Image Segmentation Patents (Class 382/173)
-
Patent number: 11340772Abstract: A device includes a memory storing a computer program and a processor configured to read the computer program to obtain a content including information acquired in a time sequence, extract characteristics at a plurality of time points from the information, determine a color or a brightness at each of the time points based on the characteristic at the time point, and generate drag bar images each corresponding to one of the time points and including a pixel having the color or the brightness at the one of the time points.Type: GrantFiled: December 13, 2019Date of Patent: May 24, 2022Assignee: SZ DJI TECHNOLOGY CO., LTD.Inventor: Ming Shao
-
Patent number: 11341690Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods that generate customized color editing options tailored to a digital image. In particular, in one or more embodiments, the disclosed systems generate an interactive editing user interface that includes selectable color controls corresponding to dominant colors extracted from the digital image and ordered by dominance. In at least one embodiment, the disclosed systems utilizes clustering techniques associated with the dominant representative colors to generate contour paths in the digital image further enabling dynamic interactive image editing.Type: GrantFiled: October 31, 2019Date of Patent: May 24, 2022Assignee: Adobe Inc.Inventors: Mrinal Kumar Sharma, Shailendra Singh Rathore, Gourav Singhal
-
Patent number: 11341758Abstract: A computer implemented method of processing an image, comprising: obtaining an input image; converting the input image into a binary image; processing the binary image using a first trained model that performs classification of image components as part of a character.Type: GrantFiled: August 6, 2021Date of Patent: May 24, 2022Assignee: Sprout.ai LimitedInventor: Niclas Stoltenberg
-
Patent number: 11336906Abstract: An image processing method for an image, a data transmission method, an image processing device for an image, a data transmission device, and a storage medium are disclosed. The image is a binarized image, and the binarized image includes a plurality of pixels, among the plurality of pixels, pixels having a first value are first pixels, and pixels having a second value are second pixels, and the image processing method includes: dividing the image to determine N rectangular regions, which include only the first pixels, of the image; and obtaining start coordinates and end coordinates of the N rectangular regions, respectively, thereby obtaining compressed data after compressing the image. N is an integer greater than or equal to 1. The image processing method can improve the compression rate of the image.Type: GrantFiled: April 17, 2019Date of Patent: May 17, 2022Assignee: BOE TECHNOLOGY GROUP CO., LTD.Inventor: Lingfeng Xu
-
Patent number: 11330171Abstract: A method for blending a left image with a right image along a seam is disclosed. The method comprises selecting collocated pixels from a collocated column of pixels in the left image and the right image; computing differences in the selected collocated pixels; and determining a correction amount to apply to seam pixels along the seam on a row-by-row basis based on the computed differences; and selectively applying the correction amount on a row-by-row basis to a neighborhood of pixels extending transversely across the seam and centered on seam.Type: GrantFiled: May 24, 2013Date of Patent: May 10, 2022Inventor: Jason Daniel Tanner
-
Patent number: 11330199Abstract: Techniques related to a method and system of adaptable exposure control and light projection for cameras are described herein.Type: GrantFiled: June 9, 2020Date of Patent: May 10, 2022Assignee: Sony Group CorporationInventors: Paul Winer, Richmond Hicks
-
Patent number: 11321555Abstract: A method includes receiving a set of video frames that correspond to a video, including a first video frame and a second video frame that each include a face, wherein the second video frame is subsequent to the first video frame. The method further includes performing face tracking on the first video frame to identify a first face resampling keyframe and performing face tracking on the second video frame to identify a second face resampling keyframe. The method further includes deriving an interpolation amount. The method further includes determining a first interpolated face frame based on the first face resampling keyframe and the interpolation amount. The method further includes determining a second interpolated face frame based on the second face resampling keyframe and the interpolation amount. The method further includes rendering an interpolated first face and an interpolated second face. The method further includes displaying a final frame.Type: GrantFiled: November 11, 2019Date of Patent: May 3, 2022Assignee: Google LLCInventor: Dillon Cower
-
Patent number: 11315353Abstract: Systems and methods for spatial-aware information extraction from electronic source documents are disclosed. Exemplary implementations may: obtain an electronic source document in electronic format, including human-readable information; obtain extracted information that has been extracted from the electronic source document; generate a character-based representation of the electronic source document that uses a grid of character positions; and present a user interface to a user to search and/or select information in the electronic source document based on the character-based representation.Type: GrantFiled: June 10, 2021Date of Patent: April 26, 2022Assignee: Instabase, Inc.Inventor: Daniel Benjamin Cahn
-
Patent number: 11308714Abstract: A computer-based method includes receiving, at a computer-based system, an aerial image of a property that includes a first visual indicator on the aerial image that follows and identifies a boundary line for the property; using a building rooftop Deep Fully Convolutional Network (DFCN), configured and trained to predict the presence of building rooftops in aerial imagery, to predict whether any building rooftops are present within the boundary line of the property based on the aerial image; and applying a second visual indicator to the aerial image to identify and outline a building rooftop in the aerial image identified by the building rooftop deep fully convolutional network. In some implementations, other Convolutional Networks (ConvNets) are used to predict other property attributes and characteristics.Type: GrantFiled: August 21, 2019Date of Patent: April 19, 2022Inventors: Christos Marios Christoudias, Ellen Dee Cousins, Ali Alhaj Darwish
-
Patent number: 11308130Abstract: The present disclosure relates to evaluating whether two data records reflect the same entity using a classifier in the absence of ground truth. Without ground truth, it is difficult to determine the precision or recall of a classifier. The present disclosure generates a list comprising a series of unique feature signatures and a set of sample record pairs for each unique feature signature. In some embodiments, users may provide labels for the set of sample record pairs for each unique feature signature.Type: GrantFiled: November 8, 2019Date of Patent: April 19, 2022Assignee: AMPERITY, INC.Inventors: Yan Yan, Stephen Meyles, Mona Akmal, Michael P. Fikes
-
Patent number: 11308492Abstract: The present disclosure involves systems, software, and computer implemented methods for transaction auditing. One example method includes determining valid pixel-based pattern(s) that are included in valid reference images. Fraudulent pixel-based pattern(s) that are included in fraudulent reference images are determined. A request to classify an image is received. A determination is made as to whether pixel values in the image match a valid pixel-based pattern or a fraudulent pixel-based pattern. In response to determining that the pixel values match a valid pixel-based pattern, a likelihood of classifying the first image as a valid image is increased. In response to determining that the pixel values match a fraudulent pixel-based pattern, a likelihood that the image as a fraudulent image is increased. The image is classified in response to the request as either a valid image or a fraudulent image based on the likelihoods.Type: GrantFiled: December 12, 2019Date of Patent: April 19, 2022Assignee: SAP SEInventors: Jesper Lind, Suchitra Sundararaman
-
Patent number: 11308625Abstract: The present invention relates to a novel method and system for crater detection in planetary data based on marked point processes (MPP), effective for various object detection tasks in Earth observation, and for planetary image registration. The resulting spatial features are exploited for registration, together with fitness functions based on the MPP energy, on the mean directed Hausdorff distance, and on the mutual information. Two different methods—one based on birth-death processes and region-of-interest analysis, and the other based on graph cuts and decimated wavelets—are included within the present framework. Experimental results confirmed the effectiveness of the present invention in terms of crater detection performance and sub-pixel registration accuracy.Type: GrantFiled: September 28, 2020Date of Patent: April 19, 2022Assignee: United States of America as represented by the Administrator of NASAInventors: Jacqueline J. Le Moigne-Stewart, David Solarna, Gabriele Moser, Sebastiano Serpico
-
Patent number: 11308605Abstract: There is a need for more effective and efficient printed circuit board (PCB) design. This need can be addressed by, for example, solutions for performing automated PCB component estimation. In one example, a method includes identifying a PCB image of a PCB; performing chromaticity-based background subtraction on the PCB image to generate a background-subtracted PCB image; performing morphological noise removal on the background-subtracted PCB image to generate a noise-removed PCB image; and performing object localization on the noise-removed PCB image to identify one or more PCB component estimations within the noise-removed PCB image.Type: GrantFiled: January 14, 2021Date of Patent: April 19, 2022Assignee: University of Florida Research Foundation, IncorporatedInventors: Navid Asadi-Zanjani, Mark M. Tehranipoor, Mukhil Azhagan Mallaiyan Sathiaseelan
-
Patent number: 11308318Abstract: An image processing apparatus generates a first image by adjusting luminance so as to preserve a signal value on a high-luminance side among signal values acquired by reading a document by a scanner, and detects a position of a region corresponding to the document in the generated first image. The image processing apparatus further generates a second image by performing adjustment so as to increase luminance of a pixel contained in the first image, and crops a document image from the generated second image based on the detected position of the region corresponding to the document.Type: GrantFiled: May 17, 2019Date of Patent: April 19, 2022Assignee: CANON KABUSHIKI KAISHAInventors: Koya Shimamura, Junya Arakawa
-
Patent number: 11295170Abstract: A system including one or more computers and one or more storage devices storing instructions that, when executed by the one or more computers, cause the one or more computers to implement a group-equivariant convolutional neural network configured to process a network input including a set of 3D points to generate an output tensor representing transformed features of the network input is described. The group-equivariant convolutional neural network includes a grouping layer, a switching layer, a pre-processing layer, a group-equivariant layer, a first subnetwork, an average pooling layer, a second subnetwork, an output subnetwork, and a final output layer.Type: GrantFiled: August 17, 2021Date of Patent: April 5, 2022Assignee: FPT USA Corp.Inventors: Thieu Ngoc Vo, Phong Xuan Nguyen, Nghia Trung Huynh, Binh Cong Phan, Tuan Anh Le
-
Patent number: 11290607Abstract: An image reading apparatus includes a reading processing unit that acquires image data including an image of a document placed on a document placement unit. A first detection processing unit detects a first line image whose width or density exceeds a predetermined first threshold value from the image data. A second detection processing unit detects a rectangular area that is defined by the first line image, included in the image data, and satisfies a predetermined detection condition. A third detection processing unit detects a second line image whose width or density exceeds a second threshold value smaller than the first threshold value from the rectangular area. A fourth detection processing unit detects, as document areas of the respective documents, the rectangular area not divided by the second line image and divided areas included in the rectangular area divided by the second line image.Type: GrantFiled: February 19, 2021Date of Patent: March 29, 2022Assignee: KYOCERA DOCUMENT SOLUTIONS INC.Inventor: Yasuyuki Yabuuchi
-
Patent number: 11281229Abstract: Embodiments of the present disclosure disclose a method and apparatus for outputting obstacle information. A specific embodiment of the method comprises: determining a candidate direction information set of a target obstacle point cloud; determining, for each piece of candidate direction information in the candidate direction information set, a target value of the target obstacle point cloud in a direction indicated by the candidate direction information based on the target obstacle point cloud and a smallest circumscribing rectangle of the target obstacle point cloud in the direction indicated by the candidate direction information; defining candidate direction information having a minimum target value in the candidate direction information set as direction information corresponding to the target obstacle cloud point; and outputting the direction information corresponding to the target obstacle cloud point, thereby improving the abundance of content of outputted obstacle information.Type: GrantFiled: September 17, 2018Date of Patent: March 22, 2022Assignee: APOLLO INTELLIGENT DRIVING TECHNOLOGY (BEIJING) CO., LTD.Inventors: Yiyuan Cheng, Yu Ma, Jun Wang
-
Patent number: 11282251Abstract: A system and method for constructing virtual radial ultrasound images from CT data includes receiving CT data, calculating a gradient of the CT data, and obtaining 2D slices by interpolation in 3D. The method further includes calculating transmission and reflection values on each radial line away from a center of each 2D slice to create a lesion image, converting the lesion image to polar, generating random noise in the polar image, and adding random comet tails to the polar image. A background image is created by computing the mean profile along the radius from a series of polar lesion images. The background image is merged with the polar lesion image and random noise is generated in the background of the merged image. Density variations are simulated in the merged image and coordinates of the merged image are transformed to Cartesian to construct a virtual radial ultrasound image.Type: GrantFiled: November 19, 2020Date of Patent: March 22, 2022Assignee: COVIDIEN LPInventors: Nicolas J. Merlet, Oren P. Weingarten
-
Patent number: 11281949Abstract: An image forming apparatus includes a medium reading section that acquires medium area data in a conveyance path, an image formation controller that sets a target pixel as a mask processing target where ink ejection is restricted when the target pixel is outside an image formable area, an anomaly estimation section that uses positions of a mask start pixel and a mask end pixel to estimate a cause of the mask processing target and generates a first text message indicating a result of estimation on the cause of the mask processing target, and an operation display section that, according to occurrence of the mask processing target, displays the first text message and an area as the mask processing target with a display generated using the position of the mask start pixel and the position of the mask end pixel superimposed on the image represented by the image data.Type: GrantFiled: September 9, 2020Date of Patent: March 22, 2022Assignee: KYOCERA Document Solutions Inc.Inventor: Takenori Yamamoto
-
Patent number: 11284155Abstract: An electronic device for controlling the broadcasting of digital audio and/or video streams that includes: a module for receiving a list of upcoming streams able to be broadcast; a module for acquiring a set of profile(s) of a group of user(s); a module for detecting, among the received list and based on the acquired set, a preferred digital stream intended to be broadcast after a current digital stream; and a module for switching from a broadcast of the current stream to a broadcast of the preferred stream. The detection module is configured to detect an event triggering a switch to the preferred stream, and the switching module is configured to operate at least according to an automatic mode, where it is able to switch from the broadcasting of the current stream to that of the preferred stream upon the detection of the triggering event.Type: GrantFiled: October 22, 2020Date of Patent: March 22, 2022Assignee: FAURECIA CLARION ELECTRONICS EUROPEInventors: Cyril Laury, Frantz Lohier
-
Patent number: 11282198Abstract: A multi-label heat map generating system is operable to receive a plurality of medical scans and a corresponding plurality of medical labels that each correspond to one of a set of abnormality classes. A computer vision model is generated by training on the medical scans and the medical labels. Probability matrix data, which includes a set of image patch probability values that each indicate a probability that a corresponding one of the set of abnormality classes is present in each of a set of image patches, is generated by performing an inference function that utilizes the computer vision model on a new medical scan. Preliminary heat map visualization data can be generated for transmission to a client device based on the probability matrix data. Heat map visualization data can be generated via a post-processing of the preliminary heat map visualization data to mitigate heat map artifacts.Type: GrantFiled: July 27, 2020Date of Patent: March 22, 2022Assignee: Enlitic, Inc.Inventors: Kevin Lyman, Eric C. Poblenz, Li Yao, Ben Covington, Anthony Upton
-
Patent number: 11276250Abstract: In an approach, data of a plurality of points is sampled in a target area, wherein the data of each point of the plurality of points comprises position information and a height value. A first area of a target area is determined, wherein the height value of each point of the plurality of points in the first area complies with a first range. A second area of the target area is determined, wherein the height value of each point of the plurality of points in the second area complies with a second range. A third area of the target area is determined, wherein the height value of each point of the plurality of points in the third area complies with a third range. A first pattern is generated, wherein the first pattern is a combination of the first area and the third area.Type: GrantFiled: October 23, 2019Date of Patent: March 15, 2022Assignee: International Business Machines CorporationInventors: Xue Mei Deng, Ang Yi, Jun Hong Zhao, Lu Huang, Ya Juan Tong, Hui Huang
-
Patent number: 11270027Abstract: The invention relates generally to systems and methods for protecting patient privacy when health care information is shared between various entities and, in particular, to systems and methods that implement a multi-stage sanitizing routine for de-identifying patient data from medical reports and diagnostic images to ensure patient privacy, while preserving the ability for sanitized medical reports and diagnostic images to be re-identified.Type: GrantFiled: September 23, 2020Date of Patent: March 8, 2022Assignee: MEDICOM TECHNOLOGIES INC.Inventors: Michael Rosenberg, Malcolm Benitz, Jason Suttles, Chris Woodlief, Brent Goldstein
-
Patent number: 11263756Abstract: A computer-implemented method for generating a semantically segmented image and a depth completion image using a convolutional neural network (CNN) from an input visible image and/or an input depth image. A central component of the CNN for semantic segmentation and depth completion is a common representation that allows both tasks to be performed when given any of these combinations of input images (i) both an input visible image and an input depth image, (ii) only an input visible image, or (iii) only an input depth image.Type: GrantFiled: December 9, 2019Date of Patent: March 1, 2022Assignee: NAVER CORPORATIONInventors: Boris Chidlovskii, Giorgio Giannone
-
Patent number: 11257270Abstract: Methods and apparatuses for generating a dynamic picture sequence are provided. The method includes generating a target vector from a target static picture; obtaining a foreground dynamic picture sequence, a foreground mask dynamic picture sequence and a background static picture by performing a prediction on the target vector; obtaining N frames of foreground static pictures by splitting the foreground dynamic picture sequence; obtaining N frames of foreground mask static pictures by splitting the foreground mask dynamic picture sequence; generating a first frame of static image by performing a prediction with inputting a first frame of the foreground static pictures, a first frame of the foreground mask static pictures and the background static picture into a first cell body of a Long Short-Term Memory network.Type: GrantFiled: December 19, 2018Date of Patent: February 22, 2022Assignee: Beijing Sankuai Online Technology Co., LtdInventors: Li Yu, Xin Jin
-
Patent number: 11258927Abstract: A noise removal device includes a noise remover, a data compressor, a data decompressor, a stillness detector, and an output selector. For a second frame succeeding a first frame, the noise remover performs noise removal on input image data of the second frame to generate noise removed data, based on the input image data of the second frame and decompressed data of the first frame. For the second frame, the output selector: selects the input image data as output image data, when the stillness detector detects that the input image data is data of a still image; and selects the noise removed data as the output image data, when the stillness detector detects that the input image data is not the data of the still image.Type: GrantFiled: August 25, 2020Date of Patent: February 22, 2022Assignee: SOCIONEXT INC.Inventor: Satoru Tanigawa
-
Patent number: 11256961Abstract: Segmentation is the identification of separate objects within an image. An example is identification of a pedestrian passing in front of a car, where the pedestrian is a first object and the car is a second object. Superpixel segmentation is the identification of regions of pixels within an object that have similar properties. An example is identification of pixel regions having a similar color, such as different articles of clothing worn by the pedestrian and different components of the car. A pixel affinity neural network (PAN) model is trained to generate pixel affinity maps for superpixel segmentation. The pixel affinity map defines the similarity of two points in space. In an embodiment, the pixel affinity map indicates a horizontal affinity and vertical affinity for each pixel in the image. The pixel affinity map is processed to identify the superpixels.Type: GrantFiled: July 6, 2020Date of Patent: February 22, 2022Assignee: NVIDIA CorporationInventors: Wei-Chih Tu, Ming-Yu Liu, Varun Jampani, Deqing Sun, Ming-Hsuan Yang, Jan Kautz
-
Patent number: 11252413Abstract: The present disclosure relates to encoding a decoding video employing texture coding. In particular, a texture region is identified within a video picture and a texture patch is determined for said region. Moreover, a set of parameters specifies luminance within the texture region (1001) by fitting the texture region samples to a two-dimensional polynomial function of the patch determined according to the set of parameters (1040); and/or motion within the texture region by fitting motion estimated between the texture region of the video picture and an adjacent picture to a two-dimensional polynomial The texture patch and the first set of parameters are then included into a bitstream which is output of the encoder and provided in this way to the decoder which reconstructs the texture based on the patch and the function applied to the patch.Type: GrantFiled: June 8, 2020Date of Patent: February 15, 2022Assignee: Huawei Technologies Co., Ltd.Inventors: Zhijie Zhao, Bastian Wandt, Yiqun Liu, Thorsten Laude, Bodo Rosenhahn, Joern Ostermann
-
Patent number: 11244195Abstract: The present disclosure relates to systems, method, and computer readable media that iteratively apply a neural network to a digital image at a reduced resolution to automatically identify pixels of salient objects portrayed within the digital image. For example, the disclosed systems can generate a reduced-resolution digital image from an input digital image and apply a neural network to identify a region corresponding to a salient object. The disclosed systems can then iteratively apply the neural network to additional reduced-resolution digital images (based on the identified region) to generate one or more reduced-resolution segmentation maps that roughly indicate pixels of the salient object. In addition, the systems described herein can perform post-processing based on the reduced-resolution segmentation map(s) and the input digital image to accurately determine pixels that correspond to the salient object.Type: GrantFiled: May 1, 2018Date of Patent: February 8, 2022Assignee: Adobe Inc.Inventors: I-Ming Pao, Zhe Lin, Sarah Stuckey, Jianming Zhang, Betty Leong
-
Patent number: 11238565Abstract: An apparatus includes a raw image pipeline comprising a first circuit and a second circuit. The first circuit of the raw image pipeline may be configured to distinguish between smooth picture noisy areas and non-smooth clean areas of an image by performing tone based non-smooth detection on data of the image to obtain an adjusted non-smoothness value for at least one area comprising a plurality of pixels of the image. The second circuit of the raw image pipeline may be configured to adjust one or more of noise reduction filtering or sharpening filtering performed on the at least one area of the image based on the adjusted non-smoothness value.Type: GrantFiled: March 10, 2020Date of Patent: February 1, 2022Assignee: Ambarella International LPInventor: Elliot N. Linzer
-
Patent number: 11238593Abstract: Techniques are disclosed for parsing a source image, to identify segments of one or more objects within the source image. The parsing is carried out by an image parsing pipeline that includes three distinct stages comprising three respectively neural network models. The source image can include one or more objects. A first neural network model of the pipeline identifies a section of the source image that includes the object comprising a plurality of segments. A second neural network model of the pipeline generates, from the section of the source image, a mask image, where the mask image identifies one or more segments of the object. A third neural network model of the pipeline further refines the identification of the segments in the mask image, to generate a parsed image. The parsed image identifies the segments of the object, by assigning corresponding unique labels to pixels of different segments of the object.Type: GrantFiled: February 12, 2020Date of Patent: February 1, 2022Assignee: Adobe Inc.Inventors: Kerem Can Turgutlu, Jayant Kumar, Jianming Zhang, Zhe Lin
-
Patent number: 11232571Abstract: The present invention discloses a method and device for quick segmentation of an optical coherence tomography image, a computing device, and a computer readable storage medium. The method includes: extracting a feature value related to a pixel in the optical coherence tomography image from the pixel to obtain a feature value representative of the pixel feature, the feature value including at least one of an intensity gradient value of the pixel, an intensity multi-scale mean value of the pixel, and an angular feature value extracted from the pixel; inputting the feature value into a Sparse Bayesian Classification model containing values related to image features corresponding to a border, to obtain a probability that the image feature belongs to features of various image region borders; and obtaining borders of various image regions contained in a predefined image region based on the probabilities obtained for various pixels in the predefined image region.Type: GrantFiled: May 31, 2019Date of Patent: January 25, 2022Assignee: SOOCHOW UNIVERSITYInventors: Lirong Wang, Meng Gan, Cong Wang
-
Patent number: 11232147Abstract: Systems, methods, and non-transitory computer-readable media are disclosed for determining multi-term contextual tags for digital content and propagating the multi-term contextual tags to additional digital content. For instance, the disclosed systems can utilize search query supervision to determine and associate multi-term contextual tags (e.g., tags that represent a specific concept based on the order of the terms in the tag) with digital content. Furthermore, the disclosed systems can propagate the multi-term contextual tags determined for the digital content to additional digital content based on similarities between the digital content and additional digital content (e.g., utilizing clustering techniques). Additionally, the disclosed systems can provide digital content as search results based on the associated multi-term contextual tags.Type: GrantFiled: July 29, 2019Date of Patent: January 25, 2022Assignee: Adobe Inc.Inventors: Ajinkya Kale, Baldo Faieta, Benjamin Leviant, Fengbin Chen, Francois Guerin, Kate Sousa, Trung Bui, Venkat Barakam, Zhe Lin
-
Patent number: 11222425Abstract: A system and methods for automatically delineating OARs in whole-volume medical images through a two-stage DCNN model, the DCNN model comprising an OAR detection network and an OAR segmentation network, and the method comprising the steps of: inputting the whole-volume medical images to the OAR detection network; extracting image features through a sequence of downsampling blocks; generating a final detection feature map via upsampling and concatenation; detecting at least one OAR candidate by branching the final detection feature map, wherein the at least one OAR candidate is defined by a predicted bounding box with a class label; inputting the predicted bounding box and corresponding class label to the OAR segmentation network; cropping the final detection feature map and a downsampling block in the OAR detection network according to the predicted bounding box; concatenating the cropped feature maps and generating a predicted binary mask delineating OARs according to the class label.Type: GrantFiled: February 11, 2020Date of Patent: January 11, 2022Inventors: Hao Tang, Yang Liu, Shanlin Sun, Narisu Bai
-
Patent number: 11222402Abstract: A method of adaptive image enhancement, comprising, receiving a low resolution image, receiving at least one application constraint, detecting at least one scene within the low resolution image, detecting a plurality of regions of interest within the detected at least one scene, prioritizing the detected plurality of regions of interest, ranking the detected plurality of regions of interest based on the prioritization, determining an enhanceable subset of the plurality of regions of interest based on the ranking of the plurality of regions of interest and the at least one application constraint and enhancing the enhanceable subset of the plurality of regions of interest.Type: GrantFiled: April 24, 2020Date of Patent: January 11, 2022Inventors: Narges Afsham, Yu Feng Liao
-
Patent number: 11216849Abstract: An information processing system includes an imaging device, a display, a memory, and a processor configured to acquire a customer ID for identifying a customer, generate a screen for the display for showing promotional content, control the imaging device to capture an image of the customer's face, then analyze the captured image to determine the degree of the customer's interest in the displayed content. The displayed promotional content is then stored in the memory in in association with the customer ID and the determined degree of the customer's interest.Type: GrantFiled: June 8, 2020Date of Patent: January 4, 2022Assignee: TOSHIBA TEC KABUSHIKI KAISHAInventor: Toshihiko Nannichi
-
Patent number: 11216953Abstract: An image processing apparatus includes a memory configured to store first region detection information corresponding to at least one object of a first frame, and a processor configured to perform region growing processing based on a seed region identified from a second frame and the first region detection information, obtain second region detection information corresponding to at least one object of the second frame, and perform image processing on the second frame based on the second region detection information.Type: GrantFiled: September 6, 2019Date of Patent: January 4, 2022Assignee: SAMSUNG ELECTRONICS CO., LTD.Inventor: Hyungjun Lim
-
Patent number: 11210531Abstract: An information processing apparatus includes a first setting unit configured to set a plurality of observation necessity degrees of positions in a real space and times, a first display unit configured to display the plurality of observation necessity degrees mapped based on a position in the real space and a time, a second display unit configured to display a plurality of targets detected from a captured image, based on observation necessity degrees each corresponding to a different one of the plurality of targets, and a receiving unit configured to receive an input of information corresponding to at least one target of the plurality of targets. The first setting unit resets at least one observation necessity degree of the plurality of observation necessity degrees based on the input information.Type: GrantFiled: August 8, 2019Date of Patent: December 28, 2021Assignee: Canon Kabushiki KaishaInventor: Shunsuke Sato
-
Patent number: 11206403Abstract: A method for decoding an image by a decoding apparatus according to the present invention comprises the steps of: deriving predicted samples by performing prediction on a current block; determining whether the current block is to be divided, on the basis of a size and a maximum transform size of the current block; when the current block is determined to be divided, dividing the current block into transform blocks; deriving transform coefficients of the transform blocks; deriving residual samples by performing an inverse transform procedure for the transform coefficients; and generating a reconstructed picture on the basis of the residual samples and the predicted samples, wherein, when the size of the current block is larger than the maximum transform size, the current block is determined to be divided.Type: GrantFiled: April 20, 2020Date of Patent: December 21, 2021Assignee: LG ELECTRONICS INC.Inventors: Sunmi Yoo, Jangwon Choi, Jin Heo
-
Patent number: 11200657Abstract: State of the art image processing techniques such as background subtraction, and Convolutional Neural Network based approaches, when used for change detection, fail to support certain datasets. The disclosure herein generally relates to semantic change detection, and, more particularly, to a method and system for semantic change detection using a deep neural network feature correlation approach. An adaptive correlation layer is used by the system, which determines extent of computation required at pixel level, based on amount of information at pixels, and uses this information in further computation done for the semantic change detection. Information on the determined extent of computation required is then used to extract semantic features, which is then used to compute one or more correlation maps between the at least one feature map of a test image and corresponding reference image. Further the semantic changes are determined from the one or more correlation maps.Type: GrantFiled: August 28, 2020Date of Patent: December 14, 2021Assignee: Tata Consultancy Services LimitedInventors: Jayavardhana Rama Gubbi Lakshminarasimha, Akshaya Ramaswamy, Balamuralidhar Purushothaman, Ram Prabhakar Kathirvel, Venkatesh Babu Radhakrishnan
-
Patent number: 11200822Abstract: A non-transitory computer-readable medium encoded with a computer-readable program, which when executed by a processor, will cause the processor to execute an image processing method, the image processing method including establishing a zero crossing region from a target image onto a screen, wherein the zero crossing region comprises a corresponding value. The method further includes receiving a variable input and an integration time input, wherein the variable input comprises a frequency number for an image accumulation procedure, and wherein the integration time comprises a time period for which an aperture for a sensor receives incoming signals. Additionally, the method includes performing the image accumulation procedure. Moreover, the method includes producing a final greyscale image by adding a plurality of sets of vertically shifted pixel values, wherein the frequency number for the image accumulation procedure ranges from 3 to 20.Type: GrantFiled: July 14, 2020Date of Patent: December 14, 2021Assignee: Purdue Research FoundationInventors: Euiwon Bae, Huisung Kim
-
Patent number: 11182599Abstract: A motion state evaluation system is provided with a motion analyzer obtains a value representing a motion state of the subject based on a ratio, to a reference length, of a distance between predetermined joints estimated on an image; a reference storage stores a height of a subject as the reference length; a motion analyzer obtains, as a value for use in evaluation of motion of the subject, the value representing the motion state of the subject from a feature point distance on the image of a plurality of anatomical feature points based on a ratio, to the height, of a distance on the image that is determined from the plurality of anatomical feature points and corresponds to the height of the subject; and an evaluation processor evaluates the motion of the subject based on the value representing the motion state.Type: GrantFiled: April 24, 2020Date of Patent: November 23, 2021Assignee: ASICS CORPORATIONInventors: Kazunari Takeichi, Takehiro Tagawa, Ryota Shinayama
-
Patent number: 11182635Abstract: A personal information separation unit separates a document image containing personal information into a personal information image containing the personal information and a general information image that does not contain the personal information on the basis of the document image, and transmits the general information image to a cloud server. A recognition result integration unit receives a general recognition result that is the recognition result of the character recognition processing for the general information image from the cloud server, and acquires a target recognition result that is the recognition result of the character recognition processing for the document image in accordance with the general recognition result and the information based on the personal information image.Type: GrantFiled: March 9, 2020Date of Patent: November 23, 2021Assignee: HITACHI, LTD.Inventors: Ryosuke Odate, Hiroshi Shinjo, Yuuichi Nonaka
-
Patent number: 11176046Abstract: The present invention relates to a graph-computing-oriented heterogeneous in-memory computing apparatus, comprising a memory control unit, a digital signal processing unit, and a plurality of analog signal processing units using the memory control unit.Type: GrantFiled: July 20, 2020Date of Patent: November 16, 2021Assignee: Huazhong University of Science and TechnologyInventors: Xiaofei Liao, Yu Huang, Long Zheng, Hai Jin
-
Patent number: 11170264Abstract: Subject matter regards improving image segmentation or image annotation. A method can include receiving, through a user interface (UI), for each class label of class labels to be identified by the ML model and for a proper subset of pixels of the image data, data indicating respective pixels associated with the class label, partially training the ML model based on the received data, generating, using the partially trained ML model, pseudo-labels for each pixel of the image data for which a class label has not been received, and receiving, through the UT, a further class label that corrects a pseudo-label of the generated pseudo-labels.Type: GrantFiled: May 31, 2019Date of Patent: November 9, 2021Assignee: Raytheon CompanyInventors: Philip A. Sallee, Stephen J. Raif, James Talamonti
-
Patent number: 11164319Abstract: Certain aspects provide a method, including: receiving a depth image from a depth sensor; receiving a segmentation mask corresponding to the depth image and segmenting the depth image into a set of foreground pixels and a set of background pixels; determining a set of seed pixels in the depth image; for each respective seed pixel of the set of seed pixels: determining a sampling line in the depth image that starts at the respective seed pixel and passes through a portion of the depth image; for each respective sampling line pixel in the sampling line having a value in the segmentation mask indicating a foreground object in the depth image: determining one or more data attribute values based on a depth value for the respective sampling line pixel in the depth image; and adding the one or more data attribute values to a feature vector.Type: GrantFiled: December 11, 2019Date of Patent: November 2, 2021Assignee: Smith & Nephew, Inc.Inventor: Kevin Steele
-
Patent number: 11164308Abstract: A method for using machine learning to perform classification of anatomical coverage of images includes acquiring a series of medical images of a subject. The method also includes automatically, with a computer system, analyzing each image in the series of medical images using a machine-learning technique to classify each image in the series of medical images based on anatomical structures reflected in each image in the series of medical images.Type: GrantFiled: February 27, 2018Date of Patent: November 2, 2021Assignee: The Regents of the University of CaliforniaInventors: Xiaoyong Wang, Pechin Lo, Matthew Brown, Bharath Ramakrishna, Jonathan Goldin
-
Patent number: 11159716Abstract: A computer program in a computer system with one or more processors receives digital image data from a smart camera and sends the digital image data from the smart camera to smart contact lenses worn by a first user in the digital image data. The approach includes one or more processors receiving from a user interface in the smart camera, an instruction, from a second user, for at least one change to the first user and sending the instruction for the at least one change to the smart contact lenses worn by the first user.Type: GrantFiled: September 21, 2020Date of Patent: October 26, 2021Assignee: International Business Machines CorporationInventors: Siddique M Adoni, Sarbajit K. Rakshit
-
Patent number: 11157724Abstract: An image recognition device executes a Hilbert scan of frame image data constituting moving-image data to generate one-dimensional spatial image data, and further arrays the one-dimensional spatial image data in a time direction to generate two-dimensional spatio-temporal image data that holds spatial information and temporal information. The image recognition device converts the moving-image data into the two-dimensional spatio-temporal image data while holding the spatial and temporal information. By means of a CNN unit, the image recognition device executes a convolution process wherein a two-dimensional filter is used on the spatio-temporal image data to image-recognize a behavior of a pedestrian who is a recognition object.Type: GrantFiled: July 31, 2018Date of Patent: October 26, 2021Assignees: EQUOS RESEARCH CO., LTD., KYUSHU INSTITUTE OF TECHNOLOGYInventors: Hideo Yamada, Ryuya Muramatsu, Masatoshi Shibata, Shuichi Enokida
-
Patent number: 11158109Abstract: A machine can be specially configured to generate one or more atlases that include two-dimensional texture maps and their corresponding UV maps from a three-dimensional object, compress the atlases, decompress the atlases, store the atlases, access the atlases, communicate the atlases, apply the texture maps from the atlases to a three-dimensional model, or otherwise process the atlases, the texture maps, the UV maps, or any suitable combination thereof. The atlases, texture maps, UV maps, or any suitable combination thereof can be generated, compiled or otherwise created by the machine in a manner that is computationally efficient to compress and decompress using video compression and decompression techniques.Type: GrantFiled: September 15, 2020Date of Patent: October 26, 2021Assignee: 8i LimitedInventors: Gianluca Cernigliaro, Philip A. Chou, Eugene Joseph d'Eon