유율이 높은 가운데 수(水), 사(沙), 야(野), 우(雨), 상(狀) 등이 전국적인 분포를 보인 반면 둠벙, 구렁, 여울, 탄(灘), 방죽, 제(提), 지(池) 등은 지역차가 커서 입지 및 지형적 특성이 지명에 투영되어 있음을 확인할 수 있었다. 사례지역 조사에서 우리나라는 기상현상과 지형조건이 결합된 범람 및 침수 재해의 가능성이 높아 수계와 곡지형에 대한 관리의 필요성이 제기되었다.

• 대한한방내과학회지
• /
• 제38권3호
• /
• pp.336-352
• /
• 2017

Objective: This study investigated the perception, utilization, and satisfaction (in the general population), of the common cold treatment at the Korean Medical Clinic, to provide basic information for the development of Korean medicine clinical practice guidelines for the common cold. Method: A questionnaire was developed that consisted of questions about the general perception, utilization status, degree of satisfaction, willingness to use, and the improvement of the common cold treatment at the Korean Medical Clinic. An online survey was conducted using this questionnaire. Results: Three-hundred subjects responded to the survey. The results of the analysis were as follows. 1. 73.7% of the subjects recognized the common cold treatment at the Korean Medical Clinic. 2. 72% of the subjects had a positive perception of Korean medicine for common cold treatment. The major reason for the positive perception was the expectation about improving immunity and preventing recurrence. 3. Only 20% of the subjects had visited the Korean Medical Clinic for common cold treatment. The expensive cost was the major reason for not visiting the Korean Medical Clinic for common cold treatment. 4. The ratio of subjects who were willing to visit the Korean Medical Clinic for common cold treatment was 70%. 5. The expansion of health insurance coverage (67.7%), the activation of public relations (54.7%), and the development of a new herb medicine preparation that was easy to take (43.3%) were found to be necessary for improving the Korean medical service for common cold treatment. Conclusion: In spite of high recognition and positive perception, actual utilization of common cold treatment at the Korean Medical Clinic was relatively low. Multifaceted efforts are necessary to enhance the competitiveness of Korean medical service for treatment of the common cold.

• 한국컴퓨터정보학회논문지
• /
• 제21권5호
• /
• pp.167-175
• /
• 2016

The volatile component analysis of 82 commercialized liquors(44 samples of single malt whisky, 20 samples of blended whisky and 18 samples of brandy) was carried out by gas chromatography after liquid-liquid extraction with dichloromethane. Pattern recognition techniques such as principle component analysis(PCA), cluster analysis(CA), linear discriminant analysis(LDA) and partial least square discriminant analysis(PLSDA) were applied for the discrimination of different liquor categories. Classification rules were validated by considering sensitivity and specificity of each class. Both techniques, LDA and PLSDA, gave 100% sensitivity and specificity for all of the categories. These results suggested that the common characteristics and identities as typification of whiskies and brandys was founded by using multivariate data analysis method.

• ETRI Journal
• /
• 제22권2호
• /
• pp.1-10
• /
• 2000

Pattern recognition is one of the most common problems encountered in engineering and scientific disciplines, which involves developing prediction or classification models from historic data or training samples. This paper introduces a new approach, called the Representational Capability (RC) algorithm, to handle pattern recognition problems using radial basis function (RBF) models. The RC algorithm has been developed based on the mathematical properties of the interpolation and design matrices of RBF models. The model development process based on this algorithm not only yields the best model in the sense of balancing its parsimony and generalization ability, but also provides insights into the design process by employing a design parameter (${\delta}$). We discuss the RC algorithm and its use at length via an illustrative example. In addition, RBF classification models are developed for heart disease diagnosis.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2008년도 추계학술대회논문집
• /
• pp.675-680
• /
• 2008

This paper proposes a way to measure the displacement of a multi-point by using a pattern recognition from video signal. Generally in measuring displacement, gab sensor, which is a displacement sensor, is used. However, it is difficult to measure displacement by using a common sensor in places where it is unsuitable to attach a sensor, such as high-temperature areas or radioactive places. In this kind of places, non-contact methods should be used to measure displacement and in this study, images of CCD camera were used. When displacement is measure by using camera images, it is possible to measure displacement with a non-contact method. It is simple to install and multi-point displacement measuring device so that it is advantageous to solve problems of spatial constraints.

• 대한전자공학회논문지
• /
• 제27권8호
• /
• pp.1275-1283
• /
• 1990

Perception of surfaces from range images plays a key role in 3-D object recognition. Recognition of 3-D objects from range images is performed by matching the perceived surface descriptions with stored object models. The first step of the 3-d object recognition from range images is image segmentation. In this paper, an approach for segmenting 3-D range images into symbolic surface descriptions using spatial Gabor filter is proposed. Since the phase of data has a lot of important information, the phase information with magnitude information can effectively segment the range imagery into regions satisfying a common homogeneity criterion. The phase and magnitude of Gabor filter can represent a unique featur vector at a point of range data. As a result, range images are trnasformed into feature vectors in 3-parameter representation. The methods not only to extract meaningful features but also to classify a patch information from range images is presented.