Abstract: This disclosure provides methods, devices, and systems for low-light imaging. In some implementations, an image processor may be configured to reduce or remove noise associated with an image based, at least in part, on a neural network. For example, the neural network may be trained to infer a denoised representation of the image. In some aspects, the image processor may scale the brightness level of the image to fall within a normalized range of values associated with the neural network. In some other aspects, a machine learning system may scale the brightness levels of input images to match the brightness levels of ground truth images used to train the neural network. Still further, in some aspects, the machine learning system may scale the brightness levels of the input images and the brightness levels of the ground truth images to fall within the normalized range of values during training.

Abstract: Methods and systems described herein generate content that quickly cycles between different-colored frames at a rapid rate that may be imperceptible to human viewers. The human eye tends to blend nearby frames together, so that a human viewer will see a full color spectrum even though only a single color is displayed at any given time. The video display may repeatedly cycle between individual color frames, thus providing video content that appears normal to a human viewer. However, a video recording device may capture only some of the color frames. Therefore, the video recorded by video recording devices may include undesirable color flicker or other color artifacts, which may deter unauthorized copying.

Abstract: Disclosed are methods and devices useful in the field of image processing and display Disclosed are also methods and devices useful in the field of ophthalmology and, in some particular embodiments, useful for the non-invasive treatment and/or prevention of refractive errors.

Abstract: A layer of additive material is formed in a circular printing area on a substrate using additive sources distributed across a printing zone. The additive sources form predetermined discrete amounts of the additive material. The substrate and the additive sources are rotated with respect to each other around a center of rotation, so that a pattern of the additive material is formed in a circular printing area on the substrate. Each additive source receives actuation waveforms at an actuation frequency that is proportional to a distance of the additive source from the center of rotation. The actuation waveforms include formation signals, with a maximum of one formation signal in each cycle of the actuation frequency. The formation signals result in the additive sources forming the predetermined discrete amounts of the additive material on the substrate.

Abstract: A method is provided that includes receiving points of an image in which objects are depicted. The method includes performing a greedy nearest-neighbor (GNN) cluster analysis of the image to group the points of the image. The GNN cluster analysis includes grouping the points into a plurality of local GNN clusters, from a greedy analysis using a k-d tree in which the points are organized. The plurality of local GNN clusters are then extended into a plurality of global GNN clusters. This evaluating similarity of the local GNN clusters, merging the local GNN clusters into a global GNN cluster when each of the defined similarity criteria is evaluated to true, and passing the local GNN clusters as global GNN clusters when any of the defined similarity criteria is evaluated to false. The method then includes detecting the objects depicted in the image based on the global GNN clusters.

Abstract: Systems and methods for detecting surface anomalies are disclosed. For example, a computer-implemented method for detecting surface anomalies on an object comprises receiving measured electromagnetic radiation (EMR) profiles for the object, generating synthetic EMR profiles for the object based on the measured EMR profiles, determining whether the object contains a surface anomaly based on the measured EMR profiles and the synthetic EMR profiles, and indicating a surface anomaly to a user via a display when a surface anomaly is detected. In another example, a system comprises a computing device comprising non-transitory memory with computer-readable instructions for receiving unpaired image data of an object of two different image types, predicting missing image data to generate paired image data of the two different image types, and determining whether the object contains a surface anomaly based on the paired image data.

Abstract: An apparatus and method for encoding objects in a camera-captured image with a deep neural network pipeline including multiple convolutional neural networks or convolutional layers. After identifying at least a portion of the camera-capture image, a first convolutional layer is applied to the at least the portion of the camera-captured image and multiple subregion representations are pooled from the output of the first convolutional layer. One or more additional convolutions are performed. At least one deconvolution is performed and concatenated with the output of one or more convolutions. One or more final convolutions are performed. The at least the portion of the camera-captured image is classified as an object category in response to an output of the one or more final convolutions.

Abstract: A method includes detecting a gradation area in an input image, and performing, based on a detection result acquired by the detecting, for an area that is not the gradation area, gain processing on the input image by using a gain based on a first gain characteristic in which different gains are set depending on luminance values, whereas performing, for the gradation area, gain processing on the input image by using a gain in which an amount of change in gain with respect to a change in luminance value is more reduced than the first gain characteristic.

Abstract: A depth measuring apparatus includes a camera assembly configured to capture a plurality of images of a target at a plurality of distances from the target. The depth measuring apparatus further includes a controller configured to, for each of a plurality of regions within the plurality of images: determine corresponding gradient values within the plurality of images; determine a corresponding maximum gradient value from the corresponding gradient values; and determine, based on the corresponding maximum gradient value, a depth measurement for a region of the plurality of regions.

Abstract: Embodiments of the present disclosure provide a method and apparatus for processing a video frame, and relates to the field of computer vision technology. The method may include: acquiring a plurality of candidate first-order radial distortion parameters preset for a to-be-processed video frame, and acquiring a specified value of a specified radial distortion parameter; performing radial distortion correction on the to-be-processed video frame to obtain a first initial corrected video frame; selecting a first initial corrected video frame in which a local region except for a center region after distortion correction includes a largest number of straight line segments; and determining a candidate first-order radial distortion parameter corresponding to the selected first initial corrected video frame for use as a target first-order radial distortion parameter of the to-be-processed video frame.

Abstract: Provided are an apparatus and method for detecting ventricular depolarization (QRS) of an electrocardiogram (ECG). The apparatus includes an input unit configured to receive an ECG signal, a memory configured to store a program for detecting R and ventricular depolarization using the ECG signal, and a processor configured to execute the program, wherein the processor detects a QRS interval and an R-peak using a first-order derivative filter and a max-filter.

Abstract: Some embodiments relate to sharpening segments of an image differently based on content in the image. Content based sharpening is performed by a content image processing circuit that receives luminance values of an image and a content map. The content map identifies categories of content in segments of the image. Based on one or more of the identified categories of content, the circuit determines a content factor associated with a pixel. The content factor may also be based on a texture and/or chroma values. A texture value indicates a likelihood of a category of content and is based on detected edges in the image. A chroma value indicates a likelihood of a category of content and is based on color information of the image. The circuit receives the content factor and applies it to a version of the luminance value of the pixel to generate a sharpened version of the luminance value.

Abstract: In order to more accurately white balance an image, weightings can be determined for pixels of an image when computing an illuminant color value of the image and/or a scene. The weightings can be based at least in part on the Signal-to-Noise Ratio (SNR) of the pixels. The SNR may be actual SNR or SNR estimated from brightness levels of the pixels. SNR weighting (e.g., SNR adjustment) may reduce the effect of pixels with high noise on the computed illuminant color value. For example, one or more channel values of the illuminant color value can be determined based on the weightings and color values of the pixels. One or more color gain values can be determined based on the one or more channel values of the illuminant color value and used to white balance the image.

Abstract: An edge-based sharpness strength control circuit include a first edge determination unit suitable for determining each of multiple regions in a pixel array to be an edge region having edge information on pixel data of the corresponding region or a flat region having flat information on pixel data of the corresponding region, the pixel array including a plurality of pixels; a second edge determination unit suitable for determining each edge region to be a step edge region having directional information within the corresponding edge region or a texture edge region having non-directional information within the corresponding edge region; and a noise removing unit suitable for removing noise from each step edge region using a first filter and for removing noise from each texture edge region using a second filter having a different gain than that of the first filter.

Abstract: In an embodiment, a method includes: providing a gray-coded time reference to a time-to-digital converter (TDC); receiving an event from an event signal; latching the gray-coded time reference into a memory upon reception of the event signal; and updating a time-of-flight (ToF) histogram based on the latched gray-coded time reference.

Abstract: Embodiments relate to lateral chromatic aberration (LCA) recovery of raw image data generated by image sensors. A chromatic aberration recovery circuit performs chromatic aberration recovery on the raw image data to correct the resulting LCA in the full color images using pre-calculated offset values of a subset of colors of pixels.

Abstract: Disclosed are an image processing method and device, and a three-dimensional imaging system. The method comprises the following steps of: acquiring a two-dimensional image to be processed; aligning the two-dimensional image to be processed to a grid template; performing mapping processing on the two-dimensional image to be processed by using a grid mapping table to acquire a first image, wherein the grid mapping table is used for representing the mapping relationship of grid images; mirroring the first image to acquire a second image; and synthesizing the first image and the second image to acquire the superimposed image of the first image and the second image.

Abstract: Devices, systems, and methods obtain an image; perform at least one nonlinear locally-adaptive mapping on the image, thereby generating a revised image; generate an edge-enhanced image based on the revised image; and perform dynamic contrast stretching on the edge-enhanced image, thereby generating an enhanced image.

Abstract: An image analyzer is configured to use an image processing technique to detect coordinates of a region of a first image frame that matches a target template and update the target template by copying a portion of the first image frame corresponding to the coordinates and a first size. The image analyzer is also configured to receive a second image frame and generate a plurality of pixel match counts based on a comparison of the target template with a plurality of regions of the first size of the second image frame. The image analyzer is further configured to determine that a region of the second image frame matches the target template based on a pixel match count of the second region. The image analyzer is configured to generate an output indicating coordinates of the region of the second image frame.

Abstract: An image signal processor includes a first sharpening filter configured to perform a first filtering operation on a raw data output from an external image device to output first sharpness data. A second sharpening filter is configured to perform a second filtering operation on the raw data in parallel with the first filtering operation to output second sharpness data. An edge determiner is configured to determine an edge strength of an image corresponding to the raw data based on the raw data and to adjust a first weight and a second weight based on the determined edge strength. A combiner configured to combine first weight data in which the adjusted first weight is reflected in the first sharpness data and second weight data in which the adjusted second weight is reflected in the second sharpness data to generate output data.

Abstract: A computer implemented method of fingertip centroid identification in real-time, implemented on a computer system comprising a processor, memory, and a camera system. The processor receives image data from the camera system; runs a first kernel comprising a set of concentric closed shapes over image data to identify an occupancy pattern in which internal closed shapes are at least nearly fully occupied, and in which a subsequent closed shape has at least a relatively low occupancy level, so as to identify one or more fingertips in the image data; for each identified fingertip, runs a second kernel over the identified one or more fingertips to establish a best fit closed shape which covers each identified fingertip; calculates a centroid for each best fit closed shape which corresponds to an identified fingertip; and stores in the memory the calculated centroids for the identified one or more fingertips.

Abstract: Devices, methods, and computer-readable media describing an adaptive approach for image selection, fusion, and noise reduction, e.g., to generate low noise and high dynamic range (HDR) images with improved motion freezing in a variety of capturing conditions. An incoming image stream may be obtained from an image capture device, wherein the image stream comprises a variety of differently-exposed captures, e.g., EV0 images, EV? images, EV+ images. When a capture request is received, a set of rules may be used to evaluate one or more capture conditions associated with the images from the incoming image stream and determine which two or more images to select for a fusion operation. The fusion operation may be designed to adaptively fuse the selected images, e.g., in a fashion that is determined to be optimal from a noise variance minimization standpoint. A fusion-adaptive noise reduction process may further be performed on the resultant fused image.

Abstract: The present disclosure provides a hybrid framework-based image bit-depth expansion method and device. The invention fuses a traditional de-banding algorithm and a depth network-based learning algorithm, and can remove unnatural effects in an image flat area whilst more realistically restoring numerical information of missing bits. The method comprises the extraction of image flat areas, local adaptive pixel value adjustment-based flat area bit-depth expansion and convolutional neural network-based non-flat area bit-depth expansion. The present invention uses a learning-based method to train an effective depth network to solve the problem of realistically restoring missing bits, whilst using a simple and robust local adaptive pixel value adjustment method in an flat area to effectively inhibit unnatural effects in the flat area such as banding, a ringing and flat noise, improving subjective visual quality of the flat area.

Abstract: In order to produce a discriminator that has higher discrimination ability, this image-processing device is provided with a synthesis unit for synthesizing a background image and an object image the hue and/or brightness of which at least partially resembles at least a portion of the background image, a generation unit for generating a difference image between the synthesized image and the background image, and a machine learning unit for performing machine learning using the generated difference image as learning data.

Abstract: Determining contextual confidence of images for associative deep learning includes receiving an image including a representation of a subject. Text data related to the image is received. One or more physical properties of the image are determined. Context information of the image is determined using natural language processing. The image is classified based upon the contextual information and the one or more physical properties using a classification model to determine a classification. An emotional state of the image is determined based upon the physical properties. A confidence of the classification and emotional state is determined.

Abstract: Methods and apparatus to determine the dimensions of a region of interest of a target object and a class of the target object from an image using target object landmarks are disclosed herein. An example method includes identifying a landmark of a target object in an image based on a match between the landmark and a template landmark; classifying a target object based on the identified landmark; projecting dimensions of the template landmark based on a location of the landmark in the image; and determining a region of interest based on the projected dimensions, the region of interest corresponding to text printed on the target object.

Abstract: A system and method for improving the detail of an input digital video signal can be implemented by computing first and second order gradients of the input digital video signal in a neighborhood size of 1,2, 4,8 and 16 increments. This gradient information can be used to construct an output digital video signal that has greater detail than the input digital video signal.

Abstract: The disclosure pertains to techniques for image processing. One such technique comprises a method for image processing comprising obtaining first light information from a set of light-sensitive pixels for a scene, the pixels including phase detection (PD) pixels and non-PD pixels, generating a first PD pixel image from the first light information, the first PD pixel image having a first resolution, generating a higher resolution image from the plurality of non-PD pixels, wherein the higher resolution image has a resolution greater than the resolution of the first PD pixel image, matching a first pixel of the first PD pixel image to the higher resolution image, wherein the matching is based on a set of correlations between the first pixel and non-PD pixel within a predetermined distance of the first pixel, and determining a disparity map for an image associated with the first light information, based on the match.

Abstract: A system and a method for searching an image within another image are disclosed. The method includes producing template edge images and target edge images, having image scales, based on determination of edge gradients of a template image and a target image in one or more directions. The template image indicates an image to be searched. The target image indicates another image within which the image needs to be searched. Further, images comprising correlation coefficient values are produced for each of the directions by computing correlation coefficients between the template edge images and the target edge images. At least one local peak is identified from each of the images comprising the correlation coefficient values. Spatial locations along with the correlation coefficients corresponding to the local peak are determined. Thereafter, a presence of the template image in the target image is identified based upon an intersection of the spatial locations.

Abstract: A camera system captures images from a set of cameras to generate binocular panoramic views of an environment. The cameras are oriented in the camera system to maximize the minimum number of cameras viewing a set of randomized test points. To calibrate the system, matching features between images are identified and used to estimate three-dimensional points external to the camera system. Calibration parameters are modified to improve the three-dimensional point estimates. When images are captured, a pipeline generates a depth map for each camera using reprojected views from adjacent cameras and an image pyramid that includes individual pixel depth refinement and filtering between levels of the pyramid. The images may be used generate views of the environment from different perspectives (relative to the image capture location) by generating depth surfaces corresponding to the depth maps and blending the depth surfaces.

Abstract: Efficient noise reduction coding may include generating, by a processor, an encoded frame by encoding an input video frame. Encoding the input frame includes determining an estimated noise level for the input video frame, determining a minimum quantization parameter level based on the estimated noise level for the input video frame, determining a quantization parameter value for encoding the input frame such that the quantization parameter value is at least the minimum quantization parameter level, obtaining the encoded frame by encoding the input frame in accordance with the quantization parameter level, including the encoded frame in an output bitstream, and outputting the output bitstream.

Abstract: Methods for partition-based image tracking are provided. In one aspect, a method includes determining a number of partitions for each image frame of image frames and dividing a first image frame of the image frames into first partitions and a second image frame of the image frames into second partitions based on the number of partitions. The method also includes detecting for a change between each of the first partitions of the first image frame and a corresponding second partition of the second partitions of the second image frame. The method also includes configuring priority settings for the second partitions based on the detection, and processing the second image frame and a third image frame based on the priority settings to detect for a change between each of the second partitions and a corresponding third partition of third partitions of a third image. Systems and machine-readable media are also provided.

Abstract: Techniques related to temporal noise reduction of images are discussed. Such techniques may include generating a noise stream corresponding to an input image and adaptively re-combining the noise stream with a reference image corresponding to the input image and a spatially noise reduced image corresponding to the input image to generate a temporal noise reduced output image.

Abstract: Disclosed is a method for processing a video signal using graph-based transform and a device for the method. A method for processing a video signal using graph-based transform, including specifying a set of available graph configurations in order to process the video signal, extracting signal structure information from samples within a processing block of the image, configuring a graph for the samples within the processing block using the signal structure information, determining transform for the processing block using the graph, and transforming the samples within the processing block using the transform. The signal structure information may be defined as information about a graph configuration which belongs to the set of specified graph configurations and which is applied to each of the samples within the processing block.

Abstract: An information processing apparatus detects a line segment from an image, determines whether four sides of a plate are detected based on the detected line segment, calculates an aspect ratio of the detected four sides in a case where it is determined that four sides of the plate are detected, and determines whether the aspect ratio satisfies a first condition. In a case where it is determined that four sides of the plate are not detected, the information processing apparatus searches a vicinity of the detected line segment for an object, and determines whether the object satisfies a second condition. The information processing apparatus judges that a recognition target area is included in the image in a case where it is determined that the first condition or the second condition is satisfied.

Abstract: To detect annotation lines in medical image data. Horizontal annotation pixel determination means obtains the color component value difference between each pixel of a predetermined number of connected adjacent pixels in a first direction of the target pixel and an adjacent pixel thereof. If the total number of pixels having color component value differences, of the predetermined number of pixels is equal to or smaller than a first threshold, the horizontal annotation pixel determination means determines that the target pixel is an annotation pixel. If annotation pixels are successive in the horizontal direction in a predetermined number, horizontal annotation line determination means determines that the annotation pixels form an annotation line. The same applies to the vertical direction. The determined annotation lines are provided to border detection means.

Abstract: One or more embodiments of the invention relate to an image processing system and method for detail enhancement and noise reduction, in which the method includes: (a) an original image is created, (b) an information measure is calculated on the basis of the original image, (c) a weighting measure is calculated on the basis of the information measure, (d) the original image is low-pass filtered with a low-pass filter to form a low-pass filtered image, (e) a high-pass filtered image is calculated by subtracting the low-pass filtered image from the original image, (f) a detail-enhanced and noise-reduced image is obtained by a high-pass image scaled with the weighting measure being added to the low-pass image. One or more embodiments of the invention additionally relate to an image processing device comprising an image recording device, an image processing unit and an image display unit.

Abstract: A super-resolution observation device includes an illumination optical system collecting a first illuminating light having a first optical frequency ?1 on a first region of an observation object, collecting a second illuminating light having a second optical frequency ?2? on a second region partially overlapping the first region, and collecting a third illuminating light having a third optical frequency ?2 on a third region containing a non-overlap region which is a region of the first region and does not overlap the second region; and an extraction unit extracting a signal light generated in accordance with a change in an energy level of a substance in the non-overlap region from a light generated in all of the first region, the second region, and the third region.

Abstract: Systems and methods are provided for image enhancement using self-examples in combination with external examples. In one embodiment, an image manipulation application receives an input image patch of an input image. The image manipulation application determines a first weight for an enhancement operation using self-examples and a second weight for an enhancement operation using external examples. The image manipulation application generates a first interim output image patch by applying the enhancement operation using self-examples to the input image patch and a second interim output image patch by applying the enhancement operation using external examples to the input image patch. The image manipulation application generates an output image patch by combining the first and second interim output image patches as modified using the first and second weights.

Abstract: Innovations in rendering of high dynamic range (“HDR”) content are described. A playback system can adjust a tone mapping function used when rendering the HDR content, thereby improving quality when rendering the HDR content. For example, the playback system includes decision logic, which is configured to receive summary information (e.g., histogram information) for sample values of a picture of HDR content, and set a tone mapping function based at least in part on the summary information. The tone mapping function can be selected from among multiple available tone mapping functions, and parameters of the selected tone mapping function can be determined based at least in part on the summary information. A tone mapper is configured to apply tone mapping to input values for the sample values of the picture of HDR content, according to the tone mapping function, and thereby produce output values.

Abstract: A method for highlighting edges of interest in an image comprising the following steps: calculating edge intensity of edges in the image using gradient operators; applying a transformation to emphasize mid-level edges and deemphasize strong edges; and if a given transformed edge is greater than a threshold value, highlighting the corresponding edge in an edge-highlighted image.

Abstract: A method for determining leading edge location in an image captured by an optical navigation sensor with auto shutter adaptation includes: plotting average pixel values per sensor column on a graph; determining an average value according to the average pixel values as a threshold value; defining a weighted location in the leading half of the captured image according to the threshold value; and determining the leading edge location according to the defined weighted location.

Abstract: An imaging system configured to display a captured image on a display having a lower dynamic range than the imaging system is provided, wherein the imaging system includes a high dynamic range imager configured to capture at least one high dynamic range image, and a processing device communicatively connected to the high dynamic range imager, wherein the processing device is configured to reduce color speckles caused by errant interpolated pixel values and reduce occurrence of errant values of alternate color while maintaining at least one of sharpness detail, edge detail, and generate smoothing color rendering in a displayed image.

Abstract: In one embodiment, a method includes adjusting the images or text rendered on a display based on the position of the viewers relative to the display.

Abstract: Efficiency in monitoring documents is promoted by quantifying the priorities of the documents to be monitored. A document analysis system (1) includes: a score calculation unit (116) that calculates a score indicating strength of linkage between a document included in document information and a classification code indicating the relevance between the document information and a lawsuit or a fraud investigation; and a score report unit (701) that reports the score to a user according to the calculated score.

Abstract: Techniques, methods, and systems for image processing may be provided. The image processing may be provided for upsampling and interpolating images. The upsampling and interpolating may include interpolating the image through at least an edge weight and a spatial weight. In various embodiments, the edge weight and/or the spatial weight may be calculated with a kernel. The kernel may be a kernel with a two dimensional (2D) distribution such as a Gaussian kernel, a Laplacian kernel, or another such statistically based kernel. The image processing may also include refining the upsampled and interpolated image through a refinement weight calculation and/or through back projection.

Abstract: A method for image processing, including: obtaining an image including a writing board and a background external to the writing board; detecting a plurality of lines within the image; determining, based on the plurality of lines, a plurality of corners of the writing board within the image; and correcting a perspective of the writing board by applying a transformation to the image based on the plurality of corners.

Abstract: Techniques are described for displaying projects of images as “collages”. Collages differ from conventional thumbnail displays of projects in that collages display an entire project as if the project were a single image. Consequently, collages better convey the characteristics of projects as a whole, while de-emphasizing the distinctiveness of individual images within the projects. When displayed as collages, side-by-side comparisons may be readily performed between projects as a whole. For example, a single display may include collages for multiple projects, thereby allowing viewers to quickly tell how the projects differ in a variety of ways, including but not limited to size of shoot or density of shoot, dominant color, mood, time of day, bracketed shots or bursts, location and subject matter. The content of the collage for a project is based on the individual images that belong to the project.

Abstract: A method includes generating, by a processor, a label map by labeling pixels in an image. A percentage of a first and second type of texture pixels is determined based on the label map to generate a texture map. The texture map is updated. A directional edge map is generated based on a data correlation analysis based on the texture map and the label map. A gain map with boundaries and smoothing is generated using localized pixels based on information from one or more of the label map, the texture map, and the directional edge map. An enhancement controlled upscaled image frame is generated using one or more of the label map, the texture map, the directional edge map and the gain map as a weight factor.

Abstract: A depth image filtering method, and a filtering threshold obtaining method and apparatus are provided. The method in the embodiments of the present invention includes: determining, for each pixel in an adjacent area of a pixel of a to-be-filtered depth image, whether the pixel meets a preset condition; determining a set of pixels meeting the preset condition; and determining a pixel value of the pixel of the to-be-filtered depth image according to pixel values of the pixels in the set. According to the embodiments of the present invention, a ringing effect at an edge of a depth image is effectively removed, and discontinuity of the depth image is reduced, thereby improving quality of a video image.