• Journal of Biomedical Engineering Research
• /
• v.20 no.1
• /
• pp.59-68
• /
• 1999

In this paper, we propose an algorithm to locate the corpus callosum automatically from midsagittal brain MR images using the statistical characteristics and shape information of the corpus callosum. In the proposed algorithm, we first extract regions satisfying the statistical characteristics of the corpus callosum and then find a region matching the shape information. In order to match the shape information, a new directed window region-growing algorithm is proposed instead of using conventional contour matching algorithms. Using the proposed algorithm, we adaptively relax the statistical requirement until we find a region matching the shape information. Experiments show promising results.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.1019-1022
• /
• 2005

본 논문에서는 현재 활발한 연구개발이 이루어지고 있는 유비쿼터스 컴퓨팅, 센서 네트워크 기술을 적용한 무인 감시/추적 시스템을 제시한다. 본 논문의 무인 감시/추적 시스템은 센서네트워크 기술, 다중센서 융합에 의한 탐지 및 위치 인식기술, 무인 감시/추적 알고리즘으로 구성되어 있다. 센서네트워크는 센싱 데이터를 실시간으로 전송하기 위해 노드의 주소를 기반으로 하는 계층적 멀티홉 라우팅 기법을 제안하였다. 침입자와 추적자의 위치 인식은 자기센서 및 초음파센서를 가진 센서모듈들로부터 얻어진 센싱 정보를 융합하고, 이를 확률적으로 침입자 및 추적자의 위치를 결정하는 Particle Filter를 적용한 위치인식 알고리즘을 통해 이루어진다. 추적 알고리즘은 무인 자율 추적을 위해 이동벡터에 기반한 알고리즘이다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 1998.04a
• /
• pp.143-147
• /
• 1998

카메라 영상에 의한 자동차 번호판 인식시스템은 영상 획득, 번호판 추출, 전처리, 문자 분리, 문자 인식 등 크게 5자기의 핵심 부분으로 구성된다. 따라서 자동차 번호판 인식시스템의 성능을 향상시키기 위해서는 이들 부분들 각각의 성능의 최적화가 필요하다. 본 연구는 자동차 번호판 인식시스템의 여러 단계 중 전처리에 해당하는 번호판 영역의 이진화에 관한 연구로서, 기존의 단일 임계치 방법과 다중 임계치 방법이 해결하지 못했던 부분을 보완하는 새로운 다중 임계치 방법을 제안한다. 본 논문에서 제안하는 다중 임계치 알고리즘(Adaptive Multi-threshold Algorithm)을 사용함으로써 gray-level 번호판 영상에 대해서 보다 깨끗한 이진 영상을 얻을 수 있었으며, 또한 이 알고리즘은 번호판 영역의 밝기값이 고르지 않은 영상에 대해서도 효율적인 알고리즘 임을 알 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.283-284
• /
• 2009

본 논문은 단일 카메라 환경에서 손 형상을 입력정보로 사용하여 손 영역만을 분할한 후 자기 조직화 특징 지도(SOFM: Self Organized Feature Map) 신경망 알고리즘을 이용하여 손 형상을 인식함으로서 수화인식을 위한 보다 안정적이며 강인한 인식 시스템을 구현하고자 한다.

• Journal of KIISE:Software and Applications
• /
• v.33 no.4
• /
• pp.440-448
• /
• 2006

Self-Organizing Map (SOM) is an unsupervised learning neural network and it is used for preserving the structural relationships in the data without prior knowledge. SOM has been applied in the study of complex problems such as vector quantization, combinatorial optimization, and pattern recognition. This paper proposes a new usage of SOM as a tool for schema transformation hoping to achieve more efficient genetic process. Every offspring is transformed into an isomorphic neural network with more desirable shape for genetic search. This helps genes with strong epistasis to stay close together in the chromosome. Experimental results showed considerable improvement over previous results.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.9
• /
• pp.731-734
• /
• 2018

Many studies have been conducted on the accurate detection and identification of targets from ground clutter, in order to improve the accuracy rate of land guided weapons. Due to the variety and complicated characteristics of the ground clutter signal compared to the target, an active target identification technique is needed. In this paper, we propose a new algorithm to identify targets and divide them into different types by extracting the unique characteristics of the target through fractal dimension calculation with the characteristics of self-similarity. In the simulation using the algorithm, the probabilities of identifying the tank and truck were 100 % and 98.89 %, respectively, and the type of the target could be identified with a probability of 98 % or more.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.04a
• /
• pp.1185-1187
• /
• 2011

이 논문에서는 운전자의 운전 패턴을 분석하기 위한 멀티 센서 기반의 패턴 분석 알고리즘을 제안한다. 센서를 통해 얻어진 주행 데이터의 상관 관계를 비교, 분석하여 주행 패턴을 인식한다. 가속도 센서에 작용하는 중력값과 지자기 센서의 방향 데이터을 통해 각 운전 패턴을 인식하는 정확도를 높이는데 이용하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.5
• /
• pp.551-556
• /
• 2004

According to revealing the DNA sequence of human and living things, it increases that a demand on a new computational processing method which utilizes DNA sequence information. In this paper we propose a classification algorithm based on negative selection of the immune system to classify DNA patterns. Negative selection is the process to determine an antigenic receptor that recognize antigens, nonself cells. The immune cells use this antigen receptor to judge whether a self or not. If one composes n group of antigenic receptor for n different patterns, they can classify into n patterns. In this paper we propose a pattern classification algorithm based on negative selection in nucleotide base level and amino acid level.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.41-47
• /
• 2022

In this paper, we propose a new and simple self-supervised learning method that predicts the middle image of a face image sequence for automatic expression recognition. Automatic facial expression recognition can achieve high performance through deep learning methods, however, generally requires a expensive large data set. The size of the data set and the performance of the algorithm are tend to be proportional. The proposed method learns latent deep representation of a face through self-supervised learning using an existing dataset without constructing an additional dataset. Then it transfers the learned parameter to new facial expression reorganization model for improving the performance of automatic expression recognition. The proposed method showed high performance improvement for two datasets, CK+ and AFEW 8.0, and showed that the proposed method can achieve a great effect.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.12A
• /
• pp.2019-2026
• /
• 2001

In this paper, we proposed the car license plate extraction and recognition algorithm using both the dynamical thresholding method and the kohonen algorithm. In general, the areas of car license plate in the car images have distinguishing characteristics, such as the differences in intensity between the areas of characters and the background of the plates, the fixed ratio of width to height of the plates, and the higher dynamical thresholded density rate 7han the other areas, etc. Taking advantage of the characteristics, the thresholded images were created from the original images, and also the density rates were computed. A candidate area was selected, whose density rate was corresponding to the properties of the car license plate obtained from the car license plate. The contour tracking method by utilizing the Kohonen algorithm was applied to extract the specific area which included characters and numbers from an extracted plate area. The characters and numbers of the license place were recognized by using Kohonen algorithm. Kohonen algorithm was very effective o? suppressing noises scattered around the contour. In this study, 80 car images were tested. The result indicate that we proposed is superior in performance.