영상에서 윈도우 배치에 따른 통계적 에지검출 비교

임동훈,Lim, Dong-Hoon 한국통계학회 2009 Communications for statistical applications and me Vol.16 No.4

In this paper we describe Wilcoxon test and T-test that are well-known in two-sample location problem for detecting edges under different window configurations. The choice of window configurations is an important factor in determining the performance and the expense of edge detectors. Our edge detectors are based on testing the mean values of local neighborhoods obtained under the edge model using an edge-height parameter. We compare three window configurations based on statistical tests in terms of qualitative measures with the edge maps and objective, quantitative measures as well as CPU time for detecting edge.

잡음영상에서 로버스트 순위-순서 검정을 이용한 효과적인 에지검출

임동훈,Lim, Dong-Hoon 한국통계학회 2007 응용통계연구 Vol.20 No.1

Edge detection has been widely used in computer vision and image processing. We describe a new edge detector based on the robust rank-order test which is a useful alternative to Wilcoxon test. Our method is based on detecting pixel intensity changes between two neighborhoods with a $r{\times}r$ window using an edge-height model to perform effectively on noisy images. Some experiments of our robust rank-order detector with several existing edge detectors are carried out on both synthetic images and real images with and without noise.

다시 격조사는 핵이다

임동훈(Lim Dong-hoon) 형태론 2008 형태론 Vol.10 No.2

This paper is a reply to Han(2003). Van Valin & LaPolla(1997) classified English prepositions into argument-marking prepositions and adjunct-marking prepositions, and argued that the former functions were case-markers. Han(2003) applied this opinion to Korean case-markers, and insisted that argument-marking case-markers in Korean should be considered as non-heads. But what's more important in determining the head of the construction is to find out which part is responsible for the distribution and function of the whole construction rather than to just identify whether the preceding word is an argument or not. For example, Han(2003) argued that uy in muncey-uy haykyel is a non-head because muncey is an argument of haykyel, and uy in kuktho-uy myencek is a head because kuktho is an adjunct of myencek. But in both constructions, X and X-uy have different distributions and functions, so in this regard it is reasonable that they both should be considered as heads. Besides, case alternations such as X-wa/lul mannata weaken Han(2003)'s argumentation.

‘은/는’과 종횡의 의미 관계

임동훈 ( Lim Dong-hoon ) 국어학회 2012 국어학 Vol.64 No.-

This paper investigates the meaning and uses of Korean particle ‘un/nun’ under the perspective of pragmatic relations. In particular it is attempted to show how ‘un/nun’ can be exploited to express informational topic, focus in collaboration with accents. The findings of this paper are as follows. First, the semantic properties of topic aren't confined to be definite, generic, and specific but contrastive non-specific elements can be topics if the contrastive set including it is construed to be definite or specific. Second, the contrastive meaning of ‘un/nun’ makes it possible that it can be used to express shift topics, derived topics from hypertheme, focus-chain topics. Third, ‘un/nun’ marked topics are realized with L+H* accents when contrastive items are activated, and otherwise with H+!H accents which simply implicate familiarity. Fourth, foci are divided into two groups where one is syntagmatic focus which means a focus in the pragmatic focus-ground articulation, the other is paradigmatic focus. And two kind of foci may be congruent or not with each other. Fifth, the main function of Korean particles is to express a paradigmatic focus. The focus element in the selectional relation generates a alternative set which provides quantificational domain for paradigmatic operations such as contrast, exclusion, addition, second best, comparison. Sixth, paradigmatic foci can take effect on truth conditions, and express the meaning of exhaustivity and scalar implicature depending on the contexts. But these properties aren't enough to define a paradigmatic focus. Seventh, ‘un/nun’ marked paradigmatic foci can be realized into contrastive topic or contrastive focus according to whether they appear in the domain of topic or in the domain of comment. Syntagmatic foci are realized with H* accents, and contrastive topics with L+H* accents, contrastive foci with H* or H*+L accents. Eighth, morphologically while case markers are exploited to express syntagmatic foci, particles are exploited to express paradigmatic foci.

장형 부정문의 생성과 ‘못’ 부정문의 분화

임동훈 ( Lim Dong-hoon ) 국어학회 2020 국어학 Vol.0 No.93

This paper starts from three research questions. First, why does Korean have asymmetric long-form negation in addition to symmetric short-form negation? What is the syntactic structure of long-form negative sentences. Second, what is the dividing factor into two negative morphemes ‘an’, ‘mos’ in Korean declarative and interrogative sentences? Third, while short-form ‘mos’ negation holds inferior meaning, long-form ‘mos’ negation does not have it. What makes this difference? As to three questions, this paper tries to answer as follows. First, long-form negative sentence is a marked device to indicate quantificational focus of negative morphemes. Second, long-form negative sentences are not complex, but simple sentences. Third, ‘mos’ negation represents event negation while ‘an’ negation represents propositional negation. Fourth, event negation developed from inferior meaning. Short-form negation holds persistent inferior meaning. But long-form negation does not have inferior meaning because its genesis is related with the need of indicating quantificational focus. Fifth, event negation has semantic shifts according to situation types which it attaches to.

보조사의 의미론

임동훈 ( Lim Dong-hoon ) 국어학회 2015 국어학 Vol.73 No.-

This paper aims to define the concept of ‘bojosa’s in Korean which differentiates them from case markers, and establish the methodology of describing their meanings, and to reveal the system within their different meanings and uses. Our findings are as follows. First, ‘bojosa’s are paradigmatic markers which indicate the semantic relation between the referents of preceding phrase and contextually contrastive entities, while case markers are syntagmatic markers which mark dependent nouns for the type of relationship they bear to their heads. Second, while case markers are used to mark the informational focus, ‘bojosa’s are used to mark the contrastive focus and invoke alternative sets. So ‘bojosa’s are defined as focus particles. Third, ‘bojosa’s and their focus aren’t always adjacent because they can be adjoined to the most important part of the focus. But the focus can’t be extended beyond the c-command domain of ‘bojosa’s. Fourth, ‘bojosa’s are subdivided into additive, exclusive, comparative, extent classes in terms of the semantic relation between the focus and their alternative sets. Exclusive ‘bojosa’s are subdivided into excluding one and excluded one. The former is again divided into weak one and strong one, and the latter represent the result of exclusion. Fifth, ‘bojosa’s have semantic shift according as their alternative sets are scalar or non-scalar. Exclusive ‘bojosa’ man show the reversal of scale according as it is on a sufficient context or a necessary context. Sixth, the semantic relation between (i)na and do is orthogonal, (i)na and man have common scalar values, (i)rado has both the properties of (i)na and do. Seventh, adverbs jom, hanbeon, da are under the process of grammaticalization into ‘bojosa’s, namely focus particles in Modern Korean in terms of their combinatorial properties and their function of invoking alternative sets.

한국어의 장소 표시 방법들

임동훈 ( Lim Dong-hoon ) 국어학회 2017 국어학 Vol.82 No.-

This paper aims to explore how the local concepts are expressed in Korean. The findings are as follows. First, there are three dimensions of local concepts, that is, that of locative vs. directional, that of source vs. goal and that of interior vs. non-interior. In addition to these, the concept of coherence works in Korean. Coherence involves contact or penetration. Local case marker eyse functions as an interior space marker, while ey without se functions as non-interior space marker. Second, the concept of goal is subdivided in terms of degree of intimacy, that is, the concept of coherence. Local case marker ey is coherent, while lo is incoherent. The concept of coherence is related to the division of point vs. area. Coherent ey presupposes the concept of area, while incoherent lo does not need to. Third, interior space marker eyse functions as ablative marker as well. Incoherent marker lo can be used as a marker of perlative case. Fourth, adverbials of distance and duration can be combined with accusative marker ul when they denote boundedness. When the noun phrases of source/goal are conceptualized into an area, they are realized with local case markers eyse, ey. However, they are realized with accusative case marker ul when they are conceptualized into a point. Fifth, when the noun phrase of local concepts is considered to be that of an affected referent in the situation type of accomplishment, they are combined with accusative marker ul. In this case, ul-marked noun phrase shows a holistic interpretation. Sixth, the above phenomena point to the fact that Korean local case systems are organized in terms of aspect category.

` - 겠 - ` 의 용법과 그 역사적 해석

임동훈(Dong Hoon Lim) 국어학회 2001 국어학 Vol.37 No.-

현대국어의 `겠-`은 다양한 의미와 용법을 가지고 있다. 본고는 이에 대해 먼저 `-겠-`이 가지고 있는 의미와 용법들을 면밀히 고찰하여 이들 간의 관계를 알아보고, 이를 바탕으로 `-겠-`의 통시적 변화 과정을 재구하는 작업을 하려 하였다. 그 다음에는 반대로 `-겠-`의 기원과 그것의 역사적 발달 과정을 탐구하여 `-겠-`의 다양한 의미와 용법들을 설명하는 작업을 진행하려 하였다. 전자는 공시적 현상을 통한 역사적 과정의 재구요, 후자는 역사적 과정을 통한 공시적 현상의 설명이라고 할 수 있다. 이러한 논의 결과 정리된 내용은 다음과 같다. 첫째, `-겠-`의 기원적 의미는 `예정`이다. 둘째, `-겠-`의 의미 중에서는 `추측`이 가장 후대의 의미이고 `의지`가 가장 앞선 의미이며, 후대의 의미일수록 제약이 적고 앞선 단계의 의미일수록 제약이 많다. 셋째, `-겠-`은 역사적으로 `-게 ㅎ·엿-` 구성에서 `-어 있-` 앞의 `-게 ㅎ·-`가 피동적으로 해석되어 사동 구성과 분기되면서 발달하였다. 넷째, `-겠-`은 `예정>의지, 능력(가능성)>미래>추측`이라는 4단계의 문법화 경로를 거쳤는데, 이러한 문법화는 작용역의 확대를 가져왔다.

한국어 조사 연구의 현황과 전망

임동훈(Dong Hoon Lim) 한국어학회 2002 한국어학 Vol.16 No.-

This paper aims at reviewing the past 50 years of studies on the particles in Modern Korean. Being based on this review. we tried to find out new directions for the study of Korean particles. As widely known. Korean Particles have lots of elements and lots of functions, and besides they have diverse distributions, so it is very difficult to define and classify them in terms of general linguistics. Korean particles are not so obligatory as Korean endings, and the combinations between them are not restricted by rules. In this respect this kind of work will be required and meaningful. This paper classify Korean particles into two groups, one is particles whose main functions are case-marking, the other is particles whose main functions are delimiting the preceding word semantically. About case marking particles, we dealt with the six issues; (1) the nature of case and its domain in Korean, (2) the sub-class of case markers, (3) the constraint on the combinations of particles, (4) the meanings and functions of particles, (5) the syntactic status of case marker phrases, (6) the constructions concerning particles. And about meanings.

현대국어 경어법의 체계

임동훈 ( Dong Hoon Lim ) 국어학회 2006 국어학 Vol.0 No.47