Abstract: A first generation unit generates a RAW image reduced in the horizontal direction from a RAW image including pixels each having single-color signal information, by limiting the band of spatial frequencies in the horizontal direction and decreasing the number of pixels in the horizontal direction. The RAW image reduced in the horizontal direction is stored in a line memory. A second generation unit generates a reduced RAW image by decreasing, at least in the vertical direction, the number of pixels of the RAW image reduced in the horizontal direction. This makes it possible to generate a reduced RAW image using little memory capacity.

Abstract: An image file representing at least a portion of a printed document is processed to highlight the differences between foreground material (e.g., text or other characters) from background. The method includes selecting a neighborhood of pixels, determining a weighted average of an attribute values (e.g., luminance) for each pixel, and modifying each pixel's value based on the weighted average. Graylevel scaling, error diffusion, and a bit level conversion are also performed each pixel ends up with either a first attribute value level (e.g., luminance of 0) or a second attribute value level (e.g., luminance of 255).

Abstract: One embodiment of the present invention sets forth a technique for improving path rendering on computer systems by efficiently representing and computing sub-pixel coverage for path objects. A stencil buffer is configured to store multiple stencil samples per pixel stored in an image buffer. The stencil samples undergo stencil testing to produce a set of Boolean values per pixel, which collectively define a geometric coverage percentage for the pixel. The coverage percentage is used to modulate a color value for the pixel. The modulated color value is then blended into the image buffer as an anti-aliased pixel. This technique advantageously enables efficient anti-aliasing for path rendering.

Abstract: To reduce flickering in driving distributed processing, an image processing apparatus which generates and outputs a plurality of sub frame images having different luminance patterns for each of a plurality of frame images included in input moving image data, includes a filtering unit configured to perform filter processing for the frame image, a setting unit configured to set a first coefficient A and a second coefficient B, a first generation unit configured to generate a first sub frame image, a second generation unit configured to generate a second sub frame image, and an output control unit configured to selectively output the first sub frame image and the second sub frame image at a predetermined timing.

Abstract: Methods and systems for displaying full-color three-dimensional imagery are provided. A first color set, having a first color spectrum, is defined to include a first set of LEDs. The first color set is assigned to a first color-coded image perspective. A second color set, having a second color spectrum, is defined to include a second set of LEDs. The second color set is assigned to a second color-coded image perspective. The full-color three-dimensional imagery is caused by activating, alternatively, at least two LEDs of the first color set or the second color set and one LED of a remaining color set and displaying the three-dimensional image based on the first image perspective and the second image perspective.

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: A non-transitory computer readable storage medium having stored thereon a computer program comprising instructions, which when executed by a computer, cause the computer to acquire an image comprising image data reconstructed from an imaging device. The instructions further cause the computer to apply a bloom reduction algorithm to the image configured to iteratively reduce a blooming of high-density objects in the image.

Abstract: A method for producing high-quality photographic images in low-light conditions and in the absence of large-aperture optics. The method includes, upon photographing, first obtaining a plurality of frames of the image with exposures which either partially overlap in time or with an insignificant pause between them. The best result can be obtained in the case when the pause between the exposures represents less than 1/20 of the overall exposure time. The method further includes separating out the initial images from a group of exposures and filtering the images having the smallest exposure interval using the images having the largest exposure interval. The final image is obtained by combining initial images having different exposure intervals from the same group.

Abstract: In general, in one embodiment, low-light noise is removed from an image by separately filtering luma and chroma components of the image, by adaptively filtering the image based at least in part on a Gaussian distribution of the image, and/or by dividing the image into separate regions and filtering each region separately.

Abstract: A method to enhance detail of an image based on noise elimination includes calculating a noise weight corresponding to a probability that a center pixel, located in a block of pixels of a region of the image, is noise by using a difference between the center pixel and a surrounding pixel located in the block of pixels, calculating a first substitution value for the center pixel based on the noise weight; and calculating a second substitution value for the center pixel by using the noise weight and a sharpen filter.

Abstract: A set of pixels of a background element is identified according to a mask that defines a shape of a foreground element. A color value for a pixel of the foreground element is determined. The determining includes ascertaining a value of a measure of brightness of one or more pixels of a set of pixels of the background element and calculating the color value for the pixel of the foreground element based on the value of the measure of brightness and a value of an adjustable contrast variable. The calculating the color value for the pixel of the foreground element preserves in the foreground element a color component of the one or more pixels of the set of pixels of the background element and increases contrast with the value of the measure of brightness according to the value of the adjustable contrast variable.

Abstract: An image processing device is provided. The image processing device includes a weighted low-pass filter which performs weighted low-pass filtering on illumination of sub-sampled pixel signals, and an illumination interpolation circuit which compares illumination of the weighted low-pass filtered pixel signals with illumination of current pixel signals and performs interpolation while applying a weight to illumination of the weighted low-pass filtered pixel signals according to the comparison result.

Abstract: Provided is an image processing device capable of performing appropriate noise reduction without causing blurring of edges and details of the image.

Abstract: It is an object to provide an image processing device capable of obtaining a high precise image while aberration asymmetry is corrected. An image processing device according to the present invention includes image obtaining means configured to obtain an input image, image restoration means configured to restore the input image using a generated or selected image restoration filter in accordance with a transfer function of an image pickup system used for forming the input image from an object image. The image restoration filter makes a difference between absolute values of transfer functions of two azimuthal directions at a time when the restoration image is obtained from an object smaller than a difference between absolute values of transfer functions of the two azimuthal directions of the image pickup system.

Abstract: There are provided an apparatus and a method for estimating edge areas of pixels in a digital image to thereby prevent an edge sharpening algorithm from being applied to non-edge area of the digital image. Therefore, the apparatus can accurately determine whether each pixel is in an edge area or in a non-edge area, by generating a binary mask obtained by using a luminance difference average between each pixel and each of neighboring pixels in the digital image. Moreover, the determination of an edge-area or a non-edge area may be applied to various digital images, by adjusting a level of edge-area determination.

Abstract: This disclosure describes an apparatus and method of multi-spectral imaging to obtain and analyze information contained in the spectral distribution (reflection, absorption, or emission) of components within the image. The spectral information is captured in a series of images from differing spectral regions. This series of images are then combined into a composite image using re-colorization and image stabilization algorithms for display in real time. The process can be repeated continuously allowing spectral changes over time to be captured and analyzed. In the alternative one sequence of images can be captured for use as a still image.

Abstract: This invention obtains a high-quality output image in which a jaggy or a disconnection of a thin line hardly stands out. An edge detection unit detects whether the pixel of interest is an edge pixel or a non-edge pixel. A screen processing unit performs screen processing for the pixel of interest to determine the output value of the N-level tone. A calculation unit performs filtering processing having a preset visual sense characteristic for an area including the pixel of interest to calculate the target value of the pixel of interest, and corrects the value of the pixel of interest so as to come close to the target value, thereby determining the output value of the N-level tone. The selector selects, according to the detection result by the edge detection unit, either the result from the calculation unit or that from the screen processing unit.

Abstract: Embodiments provide a method for merging a plurality of first digital pictures, wherein each first digital picture includes a plurality of pixel values. The method includes extracting details from the plurality of first digital pictures. The method further includes combining the pixel values of the plurality of first digital pictures to generate a combined digital picture. The method further includes modifying the combined digital picture with the extracted details to generate a merged digital picture.

Abstract: A parallel scaler unit for simultaneously scaling multiple pixels from a source image. The scaler unit includes multiple vertical scalers and multiple horizontal scalers. A column of pixels from the source image is presented to the vertical scalers, and each vertical scaler selects appropriate pixels from the column of pixels for scaling. Each vertical scaler scales the selected pixels in a vertical direction and then conveys the vertically scaled pixels to a corresponding horizontal scaler. Each horizontal scaler scales the received pixels in a horizontal direction.

Abstract: A method for sharpening a captured image includes the steps of (i) identifying a plurality of edge pixels in the captured image, (ii) reviewing the plurality of edge pixels to identify one or more line pixels, and one or more non-line pixels in the captured image (354), and (iii) sharpening the captured image utilizing a first level of overshoot control for the non-line pixels (362), and utilizing a second level of overshoot control for the line pixels (360) The method also includes the steps of (i) identifying an intensity value for each of a plurality of neighboring pixels that are positioned near a selected pixel in a predetermined pixel window that have the highest intensity values and the lowest intensity values.

Abstract: Techniques to sample texels efficiently for an image effect may include determining a number of texels (kernel size) needed to compute a weighted average for an image effect on an image. The technique may further include selecting at least one mipmap generated by a graphics processing unit (GPU) according to a function of the determined kernel size. The function may also consider a threshold kernel size. The technique may further sampling texels, with the GPU, from the selected mipmap(s), and calculate the weighted average of the sampled texels to produce the image effect.

Abstract: An image filtering method and apparatus that filter a block of a plane area by using a pseudo-random filter are provided, whereby artifacts generated in image encoding and decoding procedures are removed, the method including: determining a similarity between pixel values of a the first block; and generating a second block by selectively applying a pseudo-random filter on the first block according to a result of based on the determining the determined similarity.

Abstract: A method for reconstructing an image from emission data includes generating a compressed point-spread function matrix, generating an accumulated attenuation factor; and performing at least one image projection operation on an image matrix of the emission data using the compressed point-spread function matrix and the accumulated attenuation factor. The image projection operation can include rotating an image matrix and an exponential attenuation map to align with a selected viewing angle. An accumulated attenuation image is then generated from the rotated image matrix and rotated exponential attenuation map and a projection image is generated for each voxel by multiplying the accumulated attenuation image and point spread function matrix for each voxel. The rotating and multiplying operations can be performed on a graphics processing unit, which may be found in a commercially available video processing card, which are specifically designed to efficiently perform such operations.

Abstract: A method for reducing blurring in a blurred image (14) of a scene (12) includes the steps of: (i) creating an edge mask (362) from the blurred image (14); (ii) extending the edge mask (362; (iii) forming an initial array (e.g. extending the blurred image (360); and (iv) performing Lucy-Richardson iterations, with masking. With the deblurring method disclosed herein, the reconstructed adjusted image (16) (i) does not have (or has significantly less) ringing artifacts around the edges (22) of the captured object(s) (20C), and (ii) does not have (or has significantly less) ringing artifacts around border (24). As a result thereof, the adjusted image (16) is more attractive and more accurately represents the scene (12).

Abstract: Methods and a module for use with a microscope for enhancing image contrast in an image of a phase object. A transmission image of a specimen is formed in an image plane, of which a Fourier transform is generated in a Fourier plane. An amplitude filter is applied to the Fourier transform field, which is then transformed back to an image at a focal plane of a detector array. An image signal from the detector array is processed to generate a contrast-enhanced image of the specimen.

Abstract: An image processing apparatus includes a filter unit which filters image signals; a sampling unit which generates first digital image signals having a first resolution by sampling the filtered image signals at a predetermined sampling frequency; and a super-resolution unit which reconstructs a second digital image signal having a second resolution which is higher than the first resolution by performing super-resolution on the first digital image signals generated by the sampling unit, wherein the filter unit passes frequency components corresponding to or lower than the Nyquist frequency which is half the sampling frequency, and passes a part of frequency components within a range from the Nyquist frequency to the highest frequency which can be represented by the second resolution.

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 and apparatus for processing images, and an image display apparatus are provided that can produce favorable images even when digitized image are partially magnified (enlarged) or reduced (shrunk). The apparatus for expanding or reducing input image data supplied thereto, for each area of the image data, comprises a high frequency (HF) component smoothing processor that generates smoothed-HF-component image data by smoothing HF components of the input image data; a partial magnification/reduction controller that generates partial magnification/reduction control information that designates positions of pixels in image data obtained after expanding or reducing the input image data for each image area; and a pixel data generator that generates pixel data of pixel positions designated by the partial magnification/reduction control information in the smoothed-HF-component image data, by using pixel data in a neighborhood of the designated pixel positions.

Abstract: An image encoding method using color space estimation is provided. A method for encoding a second color space pixel corresponding to coordinates of a first color space pixel using color spaces, includes encoding a pixel of a first color space; calculating a color space estimation value of a current pixel of a second color space by referring to the encoded pixel of the first color space; and encoding a differential value of the calculated color space estimation value and the current pixel value. Thus, the image compression efficiency can be enhanced by using the information between the color spaces in the image encoding, and the lossless image compression can be accomplished by ruling out the similarity of the pixels.

Abstract: Methods and systems for creating an image filter is described. In one embodiment, a method includes receiving a first plurality of images captured by an at least one camera with a same setting as a first camera. The method further includes creating an averaged image from the plurality of captured images, wherein the averaged image comprises a measurement value of intensity for each pixel in the averaged image. The method also includes determining an image imperfection in the averaged image. The method further includes creating the image filter to reduce the image imperfection.

Abstract: Example embodiments disclosed herein relate to feature based image registration. Feature based image registration determines correspondence between image features such as points, lines, and contours to align or register a reference or first image and a target or second image. The examples disclosed herein may be used in mobile devices such as cell phones, personal digital assistants, personal computers, cameras, and video recorders.

Abstract: Gender recognition is performed using two or more modalities. For example, depth image data and one or more types of data other than depth image data is received. The data pertains to a person. The different types of data are fused together to automatically determine gender of the person. A computing system can subsequently interact with the person based on the determination of gender.

Abstract: An image processing server performs haze-removal from images. Global atmospheric light is estimated and an initial transmission value is estimated. In one embodiment, a solver is applied to an objective function to recover a scene radiance value based on the estimated atmospheric light and estimated transmission value. The scene radiance value is used to construct an image without haze. In a simplified method that avoids using a solver, bilateral filtering is performed on the transmission image in order to construct an image without haze.

Abstract: In an image processing method of processing an image obtained from an image sensor, the resolution distribution of the image sensor is held in a memory, and a resolution corresponding to a pixel of interest of the image is acquired from the resolution distribution of the image sensor which is held in the memory. Sharpening processing having the sharpening level selected on the basis of the acquired resolution is executed for the pixel of interest.

Abstract: A video processing device may compare a pixel or group of pixels of a current video picture to a pixel or group of pixels of a previous video picture. The video processing device may generate a motion vector for the pixel or group of pixels of the current video picture based on the comparison. The video processing device may determine an amount of filtering to be applied to the pixel or group of pixels of the current video picture. The video processing device may adjust the determined amount of filtering to be applied to the pixel or group of pixels based on the generated motion vector and based on a brightness value.

Abstract: An image processing module performs efficient image enhancement according to a multi-scale energy minimization process. One or more input images are progressively downsampled to generate a pyramid of downsampled images of varying resolution. Starting with the coarsest downsampled image, a label map is generated that maps output pixel positions to pixel positions in the downsampled input images. The label map is then progressively upsampled. At each upsampling stage, the labels are refined according to an energy function configured to produce the desired enhancements. Using the multi-scale energy minimization, the image processing module enhances image via hole-filling and/or super-resolution.

Abstract: A method for operating on images is described for interest point detection and/or description working under different scales and with different rotations, e.g. for scale-invariant and rotation-invariant interest point detection and/or description.

Abstract: An image capturing device includes a photodetector including a plurality of pixels, each having a photoelectric conversion unit outputting electric charges depending on received light, charge accumulation units accumulating the electric charges, and a drain electrode that discards the electric charges, a finite impulse response filter which is applied to images, a controller controlling the pixels so as to distribute the electric charges that are output by the photoelectric conversion units to the charge accumulation units and to the drain electrodes at a distribution ratio depending on a weight coefficient of a corresponding delay period of the finite impulse response filter at each delay period, and reading the electric charges accumulated in the charge accumulation units after a whole delay period of the finite impulse response filter elapses, and an image processor generating the images based on the electric charges that are read by the controller.

Abstract: An image processing apparatus is provided. The image processing apparatus includes a local brightness adjustment circuit and an image enhancement circuit. The local brightness adjustment circuit is arranged to obtain a local brightness gain relation between an input image before a local brightness adjustment and the input image after the local brightness adjustment. The image enhancement circuit is coupled to the local brightness adjustment circuit, and is arranged to perform at least one of a noise reduction and an edge enhancement upon an image to be processed according to the local brightness gain relation to generate an output image.

Abstract: An image filter and method of smoothing pixel values. A pixel value of a pixel to be smoothed is compared with block average pixel values of each of a plurality of pixel blocks. The pixel to be smoothed may be downstream from each of the pixel blocks. If the difference between the pixel value and each of the block average pixel values is less than a corresponding sigma threshold value for each of the pixel blocks, a first sigma filter utilizing the block average pixel values is applied to the pixel to be smoothed. If the difference between the pixel value and any one of the block average pixel values is not less than a corresponding sigma threshold value, a second sigma filter is applied to the pixel to be smoothed.

Abstract: Embodiments of an apparatus and method for decoding an encoded bitstream to generate a video signal are taught. A decoder receives the bitstream and decodes a portion thereof to obtain at least a portion of a predictive reference frame. The reference frame is generated by, for example, selecting a filter set including a target frame with a block having a plurality of pixels and at least one adjacent frame, determining a coding mode associated with the block, determining a block-specific factor derived from the coding mode, determining weighted averages of pixels in the filter set and using the weighted averages to generate the predictive reference frame. The decoder also generates a residual from another portion of the encoded bitstream that represents a portion of a frame of the plurality of frames and reconstructs the portion of the frame by adding the residual to the predictive reference frame.

Abstract: A computer-implemented method for identifying the edges of web media transported on a movable transport surface includes sensing, using a linear array sensor positioned along a process path of a web, the web media and the movable transport surface to obtain image data representative of variations in optical textural properties of the web media and variations in optical textural properties of the movable transport surface, wherein the variations in the optical textural properties of the movable transport surface are different from the variations in the optical textural properties of the web media; and processing the image data to determine differences between the variations in the optical textural properties of the web media and the variations in the optical textural properties of the movable transport surface to identify an edge of the web media.

Abstract: Input image data may be subjected to a spatial convolution to produce a plurality of different filter matrices or kernels. Thus, a single system may be capable of producing a plurality of different filter sizes which may be selected based on particular circumstances, uses and described precision. Thus, a single system may provide variable spatial filtering. By reducing the input data matrix size and by reusing components, in some cases, the computational complexity of producing a plurality of symmetrical filters is not significantly greater than that involved in producing only a single filter.

Abstract: One technique performs noise removal substantially removing dot noise based upon pseudo-standard deviation (PSD). Another technique performs noise removal including the substantial removal of dot noise based upon Z-score with a punctured approach. The techniques are not limited to particular kernels and include both uniform and non-uniform kernels.

Abstract: Method and system for determining a measure of quality for images are presented. Multi-level decomposition of images in the wavelet domain using a variable number of levels of decomposition and aggregation of selected subbands is performed to obtain an accurate measure of quality. The processing time is reduced in comparison to that required by other methods for generating measures of quality.

Abstract: Method and system for determining a measure of quality for images by using multi-level decomposition are presented. Multi-level decomposition of images is performed in the wavelet domain producing subbands at each level of decomposition. Aggregation of subbands is performed across multiple levels to produce an accurate measure of image quality. By aggregating only selected subbands the computational complexity of the method is greatly reduced.

Abstract: An embodiment of the invention provides an apparatus and a method. An intensity gradient between first and second optical image data of an optical image produced by an optical scanning device is determined. The first optical image data is a first optical image pixel and the second optical image data is a second optical image pixel. An intensity gradient image comprising a plurality of intensity gradient image pixels is determined, each intensity gradient image pixel representing the magnitude of an intensity gradient. One or more anomalies in the optical image in dependence on the magnitude of the intensity gradient being greater than or equal to a threshold value is identified.

Abstract: The technology described in this document relates to filters and functions that are based on exponential decay functions. In one aspect, the technology is embodied in a method that includes using a computing device to compute a first function as a combination of (i) an exponential decay function, a decay factor for which is chosen based on a Gaussian function, and (ii) at least a second function that is obtained by one or more convolution operations on the decay function. The first function provides an approximation of at least a portion of the Gaussian function.

Abstract: Computer implemented methods for compressing 3D hyperspectral image data having a plurality of spatial pixels associated with a hyperspectral image, and a number of spectral dimensions associated with each spatial pixel, include receiving, using a processor, the 3D hyperspectral image data, a set of basis vectors associated therewith, and either a maximum error amount or a maximum data size. The methods also include partitioning the 3D hyperspectral image data into a plurality of 2D images, each associated with one of the number of spectral dimensions, and an associated one of the set of basis vectors. The methods additionally include ranking the set of basis vectors if not already ranked. The methods may further include iteratively applying lossy compression to the 2D images, in an order determined by the ranking. Other embodiments and features are also disclosed.

Abstract: In a first exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, generating intrinsic images corresponding to the image, and performing a relationship maintenance process to correct artifacts in the intrinsic images.