Abstract: Tunable Gaussian filters enable imaging effects to be applied to images and videos in orientations other than standard symmetric, 0 degree orientations and 90 degree orientations. The tunable Gaussian filters are able to be applied in any orientation such as 45 degrees, slightly less than 45 degrees and slightly more than 0 degrees.

Abstract: A method and an apparatus for constructing an image blur pyramid, and an image feature extracting circuit are disclosed. The image blur pyramid construction apparatus comprises a first image blur circuit, a second image blur circuit and an image sub-sampler. The first image blur circuit and the second image blur circuit simultaneously generate a first interval and a second interval in the same octave according to an input image, a first filter and a second filter. The dimension of the second filter is greater than that of the first filter. The image sub-sampler couples with the second image blur circuit and down samples the second interval to generate a sub-sample image.

Abstract: A method of providing a blurred image includes receiving a target image to which a blur effect is to be applied; receiving blur effect information for generating the blur effect in the target image; calculating a sum image for respective pixels of the target image; applying a mean filter to the target image by using the sum image based on the blur effect information; and providing a blurred image formed by applying the mean filter.

Abstract: A system is provided for processing video images for display. In one example, the system comprises a stream parser for retrieving a compressed main video data and a sub-picture identifier. A video decoder generates a main video image having a main video intensity value for each pixel in the main video. A sub-picture decoder determines a sub-picture image based on the sub-picture identifier. The sub-picture image includes a pixel data for each pixel in the sub-picture image indicative of a sub-picture pixel type including a background pixel type and a pattern pixel type. Each pixel includes a command code indicative of a background contrast level, a pattern contrast level, a background color level and a pattern color level. An intensity formatter generates a sub-picture pixel intensity for each pixel in the sub-picture image.

Abstract: An image processing apparatus for correcting a positional deviation between a plurality of images obtained in time series and performing synthesis processing on the plurality of corrected images determines a motion vector between the plurality of images, corrects the positional deviation between images on the basis of the determined motion vector, and generates a synthesized image by synthesizing the plurality of images subjected to the positional deviation correction. The image processing apparatus extracts a filter area in which filter processing is to be performed on the basis of a degree of inconsistency occurring when the positional deviation between the plurality of images is corrected, determines a filter kernel on the basis of an imaging condition, and performs filter processing based on the filter kernel on the filter area of the synthesized image.

Abstract: A noise reducing apparatus and associated method and program for improving image quality in an image are provided. The noise reducing apparatus detects a flat area from an image and analyzes the noise from the flat area, and then it suppresses a noise component of the image based on the noise analysis result.

Abstract: A method of generating a multiscale contrast enhanced image is described wherein the shape of edge transitions is preserved. Detail images are subjected to a conversion, the conversion function of at least one scale being adjusted for each detail pixel value according to the ratio between the combination of the enhanced center differences and the combination of the unenhanced center differences. Several adaptive enhancement measures are described.

Abstract: A super-resolution processor includes an N enlargement unit that generates an N-enlarged image by enlarging the input image by a factor N; an M enlargement unit that generates an M-enlarged image by enlarging the input image by a factor M; and a high-pass filter unit that extracts a high-frequency component of the M-enlarged image, as an M-enlarged high-frequency image. Additionally, a patch extraction unit extracts an estimated patch of a predetermined size from the M-enlarged high-frequency image, the estimated patch being a part of the M-enlarged high-frequency image; and an addition unit adds the estimated patch to a processing target block of the predetermined size in the N-enlarged image, to generate the output image, where M is smaller than N.

Abstract: A CPU generates an image by performing inverse projection transformation for an input image by using an inverse matrix of a matrix which enlarges a size of the input image. When a position A on the input image is transformed into a position B on the generated image by the inverse projection transformation, the CPU obtains a ratio of a distance between pixels around the position B to a distance between pixels around the position A. The CPU calculates filter coefficients for a low-pass filter using a product of a Nyquist frequency of the input image and the ratio as a cutoff frequency. The CPU obtains a sum of results obtained by multiplying pixel values of pixels in a pixel array including the position A in the input image by the filter coefficients, as a pixel value at the position A in an image having undergone filter processing.

Abstract: An image processing equipment generates a reduced image corresponding to an obtained image in parallel with a processing of storing the generated reduced image in a reduced image storing part, and pipelines a processing of extracting a low-frequency noise component of each of pixels included in the obtained image and a processing of sequentially subtracting the low-frequency being extracted from pixel data corresponding to one of the obtained image stored in the image storing part and an adjusted image generated from the obtained, so as to achieve a pipeline processing of a multi-resolution noise filtering with a few line memories.

Abstract: A hierarchical motion deblurring method for a single image is provided. In the method, a blur kernel of a target image is calculated and a multi-scale representation for representing the target image and the blur kernel is constructed. Then, a gradient attenuation function and a strong edge suppression function are applied to a residual Richardson-Lucy algorithm, so as to iteratively calculate a residual image between the blur kernel and the target image represented by the representation in each scale and restore the residual image to obtain a first restored image and a second restored image. Finally, the two restored images are compared so as to obtain a motion deblurring image.

Abstract: A mask correcting unit corrects an externally set mask pattern. A depth map processing unit processes a depth map of an input image for each of a plurality of regions designated by a plurality of mask patterns corrected by the mask correcting unit. An image generation unit generates an image of a different viewpoint on the basis of the input image and depth maps processed by the depth map processing unit. The mask correcting unit performs a blurring process on an object boundary part of the mask pattern.

Abstract: There is provided an algorithm that allows one, using a sequence of object images provided at different angles, to generate a 360 degree view model of an object. To generate such a model, the inventive algorithm may involve the following tasks: 1) provide a set of images (the number and size are unlimited), 2) reduce their features to the same brightness and contrast, 3) separate an object from a complex (heterogeneous) background in the images, 4) stabilize the objects in every image with respect to each other, and 5) process the resulting sequence of images to generate a 360 degree view.

Abstract: An image forming apparatus capable of reducing a stitch effect, an image forming system including the same, and a printing method thereof. The printing method of the image forming system includes inputting image data to be printed; and, if adjacent first and second objects included in the input image data have different LPIs (Lines Per Inch), emphasizing at least one of boundaries of the first and second objects and printing the image data with an emphasized boundary.

Abstract: A method for enhancing a digitized image of a document containing text or graphics. The digitized image is decomposed into a low-frequency pedestal image and a high-frequency residual image. A first gain factor image is determined by applying a first tonescale function to the residual image, and a second gain factor image is determined by applying a second tonescale function to the residual image pixel values, wherein parameters defining the tonescale functions are determined by analyzing the pedestal and residual images. An enhanced document image is formed by combining a scaled pedestal image determined by multiplying the pedestal image by the first gain factor image, and a scaled residual image determined by multiplying the residual image by the second gain factor image.

Abstract: A computer-implemented method for applying a filter to an image comprising: receiving a filter setting, selecting a selected filter from a family of filters stored on a computer-readable storage medium, based at least on the filter setting, wherein the family of filters comprises a plurality of filters each being an approximation of a function on different orders, and applying the selected filter to a stored image with a processor.

Abstract: Provided are a method of interpolating an image by determining interpolation filter coefficients, and an apparatus for performing the same. The method includes: differently selecting an interpolation filter, from among interpolation filters for generating at least one sub-pel-unit pixel value located between integer-pel-unit pixels, based on a sub-pel-unit interpolation location and a smoothness; and generating the at least one sub-pel-unit pixel value by interpolating, using the selected interpolation filter, pixel values of the integer-pel-unit pixels.

Abstract: An image processing apparatus and an image processing method blurs a background portion of an image. The image processing apparatus includes an image acquiring unit for acquiring original image data; an image processing unit for separating background pixels from foreground pixels in the original image; an image filtering unit for filtering the background pixels; and an image merging unit for merging the foreground pixels with the filtered background pixels in order to generate a new image having a blurred background.

Abstract: An image processing apparatus includes a motion-blur adding mechanism performing filter processing on moving image data to be subjected to coding processing in accordance with motion information indicating motion of an image between unit images included in the moving image data as pre-processing of the coding processing so that motion-blur addition processing is performed.

Abstract: An image processing apparatus sets the pixels of an input image as process target pixels by using low pass filters (LPF #1—1 to LPF #1—n) each having different level value ranges, computes the average values of the pixels included in the level value ranges from the pixels of the input image including process target pixels included within the filter sizes of the low pass filters, generates level-value-limited smoothed images #1 limited by the level values, generates level-value-limited smoothed images #2 by using low pass filters (LPF #2—1 to LPF #2—n) each having level value ranges or filter sizes different from those of LPF #1—1 to LPF #1—n, selects either of the level-value-limited smoothed images #1 or the level-value-limited smoothed images #2 on the basis of the process target pixels and level values different from the level values utilized by LPFs, synthesizes the selected level-value-limited smoothed images, and generates a smoothed image.

Abstract: An edge-preserving diffusion filter maintains the sharp edges in images while smoothing out image noise. An edge-preserving diffusion filter applies an edge-preserving smoothing filter to an image to form a filtered image. The modified image is then blurred by a blurring filter to form a blurred image. The modified image and the blurred image are blended together to form an output image based on an error metric associated with each pixel. The edge-preserving diffusion filter may be utilized to perform a multilevel decomposition of the image. The edge-preserving diffusion filter may be applied to an unfiltered image to produce a base image. The difference between the unfiltered image and the base image defines a detail image. The detail image may be used as the input for recursively generating additional levels of detail. The multilevel decomposition may utilize filter kernels associated with different contrast levels for each iteration.

Abstract: A method for producing a noise-reduced digital image captured using a digital imaging system having signal-dependent noise characteristics, comprising: capturing one or more noisy digital images of a scene, wherein said at least one noisy digital image has signal-dependent noise characteristics; defining a functional relationship to relate the noisy digital images to a noise-reduced digital image, wherein the functional relationship includes at least two sets of unknown parameters, and wherein at least one of the sets of unknown parameters relates to the signal-dependent noise characteristics; defining an energy function responsive to the functional relationship which includes at least a data fidelity term to enforce similarities between the noisy digital images and the noise-reduced digital image, and a spatial fidelity term to encourage sharp edges in the noise-reduced digital image; and using an optimization process to determine a noise-reduced image responsive to the energy function.

Abstract: An image inspection apparatus determines whether a test image is a copy of an original image. The image inspection apparatus includes an image group generation unit that determines one or more types of image transform, and transforms the original image according to the determined types of image transform to generate an original image group; a spatial color difference computing unit that transforms a color space of the generated original image group, and calculates first spatial color differences between the original image and images in the original image group to provide color difference data; an image inspection unit that receives the test image and the original image group to calculate second spatial color differences between the test image and the images in the original image group, and compares the second spatial color differences to the color difference data; and an inspection result output unit that outputs the comparison result.

Abstract: An image processing method for removing a noise component contained in an original image, includes: separating an original image into a temporary noise-free image and a temporary noise component; extracting an edge component in the temporary noise-free image by executing edge detection on the temporary noise-free image; determining a fine edge component in the original image contained in the temporary noise component based upon a level of the extracted edge component; extracting an actual noise component by excluding the fine edge component from the temporary noise component; and removing noise from the original image based upon the extracted actual noise component.

Abstract: A method for generating a depth map for a 2D image includes receiving the 2D image; analyzing content of the received 2D image; determining a depth map based on a result of the content analysis; refining the determined depth map using an edge-preserving and noise reducing smoothing filter; and providing the refined depth map.

Abstract: A method for estimating a blur direction (20) of motion blur (16) in a blurred image (14) includes the steps of blurring the blurred image (14) in a number of different test directions (360A) (362A) (364A), and finding the test direction (360A) (362A) (364A) for which the blurred image (14) changes the least by the additional blurring (366). With this design, when more blur (366) is applied to the blurred image (14) in a test direction (360A) (362A) (364A) that is similar to the blur direction (20), the difference in the image appearance is relatively small. However, when more blur (366) is applied to the blurred image (14) in a test direction (360A) (362A) (364A) that is very different to the blur direction (20), the difference in the image appearance is relatively large. In one embodiment, a blur difference is determined for each test direction (360A) (362A) (364A). Subsequently, the test direction (360A) (362A) (364A) with the smallest blur difference is selected as the blur direction (20).

Abstract: A method is provided in one example embodiment that includes receiving a plurality of depth values corresponding to pixels of an image; and filtering the image as a function of a plurality of variations in the depth values between adjacent pixels of a window associated with the image. In more detailed embodiments, the method may include encoding the image into a bit stream for transmission over a network. The filtering can account for a bit rate associated with the encoding of the image.

Abstract: A method for removing noise of an image. The method includes: (a) detecting a horizontal edge by applying a horizontal edge detection filter to a predetermined pixel field including a notice pixel and neighboring pixel in a vertical direction in image data; (b) judging whether horizontal line noise exists in the predetermined pixel field through the horizontal edge; (c) calculating the number of pixels determined as the horizontal line noise for each horizontal line of the image data by applying steps (a) and (b) to all horizontal lines of the image data; and (d) removing the horizontal line noise by applying a low pass filter to the horizontal line judged to have the horizontal line noise according to the calculation result of step (c).

Abstract: An image processing apparatus including: an error determination processing unit inputting a distance image having distance information or parallax information up to the subject, detecting an error signal included in the distance image, and outputting a synthesis coefficient according to an error degree of an image region unit; a low pass filter processing unit applying a low pass filter with respect to the distance image and generating a noise reduction distance image; and a distance image synthesis unit generating a corrected distance image using a synthesis process of the distance image and the noise reduction distance image, in which, the distance image synthesis unit follows a synthesis coefficient of an image region unit output by the error determination processing unit, performs synthesis processing in which the synthesis ratio of the distance image and the noise reduction image is set in the image region units, and generates the corrected distance image.

Abstract: An image processing apparatus includes a reducing unit that reduces an input image chronologically input as moving images and accumulates a reduced image of the input image as a first reduced image in a predetermined memory. The image processing apparatus further includes a generating unit that generates, as a second reduced image, one reduced image from multiple first reduced images accumulated in the memory. The image processing apparatus further includes a smoothing unit that smooths level values of a pixel included in the second reduced image and pixels surrounding the pixel with respect to each of multiple ranges set according to the width of level values of pixels. The image processing apparatus further includes a synthesizing unit that synthesizes multiple range-by-range results of the smoothing on the basis of the input images. The image processing apparatus further includes an output unit that outputs a synthesized image as an output image.

Abstract: A plurality of image signals having different exposure levels is acquired; low-frequency components are extracted from each of the image signals to generate a plurality of low-frequency image signals; and the plurality of low-frequency image signals are combined to generate a combined low-frequency image signal. A first image signal is extracted from the plurality of image signals; and high-frequency components are extracted from the first image signal to generate a high-frequency image signal. Then, a combined image signal is generated by combining the combined low-frequency image signal and the high-frequency image signal.

Abstract: An ultrasound imaging system and method provides for extended imaging.

Abstract: Various embodiments pertain to image correction techniques that automatically estimate and correct vignetting based upon a single image. In various embodiments, a local gradient is estimated and the radial component of the gradient is ascertained. Radial components of gradients are correlated with a radius of the pixel from the optical center of the image, and can be utilized to estimate an attenuation due to vignetting. In some embodiments, a tangential component of a difference of gradients is also ascertained and correlated with the radius for use in estimating the attenuation. Attenuation due to vignetting can be calculated through the application of a polynomial regression to a plot of the radial components versus radius. In various embodiments, the image can be corrected by inverting the attenuation in the image.

Abstract: Systems and methods for block noise detection and filtering are disclosed. One embodiment includes, computing difference magnitudes in pixel values for adjacent pixels in the image. The difference magnitudes can include horizontal difference magnitudes for horizontally adjacent pixels and vertical difference magnitudes for vertically adjacent pixels. One embodiment further includes using normalized sums of the difference magnitudes to determine a set of noise characteristics of the block noise and a set of image characteristics of the image and configuring inputs to the block noise filter using the set of noise and image characteristics.

Abstract: The present invention provides an image processing device and an image display system achieving image quality improvement in moving picture by suppressing a motion blur in a hold-type display device, while maintaining low cost. The image processing device processes image data provided from outside and outputting the image data to a hold type display device. The image processing device includes a correction processing section performing a correction process to correct a pixel value in the image data for each pixel through performing a spatial LPF (low pass filter) process on the image data in a frame to be displayed in the display device according to a magnitude of a motion vector in the image data, the spatial LPF process allowing a slope of a changing edge portion in the image data to be more gentle.

Abstract: A noise filter according to an embodiment includes: a first filter that functions as an edge detector to detect a high-frequency component area of an image; a second filter that performs a noise filtering function for the remaining areas of the image while conserving the high-frequency component area detected by the first filter; and a function processor that controls operations of the first filter and the second filter. According to an embodiment, since noise filtering for only a noise component area is performed by dividing the image into a high-frequency component area and a noise component area, it is possible to minimize deterioration of the high-frequency area and improve the resolution and quality of the image.

Abstract: Disclosed are embodiments for a system, method, and computer program product for performing an process on an original image, the process being implemented by a computer system performs a comprising the at least one computer: performing an process on an image that renders the processed image legible than then the original image, wherein the analysis segregates dark pixels of the image from light pixels of the image. The method can comprise: first converting the image into a grayscale image. The method comprises processing a pixel area for each pixel of the image is a dark pixel or a light pixel and determining if a pixel is proximate to an edge.

Abstract: The present invention relates to a method and device for obtaining an image of a crumpled document from an image of this document when it is crumpled. The method comprises a step of determining a three-dimensional geometric model of the surface of the crumpled document by triangulation of three-dimensional points defined from the pattern of a target extracted from the image of this document when it is crumpled; the method is characterised in that it comprises a step of determining a projection of the three-dimensional geometric model onto a so-called acquisition plane, by means of error minimisation of this projection under constraints of preserving defined geometric characteristics in the vicinity of the three-dimensional points, and in that it comprises a step of superimposing the textures associated with the three-dimensional model onto the projection of this model thus determined.

Abstract: A characteristic image detection interpolation unit performs interpolation processing for increasing the luminance value in low-luminance area which is partly present in an input image. A minimum value filter unit executes minimum value filter processing on an image subjected to the interpolation processing, and a low-pass filter performs low-pass filter processing on the image subjected to the minimum value filter processing, thereby generating a low-frequency component image. A subtracter generates a high-frequency component image by subtracting the low-frequency component image from the input image. A coefficient multiplier adds the luminance value of the high-frequency component image to the input image by a preset ratio K1, thereby generating a high-frequency component enhanced image. A selector alternately outputs the low-frequency component image and the high-frequency component enhanced image.

Abstract: The present invention relates to an apparatus and method of converting data to display smooth and vivid images and to improve the gray scale use efficiency. The apparatus of converting data includes a gray scale detector that that generates a gray scale detection signal and a gray scale change signal by detecting a same gray scale, from an M bit input data, in a plurality of pixels adjacent to one another. The apparatus of converting data further includes a gray scale corrector that generates an N bit data by correcting one of the same gray scale according to the gray scale detection signal and the gray scale change signal output from the gray scale detector, wherein N and M are integers, N being larger than M.

Abstract: A method for restoring and enhancing a space based image of point or spot objects. The method includes: 1) dividing a space-variable degraded image into multiple space-invariant image sub-blocks, and constructing a point spread function for each of the image sub-blocks; 2) correcting each of the image sub-blocks via a space-invariant image correction method whereby obtaining a corrected image {circumflex over (f)}i for each of the image sub-blocks; and 3) stitching the corrected image sub-blocks {circumflex over (f)}i altogether via a stitching algorithm whereby obtaining a complete corrected image {circumflex over (f)}.

Abstract: At least one approximation image is created of the image at one or multiple scales. Translation difference images are created by pixel-wise subtracting the values of an approximation image at scale s and the values of a translated version of the approximation image. A non-linear modification is applied to the values of the translation difference image (s) and at least one enhanced center difference image at a specific scale is computed by combining the modified translation difference images at that scale or a smaller scale with weights Wi,j. An enhanced image is computed by applying a reconstruction algorithm to the enhanced center difference images. The non-linear modification of the values of the translation difference images is steered by the values of an orientation map which comprises for each pixel a local direction of interest.

Abstract: An image processor includes a reduced image forming unit, a first reflectance value calculating unit, a normalizing parameter determining unit, a second reflectance value calculating unit, a normalizing unit, a correcting unit, and a reducing information determining unit. The reduced image forming unit reduces an original image to form a reduced image. The first reflectance value calculating unit calculates a first reflectance value for each reduced pixel. The normalizing parameter determining unit determines a normalizing parameter based on a distribution of the first reflectance values. The second reflectance value calculating unit calculates a second reflectance value for each original pixel. The normalizing unit normalizes the second reflectance values based on the normalizing parameter to obtain normalized Retinex values. The correcting unit corrects the original pixel values based on the normalized Retinex values.

Abstract: An image correction apparatus includes a gradient detection unit, a blur width detection unit, a correction area identification unit, and a correction unit. The gradient detection unit detects a gradient of a pixel value of each of pixels of an input image. The blur width detection unit detects a blur width at a position of each of the pixels of the input image. The correction area identification unit identifies a correction area based on the gradient detected by the gradient detection unit and the blur width detected by the blur width detection unit. The correction unit corrects a pixel value of a pixel in the correction area identified by the correction area identification unit.

Abstract: The noise reduction filter is intensity adaptive and incorporates directional variances of input images.

Abstract: Methods and systems for spatially aware sub-pixel rendering are described herein. Embodiments allow graphics rendering systems to convey additional spatial information in displayed graphics by utilizing sub-pixels of displays. An embodiment includes sampling pixels into a buffer, re-sampling the pixels, at sub-pixel level, at different offset positions from their positions in the buffer and displaying the re-sampled pixels. For example, individual red, green and blue sub-pixels can be re-sampled at offset positions from their original positions within pixels. Since embodiments of the invention can control respective positions of each red, green and blue sub-pixel, embodiments can produce sharp graphics with low aliasing. Furthermore, sub-pixels are effectively used for communicating additional spatial information in rendered graphics.

Abstract: A digital image signal is decomposed into a multi-scale representation comprising detail images and approximation images. Translation difference images are computed by subtracting an approximation image at scale s and a translated version of that approximation image. The values of the translation difference images are non-linearly modified. An amplification image is computed at least one scale as the ratio of a first image being computed by combining the modified translation difference images at the same or smaller scale and a second image created by combining unenhanced translation difference images at the same or smaller scale. Next, an enhanced multi-scale detail representation is computed by modifying at least one scale the detail image according to the amplification image at that scale. An enhanced image representation is computed by applying a reconstruction algorithm to the enhanced multi-scale detail representation.

Abstract: An image processor is provided with the high-frequency component detecting unit that calculates a convolution for each of regions of interest in part of an image region of input image data, to output the high-frequency component detection result for each pixel of interest; the smoothing unit that smoothes the image data, to output the smoothed image data; and the image data processing unit that combines based on the high-frequency component detection result the image data with the smoothed image data, to output the combined image data; wherein the image data processing unit varies the combination rate for the smoothed image data, based on the high-frequency component detection result. Thereby, even when a region in the image data contains such high frequency components as a checkered pattern has, an image resistant to moiré that would be generated in magnifying or reducing of the image on a region basis can be displayed without deterioration of image quality.

Abstract: A deblock method for processing at least one image including a plurality of pixels. The method comprises the step of determining a shift amount of block boundary by pixel data differences between adjacent pixels and thereby determining block boundaries between macroblocks in the image, and the step of optionally performing a deblock process on the block boundaries.

Abstract: Adjacent blocks are identified in an image. Coding parameters for the adjacent blocks are identified. Deblock filtering between the identified adjacent blocks is skipped if the coding parameters for the identified adjacent blocks are similar and not skipped if the coding parameters for the identified adjacent blocks are substantially different.