Earticle

본 연구에서는, 사전 정보 학습을 통한 사이노그램 도메인에서의 저선량 CT 재구성을 수행하였다. 해당 복원 방법은 희소 데이터의 성질과 사전에 특성에 대한 성분을 훈련 시킨 정보를 기반으로 학습된 정보를 사용한다. 제안된 재구성 알고리즘을 구현하고 희소 뷰 CT 재구성에 대한 유용성을 입증하기 위해 수학적인 팬텀을 사용하여 체계적인 시뮬레이 션을 수행하였다. 제안한 방법의 타당성을 검증하기 위해 필터링 후 역투영법 (FBP)을 통해 재구성된 영상과 제안하는 방법으로 재구성한 영상의 품질을 조사하였다. 우리의 결과는 FBP 기반의 재구성 방법에 의해서 적은 투영 데이터에서 도 줄무늬 아티팩트를 줄이는데 효과적인 것으로 나타났다.

In this study, we perform a low-dose reconstructed CT with inpainting technique usint the dictionary learning in sinogram domain. This restoration method is based upon the representation learning using a dictionary functions and property of sparse data. We implemented the proposed reconstruction algorithm and performed a systematic simulation using a numerical phantom to demonstrate its viability for sparse-view CT reconstruction. The reconstruction qualities for the two reconstruction techniques of filtered back-projection (FBP) and proposed method were investigated. Our results demonstrate that the proposed approach seems to be effective for reducing the streak artifacts considerably in sparse-view CT reconstruction.

본고는 한국어를 학습하는 중국인 학습자를 대상을 중심으로 중국내에서 출간된 이중언어 사전인 ≪新世纪韩汉词典≫의 범례와 부록에 대해서 살펴보았다. 이를 통해 한중 이중언어 학습사전의 범례와 부록의 역할과 설치 원칙을 밝혔다. 사전 외부정보구조의 일부분으로써의 범례는 책의 첫머리에 그 책의 체제 설명이고 사전 본문 앞에서 유일하게 실제 의미를 가진 부분이다. ≪新世纪≫의 범례는 독자에게 제시와 인 도라는 역할을 잘 하였지만, 부족한 부분도 많이 존재한다. 언어 면에서 보면, 말이 번잡하고 차례가 명확하지 않다. 그리고 한국어로 작성하였기 때문에 중국인 학습자에게 큰 도움이 되 지 못한다고 본다. 내용 면에서 보면 표제어 부분에서 표제어의 선정 기준과 동형이의어의 배 열 원칙을 제시하지 않았다. 뜻풀이 부분에서 분류가 혼란스럽고 일관성과 체계성이 부족하 다. 한국어 표제어의 의항과 중국어 대응어의 배열 원칙에 대해서도 설명하지 않았다. 그리고 용례, 어원, 부호, 그리고 표기법 등을 뜻풀이 부분에서 설명하는 것도 적절하지 않다. 발음표 기 부분에서 한국어 표제어의 발음에 대해서 언급하지 않았다. 이는 중국인 학습자를 대상으 로 한다는 편찬 원칙과 일치하지 않다. 부호 부분에서 부호를 누락하거나 사용이 적절하지 못 한 경우도 존재한다. 범례는 전면적으로 사전의 체제를 소개해야 하고, 사전 내용과 체제에 무관한 문자를 버려야 하고, 그림과 문자를 모두 사용해야 한다고 생각한다. 부록은 사전 본문 끝에 덧붙여 부가와 보충 역할을 하는 각종 자료이며 사전 외부정보구조 의 중요한 부분이다. ≪新世纪≫ 부록의 길이와 수량은 대체적으로 적절하다. 그리고 일부분 내용은 자료 가치가 있고 독창성도 있다. 하지만 일치성과 상관성, 필요성과 실용성이 없을 뿐만 아니라 두 언어로 작성하지도 않았다. 이는 부록을 설치하는 원칙을 제대로 준수하지 않 았고, 사전의 편찬 목적에 따라 본문 내용과 사전 거시구조를 위해 제대로 역할을 하지 못한 다는 것이다. 한중 학습사전의 부록을 작성하는 데에 각종 학습사전의 장점을 빌릴 수 있지만 중국인 학습자의 입장에서 특색 있게 하여야 한다고 본고에서 주장하였다.

In view of a series of problems regarding the settings of Guide and Appendix of external information structure of Korean-Chinese Dictionary published in Chinese domestic, based on the investigation on the functions and features of the dictionary external information structure, along with commenting on the merits and demerits of the settings of Guide and Appendix in <The New Century> which is a Learning Bilingual Dictionary specilized for Chinese people to learn Korean published in Chinese domestic, this article will explore the function and setting principles of the Guide and Appendix in Korean-Chinese Learning Bilingual Dictionarys. The Guide, as the dictionary’s external information which lies in the front of the dictionary is the explanation of the dictionary’s style and the only meaningful part of dictionary information before the whole body . The Guide of < The New Century> offers better tips, guidance and aids to the readers, but to some extent, there are still some deficiency. From the language point of view, it uses Korean as its written language, but the used language is burdensome and the language level is not clear. Also, the illustrations in this dictionary is not proper. From the content point of view, the headword section is not indicated with the basis and standards of the selected word as well as the layout principles of homographs language. As for the interpretation section, it shows confusion and lack of hierarchy, logic and systematic classification and it also lacks of explaination of corresponding word senses and organization principle of Chinese. Additionally, it puts illustration, etymology, symbols, signs, word mark methods which do not belong to the contents of the interpretation section on the interpretation section. From the phonetic aspect, the Guide of <The New Century> involves no Korean pronunciation labeling principles, which is contrary to the compilation of the dictionary. From the sigh sympol point of view, it records various symbols and abbreviations in the body of the dictionary in detail, but shows some omissions and improper use. The Guide should be built on the basis of a comprehensive introduction to the dictionary style, abandoning unrelated dictionary content and layout style text and carefully use words and illustrations. The Dictionary Appendix is attached to the back of the dictionary playing the additional and complementary role to offer various types of information. It is an organic part of the dictionary external information structure. The length of Appendix of <The New Century> is basically desirable, and some part of contents is informative and novel. But overall, it lacks consistency and relevance, necessity and practicality, as well as contrast. That may because it mechanically copys some Korean dictionaries in South Korea. It does not follow the Appendix setting principles to serve in the overall structure of the dictionary. To create a Korean-Chinese Learning Dictionary Appendix, we can draw on the strengths of the various learning dictionaries, but the most important thing is to stand on the position of Korean learners, which means us Chinese, and also have their own characteristics. The research of Guide and Appendix settings of learning Korean Dictionary only offers us a reference instead of a all-acceptance in the compilation of Guide and Appendix in Korean-Chinese Learning Dictionary. Lastly, I sincerely hope this article will do some help for the compilation of Korean-Chinese Learning Dictionary.

The rapid global expansion of Chinese language education has generated an urgent demand for learner-oriented lexicographical resources. Traditional Chinese dictionaries, primarily designed for native speakers, emphasize linguistic standardization and descriptive completeness, yet often fail to accommodate the cognitive, pragmatic, and cross-cultural needs of second language learners. As a result, their practical value for international learners remains limited. Against this background, this study examines the necessity of constructing a learning-oriented basic Chinese learner’s dictionary, defined as a dictionary designed for beginning and lower-intermediate learners with the primary aim of supporting vocabulary comprehension, acquisition, and use. Drawing on pragmatics, cognitive linguistics, and second language acquisition theory, the paper systematically reviews recent Chinese and international scholarship on learner dictionaries, identifies structural and functional limitations in existing resources, and proposes an integrated analytical framework for dictionary construction. The proposed framework emphasizes learner orientation, functional system integration, technological support, and cross-cultural adaptability. On this basis, the study outlines a set of construction strategies, including corpus-informed entry selection focused on high-frequency vocabulary, hierarchically organized definitions progressing from core to extended senses, contextualized examples with controlled difficulty, pragmatic guidance addressing register, politeness, and usage constraints, and culturally informed cross-linguistic annotation. Particular attention is given to cross-cultural adaptation, arguing that simple bilingual equivalence is insufficient to represent culturally embedded meanings in Chinese vocabulary and may lead to semantic misalignment or pragmatic misuse. Rather than focusing on technical implementation details, the paper conceptualizes technology as an enabling mechanism that supports platform-based operation, multimodal presentation, learner feedback, and iterative content optimization. From a theoretical perspective, the study contributes a coherent framework that bridges lexicography and Chinese-as-a-foreign-language pedagogy. Methodologically, it proposes an operational pathway that integrates corpus-based selection, learner-centered definition design, pragmatic enrichment, and multimodal support. Practically, the findings offer implementable guidance for dictionary compilers and educational institutions engaged in resource development for international Chinese education. Overall, the study argues that learning-oriented basic Chinese learner’s dictionaries must move beyond traditional paradigms and adopt an integrated, learner-responsive model to effectively support second language vocabulary development.

In this work, a new blind separation algorithm for permuted alias image based on dictionary learning is proposed according to a type of permuted alias image with noise diversity. Sparse representation of permuted alias image is obtained by dictionary learning method, since it has high adaptability and its representation result has higher sparsity degrees than that of parameter dictionary. An optimal permuted alias image is achieved by conducting sparse representation with K-SVD dictionary learning algorithm restrained with nonzero element number. The size and the location of permuting region is found by detecting the subtraction image, which is defined as the difference between the reconstructed permuted alias image and the original permuted alias image. The permuting region is optimized by implementing image morphological operation and is separated from the permuted alias image by the threshold. Experimental results show that the permuting sub-images can be efficiently separated from the permuted alias image, which is not affected by the size, location, number of permuting sub-images and noise level of the permuting sub-images.

Signal sparse matrix structure, the degree of relationship between signal sparse representation, which affect application of compression perception to the effect of recovery reconstruction for signal. In order to solve this problem, a variety of dictionary learning algorithm such as KSVD, OLM (Online dictionary learning method) should be put forward. These algorithms used overlapping image blocks to build a dictionary, produced a large number of sparse coefficients, resulting in a fitting and calculation too slowly, and cannot ensure convergence. Based on this, it designed a fast dictionary learning algorithm based on proximal gradient. Algorithm based on the analysis of proximal gradient multiple, on the basis of convex optimization problem, applied to the dictionary learning involved in solving optimization, reduce the complexity of each iteration, reduces the iterative overhead, at the same time to ensure the convergence. Experiments on synthetic data show that the proposed algorithm dictionary learning speed, the time is short, and obtain a better dictionary.

In allusion to the partial texture information loss during image deniosing process, an image denoising algorithm based on non related dictionary learning is proposed in this article. In this algorithm, the noise image is firstly divided into mutually overlapped image blocks, and a certain quantity of these image blocks are randomly selected for subsequent dictionary learning; then, non related dictionary learning technology is adopted to obtain the redundant dictionary with relatively strong irrelevance; finally, the sparse encoding algorithm is adopted to obtain the sparse representation coefficient of each image block in the redundant dictionary, and such sparse representation coefficients are used to recover the original image. The experiment result shows: since the redundant dictionary obtained through non related dictionary learning technology can strongly represent the image texture information, PSNR (Peak Signal to Noise Ratio) of the algorithm proposed in this article is superior to that of the existing advanced algorithm, and the algorithm can well keep the image detail and texture information, thus to improve visual effect.

For hyperspectral remote sensing image denoising, this paper proposed image denoising based on non-local low-rank dictionary learning. The basic idea of algorithm is to use strong relativity of all wave bands of hyperspectral remote sensing image with local self-similarity and local sparsity of image to improve the denoising performance. First of all, combined with the strong relativity, non-local self-similarity and local sparsity, non-local low-rank dictionary learning is established. Then iterative method is used to solve the model to get redundant dictionary and sparsity to represent coefficient. Finally, redundant dictionary and sparsity is used to express restored image of coefficient. Compared with the existing advanced algorithm, by making full use of strong relativity each band of hyperspectral image, it makes the algorithm obtain the information on details to well keep the hyperspectral remote sensing image, to improve the visual effect. Experimental results verify the effectiveness of the algorithm in this paper.

Light field 카메라를 이용하여 영상을 취득한 후 다양한 응용 프로그램으로 확장이 가능한 4차원 light field 영상은 일반적인 2차원 공간영역(spatial domain)과 추가적인 2차원 각영역(angular domain)으로 구성된다. 그러나 이러한 4차원 light field 영상을 유한한 해상도를 가진 2차원 CMOS 센서로 취득하므로 저해상도의 제약이 존재한다. 본 논문에서는 이러한 4차원 light field 영상이 가지는 해상도 제약 조건을 해결하기 위하여, 4차원 light field 영상에 적합한 딕셔너리 학습 기반(dictionary learning-based) 초해상도(superresolution) 알고리즘을 제안한다. 제안하는 알고리즘은 4차원 light field 영상으로부터 추출한 많은 수의 4차원 패치(patch)들을 바탕으로 딕셔너리를 구성 및 훈련하며, 학습된 딕셔너리를 바탕으로 저해상도 입력 영상의 해상도를 향상시키는 과정을 수행한다. 제안하는 알고리즘은 공간영역과 각영역의 해상도를 동시에 각각 2배 향상시킨다. 실험에 사용된 영상은 상용 light field 카메라인 Lytro에서 취득하였고 기존의 알고리즘과의 비교를 통해 제안하는 알고리즘의 우수성을 검증한다.

A 4D light field image is represented in traditional 2D spatial domain and additional 2D angular domain. The 4D light field has a resolution limitation both in spatial and angular domains since 4D signals are captured by 2D CMOS sensor with limited resolution. In this paper, we propose a dictionary learning-based superresolution algorithm in 4D light field domain to overcome the resolution limitation. The proposed algorithm performs dictionary learning using a large number of extracted 4D light field patches. Then, a high resolution light field image is reconstructed from a low resolution input using the learned dictionary. In this paper, we reconstruct a 4D light field image to have double resolution both in spatial and angular domains. Experimental result shows that the proposed method outperforms the traditional method for the test images captured by a commercial light field camera, i.e. Lytro.

A radioactive isotope identification algorithm is a prerequisite for a low-resolution scintillation detector applied to an unmanned radiation monitoring system. In this paper, a sparse representation with dictionary learning approach is proposed and applied to plastic gamma-ray spectra. Label-consistent K-SVD was used to learn a discriminative dictionary for the spectra corresponding to a mixture of four isotopes (¹³³Ba, ²²Na, ¹³⁷Cs, and ⁶⁰Co). A Monte Carlo simulation was employed to produce the simulated data as learning samples. Experimental measurement was conducted to obtain practical spectra. After determining the hyper parameters, two dictionaries tailored to the learning samples were tested by varying with the source position and the measurement time. They achieved average accuracies of 97.6% and 98.0% for all testing spectra. The average accuracy of each dictionary was above 96% for spectra measured over 2 s. They also showed acceptable performance when the spectra were artificially shifted. Thus, the proposed method could be useful for identifying radioisotopes in gamma-ray spectra from a plastic scintillation detector even when a dictionary is adapted to only simulated data. Furthermore, owing to the outstanding properties of sparse representation, the proposed approach can easily be built into an insitu monitoring system.

In this paper, a vehicle recognition algorithm based on deep convolutional neural network and compression dictionary is proposed. Firstly, the network structure of fine vehicle recognition based on convolutional neural network is introduced. Then, a vehicle recognition system based on multi-scale pyramid convolutional neural network is constructed. The contribution of different networks to the recognition results is adjusted by the adaptive fusion method that adjusts the network according to the recognition accuracy of a single network. The proportion of output in the network output of the entire multiscale network. Then, the compressed dictionary learning and the data dimension reduction are carried out using the effective block structure method combined with very sparse random projection matrix, which solves the computational complexity caused by high-dimensional features and shortens the dictionary learning time. Finally, the sparse representation classification method is used to realize vehicle type recognition. The experimental results show that the detection effect of the proposed algorithm is stable in sunny, cloudy and rainy weather, and it has strong adaptability to typical application scenarios such as occlusion and blurring, with an average recognition rate of more than 95%.

The purpose of this study was to investigate the effectiveness of mobile-assisted vocabulary learning with three different tasks: dictionary-based, corpus-based, and video-based tasks. The experiment was administered in the spring semester of 2017. One hundred thirty-five Korean college students participated in the current study. Participants were randomly divided into three groups, dictionary group, corpus group, and video group, and performed different vocabulary tasks for homework. They studied 10 words per week. There were a total of 160 words for sixteen weeks. To assess vocabulary learning from the different tasks, pre- and post-tests were conducted before and after the experiment. A questionnaire survey was also performed to examine group differences in their attitudes and perceptions towards vocabulary learning. Paired samples t-tests as well as ANOVA and ANCOVA were administered. Major findings are as follows: First, participants in corpus and video groups increased their vocabulary gains. Group comparison results on the post- test also revealed no significant differences, indicating that three vocabulary tasks are equally beneficial for vocabulary learning. Lastly, attitudes and perceptions towards mobile-assisted learning positively observed. In particular, video-based vocabulary tasks appeared to provide the most interesting and motivating learning environments. The present study sheds new light on different types of mobile-assisted vocabulary tasks in EFL fields.

Named entity recognition (NER) is a crucial task for NLP, which aims to extract information from texts. To build NER systems, deep learning (DL) models are learned with dictionary features by mapping each word in the dataset to dictionary features and generating a unique index. However, this technique might generate noisy labels, which pose significant challenges for the NER task. In this paper, we proposed DL-dictionary features, and evaluated them on two datasets, including the OntoNotes 5.0 dataset and our new infectious disease outbreak dataset named GFID. We used (1) a Bidirectional Long Short-Term Memory (BiLSTM) character and (2) pre-trained embedding to concatenate with (3) our proposed features, named the Convolutional Neural Network (CNN), BiLSTM, and self-attention dictionaries, respectively. The combined features (1-3) were fed through BiLSTM - Conditional Random Field (CRF) to predict named entity classes as outputs. We compared these outputs with other predictions of the BiLSTM character, pre-trained embedding, and dictionary features from previous research, which used the exact matching and partial matching dictionary technique. The findings showed that the model employing our dictionary features outperformed other models that used existing dictionary features. We also computed the F1 score with the GFID dataset to apply this technique to extract medical or healthcare information.

This study explores whether multimedia glossing, particularly dictionary definition glossing, can influence the second language learners' vocabulary learning, vocabulary short-term retention and reading comprehension. The data was collected from 26 university students. An expository text was divided into two sections (text A & B) annotated with 14 glossed words and no glossed words for each section. The participants were to take three vocabulary tests including a retention test, and a reading comprehension test. All scores were submitted to paired samples t-tests, and the results showed that the mean scores of glossing texts were mostly higher than no-glossing texts. Significant differences were found in the post vocabulary test for matching English definitions in both texts, and the vocabulary retention test (Korean definition) for Text A. However, no statistical difference was found in reading comprehension for both texts, which led to some limitations as well as indications of likely directions for the future research.

It is well known that the language women use is different from men's use. In this paper, the researchers examine men's and women's speech through sentiment analysis. For the study of gender differences, this study uses the BNC64 corpus, whose files are selected from the British National Corpus (BNC). The BNC64 corpus is composed of 64 files, which are taken from the spoken part of the BNC. The corpus contains 32 files for male speakers and 32 files for female speakers that represent the characteristics of male vs. female differences. The study analyzed all 64 files using sentiment analysis and tried to discover the differences. It is known that there are roughly three types of approach to sentiment analysis: dictionary-based analysis, machine-learning-based analysis, and deep-learning-based analysis. This study takes the first and third kind of analyses. In the dictionary-based analysis, the researchers calculate the sentiment score (SS), sentiment word ratio (SR), and positive word ratio (PR). In the deep-learning analysis, the researchers take two different sorts of analyses: the GRU (Gated Recurrent Units) and the BERT (Bidirectional Encoder Representations from Transformers). Through the analyses, the study finds that (i) there is no significant differences in SS and SR between men and women, (ii) women usually use more positive words than men, and the differences are statistically significant, and (iii) the deep-learning-based analysis is much superior to modelling the gender differences in the sentiment analysis.

본고는 <연세한국어 사전>의 발간 20 년을 되짚어 해당 사전의 언어 교육적 함의를 진단하고, 이를 바탕으로 향후 언어교육용 사전의 방향성을 제시하는 것을 목적으로 삼았다. <연세>는 발간 당시 머리말에서 한국어를 학습하는 외국인도 활용할 수 있다고 제안한 바 있으나, 실제 외국인 학습자에게 친화적 사전인지 여부는 검증하지 못한 측면이 있다. 이에 본고는 <연세> 거시 구조와 미시 구조를 언어교육적 측면에서 분석함으로 해서 해당 사전의 언어교육적 함의를 분석하고자 했다. <연세>는 실제 사용어를 고려했을 때, 고유명사와 외래어, 신어의 비중이 적고 구 단위 표제어의 확대가 필요한 점, 파생어를 고려한 단어족 단위의 제시, 그리고 다의 항목의 구분에 있어 관련어나 문법 정보와의 밀도 높은 연계 등은 언어교육의 측면에서는 보다 보강될 필요가 있어 보인다. 특히 초급 학습자를 위한 뜻풀이의 메타언어 통제나 발음 듣기 지원, 의미나 문화적 지식을 위한 멀티미디어 정보의 제공도 최근의 사전에 비해 다소 부족한 부분이 있다. 하지만 이러한 면에도 불구하고, 실제 언어를 반영한 점, 고빈도 표제어를 대상으로 한 점, 실제 말뭉치에 근거한 뜻풀이로 의미의 이해를 도운 점, 구 단위 예문을 비롯한 풍부한 연어 정보를 제공한 점, 그리고 언어 산출에 필수적인 격틀 정보와 논항 정보를 제공한 점, 화용 정보를 제공한 점, 실제 사용 예문과 더불어 다양한 참고 정보를 제공하고 있는 점 등은 언어교육에 필수적으로 요구되는 학습에 필요한 요소들을 두루 갖추었다. 이른 시기에 출간된 사전임에도 불구하고 이러한 언어교육적 측면에서의 성과들은 결국 이후 사전의 개발에 의미 있는 역할을 했다고 볼 수 있다.

중국어 사전의 발전과 중국어 학습의 효율을 위해서는 중국어 사전의 평가와 그 연구가 필수적이다. 그러나 국내에서는 아직까지 중국어 사전 평가에 대한 기준과 모형이 없어 중국어 사전에 대해 체계적인 평가를 진행하기 어려웠다. 이에 따라 본고는 국내 온라인 중국어 학습사전의 질적 향상을 목적으로, 현재 국내 중국어 학습자가 가장 많이 사용하고 있는 『네이버 중국어 사전』의 미시구조를 평가하였다. 평가는 이미향(2019)이 제작한 평가 문항을 활용하여 『네이버 중국어 사전』이 온라인 중국어 학습사전으로서 필요한 정보를 적절히 조직하여 제공하는지 미시구조의 전체 구성을 평가하고, 미시구조의 각 정보항들이 갖춰야할 정보를 정확히 제공하는지를 평가하였다. 평가 결과 어휘 등급과 의미항 항목의 평가가 가장 좋았고, 미시구조의 전체 구성과 뜻풀이, 용례 항목도 비교적 긍정적인 평가를 내릴 수 있었다. 반면에 철자법, 발음, 품사, 문법 정보, 화용 정보 항목은 수정과 보완의 필요가 있다고 평가되었다. 『네이버 중국어 사전』은 이언어 사전으로서 미시구조가 기본적으로 갖춰야 하는 항목을 갖추었지만, 발음 항목에서 성조 표기 오류를 발견하였고, 뜻풀이 항목에서 잘못된 품사 표기와 함께 잘못 된 대역어를 제공한 오류를 발견할 수 있었다. 『네이버 중국어 사전』은 학습사전으로서 표제어에 사용자가 쉽게 인식할 수 있는 어휘 등급을 표기하였고, 중국어 학습에 유용한 선택정보를 제공하였다. 학습사전으로서의 기능은 참고어 항목에서 두드러졌다. 그러나 철자법 정보를 제공하지 않은 점, 용례를 표제어의 의미항 단위로 구분하여 제공하지 않은 점, 문법과 화용 정보가 부족한 점, 성조 불규칙 변화와 특수어휘 표기에 일관성이 없는 점 등은 보완할 필요가 있다. 『네이버 중국어 사전』은 온라인 사전으로서 미시구조가 열람하기 편리하게 구성되어 있고, 온라인의 특성을 활용한 서비스를 통해 다양한 정보를 제공한다. 특히 온라인 사전의 장점은 용례 항목에서 두드러졌다. 그러나 용량의 제한이 없는 온라인 사전임에도 연어 및 통사 정보, 문법 및 화용 상의 주의점, 담화 용례, 문화적 지식 등 학습사전으로서 필요한 정보를 더 제공하지 않은 점이 아쉽게 평가되었다. 본 연구는 기타 온라인 외국어 학습사전을 평가하고 수정하는 데 참고 활용할 수 있고, 온라인 중국어 사전을 수정하고 업데이트 하는 데 참고할 수 있다.

For the development of Chinese dictionary and the efficiency of Chinese learning, evaluation and study of Chinese dictionary are essential. However, there are no standards and models for Chinese dictionary evaluation in Korea, so it was difficult to conduct systematic evaluation of Chinese dictionary. Accordingly, this paper assessed the microstructure of 『Naver Chinese Dictionary』, which is currently used by domestic Chinese learners, for the purpose of improving the quality of domestic online Chinese learner’s dictionaries. Using the evaluation items prepared by Mi-Hyang Lee (2019), and the overall composition of the microstructure was evaluated to see if the 『Naver Dictionary』 appropriately organizes and provides necessary information as an online Chinese learner’s dictionary, and was assessed whether each information item in the microstructure provided the exact information to be provided. As a result of the evaluation, the vocabulary grade and semantic distribution item were the best to be able to answer ‘yes’ to all questions, and the overall composition, definition of a word, and example items of the microstructure were also relatively positive. On the other hand, items of spelling, pronunciation, part-of-speech, grammatical information, and pragmatic information were evaluated as necessary for modification and supplementation. The 『Naver Dictionary』 is an bilingual dictionary, which has items that must be equipped with a microstructure, but found errors in the intonation notation in the pronunciation item, and errors in providing wrong target word with wrong part-of-speech notation were found in the item of definition of a word. 『Naver Dictionary』 is a learning dictionary, which shows the vocabulary grade that users can easily recognize in the headword and provides useful information for Chinese learning. The function as a learning dictionary was prominent in the reference words item. However, it is necessary to supplement the fact that spelling information is not provided, the example is not provided separately by the semantic terms of the headword, the grammar and pragmatic information is not sufficient, and the irregular changes of the intonation and the inconsistency of the special vocabulary. 『Naver Dictionary』 is an online dictionary, which is convenient for viewing microstructures and provides various information through services utilizing online characteristics. In particular, the advantages of online dictionaries have been prominent in the example item. However, It was regrettable that the online dictionary, which has no limit on capacity, did not provide more information as a learning dictionary, such as collocation and syntactic information, precautions on grammar and pragmatic use, discourse usage, and cultural knowledge, etc. This study can be used to evaluate and modify other online foreign language learning dictionaries, and can be used to modify and update online Chinese dictionaries.

In learning a foreign language, in particular, the dictionary plays an important role in self-directed learning. However, the study and publishing of Chinese dictionaries have been in place for a long time, and the Chinese dictionary now has a lot of problems in its organization and composition, and it limits the self-directed learning of learners. The evaluation and study of the dictionary are essential for improving the quality of Chinese dictionaries and for the efficiency of Chinese learning. However, there is still a lack of research on systematic dictionary evaluation models in Korea, making it difficult to conduct objective evaluations of Chinese dictionaries. Therefore, this study set the evaluation criteria and evaluation items to evaluate the overall microstructure of the dictionary used mainly by domestic Chinese learners and created specific evaluation items. To evaluate a dictionary, it is necessary to identify the type of dictionary to be evaluated first. Provided that the type of dictionary is analyzed by using the dictionary classification method, the dictionary to study in this paper is a bilingual dictionary using Chinese and Korean, and a learners’ dictionary for Korean learners who are learning Chinese, and as an online dictionary using online media, this paper is called ‘Online Chinese learning dictionary’. Lexicographers have distinguished the structure of the dictionary from different perspectives, which in common include macrostructures and microstructures. In this paper, the discussion has been emphasized how the entries of the dictionary organize and provide information with a focus on its microstructure. Although there is still insufficient research on the dictionary evaluation in Korea, Western and Chinese lexicographers have already studied criteria for evaluating various types of dictionaries. However, since existing evaluation criteria are insufficient in system and unsatisfactory to evaluate microstructure as a whole. it was necessary to establish the appropriate evaluation criteria to evaluate the microstructure of the online Chinese learners’ dictionary. This paper refers to the evaluation criteria in the previous study and designed the criteria for the creation of evaluation items to evaluate the microstructure of the online Chinese learning dictionary from a macro perspective based on the lexicography, Chinese Linguistics and evaluation theory. And based on these criteria, specific evaluation items has been created to evaluate the microstructure of the online Chinese learners’ dictionary. First, five items have been created to evaluate the overall composition of the microstructure, and the vocabulary grade, spelling, pronunciation, semantic distribution, part of speech, definition of a word and example were set as essential information items of the microstructure, and 36 evaluation items have been created to evaluate them. The selection information items of the microstructure were made up of 18 evaluation items, divided into grammatical information, pragmatic information, reference words information, and other information. The evaluation items created in this paper can actually be a standard for evaluating the online Chinese learning dictionary in Korea, and the dictionary researcher and lexicographers can refer to compiling and updating online Chinese learners’ dictionaries, and can also be used to evaluate and study other foreign language dictionaries.

본 논문은 명시야 (bright-field) 현미경 영상를 위한 데이터 기반 세포 분할 알고리즘을 제시한다. 제시된 알고리즘은 일반적인 사전 학습 기법과 다르게 동시에 두 개의 사전과 관련된 희소 코드 (sparse code)를 통해 정의된 에너지 함수의 최소화를 진행하게 된다. 두 개의 사전 중 하나는 명시야 영상에 대해 학습된 사전이고 다른 하나는 사람에 의해 수작업으로 세포 분할된 영상에 대해 학습된 것이다. 학습된 두 개의 사전을 세포 분할 될 새로운 입력 영상에 대해 적용하여 이와 관련된 희소 코드를 획득한 후 픽셀 단위의 분할을 진행하게 된다. 효과적인 에너지 최소화를 위해 합성곱 희소 코드 (Convolutional Sparse Coding)와 Alternating Direction of Multiplier Method(ADMM)이 사용되었고 GPU를 사용하여 빠른 분산 연산이 가능하다. 본 연구는 이전에 사용된 가변형 모델 (deformable model)을 이용한 세포 분할 방식과는 다르게 제시된 알고리즘은 세포 분할을 위해 사전 지식이 필요없이 데이터 기반의 학습을 통해서 쉽고 효율적으로 세포 분할을 진행할 수 있다.

Cell segmentation is an important but time-consuming and laborious task in biological image analysis. An automated, robust, and fast method is required to overcome such burdensome processes. These needs are, however, challenging due to various cell shapes, intensity, and incomplete boundaries. A precise cell segmentation will allow to making a pathological diagnosis of tissue samples. A vast body of literature exists on cell segmentation in microscopy images [1]. The majority of existing work is based on input images and predefined feature models only - for example, using a deformable model to extract edge boundaries in the image. Only a handful of recent methods employ data-driven approaches, such as supervised learning. In this paper, we propose a novel data-driven cell segmentation algorithm for bright-field microscopy images. The proposed method minimizes an energy formula defined by two dictionaries - one is for input images and the other is for their manual segmentation results - and a common sparse code, which aims to find the pixel-level classification by deploying the learned dictionaries on new images. In contrast to deformable models, we do not need to know a prior knowledge of objects. We also employed convolutional sparse coding and Alternating Direction of Multiplier Method (ADMM) for fast dictionary learning and energy minimization. Unlike an existing method [1], our method trains both dictionaries concurrently, and is implemented using the GPU device for faster performance.