• Journal of KIISE
• /
• v.43 no.8
• /
• pp.848-856
• /
• 2016

Seeking and recognizing maritime targets are very important tasks for maritime safety. While searching for maritime targets using radar is possible, recognition is conducted without automatic identification system, radio communicator or visibility. If this recognition is not feasible, radar operator must tediously recognize maritime targets using movement features on radar base on know-how and experience. In this paper, to support the radar operator's mission of continuous observation, we propose an algorithm for automatic recognition of an unknown ship using movement features on radar and a method of detecting potential ship related accidents. We extract features from contact range, course and speed of four types of vessels and evaluate the recognition accuracy using SVM and suggest a method of detecting potential ship related accidents through the algorithm. Experimentally, the resulting recognition accuracy is found to be more than 90% and presents the possibility of detecting potential ship related accidents through the algorithm using information of MV Sewol. This method is an effective way to support operator's know-how and experience in various circumstances and assist in detecting potential ship related accidents.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.2 no.1
• /
• pp.79-85
• /
• 1998

The 9-digit MMSI(Maritime Mobile Service Identity) in the NBDP is converted to the 7-digit identity signal, used to identify to the phase of and to synchronize to the clock of the ISS(Information Sending Station) automatically for phasing procedure between radio stations. In this paper, algorithms for the automatic identification phasing procedures and for conversions, calculations between 7-signals and 9-identity, those check-numbers in the phasing procedures for calling of the NBDP are discussed and designed. And a method of the faster locking and phasing for calling than that of recommended by ITU-R is suggested and implemented.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.4
• /
• pp.635-640
• /
• 2021

In this paper, we describe a parking space identification system that can automatically identify empty parking lot spaces from a parking lot photo. This system is based on a deep learning technique, and the accuracy of the identification result is good by learning various existing parking lot images. It could be applied to the existing parking management system. This system was also developed as a smartphone application for easy testing. Therefore, if you take a picture of a parking lot through a smartphone camera, the captured image is automatically recognized and an empty parking space can be automatically identified.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.686-688
• /
• 2021

The performance assessment in real sea is very important to increase the reliability of the fishing gear automatically identification monitoring system. The concept of real sea experiment for fishing gear automatically identification monitoring system is introduced and results of communication performance of RoLa and LTE Cat.M1 modules are reported through this research.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.11b
• /
• pp.640-643
• /
• 2002

링크 관리 프로토콜에서 데이터 링크의 연결성 검증 및 링크 식별자의 자동 매핑을 위하여 교환하는 Test 메시지와 제어채널의 IP 주소를 자동으로 발견하기 위한 Bootstrap 메시지는 제어채널이 아닌 데이터 링크로 전달되므로 광 네트워크에서 불투명한 스위칭 기술을 요구한다. 본 논문에서는 불투명한 스위칭 기술을 지원하지 않는 광 네트워크에서 망 관리 기능을 이용하여 인접 노드 사이의 제어채널에 대한 IP 주소 및 인접 노드와 데이터 링크에 대한 식별자를 자동으로 구성 할 수 있는 메커니즘을 제안한다. 망 관리를 통한 자동 구성은 불투명한 스위칭 기술을 지원하지 않는 광 네트워크에서 수동 구성에 따른 오류 발생의 가능성을 개선시킬 수 있다.

• Annual Conference on Human and Language Technology
• /
• 2004.10d
• /
• pp.53-61
• /
• 2004

일반적으로 중국어의 명사구는 기본명사구(base noun phrase), 최장명사구(maximal noun phrase) 등으로 분류된다. 최장명사구에 대한 정확한 식별은 문장의 전체적인 구조를 파악하고 정확한 구문 트리(parse tree)를 찾아내는데 중요한 역할을 한다. 본 논문은 두 단계 학습모델을 이용하여 최장명사구 자동식별을 진행한다. 먼저 기본명사구, 기본동사구, 기본형용사구, 기본부사구, 기본수량사구, 기본단문구, 기본전치사구, 기본방향사구 등 8가지 기본구를 식별한다. 다음 기본구의 중심어(head)를 추출해 내고 이 정보를 이용하여 최장명사구의 식별을 진행한다. 본 논문에서 제안하는 방법은 기존의 단어레벨의 접근방법과는 달리구레벨에서 학습을 진행하기 때문에 주변문맥의 정보를 많이 고려해야 하는 최장명사구 식별에 있어서 아주 효과적인 접근방법이다. 후처리 작업을 하지 않고 기본구의 식별에서 25개 기본구 태그의 평균 F-measure가 96%, 평균길이가 7인 최장명사구의 식별에서 4개 태그의 평균 F-measure가 92.5%로 좋은 성능을 보여주었다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.3 no.2
• /
• pp.321-327
• /
• 1999

In this progressed super-impose individual recognition system, the photograph of a skull was caught by CCD-camcoder with the MPEG, and an ante-mortem photograph was read by scanner. These two images were processed and superimposed using horizontal angle and vertical angle of face using the forensic dental medicine theory. The enhancement of super-impose individual recognition by anatomical references was performed on the two superimposed images of the same angle using the facial triangle and histogram analysis scheme.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.13 no.5
• /
• pp.30-34
• /
• 1976

A system that can discriminate monosyliables automatically from sequential input of Korean character's data without space codes is proposed. Korean characters are synthesized by two to seven elements out of twenty four basic elements. Three thousands Korean characters are formalized into thirty character forms discriminates monosyllable automatically by detecting seven form features and character length. In this result, this system, compared with the input method with space codes which have been used to separate each syllable, can save about 25% of the memory capacity of computer and improves about 30% of the processing speed of Korean characters.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10a
• /
• pp.137-140
• /
• 2000

최근에 네트워크와 하드웨어 기술의 발달에 따라 디지털 컨텐츠의 지적소유권에 관한 분쟁이 많이 발생하고 있다. 본 논문에서는 디지털 컨텐츠의 전자상거래에서 지적소유권에 대한 분쟁을 해결하는데 사용될 수 있는 익명성을 제공하는 핑거프린팅 방식을 제안한다. 제안하는 방식은 판매된 디지털 컨텐츠를 재분배하는 구매자를 판매자가 등록센터의 도움 없이 재분배자를 식별하는 향상된 자동식별기능을 갖는다.

• Phonetics and Speech Sciences
• /
• v.15 no.4
• /
• pp.71-80
• /
• 2023

This study primarily aimed to develop an automated stuttering identification and classification method using artificial intelligence technology. In particular, this study aimed to develop a deep learning-based identification model utilizing the convolutional neural networks (CNNs) algorithm for Korean speakers who stutter. To this aim, speech data were collected from 9 adults who stutter and 9 normally-fluent speakers. The data were automatically segmented at the phrasal level using Google Cloud speech-to-text (STT), and labels such as 'fluent', 'blockage', prolongation', and 'repetition' were assigned to them. Mel frequency cepstral coefficients (MFCCs) and the CNN-based classifier were also used for detecting and classifying each type of the stuttered disfluency. However, in the case of prolongation, five results were found and, therefore, excluded from the classifier model. Results showed that the accuracy of the CNN classifier was 0.96, and the F1-score for classification performance was as follows: 'fluent' 1.00, 'blockage' 0.67, and 'repetition' 0.74. Although the effectiveness of the automatic classification identifier was validated using CNNs to detect the stuttered disfluencies, the performance was found to be inadequate especially for the blockage and prolongation types. Consequently, the establishment of a big speech database for collecting data based on the types of stuttered disfluencies was identified as a necessary foundation for improving classification performance.