Abstract: A method of generating a correction plan for a knee of a patient includes obtaining a ratio of reference bone density to reference ligament tension in a reference population. A bone of the knee of the patient may be imaged. From the image of the bone, a first dataset may be determined including at least one site of ligament attachment and existing dwell points of a medial femoral condyle and lateral femoral condyle of the patient on a tibia of the patient. Desired positions of contact in three dimensions of the femoral condyles of the patient with the tibia of the patient may be obtained by determining a relationship in which a ratio of bone density to ligament tension of the patient is substantially equal to the ratio of reference bone density to reference ligament tension.

Abstract: Digital image defect identification and correction techniques are described. In one example, a digital medium environment is configured to identify and correct a digital image defect through identification of a defect in a digital image using machine learning. The identification includes generating a plurality of defect type scores using a plurality of defect type identification models, as part of machine learning, for a plurality of different defect types and determining the digital image includes the defect based on the generated plurality of defect type scores. A correction is generated for the identified defect and the digital image is output as included the generated correction.

Abstract: A device for filling a hydrodynamic coupling with operating fluid and a method for filling a constantly filled hydrodynamic coupling during commissioning and/or for checking or setting a fill level with operating fluid. The device and the method provide the use of image recognition, by way of which an inspection window can be recognized and by way of which a position of the inspection window can be set.

Abstract: A method includes acquiring, from a camera, an image data sequence of a real object in a real scene and performing a first template-matching on an image frame in the image data sequence using intensity-related data sets stored in one or more memories to generate response maps. The intensity-related data sets represent an intensity distribution of a reference object from respective viewpoints. The reference object corresponds to the real object. A candidate region of interest is determined for the real object in the image frame based on the response maps, and second template-matching is performed on the candidate region of interest using shape-related feature data sets stored in one or more memories to derive a pose of the real object. The shape-related feature data sets represent edge information of the reference object from the respective viewpoints.

Abstract: Machine learning of a dental image for e-commerce is described. An aggregator server may receive dental images of an e-commerce consumer from an e-commerce provider. The dental images are matched to a probability dataset and correlated with an e-commerce consumer dataset to produce an e-commerce dataset. With different resolutions each neural network machine learns to probability map and detect different dental object probabilities.

Abstract: This invention relates to an apparatus (1) for carrying out a non-destructive inspection on a wooden board (9) or a similar object. The apparatus (1) comprises a movement system (3) comprising at least two chains or belts (32) that are at a distance from one another and substantially parallel to each other. The chains or belts (32) are slidable parallel to a movement path (30) and are and intended to support the wooden board (9). The apparatus (1) also comprises a non-destructive inspection station (2) that is positioned on the movement path (30), for carrying out a non-destructive inspection on the wooden board (9) that is supported by the chains or belts (32). The non-destructive inspection station (2) comprises at least one operating component (22) that is positioned on the same side as the chains or belts (32) and emits or receives a signal or an image, with an emission or reception field (220) that faces towards a region between two chains or belts (32).

Abstract: The present disclosure is directed toward systems and methods for generating a composite radiographic image that corrects effects of parallax distortion. A sequence of radiographic images—including a series of discrete exposures or image frames from a fluoroscopic procedure—may be acquired using a C-arm apparatus. An exemplary method may include receiving a plurality of radiographic image frames pertaining to a patient, identifying a region of interest on the image frames, cropping the region of interest from a plurality of image frames, selecting a plurality of sequential portions of cropped image frames, and stitching together the selected portions to form a composite image that corrects effects of parallax distortion and displaying a three-dimensional image of a part of the patient according to the orientation of the patient calculated from this two-dimensional intra-operative radiographic imaging information.

Abstract: There is provided a medical image processing apparatus which includes a first extraction unit configured to extract coronary arteries depicted in images of a plurality of time phases relating to the heart, and to extract at least one stenosed part depicted in each coronary artery; a calculation unit configured to calculate a pressure gradient of each of the extracted coronary arteries, based on tissue blood flow volumes of the coronary arteries; a second extraction unit configured to extract an ischemic region depicted in the images; and a specifying unit configured to specify a responsible blood vessel of the ischemic region by referring to a dominance map, in which each of the extracted coronary arteries and a dominance territory are associated, for the extracted ischemic region, and to specify a responsible stenosis, based on the pressure gradient corresponding to a stenosed part in the specified responsible blood vessel.

Abstract: A passenger detection device includes: an input unit that receives a captured image of an interior space of a vehicle imaged by an imaging device and a load measured by a load sensor provided on a seat provided in the interior space of the vehicle; a determination unit that determines that a passenger is seated on the seat, in a case where the load is equal to or more than a predetermined threshold value; a detection unit that obtains a face determination degree indicating a certainty of a face of the passenger in the captured image from the captured image; and a threshold value changing unit that changes the threshold value according to the face determination degree.

Abstract: This disclosure relates to a system for monitoring walls. The system receives first and second image data representing two-dimensional locations of multiple points of the wall associated with a first and second point in time, the second point in time being later than the first point in time. The system also receives depth data from a depth sensor indicative of a change in depth of the wall between the first point in time and the second point in time. The system then determines data indicative of a change of the two-dimensional locations of the multiple points of the wall between the first point in time and the second point in time based on the first image data, the second image data, and the depth data. This means correlations between the image data and the depth data can be considered, which allows accurate perpendicular line of sight monitoring.

Abstract: The present disclosure relates to an image composite system that employs a generative adversarial network to generate realistic composite images. For example, in one or more embodiments, the image composite system trains a geometric prediction neural network using an adversarial discrimination neural network to learn warp parameters that provide correct geometric alignment of foreground objects with respect to a background image. Once trained, the determined warp parameters provide realistic geometric corrections to foreground objects such that the warped foreground objects appear to blend into background images naturally when composited together.

Abstract: Described is an object recognition system. Using an integral channel features (ICF) detector, the system extracts a candidate target region (having an associated original confidence score representing a candidate object) from an input image of a scene surrounding a platform. A modified confidence score is generated based on a location and height of detection of the candidate object. The candidate target regions are classified based on the modified confidence score using a trained convolutional neural network (CNN) classifier, resulting in classified objects. The classified objects are tracked using a multi-target tracker for final classification of each classified object as a target or non-target. If the classified object is a target, a device can be controlled based on the target.

Abstract: Embodiments of the present disclosure disclose a lane line identification modeling method and apparatus, and a lane line identification method and apparatus. The lane line identification modeling method includes: identifying an image region of a lane line in an image based on two-dimensional filtering (S11); constructing model training data by using the identified image region (S12); and training a convolutional neural network-based lane line identification model by using the model training data (S13). The lane line identification modeling method and apparatus, and the lane line identification method and apparatus provided in the embodiments of the present disclosure can improve the accuracy of lane line detection.

Abstract: In various embodiments, computer program products for detecting, estimating, calculating, etc. characteristics of a document based on reference objects depicted on the document are disclosed. In one approach, a computer program product for processing a digital image depicting a document includes instructions executable by a computer for analyzing the digital image to determine one or more of a presence and a location of one or more reference objects; determining one or more geometric characteristics of at least one of the reference objects; defining one or more region(s) of interest based at least in part on one or more of the determined geometric characteristics; and detecting a presence or absence of an edge of the document within each defined region of interest. Additional embodiments leverage the type of document depicted in the image, multiple frames of image data, and/or calculate or extrapolate document edges rather than locating edges in the image.

Abstract: An image encoding method including: a constraint information generating step of generating tile constraint information indicating whether or not there is a constraint in filtering on boundaries between adjacent tiles among a plurality of tiles obtained by dividing a picture, and storing the tile constraint information into a sequence parameter set; and a filter information generating step of generating, for each of the boundaries, one of a plurality of filter information items respectively indicating whether or not filtering is executed on the boundaries, and storing the plurality of filter information items into a plurality of picture parameter sets, wherein, in the filter information generating step, the plurality of filter information items which indicate identical content are generated when the tile constraint information indicates that there is the constraint in the filtering.

Abstract: A movement vector calculation unit calculates, from a processing target frame in a motion picture and an immediately previous frame, a movement vector between the processing target frame and the immediately previous frame. An index value calculation unit calculates an index value indicating a degree of reliability of a movement from the processing target frame and the immediately previous frame. A correction unit corrects the movement vector using the index value to calculate a corrected movement vector, and a registration unit registers the immediately previous frame to the processing target frame on the basis of the corrected movement vector. A synthesis unit synthesizes the processing target frame and the registered immediately previous frame to generate a synthetic frame.

Abstract: The embodiments of the present disclosure provide an optical fingerprint recognition apparatus and a display panel. The apparatus comprises: a light emitting structure (1), a collimating structure (2) and a plurality of photosensitive sensors (3); the light emitting structure (1) is used for generating detecting light; the collimating structure (2) is used for filtering the detecting light reflected by a finger, such that light with a scattering angle in a predetermined range transmits through the collimating structure (2); the photosensitive sensors (3) are used for receiving the light transmitted through the collimating structure (2) and acquiring fingerprint information according to the received light. The embodiments of the present disclosure are used for manufacturing a display panel. The present disclosure can enhance accuracy of the recognized fingerprint information during an optical fingerprint recognition process.

Abstract: Embodiments provide slide navigation technology that addresses challenges in digital pathology of navigating and viewing high resolution slide images. Example systems comprise a virtual slide stage (VSS) having at least one sensor that detects user movement of a target placed on the VSS, and an input component, coupled to the VSS, which provides quick function movement control of the target via quick functions. The systems also comprise a connector component that connects the VSS to a user device and transmits output from the at least one sensor and input component to the user device. The systems further comprise a computer processor, in communication with the VSS, which processes the output using a computational model to generate data representing movement profiles of the target. The computer processor executes a software component, causing the output, translated based on the movement profiles, to be relayed via a viewing application on the user device.

Abstract: A system to treat a patient comprises a user interface that allows a physician to view an image of tissue to be treated in order to develop a treatment plan to resect tissue with a predefined removal profile. The image may comprise a plurality of images, and the planned treatment is shown on the images. The treatment probe may comprise an anchor, and the image shown on the screen may have a reference image marker shown on the screen corresponding to the anchor. The planned tissue removal profile can be displayed and scaled to the image of the target tissue of an organ such as the prostate, and the physician can adjust the treatment profile based on the scaled images to provide a treatment profile in three dimensions. The images shown on the display may comprise segmented images of the patient with treatment plan overlaid on the images.

Abstract: A medical image processing apparatus includes a weight applier configured to, when a difference between a first imaginary component of a first frame image and a second imaginary component of a second frame image, the second frame image being adjacent to the first frame image, is less than or equal to a first threshold value, apply a first weight to the second imaginary component to increase the difference; and an image generator configured to generate a movement-amplified image based on the first frame image and the second frame image to which the first weight is applied so that a movement of interest corresponding to the increased difference is amplified.