NONORTHOGONAL SIGNAL REPRESENTATION BY GAUSSIANS AND GABOR FUNCTIONS

被引：16

作者：

BENARIE, J

RAO, KR

机构：

[1] Department of Electrical and Computer Engineering, Illinois Institute of Technology, Chicago

来源：

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-ANALOG AND DIGITAL SIGNAL PROCESSING | 1995年 / 42卷 / 06期

基金：

美国国家科学基金会;

关键词：

D O I：

10.1109/82.392315

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

This paper describes a novel approach for nonorthogonal representation of signals using Gaussians and an extension of this method for Gabor representation of signals, based on the equivalence of Gabor expansion to Gaussian expansion in the frequency domain. The Gaussian expansion scheme yields an efficient representation of signals for low number of bits per pixel and is better than the corresponding Discrete Cosine Transform (DCT) representation for very low bit rates, This advantage diminishes gracefully for higher bit rates where the residual approximation error signal to be represented is more random and less structured. It is proved in this paper that a finite number of Gaussians can theoretically approximate sinusoids in a bounded region with arbitrarily small error, and therefore any finite support L(2)(R) signal as well. Two methods for Gaussian representation of signals are outlined, The first, called the Max-Energy paradigm, involves successive extraction of the highest energy Gaussian that best ''fits'' the signal. The second is a parallel approach and uses an adaptive projection algorithm to first derive the Gaussian basis set to be used in parallel, and then optimizes the coefficients for minimum squared error.

引用

页码：402 / 413

页数：12

共 21 条

[1]

Daugman J.G., Complete discrete 2D Gabor transform by neural networks for image analysis and compression, IEEE Trans. ASSP, 36, 7, pp. 1169-1179, (1988)

[2]

Rao K.R., Ben-Arie J., Lattice architectures for multiple-scale Gaussian convolution, image processing, sinusoid-based transforms and Gabor filtering, Analog Integrat. Circuits Signal Process., 4, 2, pp. 141-160, (1993)

[3]

Ben-Arie J., Multi-dimensional linear lattice for Fourier and Gabor transforms, multiple-scale Gaussian filtering, and edge detection, Neural Networks for Human and Machine Perception, pp. 231-252, (1992)

[4]

Ben-Arie J., Rao K.R., Lattice architectures for nonorthogonal representation of signals and generation of transforms using Gaussian sets, Tech. Rep. IIT ECE-TR-005-92, (1992)

[5]

A novel approach for template matching by nonorthogonal image expansion, IEEE Trans. Circuits Syst. Video Technol., 3, 1, pp. 71-84, (1993)

[6]

Rao K.R., Ben-Arie J., Multiple template matching using expansion matching, IEEE Trans. Circuits Syst. Video Technol., 4, 5, pp. 490-503, (1994)

[7]

Ben-Arie J., Rao K.R., Optimal template matching by nonorthogonal image expansion using restoration, Int. J. Machine Vision Applicat., 7, 2, pp. 69-81, (1994)

[8]

Rao K.R., Ben-Arie J., Non-orthogonal image expansion related to optimal template matching in complex images, Computer Vision, 56, 2, pp. 149-160, (1994)

[9]

Ben-Arie J., Rao K.R., A lattice network for signal representation using Gaussian basis functions and max-energy paradigm, Proc. IEEE 34th Midwest Symp. Circuits Syst., pp. 76-79, (1991)

[10]

Novel transform for image description and compression with implementation by neural architectures, Proc. SPIE Conf. Stochastic and Neural Methods in Signal Process., pp. 356-366, (1991)

← 1 2 3 →