Abstract: According to one implementation, an image rendering system includes a computing platform having a hardware processor and a system memory storing an image denoising software code. The hardware processor executes the image denoising software code to receive an image file including multiple pixels, each pixel containing multiple depth bins, and to select a pixel including a noisy depth bin from among the pixels for denoising. The hardware processor further executes the image denoising software code to identify a plurality of reference depth bins from among the depth bins contained in one or more of the pixels, for use in denoising the noisy depth bin, and to denoise the noisy depth bin using an average of depth bin values corresponding respectively to each of the reference depth bins.

Abstract: An image forming apparatus includes: an image former that forms a document image including a correction image; an image reader; and a calibrator, and performing correction, wherein the document image is designed so that main register marks and a subsidiary register mark are provided, and the calibrator includes a four-corner detector that detects four corners of a sheet of paper, a main register mark reference position detector that detects reference positions of the main register marks, a subsidiary register mark prediction position determiner that predicts and determines prediction positions of the subsidiary register marks, a subsidiary register mark reference position detector that detects a reference position of the subsidiary register mark, a correction image prediction position determiner that predicts and determines a prediction position of the correction image, a measured value obtainer, and a correction value calculator.

Abstract: A sensor data management apparatus according to an embodiment stores pieces of sensor data in data areas provided corresponding to prescribed data units and manages the pieces of sensor data. The pieces of sensor data have been input from a plurality of sensors. The sensor data management apparatus includes a group determining unit and a data storing unit. The group determining unit is configured to analyze the pieces of stored sensor data and determine groups of sensors based on a result of the analysis. The data storing unit is configured to determine a storage location of a newly input piece of sensor data to store pieces of sensor data from sensors belonging to the same one of the groups in the same one of the data areas, and store the piece of sensor data at the storage location.

Abstract: A device for decoding video data determines mode information for a current block of a current picture of the video data; derives weights for use in a bilateral filter based on the mode information for the current block; applies the bilateral filter to a current sample of the current block by assign the weights to neighboring samples of the current sample of the current block and the current sample of the current block and modifying a sample value for the current sample based on sample values of the neighboring samples, the weights assigned to the neighboring samples, the sample value for the current sample, and the weight assigned to the current sample; and based on the modified sample value for the current sample, outputs a decoded version of the current picture.

Abstract: An aerothermal-radiation correction method, including: using a Gaussian surface to approximate a thermal radiation noise, performing a Fourier transform on the Gaussian surface so as to obtain a centralized spectrum of the thermal radiation noise, constructing a filter function H based on the centralized spectrum of the thermal radiation noise; performing a Fourier transform on the aerothermal-radiation degraded image f so as to obtain a centralized spectrum F, taking dot product of F and H to obtain a filtered spectrum G; and performing an inverse Fourier transform on filtered spectrum G to obtain a modulus, and acquire a corrected image. The method effectively removes background noise generated by aerothermal radiation, greatly improves image quality and image signal-to-noise ratio. The method features reduced computational complexity and a shorter operation time, and is suited for real-time processing.

Abstract: The present disclosure provides an image recording device and an image processing method. The tone adjusting unit includes plural preset tone mapping adjusting curves. The auto exposure unit compares and obtains the nearest brightness interval from plural brightness intervals according to the image information of the original dynamic image. The tone adjusting unit chooses the corresponding tone mapping adjusting curve from the plural tone mapping adjusting curves to adjust the brightness of the original dynamic image according to the brightness interval. Therefore, the image recording device and the image processing method of the present disclosure takes different tone mapping adjusting curves to adjust the brightness of the original dynamic image according to different brightness of the original dynamic image. The image recording device of the present disclosure adjusts the brightness according to different brightness of the original dynamic image.

Abstract: A method for multi-dimensional image processing obtains a multi-dimensional image having a plurality of data elements, each data element having a corresponding data value. The method forms a reduced noise image by repeated iterations of a process that adjusts the data value for one or more data elements p of the obtained image by: for each data element k in a group of data elements in the image, calculating a weighting factor for the data element k as a function of the difference between an estimated data value for data element p and the corresponding data value of data element k, updating the estimated data value for the data element p according to the combined calculated weighting factors and the data value of the data element k; and displaying, storing, or transmitting the reduced noise image.

Abstract: An image processing pipeline may perform temporal filtering on independent color channels in image data. A filter weight may be determined for a given pixel received at a temporal filter. The filter weight may be determined for blending a value of a channel in a full color encoding of the given pixel with a value of the same channel for a corresponding pixel in a previously filtered reference image frame. In some embodiments, the filtering strength for the channel may be determined independent from the filtering strength of another channel in the full color encoding of the given pixel. Spatial filtering may be applied to a filtered version of the given pixel prior to storing the given pixel as part of a new reference image frame.

Abstract: A portable image enhancement device comprising an optical image sensor, a rangefinder, a parametric control, a central processing unit (CPU), and a user interface. The sensor receives image data for a target object in a tactical scene. The rangefinder continually measures distance from the device to the target object, which may be in motion relative to each other. The parametric control tunes a parametric variable (e.g., spatial frequency, filter shape, orientation) that the CPU uses, along with the potentially-changing target distance, to create custom spatial filters and apply them to create a filtered scene from the tactical scene. Each spatial filter is of a two-dimensional (2D) Fourier-domain type (e.g., high-pass, low-pass, band-pass, band-stop, orientation). A spatial filter, target distance, and parametric variable combination may be stored as a preset filter for subsequent retrieval and application to a newly-input tactical scene. The user interface displays dynamically-tunable filtered scene(s).

Abstract: Systems and method for refining a depth map of a scene based upon a captured image of the scene. A captured depth map of the scene may contain outage areas or other areas of low confidence. The depth map may be aligned with a color image of the scene, and the depth values of the depth map may be adjusted based upon corresponding color values of the color image. An amount of refinement for each depth value of the aligned depth map is based upon the confidence value of the depth value and a smoothing function based upon a corresponding location of the depth value on the color image.

Abstract: A user-facing front end screen display of an illustrative digital image painting application includes a bucket tool which enables a user to apply a selected color to a selected area of an uploaded digital image such as a wall of a room, a paint brush tool to fill in missed areas, an eraser tool is to remove misapplied color, masking tools to mask off selected areas, and a tolerance slider tool to assist in properly filling in painted areas. Improved image preprocessing methods enable better definition of areas of the image to be painted the same color.

Abstract: The present invention discloses an image anti-aliasing system, which comprises an edge detection unit, a directional filter, a low-angle directional filter, a directional filtering fusion unit and a result fusion unit; the edge detection unit outputs the edge direction and edge confidence according to an input image, the directional filter outputs a directional filtering result based on the input image and the edge direction, the low-angle directional filter outputs a low-angle directional filtering result according to the input image and the edge direction; the directional filtering fusion unit weights and combines the result output from the directional filter and the result output from the low-angle directional filter according to the edge direction, and outputs a result of weighted combination of two directional filtering results; the result fusion unit outputs an optimized image according to the edge confidence, the input image and the result of the directional filtering fusion unit.

Abstract: An image filtering method is provided, including: acquiring a sample characteristic value of a sample image; ranking images to be filtered in a specific chronological order; acquiring characteristic values of the images to be filtered; comparing the characteristic values of the images to be filtered with that of the sample image, selecting, from the images, filtered images conforming to preset conditions, storing the filtered images in the specific chronological order, and displaying the filtered images. Accordingly, a user is enabled to easily find identical or similar images amongst a large number of images.

Abstract: Aerial imagery may be captured from an unmanned aircraft or other aerial vehicle. The use of aerial vehicle imagery provides greater control over distance from target and time of image capture, and reduces or eliminates imagery interference caused by clouds or other obstacles. Images captured by the aerial vehicle may be analyzed to provide various agricultural information, such as vegetative health, plant counts, population counts, plant presence estimation, weed presence, disease presence, chemical damage, wind damage, standing water presence, nutrient deficiency, or other agricultural or non-agricultural information. Georeference-based mosaicking may be used to process and combine raw image files into a direct georeferenced mosaic image.

Abstract: An input HDR image may be represented using an SDR image and a modulation value. For the SDR image to be viewable and representative of the HDR image, we propose a mid-tone optimization technique to calculate the modulation value. In particular, we use two conditions when deriving the modulation value: (1) the blacks are not clipped down to zero too aggressively; and (2) the number of codewords in the SDR image used to represent the mid-tone range of the HDR image is maximized. The modulation value can further be clipped to avoid over-shooting in both very dark images and very bright images. Temporal stabilization can also be used to smooth temporal variations of the luminosity in the SDR video. After the modulation value is determined, the SDR image can be obtained based on the HDR image and the modulation value.

Abstract: The present technology relates to an image processing device, an image processing method, and an image processing system capable of improving detection accuracy of a defective pixel. Provided is an image processing device including a defective pixel estimation unit that estimates a defect of a pixel of interest in an image captured by a solid-state imaging device that contains a two-dimensional array of color pixels, and high-sensitivity pixels having higher sensitivity than sensitivities of the color pixels. The defect of the pixel of interest is estimated on the basis of a correlation between pixel values of the high-sensitivity pixels, and pixel values of the color pixels. The image processing device further includes a defective pixel correction unit that corrects the pixel of interest when it is estimated that the pixel of interest is a defective pixel. The present technology is applicable to an image processing device which corrects a defective pixel by signal processing, for example.

Abstract: In one embodiment, a method includes accessing a plurality of generative adversarial networks (GANs) that are each applied to a particular level k of a Laplacian pyramid. Each GAN may comprise a generative model Gk and a discriminative model Dk. At each level k, the generative model Gk may take as input a noise vector zk and may output a generated image {tilde over (h)}k. At each level k, the discriminative model Dk may take as input either the generated image {tilde over (h)}k or a real image hk, and may output a probability that the input was the real image hk. The method may further include generating a sample image ?k from the generated images {tilde over (h)}k, wherein the sample image is based on the probabilities outputted by each of the discriminative models Dk and the generated images {tilde over (h)}k. The method may further include providing the sample image ?k for display.

Abstract: A display apparatus is provided. The display apparatus includes a display panel comprising a plurality of pixels, a panel driver configured to drive the display panel, a storage configured to store information regarding a color gamut of each of the plurality of pixels, and a processor configured to determine a target color gamut of each of the plurality of pixels so that a difference in color gamut from at least one adjacent pixel is equal to or less than a predetermined threshold value, and to drive the panel driver for each of the plurality of pixels to have a grayscale value based on the target color gamut.

Abstract: An edge detection apparatus according to the present invention, includes: a detection unit configured to detect a dispersion degree of gradation values of input image data; a determination unit configured to determine detection sensitivity, which is edge detectability, based on the dispersion degree detected by the detection unit; and an edge detection unit configured to detect an edge from an input image based on the input image data, at the detection sensitivity determined by the determination unit, wherein in a case where the dispersion degree is low, the determination unit determines a lower detection sensitivity, compared with a case where the dispersion degree is high.

Abstract: A photograph generation apparatus that controls a light source and corrects a picture using an illuminance sensor and a transparent liquid crystal, the photograph generation apparatus includes: a camera part configured to take a photographed picture of a user using the photograph generation apparatus; a lighting part configured to provide lighting for an object to be photographed using the photograph generation apparatus; an illuminance sensor part configured to measure an illuminance within a predetermined range of the photograph generation apparatus to use the measured illuminance in controlling the lighting of the lighting part; and a transparent liquid crystal panel mounted at a front surface of the camera part and configured to control an amount of a light source radiated to the camera part when the picture is being taken.

Abstract: The present document relates an audio encoding and decoding system (referred to as an audio codec system). In particular, the present document relates to a transform-based audio codec system which is particularly well suited for voice encoding/decoding. A quantization unit configured to quantize a first coefficient of a block of coefficients is described. The block of coefficients comprises a plurality of coefficients for a plurality of corresponding frequency bins. The quantization unit is configured to provide a set of quantizers. The set of quantizers comprises a plurality of different quantizers associated with a plurality of different signal-to-noise ratios, referred to as SNR, respectively. The plurality of different quantizers includes a noise-filling quantizer; one or more dithered quantizers; and one or more un-dithered quantizers.

Abstract: A system and method for image segmentation are provided. A three-dimensional image data set representative of a region including at least one airway may be acquired. The data set may include a plurality of voxels. A first-level seed within the region may be identified. A first-level airway within the region may be identified based on the first-level seed. A second-level airway may be identified within the region based on the first-level airway. The first-level airway and the second-level airway may be fused to form an airway tree.

Abstract: A method of eliminating stripe noise for infrared images including: selecting a specific row of data from an to-be-filtered image Aim, and obtaining a stripe width N1 from information of a frequency spectrum of the data; intercepting (N1+1) new sequences of different lengths from a specific row sequence of the image according to the stripe width N1, and obtaining an optimal interception length N by analyzing information of amplitude spectrum of each new sequence; splitting the to-be-filtered image Aim with a size of M×Nn into a pair of optimum to-be-filtered sub-images Bim and Cim with a size of M×N according to the length N; obtaining two sub-images Bimf and Cimf after using a notch-comb filter to filter the two optimum to-be-filtered sub-images Bim and Cim; carrying out a brightness adjustment and a combination process to the two sub-image Bimf and Cimf to obtain an image Aimf.

Abstract: An image processing method and related apparatus. An image is decomposed into spatial frequency components images. The spatial frequency component images are normalized relative to specific noise models, remapped by a noise reduction function and are then combined to obtain a noise reduced version of the image.

Abstract: A biometric image processing apparatus including a memory and a processor coupled to the memory. The processor obtains a Y value, a U value and a V value in a YUV space from each pixel of an image, determines, for each pixel, whether or not the U value and the V value are in a range that is in accordance with the Y value, and extracts a pixel having been determined to be in the range.

Abstract: A method of performing an arithmetic operation by a processing apparatus includes determining a polynomial expression approximating an arithmetic operation to be performed on a variable; adaptively determining upper bits for addressing a look-up table (LUT) according to a variable section to which the variable belongs; obtaining coefficients of the polynomial expression from the LUT by addressing the LUT using a value of the upper bits; and performing the arithmetic operation by calculating a result value of the polynomial expression using the coefficients.

Abstract: Embodiments of the present disclosure relate to performing noise reduction on an input image by first filtering the input image based on coarse noise models of pixels and then subsequently filtering the filtered input image based on finer noise models. The finer noise models use the same or more number of neighboring pixels than the coarse noise filters. The first filtering and subsequent filtering of a pixel in the input image use Mahalanobis distances between the pixel and its neighboring pixels. By performing iterations of filtering using more refined noise models, the noise reduction in the input image can be performed more efficiently and effectively.

Abstract: A system and method for determining an object in a distorted image. The method comprises detecting a location of a first part of the object on the distorted image; determining a corresponding location of the first part of the object on an undistorted image plane corresponding to the distorted image based on a plurality of calibration parameters; predicting a location of a second part of the object on the undistorted image plane based on the determined location of the first part on the undistorted image plane; determining a corresponding location of the second part of the object on the distorted image based on the predicted location of the second part of the object on the undistorted image plane and the calibration parameters; and determining the object based on the detected location of the first part of the object and the determined location of the second part in the distorted image.

Abstract: A pair of video frames may be used for simultaneous localization and mapping (SLAM) initialization. The pair of frames may be determined according to a translation threshold. Whether the translation threshold is met may be determined by obtaining the pair of video frames and estimating the translation between the frames.

Abstract: A digital PSF for use in a dual modulation display. The invention allows the use of less than optimal point spread (PSF) functions in the optics between the pre-modulator and primary modulator of a dual modulation projection system. This technique uses multiple halftones per frame in the pre-modulator synchronized with a modified bit sequence in the primary modulator to produce a compensation image that reduces the errors produced by the sub-optimal PSF. The invention includes the application to dual modulation and dual modulated 3D viewing systems.

Abstract: A control device includes one or more processors. The processors detect a target captured in an image, and calculate likelihood indicating whether the target is concealed. The processors decide on an output method for outputting the target according to the likelihood. The processors generate a display image based on the image and the decided output method.

Abstract: A solid-state imaging device includes a first semiconductor substrate to which light is incident; a second semiconductor substrate stacked to the first semiconductor substrate; n first photoelectric conversion devices periodically arranged in the first semiconductor substrate and generating first electric charge signals; n first reading circuits arranged in correspondence with the n first photoelectric conversion devices in the first semiconductor substrate, respectively, each of the n first reading circuits accumulating the first electric charge signal outputting a signal voltage corresponding to the accumulated first electric charge signal as a first pixel signal; a driving circuit sequentially outputting the first pixel signal; m second photoelectric conversion devices periodically arranged in one of the first/second semiconductor substrates and generating second electric charge signals; and m second reading circuits sequentially outputting a second pixel signal, wherein m and n are natural numbers equal t

Abstract: The present disclosure relates to an image processing apparatus and an image processing method whereby a point spread function can be estimated with high precision. An exclusion unit excludes a repetitive pattern from a degraded image. An estimation unit estimates a point spread function by use of the degraded image from which the repetitive pattern has been excluded by the exclusion unit. The present disclosure can be applied to, for example, an image processing apparatus or the like that estimates a point spread function (PSF) from a degraded image and that performs a blind deconvolution using the point spread function to generate a restored image from the degraded image.

Abstract: A drive assist devices include a display that displays an image around a vehicle imaged by an imaging device installed on the vehicle, a setting unit that sets a target designated by a user on the image as a recognition target, a detection unit that detects a state change of the recognition target on the image in a case where the recognition target is set, and a notification control unit that controls a notification device to notify the user of the detection result in a case where the state change of the set recognition target is detected.

Abstract: A method for estimating an optics point spread function (518) of a lens assembly (20) includes (i) providing a test chart (14); (ii) providing a capturing system (22); (iii) capturing a test chart image (316) of the test chart (14) with the capturing system (22) using the lens assembly (20) to focus light onto the capturing system (22); (iv) selecting an image area (490) from the test chart image (316); (v) dividing the image area (490) into a plurality of area blocks (494); and (vii) estimating an area point spread function (518) of the image area (490) using a PSF cost function that sums the fidelity terms of at least two of the area blocks (494).

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: An image processing apparatus determines whether a feature amount in a partial region of an image satisfies a preset condition, selects a second image to be combined with a first image from a plurality of images based on the result of the determination, and performs a composition process on the aligned first image and second image based on a motion vector between these images. From the plurality of images, the image processing apparatus selects as the second image at least any one of an image having a partial region with the feature amount determined to satisfy the preset condition and an image corresponding to a partial region out of a plurality of partial regions set in the first image with the feature amount determined to satisfy the preset condition.

Abstract: There is provided a method performed by a device in a network camera system. First and second encoded image data representing the same scene are received (S02, SO4) over a network. The first and the second encoded image data are collected under the same light condition, albeit using different camera parameters. In particular, the second encoded image data is collected using camera parameters which simulate that the second encoded image data is collected under a darker light condition. A ratio of received data amount per time unit of the first encoded image data to received data amount per time unit of the second encoded image data is determined (S06) and the ratio is then used (S08) as an indication of a level of noise in the first encoded image data.

Abstract: An image processing apparatus includes a color separation unit separating image data into a luminance component and a color component; a processing unit generating second image data of the luminance component by manipulating pixel values of multi-resolution image data generated from first image data of the luminance component and reconstructing the manipulated multi-resolution image data; and a composition unit compositing the second image data of the luminance component with image data of the color component. Further, the processing unit generates the second image data of the luminance component based on pixel values which are adjusted by using parameters in accordance with pixel values of the image data of the color component.

Abstract: Disclosed in the present invention is a contrast constrained aerothermal radiation correction method. By analyzing features of images at different intensities of aerothermal radiation, it has been discovered that the stronger the aerothermal radiation effect is, the smaller the image contrast becomes, and when thermal radiation correction is performed using a gradient fitting algorithm, it has been discovered that time consumption thereof grows exponentially with an increase in a degree of a fitting surface and with an increase in an image size. The present invention can rapidly and effectively restore an aerothermal radiation image, remarkably improving a signal to noise ratio and quality of the image.

Abstract: According to various embodiments, an image processing apparatus includes an inputting unit configured to receive input of an image, a range information acquiring unit configured to acquire range information about a subject for each region of the input image, and a converting unit configured to assign a gradation to each region of the input image based on the range information and convert luminance data of the input image according to the assigned gradation.

Abstract: A method to enhance quality of an image is described. The method comprises receiving the image; identifying a region of the image having skin; performing motion analysis in the region of the image having skin; and if motion is detected, then controlling blending in the region of the image having skin to avoid blurring of texture in the skin.

Abstract: A method and system for decaying chrominance. One or more processors obtain a selected one of a series of root images of a digital video. The selected root image includes root pixels each associated with color values. The processor(s) selects one of the root pixels until each of the root pixels has been selected. The color values associated with the selected root pixel are expressible as a color vector with a plurality of elements each storing a different one of the color values. The processor(s) determines a perceptual luminance value for the selected root pixel, generates a monochromic vector for the selected root pixel, generates a biased monochromic vector by multiplying the monochromic vector with a bias, and generates new color values associated with a new pixel of a denoised image corresponding to the selected root pixel by blending the biased monochromic vector with the color vector.

Abstract: Embodiments generally relate to systems and methods for in-vehicle optical image stabilization. In one embodiment, the system includes a first motion sensor rigidly attached to a video capture device mounted in a vehicle; a second motion sensor rigidly attached to an object in the vehicle, the object being within the field of view of the video capture device; and a controller. The controller is operatively connected to the first motion sensor, to the second motion sensor, and to a movable optical element in the video capture device. In one aspect, the first sensor provides a first output signal to the controller, the second sensor provides a second output signal to the controller; and the controller provides a compensation signal to the movable element determined by the first and second output signals.

Abstract: Systems and methods are disclosed for calibrating an image sensor using a source image taken from the image sensor and comparing it to a reference image. In one embodiment, the method may involve determining the luminance and chrominance values of portions of the image at successive frequency levels and calculating a standard deviation at each frequency level for both the source image and the reference image. The standard deviation values may be compared and a difference determined. Using unit vector search vectors, noise values may be calculated to determine sensor calibration values.

Abstract: Techniques for detecting an event via social media content. A method includes obtaining multiple images from at least one social media source; extracting at least one visual semantic concept from the multiple images; differentiating an event semantic concept signal from a background semantic concept signal to detect an event in the multiple images; retrieving one or more images associated with the event semantic concept signal; grouping the one or more images associated with the event semantic concept signal; annotating the group of one or more images with user feedback; and displaying the annotated group of one or more images as a visual description of the detected event.

Abstract: A method and system can analyze, reconstruct, and/or denoise an image. The method and system can include interpreting a signal as a potential of a Schrödinger operator, decomposing the signal into squared eigenfunctions, reducing a design parameter of the Schrödinger operator, analyzing discrete spectra of the Schrödinger operator and combining the analysis of the discrete spectra to construct the image.

Abstract: In various embodiments, an image stabilization system includes a dual-path, processing pipeline in which a first path serves as a stabilization path, and a second path serves as a preview path. An image signal processor produces a stream of image frames for each path. Frames that are received by the stabilization path undergo stabilization processing to produce post processed frames that are cropped in a manner that provides for visual stabilization. Frames that are received by the preview path are processed to provide a real-time preview while corresponding frames are processed by the stabilization path.

Abstract: Methods and systems are presented for organizing images. In one aspect, a method can include generating a correlation value indicating a likelihood that a face included in a test image corresponds to a face associated with a base image, determining that a correlation threshold exceeds the correlation value and that the correlation value exceeds a non-correlation threshold, generating a similarity score based on one or more exposure values and one or more color distribution values corresponding to the test image and the base image, combining the similarity score with the correlation value to generate a weighted correlation value, and determining that the test image and the base image are correlated when the weighted correlation value exceeds the correlation threshold.

Abstract: A mechanism is described for facilitating depth image dequantization at computing devices according to one embodiment. A method of embodiments, as described herein, includes detecting a digital image of an object, the digital image including pixels having associated pixel values contaminated by noise, and side information pertaining to confidence of a value acquired in each pixel. The method may further include measuring characteristics of noise in each pixel of the digital image, and a plurality of weights relating to one or more of the pixel values, the side information, and the noise characteristics. The method may further include computing a smart filter based on a combination of the plurality of weights, applying the smart filter to filter the digital image by reducing the noise in the digital image, and outputting the filtered digital image.