Open Access. Powered by Scholars. Published by Universities.®

Engineering Commons™

Open Access. Powered by Scholars. Published by Universities.®

1998

Electrical and Computer Engineering

Electrical Engineering and Computer Science Faculty Publications

Articles 1 - 5 of 5

Full-Text Articles in Engineering

Extension Of The Generalized Hebbian Algorithm For Principal Component Extraction, Fredric M. Ham, Inho Kim Oct 1998

Extension Of The Generalized Hebbian Algorithm For Principal Component Extraction, Fredric M. Ham, Inho Kim

Electrical Engineering and Computer Science Faculty Publications

Principal component analysis (PCA) plays an important role in various areas. In many applications it is necessary to adaptively compute the principal components of the input data. Over the past several years, there have been numerous neural network approaches to adaptively extract principal components for PCA. One of he most popular learning rules for training a single-layer linear network for principal component extraction is Sanger's generalized Hebbian algorithm (GHA). We have extended the GHA (EGHA) by including a positive-definite symmetric weighting matrix in the representation error-cost function that is used to derive the learning rule to train the network. The …

Go to article

Performance Of Optimal Trade-Off And Distance-Classifier Circular Filters For Rotation Invariance, Samuel Peter Kozaitis, Sila Thangwaritorn Aug 1998

Performance Of Optimal Trade-Off And Distance-Classifier Circular Filters For Rotation Invariance, Samuel Peter Kozaitis, Sila Thangwaritorn

Electrical Engineering and Computer Science Faculty Publications

We evaluated the performance of circular optimal trade-off SDF (OTSDF) and distance classifier correlation filters (DCCFs) as rotation-invariant correlation filters. Because the filters are designed using different parameters we compared the filter's performance in terms of an equivalent effect of probability of error. The use of OTSDF and DCCF filters as circular filters allows their calculation to be greatly simplified when compared to using rotated views of an object to create filters. We found that both types of filters can be used for rotation-invariant object recognition in a noisy environment. In addition, the filters generated were real-valued so they may …

Go to article

Wavelet-Based Noise Reduction In Multispectral Imagery, Abdullah A. Basuhail, Samuel Peter Kozaitis Jul 1998

Wavelet-Based Noise Reduction In Multispectral Imagery, Abdullah A. Basuhail, Samuel Peter Kozaitis

Electrical Engineering and Computer Science Faculty Publications

We used a 3-D wavelet-based denoising method to reduce the noise from multispectral imagery. In our approach, we compared denoising of different bands of a multispectral image using a 2-D denoising technique, by which the wavelet coefficients corresponding to each band were denoised independent of each band, and a 3-D denoising technique by which the wavelet coefficients were denoised by involving all bands in thresholding the wavelet coefficients. Due to the high correlation of the multispectral imagery data along the wavelength axis, the noise can be easily reduced by applying the wavelet transform along the wavelength direction. Our results showed …

Go to article

Quantum Theory Of The Linewidth Of A Laser With A Saturable Absorber: Phase Diffusion, Michael Sokol Jul 1998

Quantum Theory Of The Linewidth Of A Laser With A Saturable Absorber: Phase Diffusion, Michael Sokol

Electrical Engineering and Computer Science Faculty Publications

A quantum theory of the spectral width of a laser with a saturable absorber is presented. We perform a density matrix calculation yielding the off-diagonal element ρ n,n+1(t) which is proportional to the average value of the electric field associated with the laser field having a large photon number n. The general linewidth theory is applied to any general laser which has saturable absorber features for a portion of the building of the laser field. We have found that the saturable absorber action contribution to the linewidth of such a laser can be substantial. © 1998 Society of Photo-Optical Instrumentation …

Go to article

Multiple-Input Joint Transform Correlator For Wavelet Feature Extraction, Samuel Peter Kozaitis, Mark A. Getbehead Apr 1998

Multiple-Input Joint Transform Correlator For Wavelet Feature Extraction, Samuel Peter Kozaitis, Mark A. Getbehead

Electrical Engineering and Computer Science Faculty Publications

We describe a joint transform correlator (JTC) that uses multiple input images encoded in the spatial domain for multiwavelet feature extraction. We extend the theory of a JTC to multiple inputs, which enables various combinations of cross-correlations between input images to be performed. Furthermore, we provide experimental results for four inputs with an optically addressed spatial light modulator in the Fourier plane. In addition, it is possible that the space-bandwidth product for multiwavelet feature extraction can be made the same as for a two-input JTC. © 1998 Society of Photo-Optical Instrumentation Engineers.

Go to article