Process outputs, as reflected in the measured signals, are generally stochastic in nature. This chapter focuses on processing of sensor outputs and the identification of faults that may occur during laser processing, for example, cracking, porosity, no weld, and so on, and is limited to classification methods. Measured signals are often contaminated by noise. The Fourier transform involves the transformation of information from the time domain to the frequency domain. Data reduction is used to reduce the initial size of the data to make subsequent computations faster, with minimal loss of information. The chapter focuses on pattern recognition and neural network analysis. Feature extraction involves transformation of the signal, along with a reduction in the data size. Neural networks represent the classifier by nodes which are interconnected, with information available in the network being contained in the magnitude of the links between the processors.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Processing of Sensor Outputs


    Contributors:


    Publication date :

    2023-02-07


    Size :

    30 pages




    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English






    Multiple target angle tracking using sensor array outputs

    Sword, C.K. / Simaan, M. / Kamen, E.W. | IEEE | 1990


    Enhanced LiDAR air data using supplementary sensor outputs

    LODDEN GRANT / FAN XIAO ZHU / KIMMEL DANNY THOMAS et al. | European Patent Office | 2019

    Free access