Constrained Neural Network for Estimating Sensor Reliability in Sensors Fusion
暂无分享,去创建一个
In sensors fusion, “Context Dependent” fusion operators represent a very powerful approach. Most often, context is related to the reliability of the sensors. This approach is efficient if the estimation of the sensors reliability is itself reliable. We show how this has been done thanks to an optimal non-linear data transformation by ANN. An application in audio-visual speech recognition will illustrate this implementation.
[1] Jeanny Hérault,et al. Curvilinear component analysis: a self-organizing neural network for nonlinear mapping of data sets , 1997, IEEE Trans. Neural Networks.
[2] A. Adjoudani,et al. On the Integration of Auditory and Visual Parameters in an HMM-based ASR , 1996 .
[3] Isabelle Bloch. Information combination operators for data fusion: a comparative review with classification , 1996, IEEE Trans. Syst. Man Cybern. Part A.
[4] Louis-Jean Boë,et al. La parole et son traitement automatique , 1989 .