• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.489-492
• /
• 2002

최근 인간형 로봇에 대한 개발이 괄목할 만한 성장을 이루고 있고, 친근한 로봇의 개발에 중요한 역할을 담당하는 것으로써 감성/감정의 인식이 필수적이라는 인식이 확산되고 있다. 본 논문은 음성의 감정인식에 있어 가장 큰 부분을 차지하는 피치의 패턴을 인식하여 감정을 분류/인식하는 시뮬레이터의 개발과 실험결과를 나타낸다. 또한, 피치뿐 아니라 음향학적으로 날카로움, 낮음등의 요소를 분류의 기준으로 포함시켜서 좀더 신뢰성 있는 인식을 할 수 있음을 보인다. 시뮬레이터의 내부 구조로는 음성으로부터 피치를 추출하는 부분과 피치의 패턴을 학습시키는 DRNN 부분, 그리고, 음향적 특성을 추출하는 음향 추출부가 주요 요소로 이루어져 있다. 그리고, 피치를 추출하는 방법으로는 Center-Clipping 함수를 이용한 autocorrelation approach를 사용하고, 학습 시 최적의 개체를 찾는 방법으로써 (1+100)-ES를 사용한다.

• Journal of Korean Society of Forest Science
• /
• v.96 no.3
• /
• pp.251-258
• /
• 2007

Airborne Lidar (light detection and ranging) can be an effective alternative in forest inventory to overcome the limitations of conventional field survey and aerial photo interpretation. In this study, we attempt to develop methodologies to identify individual trees and to estimate tree height from airborne Lidar data. Initially, digital elevation model (DEM) data representing the exact ground surface were generated by removing non-ground returns from the multiple-return laser point clouds, obtained over the coniferous forest site of rugged terrain. Based on the canopy height model (CHM) data representing non-ground layer, individual tree heights are extracted through pseudo-grid method and moving window filtering algorithm. Comparing with field survey data and aerial photo interpretation on sample plots, the number of trees extracted from Lidar data show over 90% accuracy and tree heights were underestimated within 1.1m in average at two plantation stands of pine (Pinus koraiensis) and larch (Larix leptolepis).

• Journal of Biosystems Engineering
• /
• v.27 no.2
• /
• pp.151-160
• /
• 2002

Cow image processing technique would be useful not only for recognizing an individual but also for establishing the image database and analyzing the shape of cows. A cow (Holstein) has usually the unique speckle pattern. In this study, the individual recognition of cow was carried out using the speckle pattern and the content-based image retrieval technique. Sixty cow images of 16 heads were captured under outdoor illumination, which were complicated images due to shadow, obstacles and walking posture of cow. Sixteen images were selected as the reference image for each cow and 44 query images were used for evaluating the efficiency of individual recognition by matching to each reference image. Run-lengths and positions of runs across speckle area were calculated from 40 horizontal line profiles for ROI (region of interest) in a cow body image after 3 passes of 5$\times$5 median filtering. A similarity measure for recognizing cow individuals was calculated using Euclidean distance of normalized G-frame histogram (GH). normalized speckle run-length (BRL), normalized x and y positions (BRX, BRY) of speckle runs. This study evaluated the efficiency of individual recognition of cow using Recall(Success rate) and AVRR(Average rank of relevant images). Success rate of individual recognition was 100% when GH, BRL, BRX and BRY were used as image query indices. It was concluded that the histogram as global property and the information of speckle runs as local properties were good image features for individual recognition and the developed system of individual recognition was reliable.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.4
• /
• pp.577-586
• /
• 2024

As new technologies are developed, new security threats such as the emergence of AI technologies that create ransomware are also increasing. New security equipment such as XDR has been developed to cope with these security threats, but when using various security equipment together rather than a single security equipment environment, there is a difficulty in creating numerous regular expressions for identifying and classifying essential data. To solve this problem, this paper proposes a method of identifying essential information for identifying threat information by introducing artificial intelligence-based entity name recognition technology in various security equipment usage environments. After analyzing the security equipment log data to select essential information, the storage format of information and the tag list for utilizing artificial intelligence were defined, and the method of identifying and extracting essential data is proposed through entity name recognition technology using artificial intelligence. As a result of various security equipment log data and 23 tag-based entity name recognition tests, the weight average of f1-score for each tag is 0.44 for Bi-LSTM-CRF and 0.99 for BERT-CRF. In the future, we plan to study the process of integrating the regular expression-based threat information identification and extraction method and artificial intelligence-based threat information and apply the process based on new data.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.3103-3105
• /
• 2000

본 논문은 RF/ID 카드를 이용한 도어락 시스템의 개발에 관한 것이다. 기존 기계적 도어락 시스템의 보안과 편리성을 도모하기 위해 전자식 도어락가 현재 널리 쓰이고 있다. 대부분의 도어락 시스템이 버튼식이지만 현재 RF/ID 카드를 많이 쓰는 추세이다. RF/ID카드는 비접촉식으로 보안성과 그 신뢰성이 뛰어나며 안테나 설계와 Tag의 특성에 따라서 그 통신 범위도 작게는 몇Cm에서 크게는 몇십m 까지 된다. 따라서 그 응용 범위 또한 광범위해서 사람. 개체. 물류등의 개체인식에 널리 쓰이고 있다. 본 논문은 이러한 RF/ID 카드를 소규모 도어락 시스템을 요구하는 가정, 또는 오피스텔과 같은 곳에서 쓰일 수 있는 도어락 시스템을 개발하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.922-925
• /
• 2019

기계 독해 기술은 기계가 주어진 비정형 문서 내에서 사용자의 질문을 이해하여 답변을 하는 기술로써, 챗봇이나 스마트 스피커 등, 사용자 질의응답 분야에서 핵심이 되는 기술 중 하나이다. 최근 딥러닝을 이용한 기학습 언어모델과 전이학습을 통해 사람의 기계 독해 능력을 뛰어넘는 방법론들이 제시되었다. 하지만 이러한 방식은 사람이 인식하는 질의응답 방법과 달리, 개체가 가지는 의미론(Semantic) 관점보다는 토큰 단위로 분리된 개체의 형태(Syntactic)와 등장하는 문맥(Context)에 의존해 기계 독해를 수행하였다. 본 논문에서는 기존의 높은 성능을 나타내던 기학습 언어모델에 대규모 지식그래프에 등장하는 개체 정보를 함께 학습함으로써, 의미학적 정보를 반영하는 방법을 제시한다. 본 논문이 제시하는 방법을 통해 기존 방법보다 기계 독해 분야에서 높은 성능향상 결과를 얻을 수 있었다.

• Korean Journal of Cognitive Science
• /
• v.19 no.4
• /
• pp.443-458
• /
• 2008

Since Web pages contain a lot of information today, the Web becomes an important resource to extract some information. In this paper, we proposes a set expansion system which can automatically extract named entities from the Web. Overall, the proposed method consists of three steps. First of all, Web pages, which may include many named entities of a domain, are collected by using several seed words of the domain. Then some pattern rules are extracted by using seed words and the collected Web pages, and the named entity candidates are selected through applying the extracted pattern rules into Web pages. To distinguish real named entities, we develop the new mutual importance measurement method which estimates the importance of named entity candidates. We conducted experiments for 3 domains for Korean and for 8 domains for English. As a result, the proposed method obtained 78.72% MAP in Korean and 96.48% MAP in English. In particular, the performances of English domains are better than the results of the Google set.

• Spatial Information Research
• /
• v.17 no.4
• /
• pp.445-454
• /
• 2009

Beyond simple transfer of information through sensor network, this study will provide the insights about the way to embody the real context aware location based service in an ubiquitous computing environment. In this paper, a new formal approach is introduced to derive knowledge about the scope of influence for a point object. A scope of influence is defined as the conceptual area where there is a possibility of the phenomenon or event occurring because of this point object. A point object can be spatially extended by considering this scope of influence in conjunction with the point. These point objects are called Spatially Extended Point (SEP) objects. Compositions of gradual changes of topological relations between a SEP and the environment near the SEP show how to represent the qualitative spatial behaviors of a SEP objects. These qualitative spatial behaviors will be good standards for Location Based Service (LBS) to provide more subdivided and suitable information to the users.

• Journal of Korean Library and Information Science Society
• /
• v.55 no.1
• /
• pp.215-237
• /
• 2024

This study proposes a selection plan for research entities and PIDs and a strategy for building and operating a PID consortium based on a survey of advanced cases of research entities and PID operations in major countries such as the United Kingdom, Germany, Canada, Japan, China, and Australia. The criteria for selecting research entities and PIDs are 'research life cycle' and 'PID infrastructure maturity'. Based on the two selection criteria, it is proposed to prioritize research entity-PID pairs such as 'Researcher-ORCID', 'Publication-DOI', 'Data-DOI', 'Institution-ROR', 'Grant-DOI', and 'Project-RAiD' and expand to other research entities and PIDs in the emerging stage. The strategy for establishing and operating a PID consortium should encourage the participation of various PID stakeholders, identify the latest trends through collaborative networks with domestic and international PID organizations, lead education and outreach activities to raise awareness and increase utilization of PID, and secure policy support and financial stability. This is expected to lay the foundation for domestic research entities to gain visibility and accessibility at the global level.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.4
• /
• pp.11-17
• /
• 2020

In this paper, we propose a method to build a knowledge base based on natural language processing for intelligent battlefield awareness service. The current command and control system manages and utilizes the collected battlefield information and tactical data at a basic level such as registration, storage, and sharing, and information fusion and situation analysis by an analyst is performed. This is an analyst's temporal constraints and cognitive limitations, and generally only one interpretation is drawn, and biased thinking can be reflected. Therefore, it is essential to aware the battlefield situation of the command and control system and to establish the intellignet decision support system. To do this, it is necessary to build a knowledge base specialized in the command and control system and develop intelligent battlefield awareness services based on it. In this paper, among the entity names suggested in the exobrain corpus, which is the private data, the top 250 types of meaningful names were applied and the weapon system entity type was additionally identified to properly represent battlefield information. Based on this, we proposed a way to build a battlefield-aware knowledge base through mention extraction, cross-reference resolution, and relationship extraction.