• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.349-350
• /
• 2008

지능 로봇은 표현 가능한 사물, 공간을 모델링하기 위해 주변 환경을 인지하고, 자신이 수행할 수 있는 행동을 결합하여 임무를 수행하게 된다. 이를 위해 온톨로지를 사용하여 사물, 공간, 상황 및 행동을 표현하고 특정 임무 수행을 위한 자바 기반 Rule을 통해 다양한 추론 방법을 제공하는 로봇 지식 체계를 사용하였다. 사용된 로봇 지식 체계는 생성되는 인스턴스가 자료의 클래스와 속성 값이 일관성 있고 다른 자료와 모순되지 않음을 보장해 준다. 이러한 로봇 지식 체계를 효율적으로 사용하기 위해서는 완전한 온톨로지 인스턴스의 생성이 밑받침 되어야 한다. 하지만 실제 환경에서 로봇이 Vision Sensor를 통해 사물을 인식할 때 False Positive False Negative와 같은 인식 오류를 발생시키는 문제점이 있다. 이를 보완 하기 위해 본 논문에서는 물체와 물체간의 Spatial Relation, Temporal Relation과 각 물체마다의 인식률 및 속성을 고려하여 물체 인식 오류에서도 안정적으로 인스턴스 관리를 가능하게 하는 Rule 기반의 일식오류 검출 필터를 제안한다.

• Journal of Information Processing Systems
• /
• v.13 no.2
• /
• pp.321-339
• /
• 2017

In this work, we are interested in the extraction of areas of interest from satellite images by introducing a MO-TRIBES/OC-SVM approach. The One-Class Support Vector Machine (OC-SVM) is based on the estimation of a support that includes training data. It identifies areas of interest without including other classes from the scene. We propose generating optimal training data using the Multi-Objective TRIBES (MO-TRIBES) to improve the performances of the OC-SVM. The MO-TRIBES is a parameter-free optimization technique that manages the search space in tribes composed of agents. It makes different behavioral and structural adaptations to minimize the false positive and false negative rates of the OC-SVM. We have applied our proposed approach for the extraction of earthquakes and urban areas. The experimental results and comparisons with different state-of-the-art classifiers confirm the efficiency and the robustness of the proposed approach.

• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.267-269
• /
• 2009

Incomplete knowledge instances may be registered from misidentifications of sensors, such as vision sensor. In previous study, we proposed robust object instance registration method to robot centered knowledge framework to guarantee the consistency of the registered knowledge. In real environment, a persistent update is necessary due to the objects can be moved dynamically. In this paper, we propose the way to update robot knowledge continually using the registration method. Our experiment in this paper shows that sound and complete knowledge can be registered and updated by the proposed method, even under imperfect sensing data.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.861-862
• /
• 2008

This paper presents the results of iris region localization for imprecisely taken iris images in UBIRIS ver2. UBIRIS database consisting of 500 images of 24-bit RGB typed TIFF file. They were captured by conventional digital camera in visible light environment. Experimental result showed that FP(False Positive rate) and FN(False Negative rate) were about 2.2% and 29.1%, respectively.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.15 no.5
• /
• pp.243-249
• /
• 2020

Event detection in wireless sensor networks is a key requirement in many applications. Acoustic sensors are one of the most frequently used sensors for event detection in sensor networks, but they are sensitive and difficult to handle because they vary greatly depending on the environment and target characteristics of the sensor field. In this paper, we propose a learning-based improvement of CFAR algorithm for increasing node-level event detection performance in acoustic sensor networks, and verify the effectiveness of the designed algorithm by comparing and evaluating the event detection performance with other algorithms. Our experimental results demonstrate the superiority of the proposed algorithm by increasing the detection accuracy by more than 45.16% by significantly reducing false positives by 7.97 times while slightly increasing the false negative compared to the existing algorithm.

• The Korean Journal of Cytopathology
• /
• v.6 no.2
• /
• pp.140-147
• /
• 1995

In a six-year period (1988. 5-1994. 4), fine needle aspiration cytology(FNAC) of 322 pulmonary lesions from 296 patients were performed at Soonchunhyang University Hospital. Of these 322, malignancy was diagnosed cytologically in 139(43.2%), suspicious malignancy in 7(2.2%), negative in 164(50.8%), and insufficient material in 12(3.8%). Malignant lesions consisted of 54 cases of adenocarcinoma, 50 cases of squamous cell carcinoma, 18 cases of small cell carcinoma. They were verified by histologic examination in 70 cases. There were 2(0.6%) false positive cases due to florid bronchoalveolar hyperplasia and atypical bronchial epithelial cells associated with granulomatous lesion. The overall accuracy rate was 90%, the sensitivity 84.3% and the specificity 94.7%.

• Proceedings of the Korean Society of Applied Pharmacology
• /
• 1993.04a
• /
• pp.171-171
• /
• 1993

발암물질의 조기검색법을 개발하고자 변이원성 물질의 스크리닝법으로 널리 사용되고 있는 Ames test 및 chromosomal aberration test를 본 연구에서 개발하고자 하는 DNA 및 Protein-adduct 형성시험법과 비교 연구하였다. 벤조피렌과 아플라톡신 B$_1$을 모델 발암물질로 하여 실시한 Ames test에서는 두 화합물 모두 양성을 나타냈으나 용량-반응 관계가 뚜렷하지 않았다. 또한 고농도에서는 시험물질의 독성체 의해 정상적인 Ames test의 수행이 어려됐다. Chromosomal aberration test에서도 Ames test와 비슷한 결과를 나타냈으며 특히 고농도에서 시험을 실시했을 경우 Ames test에서와 마찬가지로 세포독성의 현상이 관찰되었다. 그러나 본 연구에서 새로이 개발한 DNA 및 Protein-adduct형성 시험법은 저농도에서 고농도에 이르기까지 뚜렷한 용량-반응 관계를 나타냈으며 Ames test 및 chromosomal test에서 일어날 수 있는 false positive나 false negative의 결과를 가져다 줄 우려가 없다. 또한 시험시간이 1-2시간 정도 소요되므로 기존의 방법보다 시험시간을 약 40배 가량 단축시킬 수 있었다.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.5
• /
• pp.469-475
• /
• 2010

This paper presents a method of classifying discolored defects of metal pads using geometric texture for AFVI (Automated Final Vision Inspection) systems. In PCB manufacturing process, the metal pads on PCB can be oxidized and discolored partly due to various environmental factors. Nowadays the discolored defects are manually detected and rejected from the process. This paper proposes an efficient geometric texture feature, SUTF (Symmetry and Uniformity Texture Feature) based on the symmetric and uniform textural characteristics of the surface of circular metal pads for automating AFVI systems. In practical experiments with real samples acquired from a production line, 30 discolored images and 1232 roughness images are tested. The experimental results demonstrate that the proposed method using SUTFs provides better performance compared to Gabor feature with 0% FNR (False Negative Rate) and 1.46% FPR (False Positive Rate). The performance of the proposed method shows its applicability in the real manufacturing systems.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04a
• /
• pp.380-382
• /
• 2003

현재 네트워크 보안 취약점을 자동으로 검색해주는 다양한 도구들이 인터넷에 공개되어 있어 이러한 도구들을 이용한 취약점 정보수집 및 네트워크 검색공격으로 비롯된 해킹사고가 크게 증가하고 있다. 이와 같은 검색 공격에 대한 탐지 시스템은 "False-Positive(실제 공격이 아닌데 공격이라고 탐지)"와 "False-Negative(실제 공격인데 공격이 아니라고 탐지)"를 줄이는 것이 중요하다. 그러나 현재 공개되어 있는 실시간 스캔 탐지 시스템은 오탐율이 높을 분만 아니라 다양한 스캔 기법에 대해서 탐지를 할 수 없는 것이 사실이다. 본 논문에서는 다양한 포트스캐닝 기법기반의 공격에 대해서 탐지 가능하고 오탐율을 최소화한 실시간 스캔 탐지 시스템을 구현한다.실시간 스캔 탐지 시스템을 구현한다.

• The Korean Journal of Cytopathology
• /
• v.18 no.2
• /
• pp.112-118
• /
• 2007

Fine needle aspiration cytology (FNAC) has become a highly preferred, minimally invasive diagnostic tool of choice in the diagnosis of a palpable breast mass owing to its sensitivity, specificity, cost-effectiveness, and expediency. Although breast needle biopsies have been widely employed recently due to the increased detection rate of non-palpable early lesions, the importance of the use of FNAC cannot be underestimated. It comprises part of the diagnostic triad for the breast along with a physical examination and mammography, which together contribute to an increasing diagnostic accuracy. The differential diagnosis of a benign and malignant lesion is of the utmost importance in the diagnosis of breast lesions, and therefore the understanding of the possible diagnostic pitfalls is of great importance.