Abstract: An image processing method includes the steps of wavefront coding a wavefront that forms an optical image, converting the optical image to a data stream, and processing the data stream with a filter kernel to reverse effects of wavefront coding and generate a final image. By example, the filter set kernel may be a reduced filter set kernel, or a color-specific kernel. Methods and systems are also disclosed for processing color images, such as by separating color and spatial information into separate channels. Methods and systems herein are for example useful in forming electronic devices with reduced opto-mechanical, opto-electronic and processing complexity or cost.

Abstract: Systems and methods for aligning scanned images are provided. A pattern is included in the scanned image so that when the image is convolved with a filter, a recognizable pattern is generated in the convolved image. The scanned image may then be aligned according to the position of the recognizable pattern in the convolved image. The filter may also act to remove the portions of the scanned image that do not correspond to the pattern in the scanned image.

Abstract: A method of smoothing jagged edges in graphical data. The method detects one or more edge of a selected pixel in the graphical data, dependent upon intensities of the selected pixel and another pixel surrounding a respective site of one or more edges and predetermied gradients of at least the selected pixel and the respective site surrounding the pixel. An adapted convolution mask is applied to the selected pixel and a predetermined neighborhood of pixels containing the selected pixel. The coefficient values of the convolution mask are dependent upon one or more detected edges.

Abstract: A method (S100) for dynamic range compression of output channel data from an image sensor (2) comprising an array of sensor cells. The method (S100) comprises selecting a window (S130) in the channel data, the window having a reference pixel value and a plurality of nearby pixel values. The reference pixel value originates from a reference cell that is one of the sensor cells and the nearby pixel values originate from the sensor cells that are in close proximity to the reference cell. There is a step of multiplying (S140) the pixel values, in the window, by a respective weight value to provide weighted pixel values and then adding (S150) the weighted pixel values to provide a convolution value. Thereafter, there is a step of providing (S160) a dynamic range compression value for the window from a selected one of the pixel values and said convolution value and then an assigning step (S170) assigns the dynamic range compression value to a selected pixel location comprising part of an image.

Abstract: A convolution operator is applied to an input image to produce an output image. Image pixel data corresponding to at least a predetermined number of scan lines of the input image is provided to a buffer memory adapted to store a portion of the image. The image data may be provided from a source of such data, or alternatively it may be rendered from an object graphics environment. A finite convolution mask is applied to the image pixel data to produce a scan line of the output image. The finite convolution mask has a plurality of coefficients arranged in a predetermined number of rows and a predetermined number of columns, and the predetermined number of scan lines substantially equals at least one of the number of rows or the number of columns of the convolution mask. In a preferred implementation, a scan line of the input image is discarded and a next scan line is provided for each scan line of the output image produced by the convolution.

Abstract: An optimal filter kernel, formed by convolving a box filter with a filter of fixed integer width and unity area, is used to perform image resizing and reconstruction. The optimal filter has forced zeros at locations along a frequency scale corresponding to the reciprocal of the spacing of one or more pixels that comprise a source image to be resized. When a rescale value for a source image is selected, the optimal filter kernel is computed, mapped to the source image, and centered upon a location within the source image corresponding to the position of an output pixel to be generated. The number of pixels that lie underneath the optimal filter kernel is established by multiplying the number of pixels that comprise the width of the source image by the selected rescale value. Upon mapping the optimal filter kernel, the output pixel values that comprise the resized image are then evaluated by processing the one or more source image pixels, such as through interpolation.

Abstract: New and improved apparatus and methods for video encoding, for example, to efficiently and concurrently apply adaptive encoding techniques to convert analog data into digital formats, such as Digital Video (DV) format. A parallel system receives a block of video data and based on the computations and comparisons performed determines the best quantization factor for the block of video data. In an embodiment, the parallel system performs selected operations in parallel to save time and increase speed.

Abstract: This disclosure provides a system for classifying images, used in image detection, image recognition, or other computer vision. The system processes directory images to obtain eigenvectors and eigenvalues, and selects a set of “smooth” basis vectors formed by linear combinations of these eigenvectors to be applied against a target image. Contrary to conventional wisdom, however, a group of the eigenvectors having the weakest eigenvalues are used to select the basis vectors. A second process is then performed on this group of “weakest” eigenvectors to identify a set of candidate vectors, ordered in terms of “smoothness.” The set of basis vectors (preferably 3-9) is then chosen from the candidate vectors in order of smoothness, which are then applied in an image detection or image recognition process.

Abstract: A method of enhancing data which is rendered as a bi-tonal bit-mapped image for subsequent printing on a printer. The method involves receiving a source bit-mapped image at a low resolution for printing on a printer at a higher resolution. The bitmap is convoluted with a gradient operator to generate horizontal and vertical gradient values for each pixel. The bitmap is then expanded by a predetermined factor to the higher resolution for sending to the printer, and finally a value is assigned to each pixel in the expanded bitmap which is dependent on the value of pixels in the source bitmap and also the horizontal and vertical gradient values.

Abstract: A system and method for format conversion for use with images containing mixed video and computer generated imagery. A first intermediate scaled image is generated using a first interpolator which has a relatively wide passband and a narrow transition band to the stop band. This interpolator retains high frequency components in the interpolated image but may introduce ringing distortion adjacent to strong edges. The first intermediate scaled image is then processed through an edge extractor which extracts strong edges but ignores any associated ringing distortion. The output signal of the edge detector is a scaled edge image. Next, a second intermediate scaled image is generated using a second interpolator which has a narrower passband and a more gradual transition band. This interpolator produces an image having reduced high frequency components relative to the original image.

Abstract: A trainable object detection system and technique for detecting objects such as people in static or video images of cluttered scenes is described. The described system and technique can be used to detect highly non-rigid objects with a high degree of variability in size, shape, color, and texture. The system learns from examples and does not rely on any a priori (hand-crafted) models or on motion. The technique utilizes a wavelet template that defines the shape of an object in terms of a subset of the wavelet coefficients of the image. It is invariant to changes in color and texture and can be used to robustly define a rich and complex class of objects such as people. The invariant properties and computational efficiency of the wavelet template make it an effective tool for object detection.

Abstract: A cubic convolution interpolating apparatus and method for performing interpolation by optimizing a parameter which determines the interpolation coefficients according to the local property of an image signal, which can minimize the quantity of information loss in a scaled or resampled image signal. The cubic convolution interpolating apparatus includes an image signal divider dividing an image signal into a plurality of subblocks, and a block generating parameters which determine cubic convolution interpolation coefficients in units of subblocks, and perform cubic convolution interpolation.

Abstract: An apparatus and method for image registration of a template image with a target image with large deformation. The apparatus and method involve computing a large deformation transform based on landmark manifolds, image data or both. The apparatus and method are capable of registering images with a small number of landmark points. Registering the images is accomplished by applying the large deformation transform.

Abstract: An image processing method using memory management and pre-computed look up tables to speed up computations. Application of filters along directions other than image rows is simplified using several structured processing approaches that improve image data cache-ability. Time consuming or repeated computations are pre-computed and stored as look up tables to reduce the time required for image processing and to remove or reduce the need for special purpose image processing hardware.

Abstract: A method of analyzing the low- and high-frequency components of a digital image comprising the steps of: providing a rectangular digital image; specifying the size of a rectangular kernel; partitioning the image into nine sub-regions; processing each of said sub-regions using a unique image processing algorithm; combining said processed sub-regions into a resulting image of the low frequency component and of the original image subtracting the low frequency component from the original image to obtain the high frequency component of said original image.

Abstract: Provided is a system, and general method of use that may be applied to the specific system, that overlays appropriately selected impulse response maps of the response of a weapon system's optical detector over successive detailed supercomputer-generated imaged scenes. Each of the impulse response maps have been pre-calculated and stored in addressable memory in preparation for running a simulation. A preferred embodiment then convolves the response maps with the appropriately selected imaged scene in cooperation with a gyro model and, optionally, a model of the airframe on which the detector is carried. Errors determined from the gyro model are used to calculate and provide an address offset to modify values of the response map's base address for the appropriate map pixels. Scene pixel values that undergo convolution are selected using portions of these modified base address values.

Abstract: An image processing apparatus includes a pixel number transforming section for processing input image data composed of a plurality of regions, each being represented by a plurality of pixels with a pixel number transformation, by performing an interpolation or a decrimation on a pixel. It further includes a region decision value extracting section for extracting a region decision value of a region that a target pixel belongs to, based on characteristic amounts representing characteristics of a block composed of the target pixel and a plurality of pixels around the target pixel, each pixel of the image data being considered to be the target pixel. A controller controls a sequential order of a filtering process and pixel number transformation process to be applied to a region, based upon the transformation increasing or reducing a number of pixels in a region.

Abstract: A method for generating a high-resolution image from a plurality of low-resolution images generated from quantized coefficients in the spatial frequency domain and the uncertainties in the quantized coefficients. The method generates a set of linear inequalities relating the quantized coefficients, the uncertainties therein, and the pixel values for the high-resolution image. Each linear inequality relates one of the quantized coefficients to a linear sum of the high-resolution pixel values.

Abstract: A method and apparatus for detecting a pattern within an image. Image data (22) is received which is representative of the image. Filter values (70) are determined which substantially optimizes a first predetermined criterion (68). The first predetermined criterion (68) is based upon image data (22). A correlation output (40) is determined which is indicative of the presence of the pattern within the image data (22). The correlation output (40) is based upon the determined filter values (70) and the image data (22) via a non-linear polynomial relationship (78).

Abstract: A method and apparatus for digital watermarking to resolve multiple claims of ownership is disclosed. According to one embodiment of the invention, a first watermark requiring the host data for detection is embedded into the host data. A second watermark is also embedded into the host data. According to another embodiment of the invention, a pseudo-random sequence acting as a watermark is generated based on two random keys. One of the two random keys is related to the author of the host data into which the watermark is to be embedded, whereas the other of the two random keys is dependent on the host data itself.

Abstract: A method and apparatus for magnifying a portion of a digital image on a display screen in either of two ways. The first method includes a two pass scheme, where each of the passes represents an interpolation in x and y direction respectively, cubic interpolation in each direction is approximated using a one dimensional convolution filter followed by linear interpolation. The second method uses a two dimensional convolution filter first, followed by bilinear interpolation. All of the procedures that are used are accelerated using a hardware package which facilitates exceptionally fast execution.

Abstract: A 4-line buffer sequentially takes in image data items and temporarily stores a specific size of image data. A spatial filter in which a coefficient matrix based on the no-neighbor algorithm in a restoration process has been set performs a spatial filtering process on the image data items sequentially outputted from the 4-line buffer to produce a restored image based on the no-neighbor algorithm. This enables images to be restored in real time, which produces an image whose luminance distribution is approximate to that of the specimen.

Abstract: Image processing method wherein an image is re-sampled and sharpened by subjecting the image to a convolution with a kernel, the elements of which are determined by selecting a first set of kernel values so that a convolution of the image by means of a kernel with said first set of kernel values generates a sharpened image, and subsequently interpolating between the elements of this first kernel.

Abstract: A method, in a system for aiding the guidance of a vehicle, for identifying marking stripes of road lanes. A road image is subjected to a convolution operation with a mask matrix so as to identify discontinuities present in the image. The resulting convolved image is compared with a threshold value and a representation of the marking stripes is determined. The mask matrix is set in such a way as to eliminate at least partially the discontinuities which do not correspond to the marking stripes.

Abstract: A method of processing a sampled input image having pixels which can have missing information at pixel locations to reconstruct an output image therefrom, including providing a mask for the sampled image to identify valid pixel locations; interpolating by convolution with at least one finite impulse response (FIR) filter(s) applied to the information at valid pixel locations to produce reconstructed information for non-valid pixel locations; and adaptively normalizing the reconstructed information for the non-valid pixel locations produced by applying the FIR filter(s) to the mask so that the valid pixels and the reconstructed information at non-valid pixel locations provide a reconstructed image.

Abstract: A method of diagnostic image reconstruction from projection data is provided. It includes generating projection data followed by a convolution of the same. The convolved projection data is then scaled into unsigned, fixed precision words of a predetermined number of bits. The words are then split into a predetermined number of color channels corresponding to color channels of a multi-color rendering engine (150). Simultaneously and independently, the split words are backprojected along each of the color channels to obtain backprojected views for each color channel. The backprojected views for each color channel are accumulated to produce separate color images corresponding to each color channel. Finally, the separate color images are recombined to produce an output image. In a preferred embodiment, prior to the convolution of the projection data, a rebinning operation is performed to ensure that the projection data is in a parallel format.

Abstract: A digital image processing method for detecting when and where an image has been altered or modified includes the steps of generating an edge map of an original digital image; convolving the edge map with a carrier signal to generate a dispersed edge map; and combining the dispersed edge map with the original digital image to create a combined image. The combined digital image is checked to determine if it has been modified by correlating the carrier signal with the combined image to produce a recovered edge map; and comparing the recovered edge map with the original image to detect modifications to the combined image.

Abstract: A method for image processing combining a device-specific image processing kernel operation with a general image processing kernel operation. Device specific parameters are sent from a host computer to an imaging device. The imaging device selects appropriate device-specific convolution coefficients. The host computer selects additional kernel operations. In a first example embodiment, the device-specific coefficients are then uploaded to the host computer and the host computer convolves the host-specified operations with the device-specific operations. The combined kernel is then downloaded to the imaging device. In an alternative embodiment, the host-specified operations are downloaded to the imaging device for combining. The combined kernels are then used by the imaging device for convolution operations on an image. As a result, host computer software can modify a kernel operation within an imaging device with minimal knowledge of device-specific parameters such as native resolution.

Abstract: Signal processing and pattern recognition is efficiently accomplished by using a plurality of low degree polynomials to approximate the images. The polynomials are then differentiated to obtain impulse functions. The impulse functions may be easily and efficiently convolved and the convolution subsequently integrated to extract the desired signal information.

Abstract: Apparatus for embedding information in a digital image digitized from a developed photographic film in response to a film property and information to be embedded into the digital image includes a scanner for scanning the image on the developed photographic film to produce the digital image. The apparatus stores the film property and has input circuitry for storing information associated with the image and to be embedded in the digital image. The apparatus responds to the stored the film property and the associated information for modifying a predetermined number of pixel values with the associated information in the digital image so that the associated information is embedded in the digital image, wherein the embedded information can subsequently be extracted.

Abstract: An apparatus produces an encoded and compressed digital data stream from an original input digital data stream using a forward discrete wavelet transform and a tree encoding method. The input digital data stream may be a stream of video image data values in digital form. The apparatus is also capable of producing a decoded and decompressed digital data stream closely resembling the originally input digital data stream from an encoded and compressed digital data stream using a corresponding tree decoding method and a corresponding inverse discrete wavelet transform. A dual convolver is disclosed which performs both boundary and nonboundary filtering for forward transform discrete wavelet processing and which also performs filtering of corresponding inverse transform discrete wavelet processes. A portion of the dual convolver is also usable to filter an incoming stream of digital video image data values before forward discrete wavelet processing.

Abstract: The present invention pertains to a process which can be fully automated for accurately determining the alleles of genetic markers. More specifically, the present invention is related to performing PCR amplification on locations of DNA to generate a reproducible pattern, labeling the PCR products, converting the labels into a signal, operating on the signal, and then determining the genotype of the location of the DNA. An amplification can include multiple locations from the DNA of one or more individuals. The invention also pertains to genetics applications and systems which can effectively use this genotyping information.

Abstract: A method of embedding digital data in a source image includes the steps of: a) generating a multi-level data image representing the digital data; b) convolving the multi-level data image with an encoding carrier image to produce a frequency dispersed data image; and c) adding the frequency dispersed data image to the source image to produce a source image containing embedded data. The data is recovered from the image by: a) cross correlating the source image containing embedded data with a decoding carrier image to recover the data image; and b) extracting the digital data from the recovered data image.

Abstract: A method and apparatus for resizing digital or stored images initially retrieves a one-dimensional sample of the image, such as a line of pixels. A final image size D is determined so that the absolute value of the original sample size M-2.sup.N *D is a minimum, and where N is an integer greater than or equal to 0. The discrete series of pixels in the line are then converted to a continuous function under a cubic convolution interpolation technique. From the continuous function, intermediate pixel values are determined. Pyramid filtering is employed to filter the intermediate pixel values to a final series of pixel values D. The routine is performed along the opposite dimension so as to alter the size of a two-dimensional stored image.

Abstract: A post-processing system for an optical correlator automatically detects and ranks peaks in the correlation image. An array of input buffers each receive pixels from a predetermined region of the image from the correlator camera. An array of peak detectors search the image pixels in each region for correlation peaks exceeding a predetermined threshold figure of merit. For example, each peak detector can be convolver array and a local maximum detector. The peak detectors generate a report entry for each such correlation peak that contains the figure of merit and location of the peak within the image. Results sorting and reporting means collect and store the report entries from each peak detector. The report entries are sorted by their figures of merit and reported to the host computer system.

Abstract: A method of improving a digital image is provided. The image is initially represented by digital data indexed to represent positions on a display. The digital data is indicative of an intensity value I.sub.i (x,y) for each position (x,y) in each i-th spectral band. The intensity value for each position in each i-th spectral band is adjusted to generate an adjusted intensity value for each position in each i-th spectral band in accordance with ##EQU1## where S is the number of unique spectral bands included in said digital data, W.sub.n is a weighting factor and "*" denotes the convolution operator. Each surround function F.sub.n (x,y) is uniquely scaled to improve an aspect of the digital image, e.g., dynamic range compression, color constancy, and lightness rendition. The adjusted intensity value for each position in each i-th spectral band is filtered with a common function and then presented to a display device.

Abstract: A method and apparatus for assessing the visibility of differences between two input image sequences. The apparatus comprises a visual discrimination measure having a retinal sampling section, a plurality of temporal filters and a spatial discrimination section. The retinal sampling section applies a plurality of transformations to the input image sequences for simulating the image-processing properties of human vision. The temporal filters separate the sequences of retinal images into two temporal channels producing a lowpass temporal response and a bandpass temporal response. The spatial discrimination section applies spatial processing to the temporal responses to produce an image metric which is used to assess the visibility of differences between the two input image sequences.

Abstract: A convolver includes a plurality of multipliers for multiplying pixel values of a convolution window by corresponding coefficients of a convolution mask to provide products, a summer for summing the products to provide a result and a memory for storing intermediate results. The convolver may be used to perform an N.times.N convolution in two or more passes. A first subset of pixel values of an N.times.N convolution window and a first subset of corresponding coefficients of an N.times.N convolution mask are supplied to the multipliers during a first pass of the N.times.N convolution. The summer provides an intermediate result for the first pass and stores the intermediate result in the memory. A second subset of pixel values of the N.times.N convolution window and a second subset of corresponding coefficients of the N.times.N convolution mask are supplied to the multipliers during a second pass of the N.times.N convolution.

Abstract: A method for combining a first digital image and a second background digital image, both images including pixels having color values, wherein the first digital image includes both a foreground region having foreground color values and a key color region characterized by a key color, as well as a mixed region where the pixel color values are a mixture of the foreground color value and the key color, includes determining a first control signal that indicates the relative proportions of the foreground color value and the key color for pixels in the first digital image; and segmenting the first digital image into a key color region and a non-key color region in which the non-key color region includes pixels in the first digital image that are not in the key color region.

Abstract: Am image is reconstructed from x-ray attenuation data or other types of data using fast Fourier circular convolution (FFCC). Reference spectra and mask functions are generated for the FFCC. Also, the data is zero-padded for the FFCC. The data is transformed, multiplied by a radial filter, and transposed in order to construct the image by FFCC. The reconstructed image is converted from polar to Cartesian coordinates for viewing.

Abstract: An image recognition apparatus which is capable of mounting many function such as an image enhancement and image recognition processes into one chip, for thus processing an image processing algorithm of an image enhancement and recognition in real time by extracting a characteristic of an object, comparing the thusly extracted characteristic with a reference pattern, recognizing an image, thus enhancing an image by using a convolver which is a spatial region filter, and by selectively outputting a data which was obtaining by processing the recognized image data or the image data, providing an image signal receiving and inputting unit, and a memory in a host computer for storing the data which were obtained through an image enhancement and recognition process, whereby the image recognition apparatus according to the present invention does not need a memory having a large capacity.

Abstract: A multiresolution method and apparatus for searching of a database of images where the search is performed on compressed images, without first decompressing them. The method searches the database of compressed images first at a low resolution to obtain the relative quality of a match between a search template and a candidate image. If the match is below a particular threshold, the search is terminated without committing any further computational resources to the search. Conversely, if the match is above a particular threshold, the method enhances the resolution of the candidate image and then performs another match. As long as the relative quality of the match is above the particular threshold, the resolution of the candidate image is successively enhanced, until a match determination is made at a full resolution of the candidate image.

Abstract: A computer graphics system interpolator for generating pixel values in a destination image of an object in a destination image space. The pixel values in the destination image are generated from a source image of the object in a source image space. The destination image and the source image each typically comprise a two-dimensional array of evenly-spaced pixels. A pixel in the destination image is transformed to an associated resampled point in the source image space. Intermediate pixel values are determined by interpolation between horizontally-aligned neighbor pixels in the source image space. Neighbor pixels in the same row as the resampled point in the source image space are determined. The distance between the pixels in the source image space are normalized to a value of unity and a first distance between the resampled point and an immediately adjacent neighboring pixel is determined.

Abstract: A system and method for performing a discrete image convolution using convolution masks that are symmetric in both cardinal directions. The method is used in an image processor. The advantage of the convolution processing in accordance with the invention is the reduction in the number of multiply operations. With the convolution having symmetry in both m and n, the convolution processing minimizes the number of multiple operations in the following manner: ##EQU1## where c(x,y) is the convolved image array, f(m,n) is the convolution mask, and g(x,y) is the image array to be processed. The invention reduces the number of multiply operations to convolve g(x,y) with f(m,n) by up to a factor of four, to (m+1)(n+1)/4, in relation to conventional image convolution processing.

Abstract: Methods and apparatus are provided for processing a source image in an image processing system including an electronic digital computer and image processing software for operating the computer. The source image is processed by convolution of pixels of the source image with coefficients of a convolution kernel to provide an output image. A lookup table is generated before new pixel values are determined. The table includes products of each coefficient and each possible value of the pixel in the source image. For each of the coefficients, the product that corresponds to the pixel value in a convolution window is accessed in the table. The accessed products are summed to provide a new pixel value in the output image. The steps of accessing the products in the table and summing the accessed products are repeated for each of the source pixels being convolved.

Abstract: A method to extract automatic high level features of a gray level ridge flow image including the automatic determination of the location of the high level features including the core and delta points of a fingerprint image in the form of a gray level image, comprising the following five steps:(1) determining a ridge flow direction of each gray level pixel of a fingerprint image and assign it with a direction code;(2) finding out a block directional flow and assigning each with appropriate direction code thereof;(3) correcting block flow directions in which a detection of the flow direction is affected image an input image quality in order to be immune from noises infiltrated during the acquiring process;(4) locating the delta points which are based on a corrected block direction flow diagram to locate from zero to a few delta points; and(5) locating the core points which are based on a corrected block direction flow diagram to locate one to two core points.

Abstract: A data processing system and method for performing a wavelet-like transformation and a corresponding inverse wavelet-like transformation is disclosed. The wavelet-like transformation is performed on input data so as to produce decomposed data. For each set of decomposed data samples of the decomposed data, each decomposed data sample of the set is produced by computing a weighted sum of a predefined set of data samples selected from (A) subsets of the set of input data samples, (B) one or more spatially shifted subsets of the set of input data samples, (C) the sets of decomposed data samples, and (D) one or more spatially shifted sets of the sets of decomposed data samples. The weighted sum is computed using only add and bit shift operations. Similarly, the inverse wavelet-like transformation is performed on decomposed data so as to produce reconstructed data.

Abstract: A patient (7) is irradiated by an X-ray source (1) in a computed tomography apparatus. The radiation is subsequently detected by the detector cells (5) of a position-sensitive X-ray detection system (4) and the intensities detected are applied to a computing device (16). Absorption as well as elastic and inelastic scattering of X-rays occur within the patient (7). The data acquired is corrected for elastic (coherent) scatter by deriving a deconvolution function from the elastic scatter function, which deconvolution function is applied to the data. The elastic scatter function is determined, for example by a computer simulation.

Abstract: A technique, referred to as area-based interpolation, performs image interpolation. The system determines a curve by the pixel value at a location by taking the integral of a curve over a small area, where the size of the area is determined by the sampling size of a sampling cell. When the image is resampled with respect to a sampling cell that has a finer spacing, the system integrates the polynomial using a finer integration area. In accordance with the invention, the relation between the reintegrated, resampled high resolution image and the low resolution image is a function of an up-sampler, followed by a linear filter. The coefficients of the filter are independent of the data, but are dependent on the family of curves used. If the system models by, for example, a third degree or fourth degree polynomial everywhere in the image, then that model determines the number of coefficients that are in the filter.

Abstract: A system is provided for processing matrices of data on an accelerted basis by iterative procedures that significantly reduce the number of steps required to access, move and computationally process the data, achieving significant improvements in operational efficiency. Such acceleration is applicable to, and useful in conjunction with assembly of the data, in performing both spatial filtration and temporal filtration (24) of the assembled data, and in processing the data, once acquired. In general, this is accomplished through optimization of the procedures (i.e. routines) used to handle the data so that significant increases in signal-to-noise ratio and overall processing speed are achieved. Such optimization is accomplished by replacing procedurally intensive routines (in terms of memory movements and computational steps) with more efficient routines for achieving a similar result.