• 정보통신설비학회논문지
• /
• 제11권1호
• /
• pp.23-28
• /
• 2012

In this paper, we proposed and developed a fast and efficient real time face detection and recognition which can be run on embedded system instead of high performance desktop. In the face detection process, we detect a face by finding eye part which is one of the most salient facial features after applying various image processing methods, then in the face recognition, we finally recognize the face by comparing the current face with the prepared face database using a template matching algorithm. Also we optimized the algorithm in our system to be successfully used in the embedded system, and performed the face detection and recognition experiments on the embedded board to verify the performance. The developed method can be applied to automatic door, mobile computing environment and various robot.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(5)
• /
• pp.51-54
• /
• 2001

For performing Intelligent Transport System, I will implement the vehicle recognition system used image as well as Real time transmission module with TCP/IP. Generally, there are two kinds for vehicle recognition technology. Between two methods, the image-based method has a high dependency on environmental condition. Therefore, in this thesis I will propose vehicle recognition algorithm based on image using improved Gradient Method. Also I propose server-client modeling with TCP/IP, for the purpose of real time transmission of processed data and images. So I made the vehicle recognition and real time transmission system with TCP/IP for verification of proposed algorithm.

• 한국멀티미디어학회논문지
• /
• 제19권10호
• /
• pp.1759-1766
• /
• 2016

We can implement the virtual touch recognition system by mounting the virtual touch algorithm into an embedded device connected to a depth camera. Since the computing performance is limited in embedded system, the real-time processing of recognizing the virtual touch is difficult when the resolution of the depth image is large. In order to resolve the problem, this paper improves the algorithms of binarization and labeling that occupy a lot of time in all processing of virtual touch recognition. It processes the binarization and labeling in only necessary regions rather than all of the picture. By appling the proposed algorithm, the system can recognize the virtual touch in real-time as about 31ms per a frame in the depth image that has 640×480 resolution.

• 융합신호처리학회 학술대회논문집
• /
• 한국신호처리시스템학회 2001년도 하계 학술대회 논문집(KISPS SUMMER CONFERENCE 2001
• /
• pp.149-152
• /
• 2001

본 논문은 이미 판서된 오프라인(off-line) 영문 인쇄체를 실시간으로 인식하기 위해 인공신경망의 역전파 (Backpropagation) 학습알고리즘을 적용하여 인식 시스템의 성능을 최대화하고, 양질의 특성벡터를 추출함으로서 실시간 처리가 가능하도록 처리시간을 단축시키는 것을 목적으로 하였다. 실시간 영상을 획득하고 처리하기 위한 Genesis 실시간 영상처리 보드와 이 보드를 제어하기 위한 MIL(Matrox Image Library)패키지를 이용하여 실시간 인식시스템을 구현하였고, 인공신경망의 기대값을 ASCII형태로 변환시켜 출력벡터의 차수를 감소시키는 방법을 제시함으로서 패턴의 학습과 인식처리에 소요되는 시간, 그리고 인식시스템의 성능을 비교해 보았다.

• 제어로봇시스템학회논문지
• /
• 제12권9호
• /
• pp.935-944
• /
• 2006

EMG pattern recognition is essential for the control of a multifunction myoelectric hand. The main goal of this study is to develop an efficient feature projection method for EMC pattern recognition. To this end, we propose a linear supervised feature projection that utilizes linear discriminant analysis (LDA). We first perform wavelet packet transform (WPT) to extract the feature vector from four channel EMC signals. For dimensionality reduction and clustering of the WPT features, the LDA incorporates class information into the learning procedure, and finds a linear matrix to maximize the class separability for the projected features. Finally, the multilayer perceptron classifies the LDA-reduced features into nine hand motions. To evaluate the performance of LDA for the WPT features, we compare LDA with three other feature projection methods. From a visualization and quantitative comparison, we show that LDA has better performance for the class separability, and the LDA-projected features improve the classification accuracy with a short processing time. We implemented a real-time pattern recognition system for a multifunction myoelectric hand. In experiment, we show that the proposed method achieves 97.2% recognition accuracy, and that all processes, including the generation of control commands for myoelectric hand, are completed within 97 msec. These results confirm that our method is applicable to real-time EMG pattern recognition far myoelectric hand control.

• 정보와 통신
• /
• 제10권10호
• /
• pp.22-29
• /
• 1993

This paper describes development of a real-time voice recognition dialing system which can recognize around one hundred word vocabularies in speaker independent mode. The voice recognition algorithm is implemented on a DSP board with a telephone interface plugged in an IBM PC AT/486. In the DSP board, procedures for feature extraction, vector quantization(VQ), and end-point detection are performed simultaneously in every 10msec frame interval to satisfy real-time constraints after the word starting point detection. In addition, we optimize the VQ codebook size and the end-point detection procedure to reduce recognition time and memory requirement. The demonstration system is being displayed in MOBILAB of Korea Mobile Telecom at the Taejon EXPO '93.

• 한국컴퓨터정보학회논문지
• /
• 제27권4호
• /
• pp.19-26
• /
• 2022

최근 지능형 교통 체계의 발전에 따라 자동차 번호판 인식 시스템이 다양한 분야에서 활용되고 있다. 주행 중인 자동차의 번호판을 인식하기 위해서는 실시간성이 보장되어야 하며, 영상이 왜곡되어 뚜렷하지 않거나 번호판의 크기가 작은 저해상도 영상에서도 높은 인식률이 유지되어야 한다. 본 논문에서는 자유 앵커 방식 기반의 객체 탐지 알고리즘과 합성곱 신경망(CNN) 기반의 문자 인식 알고리즘을 이용하여 처리 속도를 향상한 실시간 자동차 번호판 인식 시스템을 제안한다. 더불어 공간 변형 네트워크를 이용하여 저해상도 및 왜곡된 영상에서의 인식률을 높였다. 제안하는 시스템의 인식률은 93.769%, 이미지 당 처리 속도는 약 0.006초로 기존 자동차 번호판 인식 시스템보다 빠른 속도로 자동차 번호판을 인식하며, 다양한 환경 및 품질의 영상에 대해 높은 인식률을 유지하는 것을 확인할 수 있다.

• 대한임베디드공학회논문지
• /
• 제10권6호
• /
• pp.363-372
• /
• 2015

Recent some SoC FPGA Releases that integrate ARM processor and FPGA fabric show better performance compared to the ASIC SoC used in typical embedded image processing system. In this study, using the above advantages, we implement a SoC FPGA-based Real-Time Object Recognition and Tracking System. In our system, the video input and output, image preprocessing process, and background subtraction processing were implemented in FPGA logics. And the object recognition and tracking processes were implemented in ARM processor-based programs. Our system provides the processing performance of 5.3 fps for the SVGA video input. This is about 79 times faster processing power than software approach based on the Nios II Soft-core processor, and about 4 times faster than approach based the HPS processor. Consequently, if the object recognition and tracking system takes a design structure combined with the FPGA logic and HPS processor-based processes of recent SoC FPGA Releases, then the real-time processing is possible because the processing speed is improved than the system that be handled only by the software approach.

• 인터넷정보학회논문지
• /
• 제6권3호
• /
• pp.147-158
• /
• 2005

나날이 심각해지는 교통문제에서 차량에 대한 정보를 이용하여 교통흐름을 개선해 줄 뿐만 아니라, 교통위반 차량을 효율적으로 적발할 수 있다. 차량 번호판은 차량정보를 인식하는데 중요하게 사용될 수 있다. 본 논문에서는 이동식 형태인 차량에 탑재한 카메라를 이용하여 촬영한 영상에서 차량의 번호판을 인식하는 새로운 기법을 제안한다. 여러 단계의 영상처리 과정과 인식 과정을 거쳐서 실시간에 처리할 수 있는 시스템으로 일반 차량뿐 아니라 특장차에 대한 인식도 가능하게 한다. 제안한 기법을 이용한 실제적 환경에서의 영상과 인식에 대한 결과가 실험결과에서 보여진다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.2148-2161
• /
• 2019

In this paper, we present a real-time cattle action recognition algorithm to detect the estrus phase of cattle from a live video stream. In order to classify cattle movement, specifically, to detect the mounting action, the most observable sign of the estrus phase, a simple yet effective feature description exploiting motion history images (MHI) is designed. By learning the proposed features using the support vector machine framework, various representative cattle actions, such as mounting, walking, tail wagging, and foot stamping, can be recognized robustly in complex scenes. Thanks to low complexity of the proposed action recognition algorithm, multiple cattle in three enclosures can be monitored simultaneously using a single fisheye camera. Through extensive experiments with real video streams, we confirmed that the proposed algorithm outperforms a conventional human action recognition algorithm by 18% in terms of recognition accuracy even with much smaller dimensional feature description.