Abstract: This invention is a method applicable to an image processing device, which includes the steps of providing a preprocess module for extracting a high-frequency portion of an image inputted into the device, extracting a gradient of the image and decomposing the image into plane and edge regions according to a predetermined fixed threshold, and providing a composite up-scaling module for executing the magnification processes on the image and the high-frequency portion thereof respectively, wherein the magnification process of plane regions of the image and the high-frequency portion is based on a simple interpolation while the edge regions of the image and the high-frequency portion is based on both a smart interpolation and the simple interpolation. The magnification results of the image and the high-frequency portion are then processed by a fusion process, so as to output an image having sharp but not blocky edges, rich details and strong contrast.

Abstract: Disclosed is a technique for processing pixel array data of a mobile terminal with a digital camera function. A device for processing pixel array data includes: an image sensor having (x×y)-sized pixel capacity, converting an optical signal received via a lens into an electric signal, and generating (x×y)-sized pixel array data; and an image signal processor for generating an (X×Y)-sized extended pixel array data area, receiving pixel values of the (x×y)-sized pixel array data from the image sensor, arranging the pixel values of the (x×y)-sized pixel array data in the (X×Y)-sized extended pixel array data area, arranging pixel values obtained by performing an arithmetical operation on pixel values of (x×y)-sized pixel array data, adjacent to remaining pixel areas of the (X×Y)-sized extended pixel array data other than the pixel areas having the pixel values of the (x×y)-sized pixel array data, in the remaining pixel areas, and outputting the (X×Y)-sized extended pixel array data.

Abstract: A method for calculating pixel values of interpolated pixel positions located between integer pixel positions in frames of a video picture by a predefined filter operation, the method including: storing the pixel values of the integer pixel position; determining to which of a number of predefined resolution formats the video picture is adapted; defining a fixed width (W) and fixed height value (H) for the frames corresponding to a resolution format determined by the determining; loading a first set of memory addresses for the memory of the video processing apparatus corresponding to a plurality of the pixel values of integer pixel positions; and executing a filter operation on the plurality of the pixel values of integer pixel positions and calculating a plurality of the pixel values of the interpolated pixel positions.

Abstract: Noise does not occur even if an image of an object region extracted from an input image and a background image are stored separately, and then displayed or printed, at a boundary portion of the object region. A boundary region is set at the inner side of the object region of the input image, and a pixel value of an adjacent pixel at the outer side of the boundary of the object region is written to a corresponding pixel in the boundary region. Even if a difference occurs between the background image and the object image, when a document file is displayed or printed, since the pixel value of the pixel at the outer side of the object region is assigned to the pixel in the boundary region, noise due to average color does not occur.

Abstract: An imager has first and second photosensitive sites and an interpolator located in a semiconductor substrate. The first photosensitive site is configured to receive light having a spectral component, and the second photosensitive site is configured to measure the level of the spectral component in light received by the second photosensitive site. The interpolator is configured to estimate the level of the spectral component in the light received by the first photosensitive site based on the measurement by the second photosensitive site.

Abstract: According to one embodiment of the invention, a method is provided for aspect ratio distortion minimization. The method includes receiving input pixels from a video source. The method then determines an input position and scale factor for each input pixel. A count value determines that a pixel should be outputted. A polyphase finite impulse response filter is centered on a particular input pixel based on the count value. An output pixel is generated using the polyphase finite response filter on a particular input pixel based on the count value and output pixels are outputted.

Abstract: A process for creating three-dimensional or relief views with the help of software running on a computer system for adjusting brightness of a scene comprising a plurality of pixels includes: determining a representative brightness of at least one predetermined area of the scene based on brightness of each pixel of the at least one predetermined area; comparing the average brightness to a reference value; and correcting the brightness of at least a portion of the scene in accordance with a result of the comparison.

Abstract: Disclosed is an imaging device provided with a moving image decoding section to decode encoded moving image data encoded by MPEG technique into moving image data, blur total motion vector estimating section to obtain motion vector of macroblock selected by macroblock selecting section that is suitable for estimating a total motion vector as well as to estimate the total motion vector of frame according to the motion vector, and a total motion vector interpolating section to perform blur correction when replaying moving image according to the estimated total motion vector.

Abstract: In accordance with this disclosure, methods, systems, and computer readable media for synthetic wavelength-based phase unwrapping in optical coherence tomography are provided. Synthetic wavelength phase unwrapping can be applied to OCT data and can correctly resolve sample motions that are larger than ?o/2. A method for phase unwrapping of an OCT signal can include acquiring raw OCT signal data, interpolating and processing the OCT signal data to obtain a DC spectrum, comparing the raw OCT signal data to the DC spectrum to generate an interference signal, applying Gaussian windows to the interference signal to generate a two separate signals, extracting phase information from the two signals, and comparing the phase information of the two signals to produce unwrapped phase data. A value of 2? can be added to the phase data if the difference between the phase information of the two signals is less than zero.

Abstract: An embodiment is a method and apparatus to provide a technique to scale up an image. A first image at a first size is expanded to a second size in first and second dimensions. The second size is larger than the first size and includes the first size and interpolating positions. The expanded image is convolved with a two-dimensional finite-impulse response (FIR) filter having coefficients determined by a two-dimensional interpolating function to generate interpolated pixels at the interpolating positions. The interpolated pixels are part of an up-scaled image at the second size.

Abstract: Missing-pixel interpolation data are obtained such that the mean value of a plurality of pixels constituting a group of pixels including a missing pixel is equal to the mean value of a plurality of pixels constituting a group of pixels not including the missing pixel. For example, missing-pixel interpolation data are obtained by (1) obtaining a sum (SL) of the values of pixels other than the missing pixel (L) in the k pixels constituting the group (LC) of pixels including the missing pixel, (2) obtaining a sum (SA) of the values of k pixels constituting a group (NA) of pixels not including the missing pixel, and (3) taking the difference between these sums. This makes it possible to reduce interpolation error when a highly periodic image with missing pixels is interpolated.

Abstract: A method of image processing for transforming an input image with a first resolution into an output image with a second resolution is provided. The input image is defined with a plurality of input pixels, and the output image is defined with a plurality of output pixels.

Abstract: A method and apparatus is disclosed for obtaining interpolation filters for use in a method or apparatus for enhancing the spatial resolution of a digital image. In the method or apparatus for enhancing the spatial resolution, input pixels from a low resolution image are classified and interpolated using the interpolation filters to form a high resolution image. The method for obtaining the interpolation filters includes obtaining a low resolution image from a high resolution source image. A plurality of pixels in the low resolution image are classified with respect to interpolation filters so that each of the plurality of pixels in the low resolution image is allocated to a respective interpolation filter. The interpolation filters are updated. The classifying and updating steps are repeated until a convergence criterion is met.

Abstract: A method and apparatus for rendering 3D graphic data is provided. The 3D graphic data is projected onto a 2D screen and points are interpolated and rendered, thereby quickly processing the 3D graphic data.

Abstract: An artificial neural network (ANN) based system that is adapted to process an input pattern to generate an output pattern related thereto having a different number of components than the input pattern. The system (26) is comprised of an ANN (27) and a memory (28), such as a DRAM memory, that are serially connected. The input pattern (23) is applied to a processor (22), where it can be processed or not (the most general case), before it is applied to the ANN and stored therein as a prototype (if learned). A category is associated with each stored prototype. The processor computes the coefficients that allow the determination of the estimated values of the output pattern, these coefficients are the components of a so-called intermediate pattern (24). Assuming the ANN has already learned a number of input patterns, when a new input pattern is presented to the ANN in the recognition phase, the category of the closest prototype is output therefrom and is used as a pointer to the memory.

Abstract: A vector interpolator optimizes the conversion of an interlaced signal to a non-interlaced signal. The vector interpolator improves the visual clarity of slanted features in a displayed image by adjusting the luminance value of each pixel such that the appearance of “steps” or “jaggies” in the features is reduced. For each pixel, the vector interpolator determines a similarity measure for the pixels within a predetermined area around the pixel. From the similarity measure, an angle for interpolation is selected. The luminance value is then interpolated along the selected vector corresponding to the angle and applied to the pixel.

Abstract: A method and a device for measuring a pulmonary nodule volume using a dual-energy subtraction image are provided. The method includes: (a) obtaining a dual-energy subtraction image of a chest; (b) setting a specific region including a pulmonary nodule region in the obtained dual-energy subtraction image; (c) interpolating a gray level trend of a region surrounding the pulmonary nodule region into the pulmonary nodule region using a fitting function; and (d) measuring the pulmonary nodule volume by subtracting the interpolation result from the gray level data of the pulmonary nodule region. Accordingly, when it is desired to measure the pulmonary nodule volume using the dual-energy subtraction image, information on the thickness of the pulmonary nodule is obtained from the characteristic of the dual-energy subtraction image, and the pulmonary nodule volume is accurately measured by correcting errors at the border of the pulmonary nodule region using the fitting function.

Abstract: Methods, systems and apparatus, including computer program products, for anti-aliased rendering. A plurality of objects is received. Each object in the plurality of objects is depth-ordered. Each object is rendered from back to front into a plurality of pixels. Each pixel in the plurality of pixels has a respective color value. The rendering includes selecting an object from the plurality of objects. Respective alpha values are determined for first pixels in the plurality of pixels based on coverage of the first pixels by the selected object. The respective color value of each of the first pixels is attenuated based on the respective alpha values. The respective color value of each of the first pixels is added to based on the respective alpha values and one or more color properties associated with the selected object.

Abstract: The invention features a method, system, and computer program product. A source bitmap image, a destination bitmap image, a source image feature outline surrounding a source region of the source bitmap image, and a destination image feature outline surrounding a destination region of the destination bitmap image are received. A point-to-point correspondence between the source image feature outline and the destination image feature outline is generated, such that the point-to-point correspondence defines a reshaping of the source image feature outline. An interpolation surface is calculated interpolating an area of the source bitmap image that includes the source region. The interpolation surface is reshaped using the point-to-point correspondence.

Abstract: A translation device has a translation processing unit that outputs character-area-character data by performing a translation process on the character-area-character data of an input image representing a figure-containing document, and outputting figure-area-character data by performing a translation process on the figure-area-character data of an input image representing a figure-containing document, a figure identification information searching unit that searches figure identification information used to identify a figure contained in a character area of the figure-containing-translation data, a linked page number searching unit that searches a page number in a character area of the figure-containing-translation data, a page number extracting unit that extracts a page number of a page included in figure area of figure identification information, and a page number rewriting unit that rewrites the searched page number linked to the figure identification information to the extracted page number of a page.

Abstract: Image data of a digital signal is inputted to a digital signal processor. A CPU reads out color correction coefficients from a ROM. Weights are added to the color correction coefficients based on information regarding transfer efficiency of a CCD image sensor such that the coefficients in a horizontal direction are set larger than those in a vertical direction. Further, an interpolation processor calculates horizontal and vertical components of R, G, and B colors of each pixel through color interpolation processing based on image data of each pixel and its peripheral pixels. A linear matrix circuit is controlled by the CPU and subjects the horizontal and vertical components to linear matrix operation using the color correction coefficients, thus subjecting each pixel data to color correction processing.

Abstract: The present invention relates to an apparatus for processing an image signal etc. that are well applicable to removal of coding noise from, for example, an image signal. Based on five consecutive frames of an image signal Va, a memory portion 121 outputs as pixel data xi of predictive taps plural items of pixel data located in a space directional and time directional peripheries with respect to a target position in an image signal Vb. In the case, frames before and after a current frame are subjected to motion compensation by using a motion vector. A class classification portion 124 obtains a class code CL indicating a class to which pixel data of the target position in the image signal Vb belongs, by using the pixel data xi and motion vectors BWV(0), BWV(?1), FWV(0), and FWV(+1). A calculating circuit 126 obtains pixel data y of the target position in the image signal Vb based on an estimation equation by using the pixel data xi and coefficient data Wi that corresponds to the class code CL.

Abstract: A recording medium having recorded thereon information related to coded data of an image sequence; wherein the information related to coded data of an image sequence is a set of coded frame information of a plurality of frames of an image sequence; wherein the coded frame information includes: information related to a difference image between an image of a current frame and a prediction image of the current frame, the prediction image being synthesized by performing motion compensation, information related to motion vectors estimated in performing the motion compensation, and rounding method information specifying either a positive rounding method or a negative rounding method for interpolating intensity values of pixels in performing the motion compensation when the input image is coded as a P frame, wherein no rounding method information is included in the coded frame information when the input image is coded as an I frame.

Abstract: A regression analysis is carried out (8) using pixel signals having a K-th spectral characteristic as the explanatory variable and pixel signals having an L-th spectral characteristic as the purpose variable in a plurality of pixel positions in an area neighboring a pixel of interest to obtain a pixel signal having the L-th spectral characteristic (9). Pixel signals obtained by low-pass filtering (7a-7c) of the output signals of an imaging device may be used as the explanatory variable and the purpose variable. The occurrence of false colors is thereby reduced when, in a group of pixel signals from pixels arrayed on a two-dimensional plane, each pixel having one of a plurality of spectral characteristics, the missing colors at each pixel position are obtained by interpolation.

Abstract: Certain embodiments of the invention may be found in a system and method for implementing graphics and a video scaling algorithm using interpolation based on symmetrical polyphase filtering. A video or graphics scaler may be utilized to scale luma, chroma, and/or alpha information in a video image. The scaler may comprise a first symmetric polyphase sub-filter with zero phase shift that generates an in-phase filtered pixel and a second symmetric polyphase sub-filter that generates an out-of-phase filtered pixel. The video scaler may also comprise an interpolator that may generate a scaled video image pixel based on the generated in-phase and out-of-phase filtered pixels and a scaling factor. The scaling factor may be determined based on an input video size (M) and a desired output video size (N). The interpolation of the generated in-phase and out-of-phase pixels in the video scaler may be implemented by utilizing a Farrow structure.

Abstract: Systems and methods are provided for optimizing visual consistency for a dataset of images by using observed and formulated display attributes. The formulated display attributes are derived from an analysis of a first set of images from the image repository. The formulated display attributes from the first set are combined with observer's display attributes to determine scaling factors. The optimized display attributes are then derived from the calculated scaling factors for a second set of images. In another aspect, a system and method is described where a digital image is processed in order to enhance viewing of the image on a display device in accordance to the optimized attributes associated with the observer.

Abstract: An image processing apparatus that converts an input image having first resolution into an output image having second resolution higher than the first resolution includes a super-resolution processing unit that converts the input image into an SR image having the second resolution through motion compensation, a motion mask generating unit that generates a motion mask indicating an accuracy of estimation, an edge direction information generating unit that generates edge direction information indicating an edge direction of each area of the input image, a weight map generating unit that generates a weight map indicating likelihood of the edge direction of each of the areas, an interpolation processing unit that converts the input image into an interpolated image having the second resolution, a weight computing unit that computes a weight, and an image generating unit that generates the output image by constructing a weighted sum of the SR image and interpolated image.

Abstract: A method of generating three-dimensional (3D) image data from first and second image data obtained by photographing the same subject at different points of time, the method including generating third image data by adjusting locations of pixels in the second image data so that the second image data corresponds to the first image data, and generating the 3D image data based on a relationship between the third image data and the first image data.

Abstract: A scene is captured. A two-dimensional matrix comprising a plurality of image detection components is generated according to the captured scene. Each image detection component corresponds to one color component. A plurality of matrixes are generated according to the image detection components. One image detection component and a portion of matrixes are calculated to obtain the missing color components.

Abstract: According to one embodiment, a first correlation calculator calculates a correlation between first pixel blocks, and detects as first reference pixels actual pixels contained respectively in the first pixel blocks with the highest correlation. A second correlation calculator calculates a correlation between second pixel blocks, and detects as second reference pixels actual pixels contained respectively in the second pixel blocks with the highest correlation. The first pixel blocks include pixels arranged in a plurality of rows and columns The second pixel blocks include pixels arranged in at least one row less than the rows of the first pixel blocks and a plurality of columns. An interpolation calculator calculates, when the first reference pixels are located perpendicular to the actual pixel lines, the pixel value of the interpolation pixel based on the second reference pixels.

Abstract: Aspects of the present invention are related to systems and methods for image demosaicing.

Abstract: A correlation value calculation circuit calculates respective correlation values of each pixel for color image or for gray image in four directions. A selection circuit determines respective correlation values (Cv, Ch, CdA and CdB) in the four directions on the basis of a chroma evaluation value. A first correlation judgment circuit determines a correlation direction of a specified pixel from the correlation values (Cv, Ch, CdA and CdB). A first interpolation circuit performs color interpolation on the basis of the determined correlation direction. A color-interpolated image is enlarged by an enlargement circuit. A correlation direction interpolation circuit performs enlargement and interpolation of the correlation direction determined by the first correlation judgment circuit. A filter performs a filtering process for each of pixels of the enlarged image by using the interpolated correlation direction for enlargement.

Abstract: A Bayesian two-color image demosaicer and method for processing a digital color image to demosaic the image in such a way as to reduce image artifacts. The method and system are an improvement on and an enhancement to previous demosaicing techniques. A preliminary demosaicing pass is performed on the image to assign each pixel a fully specified RGB triple color value. The final color value of pixel in the processed image is restricted to be a linear combination of two colors. Fully-specified RGB triple color values for each pixel in an image used to find two clusters represented favored two colors. The amount of contribution from these favored two colors on the final color value then is determined. The method and system also can process multiple images to improve the demosaicing results. When using multiple images, sampling can be performed at a finer resolution, known as super resolution.

Abstract: Systems and methods that generate or provide overlap displays of multiple sets of data in a manner that advantageously simplifies trend visualization in large sets of data. A two-dimensional, three-dimensional, or greater-dimensional occurrence count array is generated for a plurality of similar data sets. Each element in the array is a number of times a corresponding vector of data values occurs in the plurality of N data sets, wherein each array element corresponds to a discrete interval of data values. Once the array has been generated, a graphical display of overlap density may be generated by comparing a desired degree of overlap with the value of each array element. Those array elements having values that satisfy the desired degree of overlap are rendered, as a display object, using one or more colors, shades and/or patterns representing varying degrees of overlap.

Abstract: A method for enhancing image quality of motion compensated interpolation includes generating an interpolated frame according to at least two source frames by analyzing motion estimation information of the two source frames. The method further includes: regarding a pixel under consideration within the interpolated frame, selectively performing post filtering according to motion estimation information of a region where the pixel is located. Accordingly, an apparatus for enhancing image quality of motion compensated interpolation is also provided.

Abstract: The presented systems and methods are directed towards processing of image data. Various aspects are directed towards sharpening digital images while other aspects are directed towards interpolating and/or upsampling image data. The utilities have particular application to digital image processing. However, it will be appreciated that aspects of the present invention may be utilized with any pixel imaging application including seismic imaging.

Abstract: An interpolation image generating method includes dividing each of the first reference image and the second reference image into reference regions each including pixels, executing a correlation operation between the first reference image and first destination images located before and after the first reference image and a correlation operation between the second reference image and a second destination image to derive motion vectors for the first and second destination images every reference region, obtaining correlation values between the regions of the first and second destination images that are indicated by the motion vectors and the reference region to determine the reference region as a high or low correlation region, generating an interpolation image candidate between the first reference and second images using the reference region determined as the high correlation region, and mixing the interpolation image candidates using the motion vectors of the reference region to produce an interpolation image.

Abstract: An apparatus and method for enhancing the quality of reproduced images are provided. More specifically, provided are an apparatus and method for enhancing the quality of reproduced, progressive images, which are capable of obtaining more natural and smoother progressive images and preventing such problems as blurry colors and aliased images that could be brought about in the process of converting interlaced images into progressive images.

Abstract: An interpolation processing method for performing interpolation processing of image data obtained by photoelectrically converting an object image using interpolation data, includes a first step of estimating object information, a second step of generating first input image data based on the object information, a third step of generating first transformation data from the first input image data to the image data, a fourth step of generating second transformation data from the first input image data to interpolated image data after the interpolation processing, and a fifth step of generating the interpolation data using the first and second transformation data.

Abstract: A data divider divides N-dimensional digital image signals into higher-order bit signals and lower-order bit signals. Reference values corresponding to all combinations of the divided higher-order bit signals are divisionally stored in 2N?1 sub-memories without overlaps. The (N+1) reference values necessary for interpolation processing are read by a reference value reader at the same time.

Abstract: Disclosed is an image converter circuit which is capable of being to be applied to a display device having an arbitrary number of pixels, and is capable of reducing image deterioration in panoramic display. The image converter circuit comprises: a read-control section that reads image data from an image memory in accordance with read-designating information; a data interpolating section that interpolates pixels using the image data read out from the image memory, depending upon magnification-designating information, and outputs enlarged and interpolated data; and a magnification setting section that generates the read-designating information and the corresponding magnification-designating information. The magnification setting section vertically divides the image data into a plurality of areas such that a difference in a number of pixels between the areas is within a predetermined range.

Abstract: A camera includes an image capture unit which captures the image of a subject to obtain image data, a pixel interpolation unit which, on the basis of pixel data from multiple pixels arranged on a line in the image data obtained by the image capture unit, creates image data in an interpolation position on the line, and a storage unit which is stored with common coefficient data for pixel data from multiple pixels which are n-th pixels counted from the interpolation position in opposite directions on the line, the pixel interpolation unit performing interpolation operations on the pixel data from the n-th pixels using the coefficient data stored in the storage unit to thereby create the image data in the interpolation position.

Abstract: In a method and system for performing color correction for an image signal, a first set of matrix coefficients for color correction of the image signal in a 3-dimensional RGB color space is transformed to a first set of points of a two-dimensional XY plane. In addition, the first set of points is modified to a second set of points in the XY plane for tuning image quality. Furthermore, the first and second sets of points in the two-dimensional XY plane are displayed such as on a graphical user interface of a computer system.

Abstract: Images are upsampled using a knowledge base derived from a plurality of high-quality training images. The knowledge base is used to refine a high-frequency component including high-frequency aspects of a high-resolution, low-frequency image, interpolated from a low-resolution full-frequency image, into a high-frequency component. An enhancement step is performed without using a knowledge base to construct a high-compatibility component from the low-resolution, full-frequency image. The low-resolution, full-frequency image is combined with the coarse high-frequency component to yield an enhanced high-frequency component. A second knowledge base step is performed to construct an improved high-frequency component from the enhanced high-frequency component. The improved high-frequency component is blended with a high-resolution, low-frequency image to yield a high-resolution image.

Abstract: An apparatus and a method of interpolating an image via a region division are provided. To improve an image quality artifact around an edge, the apparatus to interpolate the image may determine whether an inputted image corresponds to an edge after performing a thinning process with respect to a gradient value of the inputted image, and convert a resolution of the image using optimal location information via a region division and a pixel replacement according to an interpolation direction of the image.

Abstract: An image interpolation device, which can perform image interpolation at low cost and with low power consumption. The image interpolation device includes a pixel data storage which temporarily and sequentially holds, in the form of a plurality of unit pixel data groups, data of adjacent pixels in the vertical direction of input image data. The image interpolation device also includes an image memory which can store at least three of the unit pixel data groups. Unit pixel data groups held in the pixel data storage are sequentially written to the image memory. At least two unit pixel data groups stored in the image memory are read simultaneously.

Abstract: A method of smoothing a triangulated raster image includes the steps of: triangulating the image to create triangles having vertices, each of which is centered on a pixel of the image; comparing the costs of a first and a second orientation of an edge of each pair of adjacent triangles that forms a quadrilateral; swapping the edge of the pair of triangles to the lower cost orientation; and iterating the comparing and swapping until a stopping point is reached. The cost function used may include a cost factor that causes a triangle edge orientation that aligns with a both a color and an intensity of a color intensity contour line of the image to have a lower cost than a triangle edge orientation that does not so align.

Abstract: According to the present invention, in a borderless printing mode, a printing region is determined from an original image, a region expanded outward is determined for borderless printing based on the determined region. When the thus determined region exceeds the region of the original image, an image in the surplus portion is extrapolated based on an image at end portions of the original image, and printing data is formed based on the extrapolated image and image data of the expanded region.

Abstract: A system and method for performing convolutions on image data using pre-computed acceleration data structures is disclosed. The method may include calculating intermediate convolution values for each of a plurality of blocks of pixels by performing an associative operation on the pixel values in each block. Each intermediate value may be associated with the block and indexed dependent on index values of pixels in the block. An image pyramid may include intermediate convolution values for multiple levels of acceleration by calculating intermediate convolution values for multiple block sizes. A convolution result for a kernel of an image may be produced by performing the associative operation on intermediate convolution values for non-overlapping blocks enclosed within the kernel and on pixel values associated with pixels in the kernel but not in one of the non-overlapping blocks. The methods may be implemented by program instructions executing in parallel on CPU(s) or GPUs.

Abstract: In digital data conversion apparatus and method, class data are generated in association with reference interpolated data for each of a plurality of classes on the basis of a reference high definition digital video signal which includes a reference standard definition digital video signal in addition to the reference interpolated data. The class data is stored at respective addresses in a memory. A standard definition digital video signal representing pixel values is received and then clustered so as to produce a class corresponding to the pixel values of the standard definition digital video signal. The class data is retrieved from the memory address which corresponds to the class of the standard definition digital video signal, and interpolated data is generated in accordance with the standard definition digital video signal and the retrieved class data with such interpolated data constituting a high definition digital video signal.