• 한국철도학회논문집
• /
• 제8권6호
• /
• pp.559-565
• /
• 2005

In this paper, we propose a stereoscopic video coding scheme for subway accident monitoring system. The proposed designed for providing flexible video among various displays, such ass control center, station employees and train driver. We uses MPEG-2 standard for coding the left-view sequence and IBMDC coding scheme predicts matching block by interpolating both motion and disparity predicted macroblocks. To provide efficient stereoscopic video service, we define both temporally and spatially scalable layers for each eye's-view by using the concept of Spatio-Temporal scalability. The experimental results show the efficiency of proposed coding scheme by comparison with already known methods and the advantages of disparity estimation in terms of scalability overhead. According to the experimental results, we expect the proposed functionalities will play a key role in establishing highly flexible stereoscopic video codec for ubiquitous display environment where devices and network connections are heterogeneous.

• 한국통신학회논문지
• /
• 제22권5호
• /
• pp.967-973
• /
• 1997

In this paper, the designed and fabrication of a K-band push-push oscillator using miniaturized hairpin resonator have been presented. One experimenal oscillator has been designed and fabricated for K-band point-to-point operation. the miniaturized harpin resonator has been analyzed theoretically and simulated by MPIE(Mixed Potential Integral Equation) method. With this results, the analysis of hairpin resonator which coupled microstrip line has been carried out with transmission-mode using this results. an optimized output matching network for the suppression of the fundamental and the 3rd order harmonic was acquired by using a nonlinear analysis method. The fabricated oscillator shows the output power of -2.28dBm, the fundamental frequency suppression of -19dBc, the 3rd order harmonic suppressionof -24dBc and 0.33 percent effiiency at 22.8GHz. The experimental outputs are in good agreement with the theoretical and simulated results.

• 한국안전학회지
• /
• 제11권3호
• /
• pp.143-153
• /
• 1996

Shutdown of boiler plants is a dynamic, complicated, and hazardous operation. Operational error is a major contributor to danserous situations during boiler plant shutdowns. It is important to develop an automatic system which synthesizes operating procedures to safely go from normal operation to complete shutdown. Knowledge representation for automatic shutdown of boiler plants makes use of the hierarchical, rule-based framework for heuristic knowledge, the semantic network, frame for process topology, and AI techniques such as rule matching, forward chaining, backward chaining, and searching. This knowledge representation and modeling account for the operational states, primitive operation devices, effects of their application, and planning methodology. Also, this is designed to automatically formulate subgoals, search for positive operation devices, formulate constraints, and synthesize shutdown procedures in boiler plants.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2004년도 추계학술발표논문집(상)
• /
• pp.513-516
• /
• 2004

젤 영상에서 스팟을 탐지한 후, 스팟 사이의 일치 여부를 판단하여 새로운 단백질의 생성되었는지 없어진 단백질이 있는지 알아내게 된다. 젤 영상은 만들어지는 과정에서 같은 단백질이라도 스팟의 위치가 조금씩 다르게 된다. 스팟 사이의 관계는 비선형 변환에 해당하고, 각 스팟 사이의 매치는 NP 문제임이 증명되었고, 이를 해결하기 위한 휴리스틱 방법이 보고되었다. 최적화에 좋은 성능을 보이고 있는 홉필드 신경회로망을 젤 매치에 적용하는 방법을 연구하였다. 홉필드 신경망의 각 뉴런은 뉴런이 대표하는 두 스팟이 일치할 때 활성화되고, 일치하지 않을 때 활성화되지 않도록 하였다. 각 뉴런의 상태를 전체 에너지가 줄어드는 방향으로 변경하면 결국 안정된 상태에 도달하게 되고, 이 때 각 뉴런은 가능한 매치를 표현하게 된다.

• Journal of Astronomy and Space Sciences
• /
• 제33권4호
• /
• pp.313-318
• /
• 2016

Communication, Ocean And Meteorological Satellite (COMS) Meteorological Imager (MI) images are processed for radiometric and geometric correction from raw image data. When intermediate image data are matched and compared with reference landmark images in the geometrical correction process, various techniques for edge detection can be applied. It is essential to have a precise and correct edged image in this process, since its matching with the reference is directly related to the accuracy of the ground station output images. An edge detection method based on neural networks is applied for the ground processing of MI images for obtaining sharp edges in the correct positions. The simulation results are analyzed and characterized by comparing them with the results of conventional methods, such as Sobel and Canny filters.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2018년도 추계학술발표대회
• /
• pp.617-620
• /
• 2018

딥러닝 기술에 있어서 대량의 학습 데이터가 필요하다는 한계점을 극복하기 위한 시도로서, 적은 데이터 만으로도 좋은 성능을 낼 수 있는 few-shot 학습 모델이 꾸준히 발전하고 있다. 하지만 few-shot 학습 모델의 가장 큰 단점인 적은 데이터로 인한 과적합 문제는 여전히 어려운 숙제로 남아있다. 본 논문에서는 모델 압축에 사용되는 distillation 기법을 사용하여 few-shot 학습 모델의 학습 문제를 개선하고자 한다. 이를 위해 대표적인 few-shot 모델인 Siamese Networks, Prototypical Networks, Matching Networks에 각각 distillation을 적용하였다. 본 논문의 실험결과로써 단순히 결과값에 대한 참/거짓 뿐만 아니라, 참/거짓에 대한 신뢰도까지 같이 학습함으로써 few-shot 모델의 학습 문제 개선에 도움이 된다는 것을 실험적으로 증명하였다.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2005년도 하계학술대회 논문집 Vol.6
• /
• pp.568-569
• /
• 2005

한빛 자기거울 장치는 고온 플리즈마 물성을 연구하기 위한 장치로서 플러즈마 밀도 형성을 위한 slot 형 안테나 고주파 가열 시스템이 중앙 진공용기에 설치되어 있다. 본 연구에서는 이러한 고주파 전송선로, 임피던스 정합 network. 장치 임피던스를 포함하는 한빛 장치의 고주파 가열 시스템에 대하여 기존에 정립된 고주파 가열 이론[1]을 기반으로 하여 이론적인 해석만으로 구성된 회로모델을 완성하였다. 임피던스 정합 소자 값들은 임피던스 정합 조건으로 결정함으로써 다양한 장치 및 플라즈마 변수들의 함수로 표현하여 그 의존성을 조사하였다.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제11권4호
• /
• pp.19-38
• /
• 2007

Graph theory is becoming increasingly significant as it is applied of mathematics, science and technology. It is being actively used in fields as varied as biochemistry(genomics), electrical engineering(communication networks and coding theory), computer science(algorithms and computation) and operations research(scheduling). The powerful results in other areas of pure mathematics. Rhis paper, besides giving a general outlook of these facts, includes new graph theoretical proofs of Fermat's Little Theorem and the Nielson-Schreier Theorem. New applications to DNA sequencing (the SNP assembly problem) and computer network security (worm propagation) using minimum vertex covers in graphs are discussed. We also show how to apply edge coloring and matching in graphs for scheduling (the timetabling problem) and vertex coloring in graphs for map coloring and the assignment of frequencies in GSM mobile phone networks. Finally, we revisit the classical problem of finding re-entrant knight's tours on a chessboard using Hamiltonian circuits in graphs.

• 스마트미디어저널
• /
• 제7권4호
• /
• pp.61-69
• /
• 2018

Multi-tracking of general objects and specific faces is an important topic in the field of computer vision applicable to many branches of industry such as biometrics, security, etc. The rapid development of deep neural networks has resulted in a dramatic improvement in face recognition and object detection problems, which helps improve the multiple-face tracking techniques exploiting the tracking-by-detection method. Our proposed method uses face detection trained with a head dataset to resolve the face deformation problem in the tracking process. Further, we use robust face features extracted from the deep face recognition network to match the tracklets with tracking faces using Hungarian matching method. We achieved promising results regarding the usage of deep face features and head detection in a face tracking benchmark.

• International journal of advanced smart convergence
• /
• 제10권3호
• /
• pp.163-171
• /
• 2021

Recently, Machine Learning-based visualization approaches have been proposed to combat the problem of malware detection. Unfortunately, these techniques are exposed to Adversarial examples. Adversarial examples are noises which can deceive the deep learning based malware detection network such that the malware becomes unrecognizable. To address the shortcomings of these approaches, we present Block-matching and 3D filtering (BM3D) algorithm and deep image prior based denoising technique to defend against adversarial examples on visualization-based malware detection systems. The BM3D based denoising method eliminates most of the adversarial noise. After that the deep image prior based denoising removes the remaining subtle noise. Experimental results on the MS BIG malware dataset and benign samples show that the proposed denoising based defense recovers the performance of the adversarial attacked CNN model for malware detection to some extent.