Abstract: A signal processing device first decomposes an input signal into a plurality of frequency components in different frequency ranges, then causes a higher harmonic wave generating section to generate a higher harmonic for each of a part or all of frequency components obtained by removing a frequency component in the lowest frequency range from frequency components obtained by the decomposition, and finally composes (i) the higher harmonic thus generated and (ii) a frequency component for which no higher harmonic has been generated. The higher harmonic wave generating section adds the frequency component and a nonlinear process signal (i) in which positive and negative signs of a frequency component for which the higher harmonic is to be generated are retained and (ii) which broadly monotonically increases nonlinearly with respect to the frequency component when the frequency component is located at least in the vicinity of zero.

Abstract: An image processing method includes configuring a noise reduction filter for each of pixels in an image in accordance with a linear noise model of the image, based on different levels of a noise effect caused to a corresponding pixel, among the pixels, by other pixels, among the pixels and adjacent to the corresponding pixel. The method further includes performing noise reduction filtering on each of the pixels, using the noise reduction filter for each of the pixels, to obtain a noise reduced image.

Abstract: A recovery processing unit performs an image quality recovery processing of reducing a blur in an image using a filter to reduce a blur at an image height position for each image height on a line from the center position of the image to an end of the image.

Abstract: An adaptive motion estimation and deblurring technique for acquired digital images includes acquiring multiple digital images with a moving digital image acquisition device that includes an image sensor, including a relatively sharp, underexposed reference image and a blurred image. Anb initial approximate point spread function (PSF) is estimated corresponding to the moving of the device. A different DC offset point is determined and a second PSF is calculated based on the different DC offset point.

Abstract: A weighting coefficient sequence defining unit generates an array in which weighting coefficients are held at positions defined by the directional components of blur vectors, and adjusts the weighting coefficients to increase the sum of the frequency components of the array or reduce variations of the frequency components. A weighting unit multiplies respective captured images by corresponding weighting coefficients, generating L weighted captured images. The weighting unit then generates the synthesized image of the L weighted captured images. A corrected image generation unit performs deconvolution processing using the frequency component of the array in which the weighting coefficients are held at positions defined by the directional components of blur vectors, and the frequency component of the synthesized image. The corrected image generation unit performs an inverse frequency transform for the deconvolution processing result, generating an output image.

Abstract: A system including a processor for adjusting the dynamic range of an image including a plurality of pixels. The processor segments the pixels into blocks, and computes statistical values for each block based on intensity values of the pixels. The processor also adjusts the dynamic range of the image by controlling the intensity values of the pixels based on the statistical values.

Abstract: Methods and systems are provided to reduce noise in thermal images. In one example, a method includes receiving an image frame comprising a plurality of pixels arranged in a plurality of rows and columns. The pixels comprise thermal image data associated with a scene and noise introduced by an infrared imaging device. The image frame may be processed to determine a plurality of column correction terms, each associated with a corresponding one of the columns and determined based on relative relationships between the pixels of the corresponding column and the pixels of a neighborhood of columns. In another example, the image frame may be processed to determine a plurality of non-uniformity correction terms, each associated with a corresponding one of the pixels and determined based on relative relationships between the corresponding one of the pixels and associated neighborhood pixels within a selected distance.

Abstract: One or more implementations access a digital image containing one or more bands. Adjacent bands of the one or more bands have a difference in color resulting in a contour between the adjacent bands. The one or more implementations apply an algorithm to at least a portion of the digital image for reducing visibility of a contour. The algorithm is based on a value representing the fraction of pixels in a region of the digital image having a particular color value.

Abstract: A block noise reduction system and method which may determine a feature value of a local texture at each location of input pixel, may selectively determine 1D filtering mode or 2D filtering mode as a filtering mode based on the determined feature value, and may perform filtering to preserve an edge according to the determined filtering mode.

Abstract: The invention is related to decoding of block wise coded video pictures. The determination of using de-blocking filtering between coded blocks is based on alternative characteristics compared to the characteristics used in H.264/AVC.

Abstract: A method according to the invention enhances contrast of an image by accessing an original image, and establishing, assigning, or initializing pixel values and pixel coordinates of pixels of the image. Then, a Laplacian pyramid having an integral number of levels is generated for the image. A contrast boost pyramid is generated and applied to the Laplacian pyramid to obtain a modified Laplacian pyramid wherein values in the levels of the Laplacian pyramid are enhanced. An enhanced image is then constructed from the modified Laplacian pyramid.

Abstract: A correlation judgment part judges a correlation direction on each pixel. In a case where a correlation direction of a specified pixel is a vertical direction and the correlation thereof is small in any other direction, it is judged that the specified pixel is a pixel on an edge in the vertical direction. Then, a noise removal filtering operation is performed on the specified pixel by using pixels on a line in the vertical direction and an edge enhancement operation is performed by using pixels on a line in a horizontal direction.

Abstract: A method for calculating an image quality metric for evaluating the quality of a digital image including the steps of denoising the data of the image, identifying edges in the denoised data, determining an edge profile of the edges, determining a grayscale angle for each identified edge in the edge profile that is associated with the edge, and calculating the image quality metric based on a weighted average of the grayscale angles for all the edges.

Abstract: A method for removing color noise on a slowly varying component contained in color difference component image data of image data which is imported from an image sensor and converted to brightness and the color difference component image data, includes the steps of: sampling pixels of said color difference component image data by thinning out according to a first defined sampling format when not performing a color noise removal process on the slowly varying component; determining if the color noise removal process is necessary to be performed or not; producing the color difference component image data, corresponding to a compressed image data size smaller than an image data size without said color noise removal process, by thinning out according to a second defined sampling format when performing said color noise removal process; and recording the color difference and brightness component image data.

Abstract: Images provide rich information regarding what they depict. For example, an image may have additional information, such as depth and/or 3D location values, for some points within the image. It may be advantageous to extrapolate the values from the valued points to the entire image because a new view of the image may be generated based upon values of points. Accordingly, an interpolated image may be generated by interpolating values for unvalued points based upon values of valued points. In particular, a set of valued points having desired cost paths may be determined for an unvalued point. A model may be applied to the set of valued points to interpolate a value for the unvalued point. One or more interpolated images may be projected onto a new view. In particular, points within an interpolated image may be projected onto locations within the new view based upon values of the points.

Abstract: In a methods and apparatus for generating an image for display from medical image data of a subject, image data is processed to reconstruct a pre-image data set, and a filter applied to the pre-image data set to produce a filtered image for display, while a value of a variable is obtained from the pre-image data set, for display with the filtered image. The value obtained from the pre-image data can be used for quantification of a feature of the medical image data.

Abstract: A method and apparatus for processing an image for enhancing an image quality captured in a low illumination environment is disclosed. The method for processing the image may include estimating motion information based on a base frame among input frames captured using a short exposure time and high ISO sensitivity conditions, removing noise of the base frame using the motion information, and enhancing an image quality of the base frame from which the noise has been removed using a reference frame captured under a long exposure condition.

Abstract: Signal processing techniques are applied to digital image data to remove the distortion caused by motion of the camera, or the movement of the subject being photographed, or defective optics, or optical distortion from other sources. When the image is captured, the effect of relative motion between the camera and the subject is that it transforms the true image into a blurred image according to a 2-dimensional transfer function. The 2-dimensional transfer function representing the motion is derived using blind estimation techniques or by using information from sensors that detect the motion. The transfer function is inverted and used to define a corrective filter. The filter is applied to the image and the blur due to the motion is removed, restoring the correct image. Another embodiment uses the transfer function to avoid blur by combining multiple consecutive images taken at a fast shutter speed.

Abstract: A target digital image is received from an image sensor. The image is contaminated by noise of unknown magnitude that is represented by a reference digital image. A process is applied that uses statistical analysis of the target digital image and of the reference digital image to estimate a magnitude of the noise for at least some pixels of the target digital image.

Abstract: A low frequency component image L[i] and high frequency component-emphasized image SH[i] are generated from an image A[i]. Lossy compression processing is performed for the low frequency component image L[i] to generate a compressed image C[i] and store it in a memory. A compressed image C[i?1] is decoded to generate a decoded image L?[i?1]. The compressed image C[i] is decoded to generate a decoded image L?[i]. A difference image E[i] between the decoded image L?[i] and the low frequency component image L[i] is generated. The low frequency component image L[i], decoded image L?[i?1], and difference image E[i] are composited at a predetermined ratio to generate a composite image SL[i]. The high frequency component-emphasized image SH[i] and composite image SL[i] are output as subframe images of the i-th frame.

Abstract: A method for filtering a digital image, comprising segmenting the digital image into a plurality of tiles; computing tile histograms corresponding to each of the plurality of tiles; deriving a plurality of tile transfer functions from the tile histograms preferably using 1D convolutions; interpolating a tile transfer function from the plurality of tile transfer functions; and filtering the digital image with the interpolated tile transfer function. Many filters otherwise difficult to conceive or to implement are possible with this method, including an edge-preserving smoothing filter, HDR tone mapping, edge invariant gradient or entropy detection, image upsampling, and mapping coarse data to fine data.

Abstract: One embodiment of the present invention sets forth a technique for dynamically specifying a texture header and texture sampler using an index. The index corresponds to a particular register value that may be static or computed during execution of a shader program. Any texture operation instruction may specify an index value for each of the texture header and the texture sampler.

Abstract: An inspection apparatus specifies a region for displaying an advertisement from an image of a virtual space. The inspection apparatus comprises: an image mapping unit that maps an image, which uses a color that is not used in the virtual space and where the display status of the advertisement varies in accordance with a viewpoint of a user, on a region where the advertisement is to be displayed in the virtual space; an inspection image acquiring unit that acquires a virtual space that is actually displayed on a screen on a frame-by-frame basis; and an advertisement region extracting unit operative to extract, by color-filtering the acquired image of the virtual space, a region that is rendered with a color that is not used in the virtual space, as a region for displaying the advertisement perceived by a user.

Abstract: There is provided an image processing system and an image processing method able to suppress block distortion in the case of decoding image data encoded in unit of blocks. A controlling unit selects a filtering content to be applied to the block image data based on the encoding types of the block image data to be filtered, and a filtering unit applies filtering to the block image data to be processed according to the filtering content selected by the controlling unit.

Abstract: There is provided an image processing device including an acquisition portion that acquires image data of an image, a dividing portion that divides the acquired image into a number of blocks N (N>1), a specification portion that sequentially specifies, each time the image data of the image is newly acquired, a number of the blocks M (N?M>1) from among the number of the blocks N, as the blocks to be updated, a filtering portion that performs filtering using a predetermined filter on the image data of the specified number of the blocks M, a counting portion that counts a number of pixels for which a filtering result is larger than a predetermined value, a first determination portion that determines whether there is an abnormality in the blocks, and a second determination portion that determines whether sabotage has occurred.

Abstract: An alpha matte is generated from image forming elements of an image. For each of one or more of the image forming elements: a respective representative foreground value is determined from one or more of the image forming element values; the respective representative foreground value and the value of the image forming element are normalized with respect to a threshold level; and a respective value of the alpha matte is generated from an evaluation of the normalized image forming element value in relation to the normalized representative foreground value.

Abstract: There is provided an image processing device which transforms an image with a bit depth of N into an upper layer with a bit depth of M and a remaining lower layer, including a histogram unit that generates histogram indicating occurrence frequency of each pixel value of an image with a bit depth of N, a table unit that generates a table listing pixel values of which occurrence frequency in the histogram generated by the histogram unit is equal to or more than one, a reordering unit that reorders an arrangement of values in the histogram using the table generated by the table unit, an update unit that updates the table generated by the table unit and the histogram reordered by the reordering unit; and an index image unit that generates an index image with a bit depth of N using the updated table and the updated histogram.

Abstract: An image processing method includes: receiving a plurality of input images; deriving an image alignment related information from performing an image alignment upon the input images; and generating a processed image by performing a defocus operation upon a selected image selected from the input images according to the image alignment related information. For example, the image processing method may be employed by an electronic device such as a mobile device. Thus, the mobile device may capture two or more images to generate the defocus visual effect, which is similar to professional long-focus lens.

Abstract: An image processing apparatus including an image acquiring unit to acquire a first image including color information and a second image having a spectral band wider than that of the first image and a noise reduction unit to extract edge information from the second image and to reduce noise of the first image based on the extracted edge information.

Abstract: The present invention discloses a method for decomposing a target pattern containing features to be printed on a wafer, into multiple patterns, the features having a plurality of patterns within a minimum pitch for processes utilized to image the target pattern. The method includes superposing a predefined kernel over a pixel, and moving the kernel from one pixel to another, the pixels representing the sub-patterns of the target pattern. Polarity of the kernel may be reversed when the pixel has a stored intensity value that is negative.

Abstract: A method includes generating a first principle bilateral filtered image component from a source image. The first principle bilateral filtered image component corresponds to a second pixel value of a set, the second pixel value greater than or equal to a first pixel value. The method includes selectively updating a result pixel of a result image based on the first principle bilateral filtered image component and deallocating the first principle bilateral filtered image component. After deallocating the first principle bilateral filtered image component, a second principle bilateral filtered image component is generated from the source image. The second principle bilateral filtered image component corresponds to a third pixel value. The third pixel value is greater than the second pixel value. The third pixel value is less than or equal to a fourth pixel value. The result pixel is selectively updated based on the second principle bilateral filtered image component.

Abstract: An image converting device includes; a downscaling unit which downscales a two-dimensional image to generate at least one downscaling image, a feature map generating unit which extracts feature information from the downscaling image to generate a feature map, a visual attention calculating unit which generates a low-level attention map based on a visual attention of the feature map, and an image expansion unit which up-scales the low-level attention map, wherein an increasing gray value may be added to image data disposed on the upper portion in the low-level attention map, and 0 or a decreasing gray value may be added to the image data disposed on the lower portion.

Abstract: In order to prevent illegal copying more effectively in an image forming apparatus, when a specified ground pattern for preventing copying is detected in input image data, the detected ground pattern is emphasized in the image data. Alternatively, information for detecting the specified ground pattern is stored in a storage device, in order to detect the ground pattern with the stored information. When the ground pattern is not detected, a ground region in the image data is analyzed further to determine whether a different ground pattern exits or not. When a ground pattern is detected, information for detecting the ground pattern is stored in the storing device.

Abstract: Image completion using scene geometry is described, for example, to remove marks from digital photographs or complete regions which are blank due to editing. In an embodiment an image depicting, from a viewpoint, a scene of textured objects has regions to be completed. In an example, geometry of the scene is estimated from a depth map and the geometry used to warp the image so that at least some surfaces depicted in the image are fronto-parallel to the viewpoint. An image completion process is guided using distortion applied during the warping. For example, patches used to fill the regions are selected on the basis of distortion introduced by the warping. In examples where the scene comprises regions having only planar surfaces the warping process comprises rotating the image. Where the scene comprises non-planar surfaces, geodesic distances between image elements may be scaled to flatten the non-planar surfaces.

Abstract: There is provided an apparatus for improving the sharpness of an image, which may prevent occurrence of distortion of an image in an edge enhancement process of the image by applying an active weight in accordance with a two-dimensional (2D) high pass filtering value of the image. The apparatus may include: a 2D high pass filter outputting a high frequency element value for the luminance values of pixels of an input image; a weight generating unit generating a weight changed depending on a magnitude of the high frequency element value; a weight applying unit applying the weight to the high frequency element value; and an edge enhancement image generating unit adding, to the luminance values of the pixels of the input image, the high frequency element value to which the weight is applied to thereby output an image of which an edge is enhanced.

Abstract: A signal processing apparatus including: a first noise reduction processing section that performs first noise reduction processing on an image, in which each of multitudes of pixels has one of a plurality of color components and the color components are distributed regularly, based only on pixel arrangement to obtain a first processed image; a color component separation section that separates the first processed image into each of the color components to obtain a plurality of color component images; and a signal classification section that compares a signal value of a target pixel for processing with a signal value of each pixel included in a predetermined range of area around the target pixel and classifies each pixel within the predetermined range of area into one of a plurality of groups based on the comparison result.

Abstract: An edge detection apparatus includes a computing circuit and a determining circuit. The computing circuit includes a first multiplier block and a first adder unit. The first multiplier block includes n×m first multiplier units, wherein each first multiplier unit has a first multiplication factor. The n×m first multiplier units respectively perform multiplications on n×m pixels which are arranged as an n×m matrix to generate n×m first product values based on the corresponding first multiplication factors. The n×m pixels include a target pixel, where n is not equal to m. The first adder unit generates a first computation result according to the n×m first product values. The determining circuit determines if the target pixel is an edge pixel according to at least the first computation result.

Abstract: X-ray beam detectors in one model can individually differ from one another. This can lead to differences in the amount of noise in an image recorded with the aid of the respective X-ray beam detector. In the present case, a variable is derived using an empty image, which variable reproduces the amount of noise, and this variable then determines the type and extent of a filtering process. Hence the image processing is adapted to the respective individual noise behavior of the respective X-ray beam detector. This is particularly suitable if the X-ray beam detector is a flat-panel detector (100) with a scintillator (22) and photodetector elements (12).

Abstract: A method includes detecting one of an application access or a file type access, and configuring, in response to detecting the application or file type access, automatically without user interaction, a display system in an image quality configuration for the application or the file type where the image quality configuration is based on providing best image quality with respect to the application or the file type. Configuring the display system in an image quality configuration, may involve determining that a profile associated with the application or associated with the file type is stored in memory, and configuring the display system according to the profile. The method may adjust at least one anti-aliasing parameter or at least one anisotropic filter parameter. The method may monitor an operating system to obtain an indication that an application has been accessed or that a file type has been accessed.

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: A system (12) for providing an adjusted image (228) of a scene (10) from a noisy captured image (14) includes a control system (26) that provides the adjusted image (228). The control system (26) can create a de-noised image (238) from the captured image (14), determine a details layer (236) from the captured image (14), and combine information from the details layer (236) with the de-noised image (238) to provide the adjusted image (228).

Abstract: Aspects of the present invention are related to systems and methods for image enhancement, wherein an input image may be separated into two images: a first image comprising image content, from the input image, which may be sharpened; and a second image comprising image content, from the input image, which may be attenuated. One aspect of the present invention relates to determining a noise estimate associated with the input image and using the noise estimate to control the separation of the input image into the two images. Another aspect of the present invention relates to controlling the image separation based on the level of sharpening being applied to the first image.

Abstract: Method for detecting edge of fixed pattern includes receiving and analyzing a first image to obtain a first edge information. Second image and a corresponding second edge information are received, in which the second image includes an accumulation of image history information. According to the first edge information and the second edge information, a consistent number and an inconsistent number for pairs of pixels at the corresponding location of the first image and the second image are calculated, in which the consistent number represents how many pairs of pixels of which two compared pixels of each pair are both edge pixels, and the inconsistent number represents one of the two compared pixels is not the edge pixel. When the consistent number is greater than first predetermined value and meanwhile the inconsistent number is less than second predetermined value, first image and second image have a fixed pattern with fixed edge.

Abstract: Temporally filtering raw image data may include a temporal filter that determines a spatial location of a current pixel and identifies at least one collocated reference pixel from a previous frame. A motion delta value is determined based upon the current pixel and its collocated reference pixel. An index is determined based upon the motion delta value and a motion history value corresponding to the current pixel from the previous frame. Using the index, a first filtering coefficient may be selected from a motion table. Then an attenuation factor may be selected from a luma table based upon the value of the current pixel, and a second filtering coefficient may be determined based upon the selected attenuation factor and the first filtering coefficient. The temporally filtered output value corresponding to the current pixel may then be based upon the second filtering coefficient, the current pixel, and the collocated reference pixel.

Abstract: An image denoising method according to the present invention includes the steps of: sequentially selecting a pixel in an image as a current pixel; dynamically determining a current search block and a strength parameter; transferring the comparison block of each pixel in the current search block to a frequency domain; determining a current frequency basis; obtaining a similarity between each neighborhood pixel and the current pixel in the current search block according to the current frequency basis; determining a weighting of each neighborhood pixel related to the current pixel according to the strength parameter, and a distance and the current pixel in the current search block; and weighted averaging each neighborhood pixel and the current pixel in the current search block according to the weighting so as to obtain a reconstruction value of the current pixel.

Abstract: Methods, machines, and computer-readable media for processing an input image with a bi-selective filter in a transform domain are described. In one aspect, a forward transform of an input image is computed. A bi-selective filter smoothes image regions with low magnitude coefficients and sharpens image regions with high magnitude coefficients. An output image is generated from the filtered forward transform.

Abstract: One embodiment of the present invention provides a system for applying a bilateral filter to an image. During operation, the system selects a first region within the image which is associated with a first pixel. Next, the system constructs a first histogram using pixel values within the first region. The system then computes a new value for the first pixel using the current value of the first pixel and the first histogram. The system then selects a second region within the image which is associated with a second pixel. Next, the system determines a non-overlapping region between the first region and the second region. The system then constructs a second histogram using the first histogram and pixel values in the non-overlapping region. Next, the system computes a new value for the second pixel using the current value of the second pixel and the second histogram.

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 system and method are provided for determining a color palette model from an image of a document. Pixel values of the image of the document are clustered to provide image clusters. Color layers of the image are determined, each color layer corresponding to an image cluster. Aspects of the color palette model can be determined using the color layers. Aspects of the color palette model include a foreground-background color pair for a content block in the document and a background-area color of the document.

Abstract: There is provided an edge detecting method, which is capable of preventing a noise influence caused by imaging device and a color interpolation. The edge detecting method includes the steps of: setting a first kernel based on a center pixel in pixel data arranged in a mosaic structure; setting a second kernel based on the center pixel within the first kernel; detecting whether a pixel having a green value in the second kernel is a defective pixel, and correcting the pixel; converting all pixels of the second kernel into pixels having green value; calculating a slope value by using a mask for detecting an edge in the second kernel; and detecting an edge by adding the slope value to a luminance value obtained by a color space conversion.