Abstract: An apparatus for scaling images includes one or more processors and one or more computer-readable storage media on which computer-executable instructions are stored. The computer-executable instructions, upon being executed by the one or more processors, provide for execution of the following steps: capturing one or more first images by an imaging and/or image recording system, wherein the one or more captured first images are related to a first resolution; and generating, by a neural network, one or more corresponding second images based on one or more captured first images, wherein the one or more second images are related to a second resolution, the first resolution differing from the second resolution.

Abstract: An image display method, an image display device, and a mobile terminal are provided, which are applied to a display panel including a plurality of pixel units having A pixel unit rows and B pixel unit columns, including: acquiring data image information including C pixel data rows and D pixel data columns of a plurality of pixel data, wherein C<A and/or D<B, and input the plurality of pixel data to multiple data lines and enable pixel unit rows, inputting the plurality of pixel data into a plurality data lines corresponding to the B pixel unit columns, and enabling the A pixel unit rows sequentially, to make at least part of the pixel units display according to a same pixel data.

Abstract: A magnification system for magnifying an image based on trained neural networks is disclosed. The magnification system receives a first user input associated with a selection of a region of interest (ROI) within an input image of a site and a second user input associated with a first magnification factor of the selected ROI. The first magnification factor is associated with a magnification of the ROI in the input image. The ROI is modified based on an application of a first neural network model on the ROI. The modification of the ROI corresponds to a magnified image that is predicted in accordance with the first magnification factor. A display device is controlled to display the modified ROI.

Abstract: A system is adapted to perform an image processing method. The processing method includes: obtaining input image data, a first training result, a second training result, and an interpolation lookup table; segmenting the input image data into a plurality of feature blocks according to a total quantity of area interpolations; establishing a position mapping relationship to record the feature blocks corresponding to positions of all of the area interpolations; assigning corresponding area interpolations to the feature blocks according to the position mapping relationship; obtaining an interpolation parameter for each of the feature blocks according to the first training result, the second training result, and the area interpolation; performing block convolution on each of the interpolation parameters and the corresponding feature block to obtain an output feature result; and obtaining an output image by combining the output feature results according to the position mapping relationship.

Abstract: Systems and methods for down-scaling are provided. In one example, a method for processing image data includes determining a plurality of output pixel locations using a position value stored by a position register, using the current position value to select a center input pixel from the image data and selecting an index value, selecting a set of input pixels adjacent to the center input pixel, selecting a set of filtering coefficients from a filter coefficient lookup table using the index value, filtering the set of source input pixels to apply a respective one of the set of filtering coefficients to each of the set of source input pixels to determine an output value for the current output pixel at the current position value, and correcting chromatic aberrations in the set of source input pixels.

Abstract: A system for super-sampling digital images detects artifacts in an SRGAN super-sampled image, determines blocks of the image that contribute to the artifacts, and if the artifact-contributing blocks exceed a threshold, discards the SRGAN generated output image in favor of applying a super-sampled image generated by an alternate mechanism, such as a nearest neighbor algorithm.

Abstract: The present invention discloses system and method for enhancing an image based on non-local features. The invention includes an end-to-end trainable and guided method including feature extraction block, non-local feature generator and non-local feature enhancement block, to deal with low-level image problem by using the non-local feature concept. The invention also deploys a non-local feature merge block to rectify translated features and improve non-local feature further to finally reconstruct the rectified features to form an enhanced image.

Abstract: A signal processing device includes: a data receiver configured to transmit a transmission wave a position of which changes in a first direction and which spreads in a second direction orthogonal to the first direction, and generate a matrix of acquired observation data in the first and second directions, as a reception signal matrix, with a value of a position in the reception signal depending on a signal strength; a range processor configured to specify a range in the second direction in an imaging matrix, and set a sparse vector including a component in the first direction of the specified range; a reconstruction processor configured to perform reconstruction processing using the reception signal matrix and the sparse vector to calculate a component of the sparse vector; and a synthesis processor configured to synthesize the resulting sparse vector while moving the range, to generate an imaging matrix.

Abstract: Method for displaying super-resolution video of at least one-moving object without image artifacts, including the procedures of acquiring microscanned images of at least one moving object, a first and second subset of the images respectively forming a first and second data set, for each data set, analyzing at least a portion of the sub-set of the images for spatial and temporal information, determining a respective movement indication of the moving object according to the spatial and temporal information, in parallel to the procedure of analyzing, forming a respective super-resolution image from each data set and designating a respective bounded area surrounding the moving object, and repeatedly displaying each super-resolution image outside the bounded area a plurality of times at a video frame rate and displaying during those times within the respective bounded area, a plurality of consecutive microscanned images of the moving object at the video frame rate.

Abstract: Embodiments provide systems and methods which facilitate optimization of a convolutional neural network (CNN). One embodiment provides for a non-transitory machine-readable medium storing instructions that cause one or more processors to perform operations comprising processing a trained convolutional neural network (CNN) to generate a processed CNN, the trained CNN having weights in a floating-point format. Processing the trained CNN includes quantizing the weights in the floating-point format to generate weights in an integer format. Quantizing the weights includes generating a quantization table to enable non-uniform quantization of the weights and quantizing the weights from the floating-point format to the integer format using the quantization table. The operations additionally comprise performing an inference operation utilizing the processed CNN with the integer format weights.

Abstract: A method, computer program, and computer system is provided for video coding. Video data including one or more views is received. A composite depth image and a composite texture image corresponding to the one or more views are generated based on the received video data. The video data is decoded based on the generated composite depth image and composite texture image.

Abstract: Systems, methods, and software described herein manage video data processing resources for video data obtained from one or more sources. In one implementation, a management system may monitor processing requirements for the video data and computing resources available at multiple video processing locations. The management system may further allocate processing operations to the video processing locations based on the processing requirements for the video data and computing resources available at the video processing locations.

Abstract: Methods and systems for training a neural network are provided. In one example, an apparatus comprises a memory that stores instructions; and a hardware processor configured to execute the instructions to: control a neural network processor to perform a loss gradient operation to generate data gradients; after the loss gradient operation completes, control the neural network processor to perform a forward propagation operation to generate intermediate outputs; control the neural network processor to perform a backward propagation operation based on the data gradients and the intermediate outputs to generate weight gradients; receive the weight gradients from the neural network processor; and update weights of a neural network based on the weight gradients.

Abstract: Techniques are described for video encoding and decoding using reference picture resampling with switchable filters. One example involves obtaining a current picture and a reference picture, identifying filter index information for a current block of the current picture, and determining that a first picture size value of the current picture and a second picture size value of the reference picture are different. Based on the determining that the first picture size value of the current picture and the second picture size value of the reference picture are different, performing a resampling process using a default filter index in place of a current filter index identified by the filter index information. Additional examples can use the current filter index identified by the filter index information in subsequent blocks. In various examples, the current filter index can be derived or signaled.

Abstract: An electro-optic display having at least one row of display pixels, the display include a first display pixel of the at least one row of display pixels, the first display pixel coupled to a first bias line, and a second display pixel of the at least one row of display pixels, the second display pixel coupled to a second bias line, wherein the second bias line is different from the first bias line.

Abstract: Disclosed are a method and apparatus for presenting material, and a storage medium. The method includes acquiring at least two key points from a position of a presentation part of an object in an image; determining a preselected target point based on positions of the at least two key points; determining a target point of the image based on the preselected target point and target points of N continuous frames before the image, and presenting the material based on the target point.

Abstract: Systems and methods for converting low resolution images into high resolution images are provided and can include (1) processing a compressed version of images with a first set of heuristics and rules of an artificial intelligence model to produce an intermediate version of the images with an intermediate resolution that is greater than the compressed resolution by first enhancing the images and subsequently upscaling the images as enhanced from the compressed resolution to the intermediate resolution and (2) processing the intermediate version of the images with a second set of heuristics and rules of the artificial intelligence model to produce a final version of the images with a final resolution that is greater than the initial resolution and the intermediate resolution by first upscaling the images from the intermediate resolution to the final resolution and subsequently enhancing the images as upscaled to produce the final version of the images.

Abstract: In one implementation, a method includes obtaining, for a particular pixel of an image, a real pixel value and a virtual pixel value. The method includes obtaining, for the particular pixel of the image, a first alpha and a second alpha. The method includes generating, for the particular pixel of the image, a combined pixel value as a weighted average of the real pixel value and the virtual pixel value, the weighting being based on at least one of the first alpha and the second alpha.

Abstract: Methods and systems for creating TEM lamella using image restoration algorithms for low keV FIB images are disclosed. An example method includes irradiating a sample with an ion beam at low keV settings, generating a low keV ion beam image of the sample based on emissions resultant from irradiation by the ion beam, and then applying an image restoration model to the low keV ion beam image of the sample to generate a restored image. The sample is then localized within the restored image, and a low keV milling of the sample is performed with the ion beam based on the localized sample within the restored image.

Abstract: A learning model generating device includes a first image reading device and a first control device. The first control device includes a processor and functions, through the processor executing a first control program, as a first segmenter, a learning model generator, and a first compressor. The first segmenter segments each of images of training prints obtained by reading performed by the first image reading device. The learning model generator learns segmented images to generate a first learning model for use in inferring a type of an image defect. The first compressor compresses each of the images of the training prints. The first segmenter segments each of compressed images obtained by compression performed by the first compressor. The learning model generator learns compressed and segmented images to generate a second learning model for use in inferring a type of an image defect.

Abstract: A photographing method and apparatus is provided. The photographing apparatus includes a photographing unit; a sensing unit for sensing motion of the photographing apparatus; a display unit for displaying at least one guide image for panorama photographing; a controller for controlling the photographing unit to automatically photograph, if a photographing direction that changes in accordance with motion of the photographing apparatus corresponds to one of the at least one guide image; and a storage unit for storing the photographed image data.

Abstract: A method for picture processing is provided. The method includes the following. Multiple pictures are obtained through photographing. At least two pictures are selected from the multiple pictures, and according to facial information of each of the at least two pictures, feature information of a preset facial part in each of the at least two pictures is obtained. One or more target pictures are determined in response to the feature information of the preset facial part in one of the at least two pictures indicating that the preset facial part is in a preset feature state, where the one of the at least two pictures is determined as one target picture. When multiple target pictures are obtained, a multi-frame denoising processing is performed on the target pictures to obtain an output picture.

Abstract: To implement non-ejection complementation processing while securing a chance for a nozzle in which an ejection failure has occurred to recover as a normal nozzle. Positional information for specifying an abnormal nozzle in which an ejection failure has occurred among a plurality of nozzles arrayed in a nozzle column is acquired and based on the positional information, input image data is converted into halftone image data. At this time, at least part of densities that are assumed to be obtained by a pixel line of the abnormal nozzle are distributed to pixels on a pixel line of a neighboring nozzle that is located in close proximity to the abnormal nozzle and in which no ejection failure has occurred so that in the halftone image data, at least in part of tones, the pixel on the pixel line of the abnormal nozzle has a density.

Abstract: Utilizing random variation (repeated positioning error) when reciprocating operation is repeatedly performed in which a stage is moved by (+x, +y) pulses toward an arbitrary position perpendicular to an optical axis of X-rays extending from an X-ray source to an X-ray detector, and then, is moved from there by (?x, ?y) pulses, an image group of images obtained by moving in parallel to each other is acquired, and an image processing unit finds a deviation between the images, and acquires an input image group in which each of the images has the deviation at a subpixel level. The image processing unit executes a reconstruction processing, using the input image group in which each of the images has the deviation at the subpixel level to generate a super-resolution image.

Abstract: Techniques for processing video data are described. In an example, a device receives input video data having a first resolution. A first processor of the device sends, based at least in part on the input video data, first video data having the first resolution to a display. The device generates second video data from the input video data by at least down scaling the input video data to a second resolution, the second resolution being lower than the first resolution. A second processor of the device determines, while the first video data is presented, a property of the input video data based at least in part on the second video data. The second processor generates an indication of the property, where the indication is output while the first video data is presented.

Abstract: The present disclosure provides a system and method for image generation. The method may include obtaining a first image acquired by an imaging device. The first image may include a representation of a blood vessel of a subject based on a tracer. The method may further include obtaining a blood vessel model configured to provide one or more constraints regarding one or more characteristics of the blood vessel. The method may further include generating a second image including a representation of the blood vessel based on the blood vessel model and the first image. An image resolution of the second image may be higher than an image resolution of the first image. The presentation of the blood vessel in the second image may satisfy at least one of the one or more constraints.

Abstract: Disclosed is an image processing apparatus. The present image processing apparatus comprises: an input unit for inputting an image; and a processor for shrinking the inputted image to a predetermined ratio, extracting a visual feature from the shrunken image, performing an image quality enhancement process reflecting the extracted visual feature in the inputted image, repeatedly performing, for a predetermined number of times, the shrinking, the extracting, and the image quality enhancement process on the image that has undergone the image quality enhancement process. The present disclosure relates to an artificial intelligence (AI) system and an application thereof that simulate the functions of a human brain, such as recognition, judgment, etc., by using a machine learning algorithm such as deep learning, etc.

Abstract: An imaging system for a vehicle for obtaining an anti-flickering super-resolution image includes an image sensor adapted to obtain a sequence of images, and an image processor adapted to receive the sequence of images, compare image information of a most recent image of the sequence of images to a reference image to detect at least one image region of mismatch in the most recent image, remove the detected image region from image information of the most recent image to obtain adjusted image information, and add the adjusted image information of the most recent image to a super-resolution image.

Abstract: The disclosure provides an image super-resolution reconstruction method, a mobile terminal, and a computer-readable storage medium. The method includes: obtaining continuous N first YUV images; extracting N luma images and N chroma images from the N first YUV images; performing sequentially sharpness estimation, image registration, and image reconstruction based on a neural network on the N luma images; performing image reconstruction on the N chroma images; and fusing the chroma image obtained after reconstruction and the luma image obtained after reconstructions to obtain the target YUV image that has a higher resolution than the N first YUV images.

Abstract: The application shows a method, an apparatus, and an electronic device for processing an image. The method includes: obtaining an image of a first rendering style; obtaining an image of a second rendering style according to the image of the first rendering style and a first preset processing model; generating at least one first intermediate gradient image according to the image of the first rendering style, the image of the second rendering style, and a second preset processing model, the at least one first intermediate gradient image comprising an image in a process of gradient from the image of the first rendering style to the image of the second rendering style; generating a first gradient video according to the image of the first rendering style, at least one first intermediate gradient image, and an image of the second rendering style.

Abstract: A radiation image display apparatus that includes: a hardware processor that generates the moving image for preview based on the pieces of image data of the plurality of frames obtained by moving image photographing of an object with radiation; and a holder that holds the moving image, wherein the hardware processor further: performs reproduction control on the moving image, performs image adjustment on the moving image, displays the moving image on the display during photographing the moving image, and displays the moving image according to the reproduction control or the moving image subjected to the image on the display.

Abstract: An electronic apparatus including a memory, and a processor to determine arrangement of a plurality of images for a plurality of regions that are divided according to a predetermined layout, store information on the determined arrangement in the memory, and control the electronic apparatus to display the plurality of images in the plurality of regions based on the stored information on the determined arrangement, and the processor may determine arrangement of the plurality of images, based on a level of importance of at least one object region determined on a basis of a category of the at least one object region included in the plurality of images and a degree of loss of the at least one object region according to the level of importance.

Abstract: This invention describes a technique to enhance pixel resolution of high-speed laser scanning imaging by the means of sub-pixel sampling, applicable to one-dimensional (1D), two-dimensional (2D), and three-dimensional (3D) imaging.

Abstract: A photographing method and apparatus is provided. The photographing apparatus includes a photographing unit; a sensing unit for sensing motion of the photographing apparatus; a display unit for displaying at least one guide image for panorama photographing; a controller for controlling the photographing unit to automatically photograph, if a photographing direction that changes in accordance with motion of the photographing apparatus corresponds to one of the at least one guide image; and a storage unit for storing the photographed image data.

Abstract: The present disclosure relates to an image processing method and apparatus, and an electronic device. The method includes: acquiring a first face coordinate of a first image; acquiring a second face coordinate of a second image, in which, the first image has a different resolution from the second image, and an image size of the first image is greater than an image size of the second image; calculating a magnification ratio according to the image size of the first image and the image size of the second image, and calculating a second target face coordinate according to the magnification ratio and the second face coordinate; comparing the first face coordinate with the second target face coordinate to obtain a comparing result; and performing face clustering according to the comparing result.

Abstract: An imaging apparatus (20) includes an interface (50) and a processor (30). The interface is configured to receive a sequence of images of an object (38) acquired at respective acquisition times, the object is labeled with fluorescent emitters. The images depict intensity fluctuations of emitters on an acquisition grid, and the intensity fluctuations being uncorrelated among the emitters and during an overall time over which the images in the sequence are acquired. The a processor is configured to calculate a temporal correlation function of the intensity fluctuations over the sequence of images, and to estimate locations of the emitters on a super-resolution grid having a higher resolution than the acquisition grid, by applying to the temporal correlation function a recovery process in which the locations on the super-resolution grid are assumed to be sparse or compressible in a predefined domain. The recovery process includes a sparse-recovery process or a regularized process.

Abstract: Methods and systems for preventing the cropping of faces depicted by digital photographs are provided. To this end, the methods and systems may automatically detect the presence of faces depicted by the digital photographs and generate a facial boundary for the faces. The methods and system may also generate a crop boundary at which the photograph is cropped. In some scenarios, the systems and methods attempt to locate the crop boundary on the digital photograph to prevent omitting a portion of the facial boundary. Additionally or alternatively, users are provided an interface to manually set a location for the crop boundary. The user is warned if the crop boundary omits a portion of the facial boundary. The digital photographs and crop boundaries are sent to a printer to generate a physical print of the digital photographs.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for upscaling an image. One of the methods includes upscaling a low resolution image, creating first pixel subsets of the first upscaled image, creating second pixel subsets of a high resolution image, determining, for each subset in the pixel subsets, a value of a property of the pixel subset, determining, for each subset in the pixel subsets, a group of subsets to which the corresponding pixel subset belongs using the value of the property, and determining, for each of the groups of subsets, a filter to apply to each of the first pixel subsets that correspond to the pixel subsets in the group to create a final pixel subset that approximates the corresponding second pixel subset using the first pixel subset, a combination of all of the final pixel subsets representing a second upscaled image.

Abstract: System, methods, and other embodiments described herein relate to classifying semantics of a particle or other material component. In one embodiment, a method includes, in response to receiving a particle image, analyzing the particle image to identify characteristics of the particle represented in respective pixels of the particle image to produce a segmented image that groups the pixels into subregions. The method includes identifying semantics of the particle according to at least boundaries between the subregions. The semantics define expected behaviors of the particle in relation to material physics. The method includes providing the segmented image including the semantics as an electronic output.

Abstract: A real-time system and method for displaying video on a display are disclosed. Received compressed video data is decoded to produce an uncompressed first video frame, a first frame syntax element, an uncompressed second video frame, and a second frame syntax element. A computationally intensive process is applied to the uncompressed first video frame to produce an enhanced first video frame. A block having a portion of the enhanced first video frame from the enhanced first video frame is adaptively transferred to the uncompressed second video frame to produce an enhanced second video frame without applying the computationally intensive process to the uncompressed second video frame. The transferring is guided by the first frame syntax element and the second frame syntax element. The enhanced first video frame and the enhanced second video frame are displayed.

Abstract: The present disclosure discloses a method and device for image encoding and a method and device for image decoding. The encoding method comprises steps of: downscaling an input high resolution HR image into a low resolution LR image by a downscaler; compressing the LR image using a first compression method; extracting an index value from at least one of the HR image and the LR image; determining a parameter from a parameter database using the index value; compressing the parameter using a second compression method different from the first compression method; and obtaining a data stream by merging the compressed parameter and the compressed LR image.

Abstract: A request for a video may be received from a client device. A determination whether to transmit the video at a first video resolution or a second video resolution based on a quality of the video at the first video resolution when converted to the second video resolution at the client device may be made. The video may be transmitted to the client device at the determined first video resolution or the second video resolution.

Abstract: Identification medium recognition device (3) includes image input (11) that acquires a captured image imaged by an imaging device, identification medium region detector (12) that detects a region of the identification medium from the captured image, identification medium recognizer (18) that recognizes the character and/or the numeric character included in the identification medium from the region of the identification medium, super-resolution processor (17) that selectively performs super-resolution processing of the region of the identification medium, region storage (14) that stores a preset region in the captured image, identification medium region determiner (13) that determines whether the region of the identification medium is positioned within the preset region, and super-resolution processing selector (16) that selects, in a case where it is determined that the region of the identification medium is positioned within the preset region, execution of super-resolution processing by the super-resolution

Abstract: An image conversion method includes: dividing a low-resolution image to generate a plurality of low-resolution image patches having N (N being a natural number of 2 or greater) number of pixels; classifying the low-resolution image patches into a plurality of image categories; converting the low-resolution image patches by using a conversion kernel for each image patch corresponding to its image category among prestored conversion matrices, to generate high-resolution image patches having L (L being a natural number) number of pixels, which is less than the N number of pixels; and arranging the high-resolution image patches to generate a high-resolution image. According to the image conversion method, an operation amount can be reduced compared with a conventional technique, and thus the image conversion method can be implemented in hardware having low complexity. A high quality and high-resolution image can be generated regardless of characteristic information of an image signal.

Abstract: In one implementation, a method includes obtaining, for a particular pixel of an image, a real pixel value and a virtual pixel value. The method includes obtaining, for the particular pixel of the image, a first alpha and a second alpha. The method includes generating, for the particular pixel of the image, a combined pixel value as a weighted average of the real pixel value and the virtual pixel value, the weighting being based on at least one of the first alpha and the second alpha.

Abstract: The invention relates to an image preprocessing method and device for a JPEG compressed file. The method comprises the following steps: receiving a JPEG image to be processed; and de-noising the received JPEG image by using a pre-trained full convolutional network model to obtain a de-noised and resolution-improved image. The method and the device in the invention can effectively improve the definition of an image in an electronic file, thus effectively facilitate the subsequent image-based file analysis, e.g., OCR and CHART parsing.

Abstract: An imaging device includes a plurality of pixels arranged in a plurality of rows, in which each of the plurality of pixels outputs a pixel signal; a row scanning unit that scans the plurality of pixels on a row basis; and an output unit that outputs first time information corresponding to a processing timing of the pixel signal on one of the plurality of rows and second time information corresponding to the pixel signal on another of the plurality of rows and having a different value from the first time information.

Abstract: A method, device, system, and article of manufacture are provided for generating an enhanced image of a predetermined scene from images. In one embodiment, a method comprises receiving, by a computing device, a first indication associated with continuous image capture of a predetermined scene being enabled; in response to the continuous image capture being enabled, receiving, by the computing device, from an image sensor, a reference image and a first image, wherein each of the reference image and the first image is of the predetermined scene and has a first resolution; determining an estimated second resolution of an enhanced image of the predetermined scene using the reference image and the first image; and in response to the continuous image capture being disabled determining the enhanced image using the reference image and the first image, wherein the enhanced image has a second resolution that is at least the first resolution and about the estimated second resolution.

Abstract: Video interpolation is used to predict one or more intermediate frames at timesteps defined between two consecutive frames. A first neural network model approximates optical flow data defining motion between the two consecutive frames. A second neural network model refines the optical flow data and predicts visibility maps for each timestep. The two consecutive frames are warped according to the refined optical flow data for each timestep to produce pairs of warped frames for each timestep. The second neural network model then fuses the pair of warped frames based on the visibility maps to produce the intermediate frame for each timestep. Artifacts caused by motion boundaries and occlusions are reduced in the predicted intermediate frames.

Abstract: Systems and methods for down-scaling are provided. In one example, a method for processing image data includes determining a plurality of output pixel locations using a position value stored by a position register, using the current position value to select a center input pixel from the image data and selecting an index value, selecting a set of input pixels adjacent to the center input pixel, selecting a set of filtering coefficients from a filter coefficient lookup table using the index value, filtering the set of source input pixels to apply a respective one of the set of filtering coefficients to each of the set of source input pixels to determine an output value for the current output pixel at the current position value, and correcting chromatic aberrations in the set of source input pixels.