Neural Network and Wavelet Transform For Classification and Object Detection

The practical utilization of object detection and classification, in high-performance structural mine detection or proximity fuses is somewhat impeded due to some complicated phenomena such as: existence of multiple wave modes, jamming, high susceptibility to diverse interferences, bulky sampled data, clutters and difficulty in signal interpretation. An intelligent signal processing approach using the wavelet transform and artificial neural network algorithms was developed; this was actualized in a signal processing package. The intelligent signal processing technique comprehensively functions as signal filtration, data compression and pattern recognition, capable of extracting essential features from acquired raw wave signals and further assisting in structural mine detection or proximity fuses evaluation. For validation, the algorithm was applied to the detection and classification of 10 different objects. [Afshin Shaabany, Fatemeh Jamshidi. Neural Network and Wavelet Transform For Classification and Object Detection. Journal of American Science 2011;7(5):20-25]. (ISSN: 1545-1003). http://www.americanscience.org.