We present methods specially designed to be effective with moderately non-linear regression relationships. The model fitted is of the Projection Pursuit Regression (PPR) type with a smooth, non-parametric link function connecting the mean response to a linear combination of the regressors. New algorithms, close to ordinary linear regression, are developed. Considerable numerical evidence is giv...

Product models of low dimensional experts are a powerful way to avoid the curse of dimensionality. We present the "under­ complete product of experts" (UPoE), where each expert models a one dimensional pro­ jection of the data. The UPoE may be inter­ preted as a parametric probabilistic model for projection pursuit. Its ML learning rules are identical to the approximate learning rules proposed ...

The technology of hyperspectral remote sensing image improves the capability of collecting the objects such as lakes, rivers, farmlands, buildings, forest and desert in the ground surface. Since the spatial resolution is becoming higher recently, image segmentation of hyperspectral remote sensing is important to the next step of remote sensing image classification and object recognition. In thi...

We propose a convolution model for the blocking artifacts in compressed images and videos, and a cepstral filtering algorithm for reducing the artifacts. Since each of the horizontal or vertical line of the compressed image has abrupt changes at almost every 8th samples, it can be modeled as a convolution of the original sequence with an excitation signal of periodicity 8. Comparison of the cep...

A current to voltage lowpass filter using a single CCII is proposed. This filter can absorb all finite CCII resistances and capacitances and offer both non-inverting and inverting output. Convenient compensation methods exist for gain inaccuracy and excess phase effects. The sensitivities are also very low. The filter is particularly useful for data converting and sampled-data filtering systems...

In a typical disparity (or motion) estimation algorithm developed for inter-image prediction, an interpolation of intensities is applied to one of the two images used. Therefore, non-filtered intensities of the image being predicted are compared with lowpass-filtered intensities of the other image of the stereo pair. Consequently, noise and detail suppression in the two images are unequal. In t...

We are developing a forecaster for daily extremes of demand for electric power encountered in the service area of a large midwestern utility and using this application as a testbed for approaches to input dimension reduction and decomposition of network training. Projection pursuit regression representations and the ability of algorithms like SIR to quickly find reasonable weighting vectors ena...

7 processing. Conventional dimensionality reduction on-board remote devices is 8 often prohibitive due to limited computational resources; on the other hand, 9 integrating random projections directly into signal acquisition offers an alternative 10 to explicit dimensionality reduction without incurring sender-side computational 11 cost. Receiver-side reconstruction of hyperspectral data from su...