Abstract: Aspects of the disclosure provide fuel receptacle position estimation for aerial refueling (derived from aircraft position estimation). A video stream comprising a plurality of video frames each showing an aircraft to be refueled, is received from a single camera. An initial position estimate is determined for the aircraft for the plurality of video frames, generating an estimated flight history for the aircraft. The estimated flight history for the aircraft is used to determine a temporally consistent refined position estimate, based on known aircraft flight path trajectories in an aerial refueling setting. The position of a fuel receptacle on the aircraft is determined, based on the refined position estimate for the aircraft, and an aerial refueling boom may be controlled to engage the fuel receptacle. Examples may use a deep learning neural network (NN) or optimization (e.g., bundle adjustment) to determine the refined position estimate from the estimated flight history.

Abstract: An image processing method includes a step of acquiring, from an image data acquisition device and a three-dimensional data acquisition device that are calibrated, a plurality of pieces of unit image data in which image data and three-dimensional data are associated with each other, for a target object, for each angle of view of the image data acquisition device; a step of estimating a plane, based on a plurality of pieces of three-dimensional data included in the plurality of pieces of unit image data; a step of acquiring image group data by grouping a plurality of pieces of image data included in a plurality of pieces of unit image data belonging to the estimated plane; and a step of selecting image data necessary to form the plane from among the plurality of pieces of image data included in the image group data. Thus, image data can be reduced.

Abstract: Embodiments of the invention provide automatic detection of same-type objects on an inspection line and automatically capture and/or display an image of the object once it is detected. Additionally, embodiments of the invention automatically provide the outline of an object to a user, enabling the user to either confirm or easily correct the outline. The confirmed or corrected outline is then automatically applied to all images of inspected items of the same-type, allowing optimization of inspection tasks such as defect detection, for all inspected items, based on a short, possibly a one-time, input, from the user.

Abstract: An object detection method includes obtaining a first video stream having video images with object detection data of an area, identifying an object in the area by identifying one or more first object detection data of the object in the first video stream corresponding to validated object detection data, determining a position of the object in the area, obtaining a second video stream having video images with object detection data of the area, identifying second object detection data of the second video stream at the position of the object and determining whether the second object detection data is un-validated object detection data, and updating the validated object detection data of the object detection system if it is determined that the second object detection data of the object is un-validated object detection data.

Abstract: A system for verifying a machine-readable label comprises a scan table processing device comprising a first input for receiving a list of items with machine-readable labels; a second input for receiving a list of stores that have an inventory of the items in the list of items and that have at least one sensing device for capturing images of the items; and an output that includes a plurality of electronic records. The system further comprises a data repository that stores the captured images of the items and that updates the electronic records to include an association to the captured images; a graphical user interface (GUI) processing apparatus that modifies the captured images in preparation for training an artificial intelligence apparatus to identify the items in the images; and a machine language (ML) model processor that determines whether the images training the artificial intelligence apparatus are correctly identified with machine-readable labels associated with the items.

Abstract: Embodiments relate to a deep learning/AI-based product surface quality inspection system which is accurate and reliable in product quality inspection which is a core task in an injection process among various manufacturing fields. The system can provide, to a user, better performance than non-defective/defective manufactured product classification methodology which is an existing commonly used method through a method and a system considering characteristics of a factory environment and an actual product production process for all pipelines of product quality inspection by using only a non-defective manufactured product image unlike most injection process surface inspection AIs developed to date.

Abstract: This invention provides an information processing apparatus comprising an image creating unit which creates, from a first image associated with global coordinates, a partial image that is a portion of the first image, as a second image, a recognition processing unit which executes, on the second image, recognition processing of a preset characteristic, and creates a recognition result associated with local coordinates of the second image, and a coordinate converting unit which converts the coordinates of the recognition result from the local coordinates into the global coordinates.

Abstract: According to one embodiment, a defect management apparatus includes a processor. The processor acquires first information and second information, the first information including first defect positions relating to defects detected with a first device for an inspection target and corresponding first labels indicating classifications of the defects, the second information including second defect positions relating to defects detected with a second device for the inspection target. The processor determines a first defect position corresponding to a second defect position as a corresponding defect position. The processor diverts the first label corresponding to the corresponding defect position as a second label of the second defect position.

Abstract: Systems and methods are disclosed for driver monitoring. In one implementation, one or more images are received, e.g., from an image sensor. Such image(s) can reflect a at least a portion of a face of a driver. Using the images, a direction of a gaze of the driver is determined. A set of determined driver gaze directions is identified using at least one predefined direction. One or more features of one or more eyes of the driver are extracted using information associated with the identified set.

Abstract: The embodiments of the present disclosure disclose an image processing method and device. The image processing method comprises transforming a first image to obtain a plurality of second images; obtaining feature maps of each of the second images by performing feature extraction on the second images using a first machine learning unit selected from a group including at least one first machine learning unit; and inputting the feature maps of each of the second images to a second machine learning unit to obtain a processing result of the first image.

Abstract: This application discloses an image bloom processing method and apparatus, and a storage medium, and relates to the field of image processing technologies. An electronic device obtains a first brightness region class in a first image, where the first brightness region class includes one or more brightness regions in the first image; after determining that the first brightness region class in the first image is the same as a target brightness region class in a second image on which bloom processing is performed, the electronic device obtains a first intermediate image obtained after Gaussian blur processing is performed on the target brightness region class in the second image; and the electronic device generates a bloom image of the first image based on the first image and the first intermediate image.

Abstract: A system is disclosed, in accordance with one or more embodiment of the present disclosure. The system may include a controller including one or more processors configured to execute a set of program instructions. The set of program instructions may be configured to cause the processors to: receive images of a sample from a characterization sub-system; identify target clips from patch clips; prepare processed clips based on the target clips; generate encoded images by transforming the processed clips; sort the encoded images into a set of clusters; display sorted images from the set of clusters; receive labels for the displayed sorted images; determine whether the received labels are sufficient to train a deep learning classifier; and upon determining the received labels are sufficient to train the deep learning classifier, train the deep learning classifier via the displayed sorted images and the received labels.

Abstract: A method for inspecting nominally identical workpieces with manufacturing tolerances involves feeding a workpiece to an inspection device; applying a simultaneous translatory and rotational movement to the workpiece via the inspection device while illuminating the workpiece; repeatedly taking camera images of the workpiece during the simultaneous translatory and rotational movement as the workpiece moves across a camera window, the camera images consisting of pixels; setting workpiece parameters based on size and shape of the workpiece; assembling a composite image from the camera images based on the workpiece parameters; determining numerical brightness scores of the workpiece, comparing the determined numerical brightness scores with known numerical brightness scores; assigning a discrepancy score; setting a discrepancy score threshold; determining that the workpiece passes inspection when the discrepancy score is below the discrepancy score threshold; and that the workpiece fails the inspection when the di

Abstract: Systems and methods for detecting and marking quality defects on individual packaging products being processed through converting machines at high speeds are provided. An example system comprises a controller in communication with a first and second sensor as well as a first and second marker. The first and second sensors sense data corresponding to a first type and second type, respectively, of potential defects in the manufacture of the packaging product, enabling detection of a first- and second-type defect. The controller may be configured to trigger a first marker to mark the individual packaging product with a first mark for the first-type defect; and to trigger a second marker to mark the packaging product with a second mark for the second-type defect. Thus, the type of defects detected on an individual packaging product may be readily indicated on the individual packaging product itself.

Abstract: Systems and methods for real-time user verification in online education are disclosed. In certain example embodiments, user identifying information associated with a user and a request to access online education content may be received from a user device. A face template including historical facial image data for the user can be identified. Current facial image data can be compared to the face template to determine if a match exists. Biometric sensor data, such as heart rate data, may also be received for the user. The biometric sensor data may be evaluated to determine if the user is currently located at the user device. If the user is currently located at the user device and the current facial image data matches the face template, access to the online education content may be provided to the user at the user device.

Abstract: This application discloses an inspection method and an inspection machine for a display panel. The inspection method includes the steps of: taking a picture of a to-be-inspected display panel to obtain a to-be-inspected image of the to-be-inspected display panel; removing high-frequency information in the to-be-inspected image to obtain a high-frequency-removed image; calculating a difference between the to-be-inspected image and the high-frequency-removed image to obtain a difference image; making a determination regarding the difference image by comparing the difference image with a preset threshold, determining that the display panel fails the inspection if the difference image exceeds the preset threshold, and determining that the display panel passes the inspection if the difference image does not exceed the preset threshold.

Abstract: Described embodiments relate to a method comprising: determining a candidate document comprising image data and character data and extracting the image data and the character data from the candidate document. The method comprises providing, to an image-based numerical representation generation model, the image data, and generating, by the image-based numerical representation generation model, an image-based numerical representation of the image data. The method comprises providing, to a character-based numerical representation generation model, the character data; and generating, by the character-based numerical representation generation model, a character-based numerical representation of the character data.

Abstract: A basic pattern extracting unit (15) extracts a “basic pattern” for each human from detection points acquired by an acquiring unit (11). The “basic pattern” includes a “reference body region point” corresponding to a “reference body region type”, and base body region points corresponding to base body region types that are different from the reference body region type and that are different from each other. For example, the “basic pattern” includes at least one of the following two combinations. A first combination is a combination of the reference body region point corresponding to a neck as the reference body region type and two base body region points respectively corresponding to a left shoulder and a left ear as the base body region type. A second combination is a combination of body region points corresponding to a neck, a right shoulder and a right ear.

Abstract: The disclosure provides a plant blooming period broadcast method and system, and a computer-readable storage medium. The method comprises: receiving an image, and identifying a plant in the image to obtain the species of the plant; using a plant variety identification model corresponding to the species of the plant to identify the specific variety and blooming state of the plant; obtaining the photographing time and photographing position of the image, and determining the blooming period of the plant according to the photographing time and the blooming state; and marking the photographing position on a blooming period broadcast map as a viewing place of the plant, and correspondingly displaying the specific variety and blooming period of the plant.

Abstract: Provided is an apparatus for analyzing composition of electronic and electrical device part scraps which can determine a composition of part scraps in the electronic and electrical device part scraps in a short time, a device for processing electronic and electrical device part scraps, and a method for processing electronic and electrical device part scraps using those devices.