Showing papers on "Quadrature mirror filter published in 1991"

A class of fast Gaussian binomial filters for speech and image processing

Abstract: The authors present an efficient, in-place algorithm for the batch processing of linear data arrays. These algorithms are efficient, easily scaled, and have no multiply operations. They are suitable as front-end filters for a bank of quadrature mirror filters and for pyramid coding of images. In the latter application, the binomial filter was used as the low-pass filter in pyramid coding of images and compared with the Gaussian filter devised by P.J. Burt (Comput. Graph. Image Processing, vol.16, p.20-51, 1981). The binomial filter yielded a slightly larger signal-to-noise ratio in every case tested. More significantly, for an (L+1)*(L+1) image array processed in (N+1)*(N+1) subblocks, the fast Burt algorithm requires a total of 2(L+1)/sup 2/N adds and 2(L+1)/sup 2/ (N/2+1) multiplies. The binomial algorithm requires 2L/sup 2/N adds and zero multiplies. >

Adaptive notch filter for FM interference cancellation

Abstract: An improved adaptive notch filter for removing undesired cochannel FM interference includes in-phase and quadrature signal processing paths in which the undesired FM signal is translated to zero frequency, i.e. to DC. In each of the in-phase and quadrature phase signal paths, a first multiplier translates the input frequency spectrum in a dynamic manner so that the frequency of the undesired FM signal is always centered on DC. The undesired signal, now at DC, is filtered out by a high pass filter with a sharp rolloff. The output of the high pass filter is remultiplied in a second multiplier to translate the desired FM signal back to the original position in the frequency spectrum. In order to eliminate an unwanted difference frequency remodulation signal at the output of the second multiplier, the quadrature phase signal path performs a parallel operation on the same input signal, but with a quadrature phase (90 degree phase shift) control signal. The output the in-phase and quadrature signal processing circuits are combined in an adder circuit to cancel the unwanted difference frequency remodulation signal.

A practical approach to fractal-based image compression

Abstract: A technique for image compression is based on a very simple type of iterative fractal. A wavelet transform (quadrature mirror filter pyramid) is used to decompose an image into bands containing information from different scales (spatial frequencies) and orientations. The conditional probabilities between these different scale bands are then determined, and used as the basis for a predictive coder. The wavelet transform's various scale and orientation bands have a great deal of redundant, self-similar structure in the form of multi-modal conditional probabilities, so that linear predictors perform poorly. A simple histogram method is used to determine the multi-modal conditional probabilities between scales. The resulting predictive coder is easily integrated into existing subband coding schemes. >

ECG data compression by sub-band coding

Abstract: A data compression technique is presented for discrete-time electrocardiogram signals. The single lead electrocardiogram signal is decomposed into several multiresolution subsignals by using a quadrature mirror filter bank. The resultant subsignals are compressed according to their frequency contents using various coding methods, including a discrete cosine transform based technique and pulse code modulation with variable length coding. Compression ratios as high as 5.7 are obtained without introducing any visual distortion.

New results on cosine-modulated FIR filter banks satisfying perfect reconstruction

Abstract: The authors present a derivation of the necessary and sufficient condition on the polyphase components of a linear-phase prototype (length N=2 mM) such that the polyphase component matrix of the filter bank is lossless. This in turn ensures that the modulated filter bank satisfies the PR (perfect reconstruction) property. An efficient design procedure (which involves fewer parameters to be optimized than other approaches) is presented. By this approach, FIR (finite-impulse-response) PR filter banks can be designed for an arbitrary number of channels. Moreover, since both the analysis and synthesis filter banks are obtained by modulation, they can be implemented very efficiently (using the discrete cosine transform). The design procedure is outlined and a design example is presented. >

Perfect reconstruction filter banks with rational sampling rate changes

Abstract: The authors present a general, direct method for designing perfect reconstruction filter banks with rational sampling rate changes. Such filter banks have N branches, each one having a sampling factor of p/sub i//q/sub i/ and their sum equal to one. A design example showing the advantage of using the direct over the indirect method is given. Due to recent results pointing to the relationship between filter banks and wavelet theory, the regularity question is addressed as well, and a regular filter is shown for a dilation factor of 3/2. >

Multirate Filter Banks for Subband Coding

Abstract: This chapter develops the theory of multirate signal processing as used in subband coding systems. Multirate operations are reviewed, multirate filter banks are analyzed and structures and design methods for perfect reconstruction filter banks are indicated. Special attention is given to the multidimensional case. We do not consider how to actually code the subband signals, and it will be assumed that the subband signals are perfectly transmitted.

A multirate filter bank based approach to the detection and enhancement of linear features in images

Abstract: A general method based on the directional filter bank for the detection and enhancement of linear features in multidimensional data is presented. Results demonstrating the effectiveness of this new method are presented for two trial cases: the enhancement of cell boundaries in biomedical images, and the analysis and enhancement of fingerprints. >

Sub-band coding of digital images

Abstract: An efficient sub‐band coding method for encoding images is presented in this paper. In this method a frequency band decomposition of the image is carried out by means of two‐dimensional separable quadrature mirror filters (QMF's), which split the image spectrum into sixteen uniform sub‐bands. In the coding process, we employ special quantizers and encoding algorithms which are designed according to the characteristics (including the visual perceptual property) of the sub‐bands to quantize and to encode these subimages, separately. The simulation results are presented in terms of average bit rates and the quality (in subjective as well as mean squared senses) of the reconstructed images. These results indicate that a good quality reproduction can be achieved at high compression rates. The paper also shows that a seven‐band decomposition has very high performance in quality, compression rate, and economy.

Perfect reconstruction recursive QMF banks for image subband coding

Abstract: A scheme for implementing a perfect reconstruction QMF (quadrature mirror filter) bank is considered. The scheme is based on a uniform DFT (discrete Fourier transform) L-band analysis filter bank realized using a polyphase decomposition of a lowpass L-th band filter (composed of stable allpass subfilters) followed by an L-band synthesis filter bank (realized using the transpose of the analysis filter bank with the allpass filters replaced by their inverse filters). In the subband coding of finite-length sequences, the instability problem of the inverse filters can be avoided by using the original stable allpass filter in each branch of the synthesis bank with time-reversed input and output. Conditions for perfect reconstruction are derived and illustrated with examples. >

QMF-based A/D converters: overview and new results

Abstract: Provides an overview of A/D converters incorporating a quadrature mirror filter bank. The major difference existing between these and the time-interleaved architectures (both make use of an array of A/D converters) is that, in the former case, suitable bandpass filters are used before and after the quantizers, reducing the effect of quantization noise produced by the array of A/D converters, as shown here. The noise introduced by the switched-capacitor filters, which implement the analysis filter bank in the QMF-based data converters, is estimated. Simulation results obtained with images are presented.

Computationally efficient pseudo QMF filter bank for a multi-compatible HDTV codec

Abstract: A review is presented of methods usable in a coding scheme to obtain compatible images (images with lower resolution) based on the discrete cosine transform (DCT), quadrature mirror filters (QMF), conjugate quadrature filters (CQF) and pseudo quadrature mirror filters (PQMF). Compared to QMF or CQF, PQMF gives better frequency selectivity for a much lower computational cost. The theory of PQMF is reviewed and new results are described, especially regarding computational cost, hardware implementation, and constraints on the length of the prototype filter used to generate the modulated filter bank. Both cases of NF even and NF odd, where NF is the prototype filter number of taps, are then checked, and optimal implementations using fast transforms (a classical DCT for NF odd and a type IV DCT for NF even) are explained for the encoder and the compatible decoders. >

***WITHDRAWN PATENT AS PER THE LATEST USPTO WITHDRAWN LIST***Method and apparatus for canceling quadrature distortion as for video signals having in-phase and quadrature phase components

Abstract: High definition or enhanced television signals include supplementary information which is double-sideband modulated in quadrature with the picture carrier, and passed through an inverse Nyquist filter before being combined with a conventional vestigial sideband modulated main video signal. A receiver which recovers both the main video signal and the supplementary information processes the received video signal through an IF filter having a vestigial sideband frequency response characteristic. Two synchronous detectors recover in-phase and quadrature phase signal components from the filtered television signals, these recovered signals each include quadrature crosstalk distortion components from the other signal. The crosstalk from the supplementary signal into the main signal is canceled by filtering the recovered quadrature signal and adding the result to the recovered in-phase signal. This filtering also equalizes the in-phase signal by converting the quadrature distortion components in the recovered quadrature signal into reinforcing signal components in the in-phase signal. This signal recovery system is advantageously used with a ghost cancellation system.

Structural texture recognition using QMF bank based subband decomposition

Abstract: An application of quadrature mirror filter (QMF) bank based subband decomposition to texture analysis is presented. Two-dimensional four bands QMF structure is used and the QMF features are introduced. The low-low band extracts the information of spatial dependence, and the low-high, high-low and high-high bands extract the structural information. This approach has the twin advantages of efficient information extraction and computation reduction. The classification abilities of QMF features are compared to those of Haralick features. The experiments demonstrate that the QMF features have better performance than the Haralick features. >

Quadrature distortion cancellation system

Abstract: High definition or enhanced television signals include supplementary information which is double-sideband modulated in quadrature with the picture carrier, and passed through an inverse Nyquist filter (222) before being combined (224) with a conventional vestigial sideband (221) modulated main video signal. A receiver which recovers both the main video signal and the supplementary information processes the received video signal through an IF filter (228) having a vestigial sideband frequency response characteristic. Two synchronous detectors (230, 232) recover in-phase and quadrature phase signal components from the filtered television signals, these recovered signals each include quadrature crosstalk distortion components from the other signal. The crosstalk from the supplementary signal into the main signal is cancelled by filtering (246) the recovered quadrature signal and adding (244) the result to the recovered in-phase signal. This filtering also equalizes the in-phase signal by converting the quadrature distortion components in the recovered quadrature signal into reinforcing signal components in the in-phase signal. This signal recovery system is advantageously used with a ghost cancellation system.

A digital realization technique of optimum receiving matched filter

Abstract: Receiving filters are investigated for the detection of quadrature overlapped modulation signal (QOMS). The minimum energy symbol matched (MESM) filter for QOMS scheme is presented. BER performances with different kinds of receiving filter are studied and compared. The digital realization of the filters is presented using a memory-based FIR digital filter structure, which is versatile and flexible. >

On the problem of reconstructing a segment of a wideband signal using a digital filter bank

Abstract: The design process is reduced to designing the filter bank to satisfy a set of conditions and it is known that the reconstructed signal always possesses some alias and distortion. It is the present objective to find the necessary and sufficient conditions on the filters so that the resulting QMF (quadrature mirror filter) bank cancels most alias components. Having found these conditions, the author suggests an algorithm to design the PAC (partial alias cancellation) QMF bank. Examples are given to demonstrate the theory. >

On bit allocation strategies for paraunitary subband coders

Abstract: Quadrature mirror filters have been employed to decompose a signal into subbands. The signals in each subband are quantized before transmission. The authors derive conditions on the individual quantizer error variances so as to minimize the overall reconstruction error variance. The present results hold for filter banks having paraunitary polyphase matrices. Under the assumption that the total number of bits over all channels is fixed, it is shown that their distribution which minimizes the reconstruction error variance is the one that makes the individual quantizer variances equal. This analysis is extended to tree-structured filter banks, such as those used for generating orthonormal wavelets. In this case, it is shown that the quantizer error variance in a particular branch should be proportional to the corresponding decimation factor. >

The Design Of Multidimensional Exact/Aliasing-Free Reconstruction Multirate Filter Banks Using McClellan Transformations

Abstract: In [l], the use of McClellan transformations for the design of multidimensional filter banks from one-dimensional analysis/synthesis filters is presented for the case of is two-band, diamond-shaped pair. In [2], these concepts are expanded for the design of two-band filter banks with general parallelogram-shaped passband geometries. However, in [l, 21, the dlesign technique is restricted to filters which had zero-phase characteristics, which results in zero-phase two-dimensional filters. In this paper, we note that if the signal being processed is periodic, (which naturally arises when implementing filter banks in a size-limiting manner [a] for finite extent inputs), then the zero-phase restriction can be relaxed.

A class of efficient recursive quadrature mirror filters for subband coding

Abstract: A class of recursive quadrature mirror filters (QMFs) of the form H(Z)=N(Z)/D(Z/sup 2/) is proposed for subband coding. These filters can be efficiently designed with the iterative Remez exchange algorithm. Simple modifications are made to the Pei (see IEEE Trans. Circuits Syst., vol.CAS-36, no.2, p.311, 1989), Liang, McClellan, and Parks computer program for designing these recursive QMF filters efficiently. The resulting filters totally eliminate the aliasing terms and can achieve almost perfect magnitude reconstruction, with -131.5-dB distortion. A design example is presented to show the effectiveness of this approach. >

Subband coding of images using trellis coded quantization

Abstract: An image coding scheme combining subband coding and the established energy compaction technique of the discrete cosine transform (DCT) with trellis coded quan­ tization (TCQ) is introduced. Image spectrums are split into 16 subband images using a quadrature mirror filter bank, and the DCT is performed on the lowest subband. The data is quantized using TCQ, transmitted and recombined at the receiver. It is shown that quantizing the subband data with TCQ decreases the mean-squared error (MSE) incurred in the quantization step, versus that of a Lloyd-Max scalar quantizer. The performance of the system at an average bit rate of 0.5 bits/pixel is subjectively poor; many false contours are visible and the images have a "blurry" look. The system produces good subjective performance at a bit rate of 1 bit/pixel with almost no visible false contours and a much sharper look. At 1.5 and 2 bits/pixel the performance of the system is excellent with no visible false contours and no blurriness.

Picture coding system

Abstract: PURPOSE:To improve coding efficiency by using information for one picture in advance so as to predict a mean coding parameter thereby increasing/ decreasing the coding parameter in response to the quantity of generated information for each block. CONSTITUTION:A picture coder consists of an input terminal 1, a QMF (Quadrature Mirror Filter) 2, a mean quantization step width prediction circuit 3, a delay 4, a subtractor 5, a prediction device 6, a quantizer 7, an adder 8, a quantization step width setting circuit 6, a memory 10, a coding device 11, a code quantity monitor section 12, a multiplexer 13 and an output terminal 14. Then the information by one picture is used to calculate a coding parameter to a setting code rate and the deviation from a coding parameter is set in response to the generated information quantity for each block resulting block processing of one pattern and the succeeding block is coded by using a coding parameter deviated in response to the deviation. Thus, the picture is coded efficiently with the setting coding rate.