http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : 최태완 (검색결과 50 건)
- 무료
- 기관 내 무료
- 유료
-
-
최태완,권율,김재창,남기곤,윤태훈,Choi, Tae-Wan,Kwon, Yool,Kim, Jae-Chang,Nam, Ki-Gon,Yoon, Tae-Hoon 대한전자공학회 1995 전자공학회논문지-B Vol.b32 No.1
The diffusion neural network can be cfficiently applied to the Gaussian processing. For example, a difference of two Gaussians(DOG) is performed by this network with ease. In this paper, we model a neural network to perform the function /t(.del.${\Delta}^{2}$G) by using the diffusion neural network. This model is used to detect the edges of moving target in image. By this model not only moving target is separated from stationary background but also their trajectories are obtained using accumulated past information in the diffusion neural network. Furthermore this model needs a small number of connections per cell and the connection weights are fixed-valued. Therefore its hardware can be easily implemented with simple structure.
-
-
방향성 계층적 시공간 필터에 의한 움직이는 물체의 검출
최태완,김재창,윤태훈,남기곤 대한전자공학회 1996 전자공학회논문지-B Vol.b33 No.9
In this paper, we popose a neural network that detects edges of moving objects in an image using a neural network of hierarchical spatial filter with orientation selectivity. We modify the temporal difference network by adding a self loop to each neuraon cell to reduce the problems of phantom edge detected by the neural network proposed by kwon yool et al.. The modified neural network alleviates the phantom edges of moving objects, and also can detect edges of miving objects even for the noisy input. By computer simulation with real images, the proposed neural netowrk can extract edges of different orientation efficiently and also can reduce the phantom edges of moving objects.
-
최태완 진주산업대학교 1997 論文集 Vol.36 No.-
In this paper, the design technique of a rotary transformer for VCR is reported. Through the analysis of equivalent circuit, the primary coil inductance and the turn ratio are chosen to maximize the bandwidth and the gain of the reproduction system. The magnetic circuits and the crosstalk of a rotary transformer are theoretically analyzed. The coil turn ratio and the width between the two channels are determined to satisfy each channel pairing and the crosstalk.
-
실시간 처리를 위한 컨테이너 ISO코드 인식시스템의 구현
최태완,Choi Tae-Wan 한국정보통신학회 2006 한국정보통신학회논문지 Vol.10 No.8
컨테이너 ISO코드 인식시스템은 ISO코드 검출 및 영상 획득, ISO코드 영역 추출, 개별 문자 추출, 문자인식 및 데이터베이스의 5가지 핵심부분으로 구성된다. 이 중에서도 ISO코드 추출의 정확성은 전체 시스템 인식률에 지대한 영향을 줄 수 있는 부분이며, 다양한 컨테이너 종류 및 주위 환경 변화에서도 정확한 추출을 요구한다. 본 논문에서는 획득된 영상을 주위 환경 변화에도 적응 가능한 이 진화 방법을 사용하여 ISO코드 템플릿의 영역을 이 진화하고 ISO코드의 분포를 가지는 후보 영역을 추출한다. 추출된 후보 영역 중에서 ISO코드 문자 분포의 특성을 이용한 검증과정을 통해 최종 영역을 추출하여 ISO코드를 인식하는 시스템을 설계 및 구현하였다. 구현된 시스템을 실시간으로 컨테이너에서 획득한 영상에 적용한 결과 다양한 컨테이너 종류 및 주위 환경변화에서도 ISO코드 영역이 정확히 추출됨을 확인하였다. This paper describes system to extract ISO codes in container image. A container ISO code recognition system for real-time processing is made of 5 core parts which are container ISO code detection and image acquisition, ISO code region extraction, individual character extraction, character recognition and database. Among them, the accuracy of ISO code extraction can affect significantly the accuracy of system recognition rate, and also the more exact extraction of ISO code is required in various weather and environment conditions. The proposed system produces binary of the ISO code's template lesions using an adaptive thresholding, extracts candidate regions containing distribution of ISO code, and recognizes ISO codes as detecting a final region through the verifications by using character distribution characteristics of ISO code among the extracted candidates. Experimental results reveal that ISO codes can be efficiently extracted by the proposed method.
-
崔泰完 진주산업대학교 1998 論文集 Vol.37 No.-
This paper presents a method using a diffusion neural network for computing the convolution of an image with LoG(Laplacian-of-Gaussian) masks. The diffusion neural network can be efficiently applied to the Gaussian processing. For example, a DOG(difference of two Gaussians) is performed by this network with ease. The neural network has band-pass filter characteristics. This paper takes advantage of the specific spectral characteristic of a LoG filter in computation. The proposed method leads to the result that the computation time decreases when σ increases. This paper illustrates the proposed method on an image.
-
최태완,Choi Tae-Wan 한국정보통신학회 2006 한국정보통신학회논문지 Vol.10 No.5
교통제어나 차량에 연관된 범죄 등에서 자동차의 인식에 관한 연구의 중요성 때문에 이에 관련된 연구는 오래전부터 수행되어 왔다. 본 논문에서는 차량이 주행할 때의 정보와 영상을 획득하여 제조회사별 차종을 인식하는 방법을 제안하고자 한다. 본 논문의 차종인식은 차량의 압력을 이용한 차폭 검출방법, 그리고 보다 더 정확한 인식률을 얻기 위한 레이저 거리계를 이용한 차고 검출방법, $3\sim5$종의 구별을 위 한 차량의 번호판 인식 방법을 조합함으로써 차량 인식의 오류를 줄이는 시스템을 구현하였다. 구현된 차종인식 시스템은 2차원 CCD에 의한 차량의 영상획득과 이를 통한 다양한 영상처리 알고리즘에 의해서 국내의 전 차종에 적용할 수 있으며, 실제의 실험 결과는 높은 인식률을 나타내었다. This paper introduces a simple but effective method for recognizing vehicle models corresponding to each maker by information and images for moving vehicles. The proposed approach is implemented by combination of the breadth detection mechanism using the vehicle's pressure, exact height detection by a laser scanning, and license plate recognition for classifying specific vehicles. The implemented system is therefore capable of robust classification with real-time vehicle's moving images and established sensors. Simulation results using the proposed method on synthetic data as well as real world images demonstrate that proposed method can maintain an excellent recognition rate for moving vehicle models because of image acquisition by 2-D CCD and various image processing algorithms.
-
-
최태완,류대현 진주산업대학교 2004 論文集 Vol.43 No.-
In this paper, a target tracking method using the diffusion neural network is proposed. The proposed neural network can more easily segment the moving target than the previous work in images with complex and noisy background. Moreover, it can well trace the target when those are occluded by an object. Simulation results show that by the neural network not only moving target is separated traced from stationary background but also their trajectories are obtained using accumulated past information in the diffusion neural network. Furthermore this neural network need a small number of connections per cell and the connection weights are fixed-valued. Therefore its hardware can be easily implemented with a simple structure.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
