Abstract: Systems, methods, and techniques are provided for performing any one or more of edge-preserving image sharpening, edge-preserving image smoothing, edge-preserving image dynamic range compression, and edge-aware data interpolation on digital images, wherein a pixel prediction module is adapted for coupling to a memory storing pixel data representative of a digital image and extracts from the image predicted pixel values using robust smoothing. The predicted pixels are stored in a memory and respective detail values equal to the difference between respective original and predicted values are computed. A pixel update module computes approximation values by averaging the respective detail values with original pixel values using robust smoothing, and stores the approximation values for subsequent rendering.

Abstract: A method for reducing noise in a color image captured using a digital image sensor having pixels being arranged in a rectangular minimal repeating unit. The method comprises, for a first color channel, determining noise reduced-pixel values using a first noise reducing process that includes computing weighted pixel differences by combining the pixel differences with corresponding local edge-responsive weighting values. The method further comprises a second noise reducing process that includes computing weighted chroma differences by combining chroma differences with corresponding local edge-responsive weighting values.

Abstract: An information processing apparatus includes: a synthesis filter unit configured to execute synthesis filtering for coefficient data of each subband, obtained by decoding a code stream of a reversibly-encoded image and performing band division, in each predetermined number of lines or in each predetermined block to obtain baseband image data on a line basis; a storage unit configured to store the baseband image data obtained by the synthesis filter unit; an analysis filter unit configured to read the baseband image data stored in the storage unit and execute analysis filtering for the image data to divide the image data up to a predetermined division level; and a control unit configured to control the analysis filter unit to execute analysis filtering as soon as the amount of the baseband image data stored in the storage unit reaches a data amount allowing analysis filtering.

Abstract: A video scaler is disclosed. A polyphase filter can be used to generate interpolated pixels. The values of pixels adjacent an interpolated pixel are examined to determine variation in values among the adjacent pixels to determine minimum and maximum value variations. The value of the interpolated pixel is limited based on the minimum and maximum value variations. Ringing artifacts can be reduced by limiting the color range of an interpolated pixel.

Abstract: An image processing device to process a moving image, which is shot by a video shooting device, in increments of access units, includes: a correcting unit to correct the access unit to be processed by changing the properties of a low-pass filter which indicates imaging blur according to parameter values showing the properties of imaging blur which occurs at the time that the moving image is shot with the video shooting device, generating an inverse filter having inverse properties as to the low-pass filter, and performing processing to apply the inverse filters as to the access unit to be processed; wherein, with a frequency of which the frequency property of the low-pass filter indicating the imaging blur is zero being zero points, performing processing to apply the inverse filter is forbidden for predetermined frequency components of the frequency components of the access unit to be processed including the zero points.

Abstract: System, method, and computer program product to adaptively blend the interpolation results from an 8-tap Lanczos filter and the interpolation results from a bilinear filter, according to the local transitions of the input content. Artifacts may occur, which may be identified as such and corrected. Pixels that represent artifacts in the blended image may be replaced with the pixel for that location taken from the bilinear interpolation.

Abstract: A computer-implemented method for generating depth-of-field effects includes receiving information specifying an input image rendered using a pinhole camera. A first set of circles of confusion associated with portions of the input image are determined based a set of parameters associated with the pinhole camera. The first set of circles of confusion are then associated with heat conductivity of a medium. The portions of the input image are blurred according to heat diffusion in the medium to generate a first output image having a first depth-of-field effect.

Abstract: An image noise reduction method is provided. An image is received. A first-stage process is performed to the image to obtain a luminance information Y and a color information Cb and/or Cr corresponding to a pixel array in an YCbCr domain. A second-stage process is performed to the luminance information Y to reduce at least a luminance noise. A third-stage process is performed to the color information Cb and/or Cr to reduce at least a color noise. The luminance information Y and the color information Cb and/or Cr are then combined.

Abstract: Video processing method and means for enhancing a video stream, by computing transform coefficients using a spatio-temporal transform comprising a spatial subband transform and a causal time wavelet transform performing filterings with multiscale causal wavelets, modifying the transform coefficients with a nonlinear processing, and computing a processed video stream from the modified transform coefficients using a spatio-temporal reconstruction transform comprising an inverse subband transform and a short delay inverse time wavelet transform, where the short delay inverse time wavelet transform is implemented with wavelet filters modified with window functions to control the processing delay of the entire video processing method.

Abstract: A degraded-image restoration method includes inputting a degraded image degraded by a predetermined degradation process; reading from a storage unit a filter coefficient in a filtering process of restoring the degraded image, in which an iteration formula for minimizing a square error between the degraded image and an image obtained by degrading a restored image by the degradation process is expanded corresponding to a predetermined number of iterations; and performing the filtering process using the filter coefficient with respect to the degraded image to obtain a restored image of the degraded image. Accordingly, a throughput in an apparatus that restores the degraded image can be reduced, thereby obtaining a preferable restored image.

Abstract: A tone adjustment method for a digital image and an electronic apparatus using the same are presented. A digital image is loaded. A detail image layer and a primary image layer are generated according to the digital image. A tone mapping procedure is performed on the primary image layer, for generating a tone mapping layer. A detail gain lookup table is loaded and then a corresponding gain is looked up according to each pixel value in the primary image layer, for generating a detail gain adjustment layer. A detail gain control procedure is performed and then a detail gain adjustment is performed on the detail gain adjustment layer and the detail image layer, for generating a gain correction layer. The gain correction layer and the tone mapping layer are combined, and then a combined layer is output, thereby completing the tone adjustment for the digital image.

Abstract: A first sigma filtering circuit sigma filters an image to produce a filtered image. An analysis circuit processes the sigma filtered image to produce an approximation part and a detail part. A second sigma filter circuit filters the approximation part to produce a sigma filtered approximation part. Another analysis circuit process the sigma filtered approximation part to produce a second approximation part and a second detail part. A third sigma filter circuit sigma filters the second approximation part to produce a sigma filtered second filtered approximation part. A first synthesizer synthesizes the sigma filtered second filtered approximation part and the second detailed part to produce a first reconstructed image, and a second synthesizer synthesizes the first reconstructed image and the first detail part to produce a final filtered image.

Abstract: Tomographically reconstructing a 3D image object corresponding to a data set includes for each step in a series of iteration steps, determining an updated object by performing a combined operation, which includes performing an update operation for updating an input object and performing a pixon smoothing operation, and following a last iteration, outputting one of the updated objects as the 3D image object.

Abstract: A method and apparatus is provided for adjusting the contrast of an input image. A blurred image corresponding to the input image is obtained, the brightness level of at least some of the pixels in the input image being varied to provide the blurred image. A ring likelihood for pixels in the input image is obtained, the ring likelihood providing a measure of the likelihood that said pixels are non-edge pixels in the neighborhood of an edge pixel. The output image is produced as a sum of the brightness levels in the input image and the blurred image in dependence on the ring likelihoods.

Abstract: Recursive filtering that multiplies image data of a previous frame read out from a memory by a, multiplies image data of the present frame by (1?a), adds the resultants together, and stores the resultant of the addition in the memory is performed. Here, a is a coefficient in the range of 0<a<1. Then, spatial filtering is performed on the recursively filtered image data using a spatial filter changed in accordance with a coefficient of the recursive filtering and the number of times the recursive filtering has been performed.

Abstract: Tomographically reconstructing a 3D image object corresponding to a data set includes reconstructing a first reconstructed object from the data set, receiving a smoothing map, smoothing the first reconstructed object based on the smoothing map thereby creating a first smoothed object, and outputting the first smoothed object as the 3D image object. Smoothing a first object thereby creating a smoothed object having a smoothed value associated to each object point in object space includes receiving the first object, determining, in a series of steps, single-kernel-smoothed objects, wherein each iteration step is associated with a kernel function and includes, determining a start object based on the first object, and smoothing the start object using the kernel function of the iteration step, thereby creating the single-kernel-smoothed object having single-kernel-smoothed values associated to each object point, and constructing the smoothed object from the single-kernel-smoothed values.

Abstract: In an aspect, tomographically reconstructing a 3D image object corresponding to a data set includes for each step in a series of iteration steps, determining an updated object by performing a combined operation, which includes performing an update operation for updating an input object and performing a smoothing operation, and following a last iteration, outputting one of the updated objects as the 3D image object.

Abstract: Recursive filtering that multiplies image data of a previous frame read out from a memory by a, multiplies image data of the present frame by (1?a), adds the resultants together, and stores the resultant of the addition in the memory is performed. Here, a is a coefficient in the range of 0<a<1. Then, spatial filtering is performed on the recursively filtered image data using a spatial filter changed in accordance with a coefficient of the recursive filtering and the number of times the recursive filtering has been performed.

Abstract: The invention distinguishes not only character areas and halftone dot areas in an image, but also makes highly accurate decisions concerning characters present in halftone dot areas. A decision signal generating unit generates a decision signal having luminance as a main component from a signal indicating inputted image data, and supplies the decision signal to a character decision unit, a halftone dot decision unit, and a character-in-halftone dot decision unit. The character decision unit generates data indicating whether a pixel of interest is inside a character image area. The halftone dot decision unit generates data indicating whether a pixel of interest is inside a halftone dot area. The character-in-halftone dot decision unit generates data indicating whether there is a character image inside the halftone dot area. Based on these three signals, an attribute flag generating unit generates attribute data of the pixel of interest.

Abstract: A method and system for retouching digital images for a motion picture removes semi-transparent artifacts or ‘blotches’ caused by contaminates in the optical path of the camera. This approach provides the benefit of only having to retouch a single average image that is than automatically applied via a correction power map to the entire sequence of images for the affected scene. The formation of an average image tends to reinforce the artifacts making them easier to identify and reduce background detail making it easier to retouch the artifact.

Abstract: An information processing apparatus comprises: a spatial filtering unit configured to perform spatial filtering in a frequency range based on a spatial frequency of an object for image data of a current frame; and a recursive filtering unit configured to perform recursive filtering by obtaining image data, which has been processed prior to the current frame, from a memory, multiplying the obtained image data by a coefficient ? (?<1), adding the image data multiplied by the coefficient ? to the image data of the current frame after the spatial filtering, and storing the image data after the addition in the memory.

Abstract: Apparatus and method of optically detecting click events are described. Images of a contact surface are captured at respective capture times. The captured images are convolved with a two-dimensional circularly symmetric spatial bandpass filter to produce corresponding filtered images each including a set of pixels with respective pixel value magnitudes. Based on the pixel value magnitudes of the corresponding filtered images, each of the capture times is assigned to one of an in-contact time class during which the contact surface is determined to be in-contact with a user's finger and an out-of-contact time class during which the contact surface is determined to be out-of-contact with the user's finger. A select signal indicating that the contact surface has been touched to make a selection is generated based on a click event predicate defining at least one condition on the time classes respectively assigned to successive ones of the capture times.

Abstract: A device for eliminating a block artifact phenomenon of a decoded image frame, and an eliminating method thereof. A method for eliminating a block artifact phenomenon according to the present invention includes: extracting pixel information including a magnitude of edge response, an edge direction, and an edge position, by applying an edge detector to an arbitrary pixel in an image frame; distinguishing the pixel into a non-edge pixel, an object edge pixel, or a block edge pixel by using the extracted pixel information; and performing filtering according to the distinguishing result. In this manner, an object edge information is selectively protected, so that an improved image quality may be provided.

Abstract: An image processing apparatus includes a feedback unit configured to perform first nonlinear conversion of a signal obtained by subtracting a value of a subtraction image signal from that of a current-frame image signal, and to obtain the next subtraction signal by subtracting a signal obtained by performing the first nonlinearly conversion from the current-frame image signal, and a noise reduction unit configured to perform second nonlinear conversion of a signal obtained by subtracting a value of the subtraction image signal from that of the current-frame image signal, and to obtain an output image signal by subtracting a signal obtained by performing the second nonlinear from the current-frame image signal.

Abstract: An imaging acquisition system that generates a depth map from two pictures of a three dimensional spatial scene is described. According to one aspect of the invention, the system generates the depth map based on the relative blur between the two pictures and the absolute blur contributed by the system. According to another aspect of the invention, the system calculates the depth map directly from the relative blur between the two pictures.

Abstract: A recursive filter system and method is provided. A weighted value of luminance and chrominance of a stored picture element is generated by summing weighted values of luminance and chrominance of a picture element and neighboring picture elements. A first proportional parameter for the stored picture element is generated with a sum of differences between the luminance and chrominance of the stored picture element and surrounding picture elements. A weighted picture element is generated from a proportion of the stored picture element and a complimentary proportion of the weighted value. Picture elements of the next image and a second proportional parameter are inputted. A filtered image is generated by combining a proportion of each input picture element of the next image with a complementary proportion of the weighted stored picture element of the first image. The filtered image is stored for corresponding combination with a succeeding image.

Abstract: A computer-implemented method for generating depth-of-field effects includes receiving information specifying an input image rendered using a pinhole camera. A first set of circles of confusion associated with portions of the input image are determined based a set of parameters associated with the pinhole camera. The first set of circles of confusion are then associated with heat conductivity of a medium. The portions of the input image are blurred according to heat diffusion in the medium to generate a first output image having a first depth-of-field effect.

Abstract: Disclosed herein is an improved blur computation algorithm. The proposed algorithm accomplishes a blur of an image using fragment programs on a GPU. Alternatively, the blur may be computed on a CPU through emulation or directly programmed. Modifications of the program are possible that accomplish motion blur, zoom blur, radial blur, and various other forms of blur that vary across an image computed for the purpose of simulating depth-of-field.

Abstract: The present invention relates to a method for processing digital image data (ID), wherein the process of multi rate, multi resolution and/or multi scale signal processing is involved in order to realize a respective multi rate, multi resolution and/or multi scale sharpness enhancement with respect to said image data (ID).

Abstract: The invention relates to an iterative method for determining an interpolated image information value in an image that comprises a number of image regions arranged in a matrix-like fashion having original image regions to which one image information value has been assigned, and having at least one image region to be interpolated. In order to determine an interpolated image information value, the method specifies a starting interpolation direction and determines a quality measure for the starting interpolation direction. At least one image direction is selected different from the starting interpolation direction and determining a quality measure for this at least one image direction.

Abstract: A radiographic apparatus according to this invention, when carrying out recursive computation, pixel groups consisting of detection pixels respectively corresponding to positions on a radiation detection device are sorted into locations subjected to the recursive computation and locations exempted from the recursive computation. For the locations subjected to the recursive computation, lag-behind parts are removed by the recursive computation to obtain corrected radiation detection signals. The recursive computation is not carried out at least for the locations exempted from the recursive computation. The lag-behind parts can be removed from the radiation detection signals, with a calculation amount for the recursive computation reduced by an amount corresponding to the recursive computation excluded.

Abstract: Representations of an input signal at different lower resolutions are obtained, where n denotes a level of resolution, and where 0?n?N. Level n=N corresponds to the representation having the lowest resolution, and level n=0 corresponds to the input signal. The resolution increases as level n decreases. The nth level representation is used to estimate and remove low frequency error in the n?1th level representation. Then application-specific processing is performed on the n?1th level representation.

Abstract: A method of reducing noise levels in an input video signal comprising pixel data includes receiving input pixel data of the input video signal, where the input pixel data comprises luminance data and chrominance data, and estimating a noise level of the input video signal using the luminance data of the input pixel data. A plurality of filter parameters are identified based on the estimated noise level, and the input pixel data is filtered using a three dimensional spatiotemporal noise reduction filter that is controlled by a first set of the plurality of filter parameters.

Abstract: A lane marker recognition apparatus which recognizes a lane marker based on a captured image of a road surface in the direction in which a vehicle is traveling, includes recognizing means for recognizing at least one left lane marker and at least one right lane marker captured in the image, and generating lane marker information indicative of the recognized lane markers; calculating means for calculating, based on the lane marker information, at least one type of control value for each potential running lane demarcated; first selecting means for selecting a control value of one of the running lanes that is to be indicated as information; generating means for generating information indicative of the selected control value; and second selecting means for, after the information has been generated, selecting the control value of the potential running lane that is closest to the control value indicated by the information.

Abstract: A segmentation operation is applied to an input image to identify foreground objects of interest, and then a shadow removal operation is applied to remove any detected shadows from the foreground segmentation. The shadow removal algorithms can leave holes and bisections in the segmentation map, however, which will then subsequently impact on an object detection step performed using connected component analysis. To get around this problem, a conditional morphological dilation operation is applied to the segmentation map to ‘grow’ the segmented blobs to fill in any holes and bisections, without re-growing shadow pixels in the segmentation. The result is an object detection method and system which is robust to illumination changes causing shadows and/or highlights.

Abstract: Analysis filter processing is recursively repeated on horizontal and vertical low frequency component coefficients obtained as a result of the analysis filter processing, until a predetermined decomposition level is reached. Coefficients obtained during a process of computation in the analysis filter processing and, except for a preset decomposition level, horizontal and vertical low frequency component coefficients obtained as a result of the computation in the analysis filter processing are stored in a first storage section independently for each decomposition level. The coefficients stored in the first storage section are read as appropriate and the read coefficients are supplied for the analysis filter processing. The horizontal and vertical low frequency component coefficients of the preset decomposition level are stored in a second storage section. The coefficients stored in the second storage section are read as appropriate. The read coefficients are supplied for the analysis filter processing.

Abstract: A degraded-image restoration method includes inputting a degraded image degraded by a predetermined degradation process; reading from a storage unit a filter coefficient in a filtering process of restoring the degraded image, in which an iteration formula for minimizing a square error between the degraded image and an image obtained by degrading a restored image by the degradation process is expanded corresponding to a predetermined number of iterations; and performing the filtering process using the filter coefficient with respect to the degraded image to obtain a restored image of the degraded image. Accordingly, a throughput in an apparatus that restores the degraded image can be reduced, thereby obtaining a preferable restored image.

Abstract: Tomographically reconstructing a 3D image object corresponding to a data set includes for each step in a series of iteration steps, determining an updated object by performing a combined operation, which includes performing an update operation for updating an input object and performing a pixon smoothing operation, and following a last iteration, outputting one of the updated objects as the 3D image object.

Abstract: A method and apparatus for processing a line pattern using a convolution kernel. An apparatus for processing a line pattern, the apparatus including a light source module radiating predetermined light and a camera module capturing an image of an object, onto which the light is radiated, the apparatus processing a line pattern included in the captured image, includes: a kernel learning unit determining a kernel shape and an optimal threshold value with respect to each vertical interval of the image using a simple line pattern formed at a predetermined location; a convolution application unit applying convolution to a captured test image using the determined kernel shape and optimal threshold value; and a region selection unit scanning each pixel column in the image, to which the convolution is applied, in a vertical direction, and setting pixel values of pixel groups other than a selected pixel group to 0 when a plurality of pixel groups having pixel values other than 0 exists.

Abstract: An apparatus to measure noise in an image signal and a method thereof. The apparatus includes an area division unit to divide an image signal having noise therein into a plurality of areas and to output area information corresponding to the plurality of areas, a variance calculation unit to calculate variances of the plurality of areas using the area information, the image signal and a noise-attenuated image signal obtained by attenuating the noise from the image signal, and a noise determination unit to determine an amount of noise in the image signal by comparing the calculated variances of the plurality of areas. Accordingly, the noise measurement apparatus can measure the amount of noise irrespective of characteristics of the image signal, and thus noise attenuation can be performed more accurately.

Abstract: Disclosed herein is an improved blur computation algorithm. The proposed algorithm accomplishes a blur of an image using fragment programs on a GPU. Alternatively, the blur may be computed on a CPU through emulation or directly programmed. Modifications of the program are possible that accomplish motion blur, zoom blur, radial blur, and various other forms of blur that vary across an image computed for the purpose of simulating depth-of-field.

Abstract: Unnecessary flare correction at the boundary between a blanking area and video signal area is suppressed. An image quality improving apparatus includes 2-dimensional low-pass filter circuit 2 that extracts low-frequency components from an input video signal; subtractor circuit 4 that obtains high-frequency components by subtracting the low-frequency components extracted by 2-dimensional low-pass filter circuit 2 from the input video signal; and adder circuit 6 that adds the high-frequency components obtained by subtractor circuit 4 as a correction signal to the input video signal. The input video signal contains blanking signals without any video information. 2-dimensional low-pass filter circuit 2 starts extraction of the low-frequency components after a lapse of a predetermined unit time from when the input video signal changed from the blanking signal to the signal that contains the video information.

Abstract: A method and apparatus for processing image data are described. In one embodiment, a method of processing image data comprises decomposing the image data into multiple decomposition levels by applying a wavelet transform to the image data, and modifying coefficients in at least two of the decomposition levels by scaling coefficients in theses decomposition levels using different scale dependent parameters for each of the decomposition levels.

Abstract: An edge preservation type of recursive filter 12 outputs an unsharpened image (i.e., low-frequency components) which is smoothed while preserving an edge of an input image in at least one direction (e.g., a vertical direction of the image). The low-frequency components are compressed into a suitable amount by a LUT 14 provided as low-frequency component compression amount setting means and then output. An arithmetic unit 18 provided as low-frequency component compression means compresses the low-frequency components of the input image by subtracting from the input image the output value from the LUT 14. The edge-preservation-type recursive filter 12 can extract the low-frequency components of the input image with an input time delay corresponding to one line of the input image. Therefore, DRC processing has only a short delay due to image processing and is capable of real-time processing by a small amount of computation.

Abstract: Locations of reference blocks of pixels of decoded interlaced video reference fields decoded from a compressed stream of video data are computed using motion vectors decoded from the same stream of video data. The reference block computed locations are de-interlaced with current blocks from current video fields decoded from the compressed streams of video data by low pass filtering light intensities of adjacent pixels of the reference and current blocks using low pass filter with coefficients adjustable according to discrete cosine transform (DCT) high frequency coefficients decoded from the coded stream. In addition, moving edges or objects of the de-interlaced blocks of pixels can be identified by comparing the decoded motion vector of the reference blocks to a threshold value. Then proper or more appealing light intensities for the moving edge pixels can be interpolated.

Abstract: Regions within a scanned document are labeled according to content. These labeled regions allow for particular enhancement and sharpening methods to be applied. The labeled regions may be further leveraged in subsequent rendering of the scanned document on a printer or other output device and/or in storing the scanned document.

Abstract: A method and system for making information available of a computer system and compensating information owners or creators for access to said information. In various aspects, the invention provides a mechanism for giving users meaningful access to information via a computer system and network while protecting the interests of publishers and creators in information. The invention provides a solution for information including, but not limited to: text, graphics, photos, executable files, data tables, audio, video, and three dimensional data. In a further aspect, the invention comprises a new method for allowing a user to review a document while connected to a network but prevents the user from downloading, printing, or copying the document unless a fee is paid. In a further aspect, the invention comprises a new method for allowing a user to review documents at a first cost basis (which may be free), but only provides other access to documents, such as copying, printing, or downloading on a second cost basis.

Abstract: Determining a pixon map for pixon smoothing of an object based on a data set includes receiving the data set and an input object associated to the data set. Determining a pixon map further includes determining, in a series of steps, statistical objects for a set of pixon kernel functions, wherein each step includes selecting a pixon kernel function from the set of pixon kernel functions, smoothing the input object on the basis of the selected pixon kernel function, thereby creating a smoothed object, and determining the statistical object for the selected pixon kernel function on the basis of the smoothed object, the data set, and a Mighell-like statistical weight. Determining a pixon map further includes determining contributions of the pixon kernel functions to the pixon map based on the statistical objects and assigning values to the pixon map corresponding to the contributions of the pixon kernel functions.

Abstract: A vehicle restraint system has a vision-based occupant classification system for control of airbag deployment during a crash scenario. The classification system utilizes two imaging sensors which together create a stream of paired images received and stored by an occupant classification controller. A computer program product of the controller utilizes the paired images to extract disparity/range features and stereo-vision differential edge density features. Moreover, the controller extracts wavelet features from one of the two paired images. All three features or maps are classified amongst preferably seven classifications by algorithms of the computer program product producing class confidence data fed to a sensor fusion engine of the controller for processing and output of an airbag control signal input into a restraint controller of the vehicle restraint system.

Abstract: A method and system for de-screening an image signal utilizing a bank of filters to provide several increasingly blurred versions of the original image signal is disclosed. At any given time, only two of these blurred versions are created, on a pixel-by-pixel basis. The outputs from the selected pair of blurred signals are then blended together to create a variable blending output that can vary smoothly from no blurring to maximum blurring in a smooth and continuous manner. In addition, the method provides the capability to enhance text and line art by using a variable un-sharp masking mechanism with independent post-blur sharpening control, and the capability to detect and enhance neutral (no-color) output pixels.