• Koreanishche Zeitschrift fur Deutsche Sprachwissenschaft
• /
• v.9
• /
• pp.217-236
• /
• 2004

Das Ziel dieser Arbeit ist es, als eine methodische Grundlage zur Erstellung des koreanisch-deutschen WordNets das Grundwissen $\"{u}ber$ das WordNet und einige bisherige Untersuchungen des WordNets darzulegen. Ais erster Schritt wurde einige grundlegende Punkte $f\"{u}r$ das WordNet im Rahmen des WordNets fur Englisch in Betracht gebracht. Dabei ging es um lexikalische Hierarchie, und um semantische Relationen zwischen den Synsets(Zusammensetzen der synonymen $W\"{o}rter$) wie Synonymy, Antonymy, Hyponymy, Mronymy, Troponomy und Entailment. $Anschlie{\ss}end$ wurden EuroNet und GermaNet in kurzer Form vorgestellt, die auf dem Princeton WordNet basierten. EuroNet ist eine multilinguale Datenbasis mit WordNets $f\"{u}r$ einige europaische Sprachen (hollandisch, italienisch, spanisch, deutsch, franzasisch, tschechisch und estnisch). Dieses auf das Deutsch bezogenen WordNet kann wichtige Hinweise $f\"{u}r$ die Erstellung des koreanisch-deutschen WordNets geben. In Korea wurden auch verschiedene Untersuchungen uber das WordNet $f\"{u}r$ Koreanisch unternommen. Darunter kann insbesondere KORTERM WordNet $f\"f{u}r$ Koreanisch als ein umfassendes System $erw\"{a}hnt$ werden, in dem Nomen, Verben, Adjektive und Adverbien miteinander interagieren. KORTERM WordNet fur Koreanisch ist eine multilinguale Datenbasis mit WordNets $f\"{u}r$ einige asiatische Sprachen (koreanisch, japanisch und chinesisch) und versucht noch die weiteren Sprachen in diese multilinguale Datenbasis hineinzubringen. Nach diesem WordNet wird das koreanisch-deutsche WordNet erstellt.

• The KIPS Transactions:PartB
• /
• v.15B no.3
• /
• pp.219-228
• /
• 2008

For real-time social networking service, mobile interaction in a usable-unified-ubiquitous (U3) web service was studied. Both as a convenient mobile HCI for real-time social networks and as indexing keys to metadata information in ubiquitous web service, the multi-lingual single-character domain names (e.g. 김.net, 이.net, 가.net, ㄱ.net, ㄴ.net, ㅎ.net, ㅏ.net, ㅔ.net, ㄱ.com, ㅎ.com) are convenient mobile interfaces when searching for social information and registering information. We introduce the sketched design goals and experience of mobile interaction in Korea, Japan and China, with the implementation of real-time social networking service as an example of U3 Web service. We also introduce the possibility of extending the application to the metadata directory service in IP-USN (IP-based Ubiquitous Sensor Network) for a unified information management in the service of social networking and sensor networking.

• Koreanishche Zeitschrift fur Deutsche Sprachwissenschaft
• /
• v.6
• /
• pp.323-344
• /
• 2002

EuroNet ist eine multilinguale Datenbasis mit WordNets $f\"{u}r\;einige\;europ\"{a}ische$ Sprachen ($holl\"{a}ndisch$, italienisch, spanisch, deutsch, $franz\"{o}sisch$, tschechisch und estnisch). Die WordNets werden genauso wie das amerikanische WordNet $f\"{u}r$ Englisch (Princeton WordNet, Miller et al. 1990) in Synsets (Zusammensetzen der synonymen $W\"{o}rter$) mit grundlegenden lexikalisch-semantischen Relationen zwischen ihnen $ausgedr\"{u}ckt$ strukturiert. Jedes WordNet stellt also ein einzigartiges innersprachliches System $f\"{u}r$ die lexikalischen und konzeptuellen Relationen dar. $Zus\"{a}tzlich$ werden diese auf dem Princeton WordNet basierten WordNets (z.B. GermaNet) mit einem Inter-Linguale-Index (kurz, ILI) verbunden. $\"{U}ber$ diesem Index werden die Sprachen zusammengeschaltet, damit zu gehen ist $m\"{o}glich$, von den $W\"{o}rtern$ in einer Sprache zu den $\"{a}hnlichen\;W\"{o}rtern$ in jeder $m\"{o}glicher$ anderen Sprache. Der Index gibt auch Zugang zu einer geteilten Top-Ontologie von 63 semantischen Unterscheidungen. Diese Top-Ontologie stellt einen allgemeinen semantischen Rahmen $f\"{u}r$ aile Sprachen zur $Verf\"{u}gung,\;w\"{a}hrend$ sprachspezifische Eigenschaften in den einzelnen WordNets beibehalten werden. Die Datenbasis kann, unter anderen, $f\"{u}r$ einsprachige und multilinguale Informationsretrieval benutzt werden. In der vorliegenden Arbeit handelt sich also um eine methodologische Untersuchung der koreanisch-deutschen ILI-Verbindung zur Anwendung der auf dem EuroNet basierten lexikalischen, semantischen Datenbasis. Dabei werden einzelnen Lexeme in koreanischen, deutschen WordNets $zun\"{a}chst$ mit Hilfe der Sense-Analyse semantisch differenziert, und dann durch lexikalische und konzeptuelle Relationen(ILI) miteinander verbunden. Die Equivalezverbindungen dienen, sprachspezifische Konzepte zum ILI abzubilden. Sie werden von einem anderen Synset der moglichen Relationen aus der Euronet-Spezifikation genommen. Wenn es keinen ILI-Rekord gibt, der ein direktes Equivalenz zu einem gegebenen Konzept darstellt, kann das Konzept in der Frage $\"{u}ber$ EQ-Near-Synonymie, EQ-Hyperonymie oder EQ-Hyponymie Relationen verbunden werden.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.15 no.5
• /
• pp.227-234
• /
• 2020

In this paper, we propose wav-U-Net to improve speech enhancement in heavy noisy environments, and it has implemented three principal techniques. First, as input data, we use 128 modified Mel-scale filter banks which can reduce computational burden instead of 512 frequency bins. Mel-scale aims to mimic the non-linear human ear perception of sound by being more discriminative at lower frequencies and less discriminative at higher frequencies. Therefore, Mel-scale is the suitable feature considering both performance and computing power because our proposed network focuses on speech signals. Second, we add a simple ResNet as pre-processing that helps our proposed network make estimated speech signals clear and suppress high-frequency noises. Finally, the proposed U-Net model shows significant performance regardless of the kinds of noise. Especially, despite using a single channel, we confirmed that it can well deal with non-stationary noises whose frequency properties are dynamically changed, and it is possible to estimate speech signals from noisy speech signals even in extremely noisy environments where noises are much lauder than speech (less than SNR 0dB). The performance on our proposed wav-U-Net was improved by about 200% on SDR and 460% on NSDR compared to the conventional Jansson's wav-U-Net. Also, it was confirmed that the processing time of out wav-U-Net with 128 modified Mel-scale filter banks was about 2.7 times faster than the common wav-U-Net with 512 frequency bins as input values.

• Korean Chemical Engineering Research
• /
• v.57 no.2
• /
• pp.274-282
• /
• 2019

In recent years, big data analysis has been expanded to include automatic control through reinforcement learning as well as prediction through modeling. Research on the utilization of image data is actively carried out in various industrial fields such as chemical, manufacturing, agriculture, and bio-industry. In this paper, we applied NASNet, which is an AutoML reinforced learning algorithm, to DeepU-Net neural network that modified U-Net to improve image semantic segmentation performance. We used BRATS2015 MRI data for performance verification. Simulation results show that DeepU-Net has more performance than the U-Net neural network. In order to improve the image segmentation performance, remove dropouts that are typically applied to neural networks, when the number of kernels and filters obtained through reinforcement learning in DeepU-Net was selected as a hyperparameter of neural network. The results show that the training accuracy is 0.5% and the verification accuracy is 0.3% better than DeepU-Net. The results of this study can be applied to various fields such as MRI brain imaging diagnosis, thermal imaging camera abnormality diagnosis, Nondestructive inspection diagnosis, chemical leakage monitoring, and monitoring forest fire through CCTV.

• Journal of the Korean Society of Radiology
• /
• v.17 no.5
• /
• pp.663-670
• /
• 2023

In this study, we aimed to perform a comparative evaluation using quantitative factors between a region-growing based segmentation with noise reduction algorithms and a U-Net based segmentation. Initially, we applied median filter, median modified Wiener filter, and fast non-local means algorithm to computed tomography (CT) images, followed by region-growing based segmentation. Additionally, we trained a U-Net based segmentation model to perform segmentation. Subsequently, to compare and evaluate the segmentation performance of cases with noise reduction algorithms and cases with U-Net, we measured root mean square error (RMSE) and peak signal to noise ratio (PSNR), universal quality image index (UQI), and dice similarity coefficient (DSC). The results showed that using U-Net for segmentation yielded the most improved performance. The values of RMSE, PSNR, UQI, and DSC were measured as 0.063, 72.11, 0.841, and 0.982 respectively, which indicated improvements of 1.97, 1.09, 5.30, and 1.99 times compared to noisy images. In conclusion, U-Net proved to be effective in enhancing segmentation performance compared to noise reduction algorithms in CT images.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.65-68
• /
• 2021

본 연구에서는 Dilated U-Net 기반의 이미지 복원기법을 통해 콘크리트 균열 추출 성능 개선 방안을 제안한다. 콘크리트 균열은 구조물의 미관상의 문제뿐 아니라 추후 큰 안전사고의 원인이 될 수 있어 초기대응이 중요하다. 현재는 점검자가 직접 육안으로 검사하는 외관 검사법이 주로 사용되고 있지만, 이는 정확성 및 비용, 시간, 그리고 안전성 면에서 한계를 갖고 있다. 이에 콘크리트 구조물 표면에 대해 획득한 영상 처리 기법을 사용한 검사 방식 도입의 관심이 늘어나고 있다. 또한, 딥러닝 기술의 발달로 딥러닝을 적용한 영상처리의 연구 역시 활발하게 진행되고 있다. 본 연구는 콘크리트 균열 추개선출 성능 개선을 위해 Dilated U-Net 기반의 이미지 복원기법을 적용하는 방안을 제안하였고 성능 검증 결과, 기존 U-Net 기반의 정확도가 98.78%, 조화평균 82.67%였던 것에 비해 정확도 99.199%, 조화평균 88.722%로 성능이 되었음을 확인하였다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.7 no.3
• /
• pp.87-94
• /
• 2006

For realizing an ubiquitous home network(uHome-net), various sensors should be able to be connected to an integrated wire/wireless sensor network. This paper describes an application case of applying LonWorks technology being widely used in control network to wire/wireless sensor network in uHome-net and the design and application of LonRF device that consists of a neuron chip including LonTalk protocol, a 433.92MHz RF transceiver, a sensor, and application programs. As an application example of the LonRF device, the LonRF smart badge that can measure the 3D location of objects in indoor environment and interwork with the uHome-net was developed. LonRF device based home network services were realized on the uHome-net testbed such as indoor positioning service, remote surveillance service and remote metering service were realized. This research shows that LonWorks technology based sensor network could be applicable to the control network in an ubiquitous home network and the LonRF device can be used as a wireless node in various sensor networks.

• Smart Media Journal
• /
• v.9 no.1
• /
• pp.16-22
• /
• 2020

In this study, a neural network method performing both Detection and Classification of diseases and insects in paprika is proposed with Multi-Tasking U-net. Paprika on farms does not have a wide variety of diseases in this study, only two classes such as powdery mildew and mite, which occur relatively frequently are made as the targets. Aiming to this, a U-net is used as a backbone network, and the last layers of the encoder and the decoder of the U-net are utilized for classification and segmentation, respectively. As the result, the encoder of the U-net is shared for both of detection and classification. The training data are composed of 680 normal leaves, 450 mite-damaged leaves, and 370 powdery mildews. The test data are 130 normal leaves, 100 mite-damaged leaves, and 90 powdery mildews. Its test results shows 89% of recognition accuracy.

• Journal of the Korean Society of Radiology
• /
• v.17 no.5
• /
• pp.709-715
• /
• 2023

In this study, the performance evaluation of image quality for noise reduction was implemented using the U-net deep learning architecture in computed tomography (CT) images. In order to generate input data, the Gaussian noise was applied to ground truth (GT) data, and datasets were consisted of 8:1:1 ratio of train, validation, and test sets among 1300 CT images. The Adagrad, Adam, and AdamW were used as optimizer function, and 10, 50 and 100 times for number of epochs were applied. In addition, learning rates of 0.01, 0.001, and 0.0001 were applied using the U-net deep learning model to compare the output image quality. To analyze the quantitative values, the peak signal to noise ratio (PSNR) and coefficient of variation (COV) were calculated. Based on the results, deep learning model was useful for noise reduction. We suggested that optimized hyper parameters for noise reduction in CT images were AdamW optimizer function, 100 times number of epochs and 0.0001 learning rates.