DocumentCode :
1090347
Title :
Detection and classification of buried dielectric anomalies using neural networks-further results
Author :
Azimi-Sadjadi, Mahmood R. ; Stricker, S.A.
Author_Institution :
Dept. of Electr. Eng., Colorado State Univ., Fort Collins, CO, USA
Volume :
43
Issue :
1
fYear :
1994
fDate :
2/1/1994 12:00:00 AM
Firstpage :
34
Lastpage :
39
Abstract :
The development of a neural network-based detection and classification system for use with buried dielectric anomalies is the main focus of this paper. Several methods of data representation are developed to study their effects on the trainability and generalization capabilities of the neural networks. The method of Karhonen-Loeve (KL) transform is used to extract energy dependent features and to reduce the dimensionality of the weight space of the original data set. To extract the shape-dependent features of the data, another data preprocessing method known as Zernike moments is also studied for its use in the detector/classifier system. The effects of different neural network paradigms, architectural variations, and selection of proper training data on detection and classification rates are studied. Simulation results for nylon and wood targets indicate superior performance when compared to conventional schemes
Keywords :
data reduction; data structures; dielectric measurement; dielectric properties of solids; digital simulation; feature extraction; geophysical techniques; image processing equipment; learning (artificial intelligence); neural nets; pattern recognition; polymers; soil; transforms; wood; 2D; Karhonen-Loeve transform; Zernike moments; architectural variations; buried dielectric anomalies; data preprocessing; data representation; detector/classifier; dimensionality; energy dependent features; landmines; neural network paradigms; neural networks; nylon; shape-dependent features; simulation; soil properties; superior performance; trainability; weight space; wood targets; Data mining; Data preprocessing; Detectors; Dielectrics; Feature extraction; Helium; Landmine detection; Neural networks; Soil; Training data;
fLanguage :
English
Journal_Title :
Instrumentation and Measurement, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9456
Type :
jour
DOI :
10.1109/19.286352
Filename :
286352
Link To Document :
بازگشت