• 전자공학회논문지B
• /
• 제33B권2호
• /
• pp.197-208
• /
• 1996

A neural network-based PSRI(position, scale, and rotation invariant) feature extraction and ATR (automatic target recognition) system are proposed and an efficient parallel implementatio of the proposed system using multicomputer is also presented. In the proposed system, the scale and rotationinvariant features are extracted from the contour projection of the number of edge pixels on each of the concentric circles, which is input t the cooperative network. We proposed how to decide the optimum depth and the width of the parallel pipeline system for real time applications by modeling the proposed system into a parallel pipeline implementation method using transputers is also proposed. The implementation results show that we can extract PSRI features less sensitive to input variations, and the speedup of the proposed ATR system is about 7.55 for the various rotated and scaled targets using 8-node transputer system.

• 한국CDE학회논문집
• /
• 제17권1호
• /
• pp.7-16
• /
• 2012

One distinguishing feature of BIM(Building Information Modeling) is the objectification of spatial elements independently, which makes it easy to represent spatial network. From this perspective, this study aimed to develop the spatial network analysis tool based on open BIM technologies. From the literature review, an object model of spatial network with nodes and links and a process model from construction to visualization were established. A prototype system implementing the proposed models, named SNAT(Spatial Network Analysis Tool), was developed in Java platform with using its open source packages. SNAT can create a spatial network from IFC-BIM model, calculate the indices of spatial network analysis, and visualize it with the representing types(map, graph, matrix and table).

• ETRI Journal
• /
• 제34권4호
• /
• pp.583-590
• /
• 2012

In this paper, we present a new approach for speaker diarization. First, we use the prosodic information calculated on the original speech to resynthesize the new speech data utilizing the spectrum modeling technique. The resynthesized data is modeled with sinusoids based on pitch, vibration amplitude, and phase bias. Then, we use the resynthesized speech data to extract cepstral features and integrate them with the cepstral features from original speech for speaker diarization. At last, we show how the two streams of cepstral features can be combined to improve the robustness of speaker diarization. Experiments carried out on the standardized datasets (the US National Institute of Standards and Technology Rich Transcription 04-S multiple distant microphone conditions) show a significant improvement in diarization error rate compared to the system based on only the feature stream from original speech.

• 한국CDE학회논문집
• /
• 제5권4호
• /
• pp.312-318
• /
• 2000

In this study, an exclusive CAD/CAM system is developed for enhancing the effectiveness and productivity of CNC universal cylindrical grinding machines on which the external/facing/internal grinding cycles and the wheel dressing cycles are integratively carried out. The CAD/CAM system can manage the various processes such as geometry design, NC code generation, NC code verification, DNC operation, and so on. Especially, the feature-based modeling concept is introduced to improve the geometry design efficiency. And the NC code verification is realized by virtual manufacturing technique based on the real-time analysis of NC codes and the boolean operation between workpiece and wheel.

• 한국멀티미디어학회논문지
• /
• 제14권5호
• /
• pp.642-649
• /
• 2011

Among the various aspects of offline handwritten character patterns, it is the great variety of writing styles and variations that renders the task of computer recognition very hard. The immense variety of character shape has been recognized but rarely studied during the past decades of numerous research efforts. This paper tries to address the problem of measuring image distortions and handwritten character patterns with respect to reference patterns. This work is based on mass-spring mesh model with the introduction of simulated electric charge as a source of the external force that can aid decoding the shape distortion. Given an input image and a reference image, the charge is defined, and then the relaxation procedure goes to find the optimum configuration of shape or patterns of least potential. The relaxation process is based on the fourth order Runge-Kutta algorithm, well-known for numerical integration. The proposed method of modeling is rigorous mathematically and leads to interesting results. Additional feature of the method is the global affine transformation that helps analyzing distortion and finding a good match by removing a large scale linear disparity between two images.

• 한국멀티미디어학회논문지
• /
• 제25권6호
• /
• pp.807-818
• /
• 2022

In order to further improve the accuracy and time efficiency of behavior recognition in intelligent monitoring scenarios, a human behavior recognition algorithm based on YOLO combined with LSTM and CNN is proposed. Using the real-time nature of YOLO target detection, firstly, the specific behavior in the surveillance video is detected in real time, and the depth feature extraction is performed after obtaining the target size, location and other information; Then, remove noise data from irrelevant areas in the image; Finally, combined with LSTM modeling and processing time series, the final behavior discrimination is made for the behavior action sequence in the surveillance video. Experiments in the MSR and KTH datasets show that the average recognition rate of each behavior reaches 98.42% and 96.6%, and the average recognition speed reaches 210ms and 220ms. The method in this paper has a good effect on the intelligence behavior recognition.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2022년도 제66차 하계학술대회논문집 30권2호
• /
• pp.95-97
• /
• 2022

본 논문에서는 산업 현장에서 발생하는 불량품 탐지 분야에서 효율적으로 생산품의 불량을 탐지할 수 있는 PaDiM 구조의 Backbone 모델을 단일 Wide-ResNet 대신 두 개의 Wide-ResNet을 사용함으로써, 단일 모델에서 추출된 저차원의 Feature를 앙상블을 통해 성능 향상을 일으킬 수 있는 것을 증명하였다. 단일 Wide-ResNet 환경에서는 MVTec 데이터셋에서 생성된 다변량 가우시안 분포가 데이터셋의 적은 샘플수로 인하여 각 클래스 간 불균형이 발생하는 문제를 동종 앙상블을 통해 해결할 수 있었다. 따라서 본 논문에서는 제안하는 동종 모델의 앙상블을 사용함으로써 기존의 One-class classification 환경에서 불량품 탐지환경에서 적은 수의 데이터 샘플 환경에서 성능 향상을 나타낼 수 있음을 입증하였다.

• 정보처리학회논문지D
• /
• 제12D권4호
• /
• pp.589-600
• /
• 2005

다양하고 빠르게 변화하는 사용자의 요구사항을 만족시키기 위하여 재사용을 고려한 애플리케이션의 개발이 요구되어지고 있다. 이를 통해 대두된 것이 프로덕트 라인이다. 프로덕트 라인은 다양하게 빠르게 변화하는 시장의 요구사항과 특정 도메인 영역에 속하는 애플리케이션 간의 재사용 가능한 아키텍처 및 컴포넌트의 구성으로부터 연관된 시스템 구축 시 생산성과 품질의 향상을 제공함으로써 현재 많은 관심의 초점이 되고 있다. 또한, 이들 프로덕트 라인에서는 프로덕트들 사이의 공통성과 변화성에 초점을 두고 이들 분류 방법으로 휘처 모델링 이라는 개념을 주로 사용하여 분석하고 있다. 또한 재사용 가능한 아키텍처는 많은 변화 계획들과 메커니즘을 포함하고 있다. 그러나 지금까지 이러한 변화들이 일어나는 상황을 이해하는 것과 특별한 상황에서도 가능하게 하는 옵션들을 기록하는 것은 명확히 이루어지지 못하였다. 아키텍처가 오랜 기간 동안 많은 프로덕트 버전에서 사용되어 진다거나, 다른 프로덕트들의 설계를 위해 사용되어지는 아키텍처에서의 프로덕트 라인 문맥에서라면, 매우 중요하게 다루어진다. 즉, 명백한 변화성의 표현과, 아키텍처에서 변경이 되는 적절한 위치를 식별하는 것이 중요하다. 하지만, 아키텍처를 설계하기 위한 아키텍처에서의 변화성 관리에 대한 명확한 방법이 미흡하다. 따라서 본 논문에서는 재사용 가능한 아키텍처를 설계하기 위해 변화성의 명확한 표현과 아키텍처에서의 적절한 위치를 식별하기 위해, 다양한 변화성 타입을 정의하고, 휘처 모델을 기반으로 한 아키텍처의 변화성과 아키텍처의 컴포넌트 관련성에서의 변화성 표현 방법을 기술하고, 제시한 이론을 기반으로 웹 학습 시스템의 설계과정을 거쳐 구현하고자 한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권2호
• /
• pp.775-792
• /
• 2015

Person re-identification is an important and challenging task in computer vision with numerous real world applications. Despite significant progress has been made in the past few years, person re-identification remains an unsolved problem. This paper presents a novel appearance-based approach to person re-identification. The approach exploits region covariance matrix and color histograms to capture the statistical properties and chromatic information of each object. Robustness against low resolution, viewpoint changes and pose variations is achieved by a novel signature, that is, the combination of Log Covariance Matrix feature and HSV histogram (LCMH). In order to further improve re-identification performance, third-party image sets are utilized as a common reference to sufficiently represent any image set with the same type. Distinctive and reliable features for a given image set are extracted through decision boundary between the specific set and a third-party image set supervised by max-margin criteria. This method enables the usage of an existing dataset to represent new image data without time-consuming data collection and annotation. Comparisons with state-of-the-art methods carried out on benchmark datasets demonstrate promising performance of our method.

• 한국정보시스템학회지:정보시스템연구
• /
• 제31권4호
• /
• pp.47-73
• /
• 2022

Purpose The purpose of this study is to select the electronic architecture concept of the powertrain domain of the electronic platform to be applied to electric vehicles after 2025. Previously, the automotive electrical and electronic architecture was determined only by trend analysis, but the purpose was to determine the scenario based on the data and select it with clear evaluation indicators. Design/methodology/approach This study identified the function to be applied to the powertrain domain of next-generation electric vehicle, estimated the controller, defined the function feature list, organized the scenario candidates with the controller list and function feature list, and selected the final architecture scenario. Findings According to the research results, the powertrain domain of electric vehicles was selected as the architectural concept to apply the DCU (Domain Control Unit) and VCU (Vehicle Control Unit) integrated architecture to next-generation electric vehicles. Although it is disadvantageous or equivalent in terms of cost, it was found to be excellent in most indicators such as stability, security, and hardware demand.