• Title/Summary/Keyword: vector computer

Search Result 2,004, Processing Time 0.03 seconds

Motion Vector and Disparity Vector Prediction for Multi-view Video Coding (다시점 동영상 부호화를 위한 움직임 벡터 및 변이 벡터 예측 기법)

  • Lee, Seo-Young;Shin, Kwang-Mu;Kim, Sung-Min;Chung, Ki-Dong
    • Proceedings of the Korean Information Science Society Conference
    • /
    • 2007.10d
    • /
    • pp.560-563
    • /
    • 2007
  • 다시점 영상은 관찰자가 원하는 시점을 선택할 수 있고, 자연스러운 입체감을 제공한다. 하지만 카메라의 수가 증가함에 따라 데이터양이 늘어나는 단점이 있다. 본 논문은 시.공간적 상관성을 고려한 시점 간 움직임 벡터 예측 기법을 제안한다. 인접한 시점의 영상이 비슷한 움직임 정보를 가지는 특징을 이용해 현재 블록의 움직임 벡터의 예측간을 계산한다. 또한 시간적으로 연속한 프레임의 경우 비슷한 변이 벡터를 가지므로, 이전 프레임의 변이 벡터 정보를 현재 매크로블록의 변이벡터로 활용하는 방법을 제안한다.

  • PDF

EEG Based Brain-Computer Interface System Using Time-multiplexing and Bio-Feedback (Time-multiplexing과 바이오 피드백을 이용한 EEG기반 뇌-컴퓨터 인터페이스 시스템)

  • Bae, Il-Han;Ban, Sang-Woo;Lee, Min-Ho
    • Journal of Sensor Science and Technology
    • /
    • v.13 no.3
    • /
    • pp.236-243
    • /
    • 2004
  • In this paper, we proposed a brain-computer interface system using EEG signals. It can generate 4 direction command signal from EEG signals captured during imagination of subjects. Bandpass filter used for preprocessing to detect the brain signal, and the power spectrum at a specific frequency domain of the EEG signals for concentration status and non-concentration one is used for feature. In order to generate an adequate signal for controlling the 4 direction movement, we propose a new interface system implemented by using a support vector machine and a time-multiplexing method. Moreover, bio-feed back process and on-line adaptive pattern recognition mechanism are also considered in the proposed system. Computer experimental results show that the proposed method is effective to recognize the non-stational brain wave signal.

Appearance-based Robot Visual Servo via a Wavelet Neural Network

  • Zhao, Qingjie;Sun, Zengqi;Sun, Fuchun;Zhu, Jihong
    • International Journal of Control, Automation, and Systems
    • /
    • v.6 no.4
    • /
    • pp.607-612
    • /
    • 2008
  • This paper proposes a robot visual servo approach based on image appearance and a wavelet function neural network. The inputs of the wavelet neural network are changes of image features or the elements of image appearance vector, and the outputs are changes of robot joint angles. Image appearance vector is calculated by using eigen subspace transform algorithm. The proposed approach does not need a priori knowledge of the robot kinematics, hand-eye geometry and camera models. The experiment results on a real robot system show that the proposed method is practical and simple.

Weight-based Motion Vector Composition using Activity Information and Overlapped Area (움직임 정보 및 중첩 영역을 이용한 가중치 기반의 움직임 벡터 합성 기법)

  • Kim, Hyun-Hee;Kim, Sung-Min;Lee, Seung-Won;Jung, Ki-Dong
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2004.05a
    • /
    • pp.1573-1576
    • /
    • 2004
  • 멀티미디어 압축 및 이동 통신 기술의 발전으로 다양한 형태의 멀티미디어 서비스가 이슈화되고 있다. 비디오를 전송하기 위해서는 많은 대역폭을 필요로 하지만, 모든 네트워크가 높은 수준의 대역 및 처리 능력을 가지는 것은 아니다. 이질적인 네트워크간의 멀티미디어를 서비스하기 위해서는 네트워크 상황 또는 수신자의 처리 능력에 맞도록 재 부호화해야 하지만 그 처리비용이 높다. 트랜스코딩 기법 중에서 시간당 요구된 프레임의 개수를 조절하면 제거된 프레임의 움직임 벡터를 재 사용하여 비트율을 감소시킬 수 있다. 본 논문에서는 기존의 기법보다 향상된 움직임과 중첩 영역의 정보를 적용한 WBVC(Weight-Based Vector Composition) 기법을 제안한다. 실험을 통한 기존의 기법과의 비교 분석 결과, 비슷한 계산 복잡도에서 제안한 WBVC 기법이 높은 성능을 보였다.

  • PDF

Feature Extraction Based on DBN-SVM for Tone Recognition

  • Chao, Hao;Song, Cheng;Lu, Bao-Yun;Liu, Yong-Li
    • Journal of Information Processing Systems
    • /
    • v.15 no.1
    • /
    • pp.91-99
    • /
    • 2019
  • An innovative tone modeling framework based on deep neural networks in tone recognition was proposed in this paper. In the framework, both the prosodic features and the articulatory features were firstly extracted as the raw input data. Then, a 5-layer-deep deep belief network was presented to obtain high-level tone features. Finally, support vector machine was trained to recognize tones. The 863-data corpus had been applied in experiments, and the results show that the proposed method helped improve the recognition accuracy significantly for all tone patterns. Meanwhile, the average tone recognition rate reached 83.03%, which is 8.61% higher than that of the original method.

A Hybrid Routing Protocol Based on Bio-Inspired Methods in a Mobile Ad Hoc Network

  • Alattas, Khalid A
    • International Journal of Computer Science & Network Security
    • /
    • v.21 no.1
    • /
    • pp.207-213
    • /
    • 2021
  • Networks in Mobile ad hoc contain distribution and do not have a predefined structure which practically means that network modes can play the role of being clients or servers. The routing protocols used in mobile Ad-hoc networks (MANETs) are characterized by limited bandwidth, mobility, limited power supply, and routing protocols. Hybrid routing protocols solve the delay problem of reactive routing protocols and the routing overhead of proactive routing protocols. The Ant Colony Optimization (ACO) algorithm is used to solve other real-life problems such as the travelling salesman problem, capacity planning, and the vehicle routing challenge. Bio-inspired methods have probed lethal in helping to solve the problem domains in these networks. Hybrid routing protocols combine the distance vector routing protocol (DVRP) and the link-state routing protocol (LSRP) to solve the routing problem.

A Novel Feature Selection Approach to Classify Breast Cancer Drug using Optimized Grey Wolf Algorithm

  • Shobana, G.;Priya, N.
    • International Journal of Computer Science & Network Security
    • /
    • v.22 no.9
    • /
    • pp.258-270
    • /
    • 2022
  • Cancer has become a common disease for the past two decades throughout the globe and there is significant increase of cancer among women. Breast cancer and ovarian cancers are more prevalent among women. Majority of the patients approach the physicians only during their final stage of the disease. Early diagnosis of cancer remains a great challenge for the researchers. Although several drugs are being synthesized very often, their multi-benefits are less investigated. With millions of drugs synthesized and their data are accessible through open repositories. Drug repurposing can be done using machine learning techniques. We propose a feature selection technique in this paper, which is novel that generates multiple populations for the grey wolf algorithm and classifies breast cancer drugs efficiently. Leukemia drug dataset is also investigated and Multilayer perceptron achieved 96% prediction accuracy. Three supervised machine learning algorithms namely Random Forest classifier, Multilayer Perceptron and Support Vector Machine models were applied and Multilayer perceptron had higher accuracy rate of 97.7% for breast cancer drug classification.

차화상으로부터 이차원 이동 벡터의 추출

  • 장순화;김종대;김성대;김재균
    • Proceedings of the Korean Institute of Communication Sciences Conference
    • /
    • 1986.10a
    • /
    • pp.182-185
    • /
    • 1986
  • In this paper, the four algorithm which obtain 2D displacement vector are proposed. In corwocutive difference pictures, the characteristics of up DP boundary and region are discussed and we estimate displacement vector using the DP boundary and region, Finally, the performance of proposed algorithm for gaussian noisy image which generated by computer are discussed.

  • PDF

Detour paths algorithm using the vectors in Hypercube Networks (하이퍼큐브 네트워크에서 벡터들을 이용한 우회경로 알고리즘)

  • Jin, Ming-He;Rhee, Chung-Sei
    • The Journal of Korean Institute of Communications and Information Sciences
    • /
    • v.34 no.12C
    • /
    • pp.1123-1130
    • /
    • 2009
  • The advances in networking technology and importance of multimedia communication require real time transaction. In many applications, high reliable real-time communications are required more frequently. In this paper, we propose a reliable communication in cube-based multi -computer using the safety vector. Each node in a cube-based n dimensional multi-computer is associated with a safety vector of n bits, which is an approximated measure of the number and distribution of faults in the neighborhood. We propose an algorithm that can establish detour paths using the safety vector. The established detour paths are disjoint with the primary real-time channel. Therefore, our algorithm is more efficient than earlier proposed algorithms.