Abstract: The multi-image sharpening and denoising technique described herein creates a clean (low-noise, high contrast), detailed image of a scene from a temporal series of images of the scene. The technique employs a process of image alignment to remove global and local camera motion plus a novel weighted image averaging procedure that avoids sacrificing sharpness to create a resultant high-detail, low-noise image from the temporal series. In addition, the multi-image sharpening and denoising technique can employ a dehazing procedure that uses a spatially varying airlight model to dehaze an input image.

Abstract: The invention relates to an improved robustness in upscaling the resolution of regularly sampled multi-dimensional signals, where a single low-resolution signal is available. These methods are referred to as example-based super-resolution or single-image super-resolution. A method for super-resolving a single image comprises three stages. First, an interpolation-based up-scaling of the input image is performed, followed by a Local De-noising step in contour regions. The second stage comprises extrapolation through cross-scale block matching, wherein an extrapolated high-frequency band is obtained that is de-noised through regularization in the same contour regions. The third stage comprises adding the contributions of the low-frequency band of the high-resolution image and the extrapolated high-frequency band.

Abstract: Methods and apparatus according to various aspects take as input image data in a lower-dynamic-range (LDR) format and produce as output enhanced image data having a dynamic range greater than that of the input image data (i.e. higher-dynamic range (HDR) image data). In some embodiments, the methods are applied to video data and are performed in real-time (i.e. processing of video frames to enhance the dynamic range of the video frames is completed at least on average at the frame rate of the video signal).

Abstract: A method for sharpening a captured image (14) includes (i) selecting a pixel (240) in the captured image (14); (ii) selecting a selected high intensity value (302); (iii) selecting a selected low intensity value (302); (iv) normalizing the intensity value to establish a normalized intensity value using the selected high intensity value and the selected low intensity value (304); (v) determining an adjusted normalized intensity value for the normalized intensity value using a contrast correction function (306); and (vi) scaling the adjusted normalized intensity value to get a transformed intensity value (308). Subsequently, the adjusted image (16) can be generated using the transformed intensity value for each pixel (240). The contrast correction function can be selected that provides the desired amount of sharpening. Thus, the amount of sharpening that is applied to the image (14) can be specifically selected.

Abstract: Imaging systems with image sensors and image processing circuitry are provided. The image processing circuitry may calculate sharpness values for a window within an image captured by the image sensors. The window may be divided into zones. A first filter may be applied to each row of each zone. A first sharpness value may be calculated by averaging the absolute values of the outputs of the first filter that are greater than a first threshold. A second sharpness value may be calculated by averaging the absolute values of the outputs of the second filter that are greater than a second threshold. A final sharpness value for each zone may be calculated by dividing the second sharpness value by the first sharpness value and multiplying the result by corresponding scalar weights. A window sharpness value may be calculated from the weighted sum of the final sharpness values of each zone.

Abstract: The invention discloses an image processing system and method thereof. The image processing system includes an local variance estimator, a noise detector, a spatial de-noise filter. The local variance estimator estimates each pixel of an input image signal to separately output a local variance value of each pixel, and generates a noise threshold according to the local variance values. The noise detector determines which pixel indicates noise or image according to the noise threshold. The spatial de-noise filter filers the pixel indicating noise to generate an output image signal.

Abstract: Devices, systems, methods, and other embodiments associated with reducing digital image noise are described. In one embodiment, a method includes determining, on a per pixel basis, mosquito noise values associated with pixels of a digital image. The method determines, on a per pixel basis, block noise values associated with the digital image. The method filters the digital image with a plurality of adaptive filters. A compression artifact in the digital image is reduced. The compression artifact is reduced by combining filter outputs from the plurality of adaptive filters. The filter outputs are combined based, at least in part, on the mosquito noise values and the block noise values.

Abstract: An image processing device and a method thereof are provided. In the method, an original image and a corresponding depth image are received, wherein the depth image includes a plurality of depth values, and the depth values indicate depth of field of a plurality of blocks in the original image respectively. Further, each of the blocks is processed to obtain a corresponding smoothness and/or sharpness effect according to each of the depth values. Thereby, a stereoscopic sensation of the original image can be enhanced.

Abstract: The present disclosure relates to combination of images. A method according to an embodiment comprises: receiving a visual image and an infrared (IR) image of a scene; extracting high spatial frequency content from said visual image; and combining said extracted high spatial frequency content from said visual image with said IR image, wherein a resolution for the visual image and the IR image are substantially the same, to generate a combined image.

Abstract: A method of classifying an image taken by an image capture device, the method comprising the steps of: extracting an initial Sensor Noise Pattern (SNP) for the image; enhancing the initial SNP to create an enhanced SNP by applying a correcting model, wherein the correcting model scales the initial SNP by a factor inversely proportional to the signal intensity of the initial SNP; determining a similarity measure between the enhanced SNP for said image with one or more previously calculated enhanced SNPs for one or more different images; and classifying the image in a group of one or more images with similar or identical SNPs based on the determined similarity measure.

Abstract: A frame image of decoded image data in which an image stream generated by performing an orthogonal transformation and a lossy compression coding has been decoded is inputted so as to be stored in a frame memory for each block, and based on the stored frame image and a newly input frame image, a motion vector for each block is detected, and a motion vector variance map generator calculates a variance value map by calculating a variance value of the motion vectors. Based on the variance value map and a threshold, an outline domain included in the frame image is extracted, and a band limiting filter performs band limitation for each block in regard to the outline domain so as to accomplish removal of noise from the decoded image.

Abstract: An image signal processor for image enhancement is provided. The image signal processor receives a first, a second, and a third component of an RGB signal of each pixel of an image, and has: an image enhancer for performing image enhancement on the first component of the RGB signal to produce a first enhanced component of the RGB signal; a gain generator, coupled to the image enhancer, for producing an enhancement gain based on the first component and the first enhanced component of the RGB signal; and a gain multiplier, coupled to the gain generator, for performing image enhancement on the second and the third component based on the enhancement gain to produce a second and a third enhanced component.

Abstract: To provide a defect inspection apparatus and method adapted to easily assign threshold levels to scattered-light detectors and to appropriately acquire data detected by each scattered-light detector. The apparatus includes a stage device on which to rest a sample; a laser light irradiation device that irradiates the sample on the stage device with inspection light; scattered-light detectors, each of which detects a beam of light, scattered from the sample, and outputs an image signal; a threshold level setter formed so that an associated threshold level for judging whether defects are present is set only for an image signal selected from individual image signals of the scattered-light detectors or from image signals obtained by arithmetic processing based on the image signals, and a threshold level setting circuit that acquires the individual image signals, only if the image signal exceeds the threshold level set in the threshold level setter.

Abstract: A system for processing video data to improve image quality at low data rates comprising an encoder generating an entropy signal representing the relative entropy of the video data and a plurality of filters each receiving the entropy signal and a threshold setting and filtering the video data if the entropy signal is greater than the threshold setting.

Abstract: One or more techniques and/or systems for detecting edges in images of objects subjected to imaging using imaging apparatus are disclosed, such that the effect of image noise on the edge detection can be mitigated. Ratios of intensity values (e.g., signal values) for a subject region (e.g., a pixel) and respective adjacent regions are determined. An adaptive threshold value is determined for respective adjacent region pairs. The ratio value is compared to the adaptive threshold value for respective adjacent region pairs, to determine whether an edge is present between the respective adjacent region pairs.

Abstract: An electro-optic display device includes a display matrix of display pixels. Each display pixel may include two or more first sub-pixels and a second sub-pixel. Each first sub-pixel may have two or more display states and a color filter. The second sub-pixel may have two or more display states and a white filter. The first sub-pixels may be arranged in rows and columns in a repeating pattern, and each first sub-pixel may be horizontally adjacent and vertically adjacent to one or more second sub-pixels. In addition, a display controller may include a data switch, a color correction module, an image filtering module, a color linearization module, an image dithering module, and a saturation adjustment module. The modules may be configured to perform operations at first and second pixel resolutions.

Abstract: Systems and methods are provided for producing a rendered drawing or rendering from a detailed image of an object (e.g. photograph) resulting in a rendering that is photogrammetric and that preserves detail in the said image of said object. The combination of the metric nature and image detail preservation in a rendering resulting from the process enhances the usefulness of the rendering to users. The invention is useful in particular for large format renderings such as wire frame style drawings used for blueprints in the architecture, engineering and construction industry (AEC industry) when used for existing structures. The processes combine graphic arts techniques with photogrammetric techniques to preserve, fully or partially, information about an object as captured in image detail of said object and to present said information in photogrammetrically correct rendering, which rendering may be incorporated into drawings useful to and/or familiar to end users of said drawings.

Abstract: A method for imaging includes receiving an input image that includes a matrix of pixels (32) having input pixel values. The input pixel values are processed so as to detect a region of saturation in the input image. A first image filtering operation is applied to the input pixel values of at least the pixels that are outside the region of saturation, so as to generate first filtered pixel values. A second image filtering operation is applied to the input pixel values of at least the pixels that are within the region of saturation, so as to generate second filtered pixel value. An enhanced output image is generated by stitching together the first and second filtered pixel values.

Abstract: A method for detecting edges within an image. An edge strength indicator for each pixel within the image is determined and the edge strength indicator is classified into at least three classes to generate a map having one value associated to each class. The classified map is then separated into individual maps depending on the classification. A dilation process is applied to at least a part of the individual maps. The processed and/or individual maps are finally combined by logical operation to obtain a final map indicating edges without nearby texture.

Abstract: Provided are an image processing method and apparatus for effectively reducing noise in an image, and a recording medium storing a program for executing the method. The image processing apparatus includes a noise reduction unit configured to apply a noise reduction filter to first image data to obtain second image data; a first edge data obtaining unit configured to calculate edge data lost in the second image data compared with the first image data to obtain the first edge data; and a first synthesis unit configured to tune the first edge data and calculate third image data from the second image data and the tuned first edge data.

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

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

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

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

Abstract: The present disclosure relates to a method of improving an IR image comprising capturing a visual image and an IR image of an object, altering a resolution of at least one of said visual image and IR image, high pass filtering said visual image to generate a processed visual image, low pass filtering said IR image to generate a processed IR image, and combining information from said processed visual image and said processed IR image to generate a combined image.

Abstract: An imaging apparatus that adjusts focus based on frequency components extracted from an image signal is provided. The apparatus includes a 1st order Infinite Impulse Response (IIR)-filter including a plurality of sample delayers having a plurality of line buffers for buffering and delaying a plurality of samples to sequentially read pixels of the image signal through raster scanning and extracting frequency components of the image signal from the read pixels by using the plurality of sample delayers; and a delay sample number controller for controlling the IIR-filter to match the number of horizontal pixels sequentially read from the image signal through raster scanning to the number of the plurality of samples buffered by the plurality of line buffers.

Abstract: Method and apparatus for processing edges in an image are provided. The method in an embodiment includes the following steps. With respect to a cross-shaped patterned centered at a target pixel of an input image, a first-direction gradient along a first direction and a second-direction gradient along a second direction are calculated. According to the first-direction and second-direction gradients, it is determined whether to compensate the target pixel based on pixel values of a first plurality of pixels along the second direction or pixel values of a second plurality of pixels along the first direction within the cross-shaped pattern, or to output a pixel value of the target pixel.

Abstract: An apparatus for improving image sharpness includes a pixel detector that labels each of pixels in an image as one of an edge/texture type and a non-edge/texture type, a pixel type determiner that classifies each of the pixels that are labeled as the edge/texture type as one of a border point type pixel, a transition point type pixel, and a peak point type pixel, and a filter. The filter includes a shrinking/expanding filter and a high boost filter, the shrinking/expanding filter filtering the border point type pixels, and the high boost filter subsequently filtering the peak point type pixels and then filtering the transition point type pixels.

Abstract: A system for, and method of, improving video image sharpness and a continuous-light-emitting video display system employing the system or the method. In one embodiment, the system for improving video image sharpness includes: (1) a sub-frame generator configured to receive a frame of a video image and generate plural sub-frames therefrom and (2) a spatial filter, associated with the sub-frame generator and configured to cause the plural sub-frames to be spatially filtered with respect to one another based on a display sequence thereof.

Abstract: Input image data indicating an image is rendered to a display panel in a display device or system that is substantially configured with a three primary color or multi-primary color subpixel repeating group using a subpixel rendering operation based on area resampling techniques. Examples of expanded area resample functions have properties that maintain color balance in the output image and, in some embodiments, are evaluated using an increased number of input image sample points farther away in distance from the subpixel being reconstructed than in prior disclosed techniques. One embodiment of an expanded area resample function is a cosine function for which is provided an example of an approximate numerical evaluation method. The functions and their evaluation techniques may also be utilized in constructing novel sharpening filters, including a Difference-of-Cosine filter.

Abstract: Error-function determining means makes use of a blur function serving as a function representing the degree of blurring of a blurred input image and a geometrical-deformation matrix serving as a matrix for restricting an operation to lay out a 2-dimensional code shown by the input image on cells each serving as a configuration unit of the 2-dimensional code in order to determine an error function serving as a function satisfying a relation that, the smaller the error between the input image and an image obtained by adding geometrical deformations and blurs to a recovered image obtained from a recovering process, the more the approach to a result determined in advance. Pixel-value restricting means determines pixel-value restriction information prescribing a restriction condition for a restriction imposed on a predetermined pixel value obtained from discretization of pixel values of the 2-dimensional code to be recovered.

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

Abstract: In a method for displaying an image of a scene in front of a vehicle, recorded by a video camera, in which the scene in front of the vehicle contains a roadway which is illuminated by the vehicle's headlights, the gray values of the pixels of the image data generated by the video camera are weighted using a reduction factor, which is a function of the respective pixel within the image and of the brightness of objects in the near range in front of the vehicle, in such a way that the contrast between the display of the near range in front of the vehicle and the display of other parts of the image is reduced.

Abstract: An image processing method, for correcting a blur a due to an optical system of an image capturing apparatus, the image processing method including, storing a plurality of representative filters in a memory, selecting a subset of representative filters from the plurality of representative filters based on a pixel position of a pixel of interest in an image, applying each of the selected representative filter to a pixel value of the pixel of interest, and correcting the pixel value of the pixel of interest based on (a) a result of the application of filters, and (b) the pixel position of the pixel of interest.

Abstract: A system and method for defringing chromatic aberrations that occur in imaging devices such as scanners. The system comprises shift filters to shift lines in the various color planes together. In addition in each color plane, a spread filter is used to compensate for the unequal point spread functions of each color. Furthermore, the results can be enhanced by filtering in the luminance-chrominance space.

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

Abstract: A method includes receiving image data, generating a low-pass image and a high-pass image from the image data, applying dynamic range compression to the low-pass image and not the high-pass image, and adding the high-pass image to the low-pass image after dynamic range compression to create an output image.

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

Abstract: Systems and methods for increasing the resolution of a depth map by identifying and updating false depth pixels are described. In some embodiments, a depth pixel of the depth map is initially assigned a confidence value based on curvature values and localized contrast information. The curvature values may be generated by applying a Laplacian filter or other edge detection filter to the depth pixel and its neighboring pixels. The localized contrast information may be generated by determining a difference between the maximum and minimum depth values associated with the depth pixel and its neighboring pixels. A false depth pixel may be identified by comparing a confidence value associated with the false depth pixel with a particular threshold. The false depth pixel may be updated by assigning a new depth value based on an extrapolation of depth values associated with neighboring pixel locations.

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

Abstract: Methods and apparatus for procedural directional texture generation. A procedural directional texture generation method may, for example, be used to design hair or hairstyles. The method may obtain one or more strokes, one or more optional masks, and one or more optional user maps. One or more interpolated orientation maps may be generated from the input. The orientation maps, possibly along with one or more optional user maps, may be used to compute synthetic low-frequency lighting. A noise map may be generated at one or more frequencies and used, along with the interpolated maps, to generate high-frequency lighting. Alternatively, a flow-guided texture synthesis method may be used to generate high-frequency lighting. The low- and high-frequency lighting may then be combined to generate a desired frequency spectrum. Color may be added to the full structure, or alternatively color may be added at an earlier step.

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

Abstract: A method for simultaneously optimizing a digital image taken in or through a scattering medium and obtaining information regarding optical properties of the scattering medium is provided. Digital image data are received by a computer. The digital image is evaluated according to an objective image quality metric and a resulting image quality value is compared to a previously stored image quality value for the image. A revised optical transfer function is derived by modeling the optical properties of the medium to be used to generate a restored digital image, which is derived from the original image and the revised optical transfer function. The restored digital image is evaluated according to the objective image quality metric and an optimized restored image is identified. The optical properties associated with the optical transfer function producing the optimized restored image represent a close approximation of the true optical properties of the medium.

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 device for determining an edge histogram of an image based on information about a gradient strength and a gradient direction of a local gradient in an image content of a partial image of the image includes an allocator which is implemented to allocate the information about the gradient strength and the gradient direction based on an allocation regulation to an edge image in order to obtain edge type information. The allocation regulation is selected such that with a predetermined gradient direction at least three different allocated edge types exist mirroring the different gradient strengths. The device further includes an edge histogram determiner which is implemented to determine the edge histogram based on the edge type information so that in the edge type histogram at least three edge types having different allocated gradient strengths may be differentiated.

Abstract: A method for sharpness enhancing digital image data (ID), including: generating an intermediate processed image (IPI), in which uniform sections of the digital image data (ID) are cleaned from noise and in which at least one of a group consisting of contours, edges, wedges, structures and textures are detected, wherein the uniform section each fulfil a predefined uniformity criterion, which is descriptive for at least one of a group consisting of uniformity, flatness, structure, and noise content; deriving a blend value b for each pixel (Pbjkab) from the intermediate processed image (IPI) with b ranging between 0 corresponding to flatness and 1 corresponding to texture; two-dimensional sharpness enhancing the digital image data (ID) to generate processed digital image data (PID); and generating output data (outD) by pixelwise blending the digital image data with the processed digital image data.

Abstract: An image processing method for boundary resolution enhancement is disclosed. Firstly, an image is transferred into an image layer. Noise of the image layer is removed by a bilateral filter and crisp edges are retained at the same time. Moreover, the image layer is interpolated by an interpolation filter for resolution enhancement. The image processing method of the present invention can lower the image blur degree substantially, enhance the image resolution and be widely implemented in all sorts of image/video processing hardware devices.

Abstract: A hair image display method comprises the steps of: selecting a hair area from a hair image; performing an edge detection on the hair area to calculate an edge image; performing representation processing on the edge image to calculate a representative value image; calculating direction differences between the edge image and the representative value image; and rendering the direction differences of respective pixels in color or grayscale to display a direction difference image, or rendering the correlation lengths of respective pixels in an edge direction image in color or grayscale to display a bundle width image, or determining the curvatures of the respective pixels in the edge direction image to form a curvature image. This makes it possible to display directions straying from a flow, bundles of hairs in the same flow, and the state of curls in hair styling clearly, and to facilitate evaluations of the hair styling.

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

Abstract: A method of processing a blood vessel image is provided. The method includes (a) sharpening an original blood vessel image using a Gabor filter in consideration of various directions and thicknesses of blood vessels included in the blood vessel and (b) detecting edges according to a change in brightness in a blood vessel domain and a non-blood vessel domain of the original blood vessel image and the blood vessel image on which the Gabor filtering step is completed, using an edge extraction method based on a first-order differentiation or second-order differentiation.