Abstract: The objective of the present invention is to provide an object detection method such that an object such as a vehicle can be detected from an image captured by an on-board camera even when a lens of the on-board camera is dirty. In order to achieve the objective, in this object detection method for detecting an object included in a captured image, an original image including the object to be detected is prepared from the captured image, a processed image is generated from the prepared original image by applying predetermined processing to the original image, and learning regarding restoration of an image of the object to be detected is performed using the original image and the processed image.

Abstract: Systems and methods for use in identifying a brain condition of a subject are provided. In some aspects, a provided method includes constructing a classifier to identify a brain condition of a subject comprising steps of receiving image data obtained from a plurality of subjects, wherein the image data is acquired during an acquisition period following administration of at least one radioactive tracer. The method also includes defining a plurality of brain condition classes using the image data associated with one or more time frames during the acquisition period, and processing the image data to generate signatures corresponding to each of the plurality brain condition classes. The method further includes constructing the classifier using the signatures. The classifier can then be applied to determine a degree to which the subject expresses one or more disease states in order to determine a brain condition of the subject.

Abstract: Embodiments of the disclosure may be directed to an image processing system configured to receive a medical image of a region of a subject's body taken at a first time and to receive a surface image of an exterior portion of the region of the subject's body taken at the first time. The image processing may also be configured to receive a medical image of the region of the subject's body taken at a second time and to register the medical image taken at the first time, the surface image taken at the first time, and the medical image taken at the second time.

Abstract: The present disclosure relates to the technical field of medical image processing and, in particular, to a liver boundary identification method and a system. The method includes: obtaining liver tissue information of a liver tissue to be identified; identifying a liver tissue boundary in the liver tissue information according to a feature of the liver tissue corresponding to the liver tissue information and a feature of the liver tissue boundary corresponding to the liver tissue information using an image processing technology or a signal processing technology; and outputting position information of the identified liver tissue boundary. By using the disclosed method, the liver tissue boundary can be identified automatically, the efficiency of identifying the liver boundary can be improved, and automatic positioning of the liver boundary can thus be achieved.

Abstract: An extraneous-matter detecting apparatus including a dividing unit that divides a captured image into a plurality of regions. The captured image is output from an image capturing device; a computing unit detecting luminance values of pixels included in each of the regions divided by the dividing unit, and computes, for each of target regions, an average value or a standard deviation of luminance values of the detected luminance values. The corresponding target region corresponds to the one or more regions; and a detection unit that extracts the target regions having the average values or the standard deviations of the luminance values computed by the computing unit, which are within a predetermined range, and detects, when the extracted target regions are adjacent to each other, the adjacent target regions as an extraneous-matter region. The extraneous-matter region is a region of the image capturing device to which an extraneous matter is adhered.

Abstract: A method of detecting failure of an object tracking network with a failure detection network includes receiving an activation from an intermediate layer of the object tracking network and classifying the activation as a failure or success. The method also includes determining whether to initiate a recovery mode of the object tracking network or to remain in a tracking mode of the object tracking network, based on the classifying.

Abstract: The present disclosure provides for improved image analysis via novel deblurring and segmentation techniques of image data. These techniques advantageously account for and incorporate segmentation of biological analytes into a deblurring process for an image. Thus, the deblurring of the image may advantageously be optimized for enabling identification and quantitative analysis of one or more biological analytes based on underlying biological models for those analytes. The techniques described herein provide for significant improvements in the image deblurring and segmentation process which reduces signal noise and improves the accuracy of the image. In particular, the system and methods described herein advantageously utilize unique optimization and tissue characteristics image models which are informed by the underlying biology being analyzed, (for example by a biological model for the analytes). This provides for targeted deblurring and segmentation which is optimized for the applied image analytics.

Abstract: The present disclosure relates to a method and system for registering multi-modality images. The method may include: acquiring a first image relating to a registration model, wherein the registration model includes a plurality of reference objects; acquiring a second image relating to the registration model; determining a set of reference points based on the plurality of reference objects; determining a set of mapping data corresponding to the set of reference points in the first image and the second image; and determining one or more registration parameters based on the set of mapping data.

Abstract: A medical image processing device is disclosed. The disclosed medical image processing device may include: an input interface through which a depth adjusting command is input from a user; an image processor and controller generating a two-dimensional reconstruction image by overlapping a part or all of CT image data in a view direction, and changing a contrast of at least a part of the two-dimensional reconstruction image according to the depth adjusting command; and a display part displaying the two-dimensional reconstruction image.

Abstract: A method includes extracting a plurality of feature points from a first image and a second image of an image-capture target, acquiring, from the second image, for each of the plurality of feature points in the first image, a set of points having a correlation with the feature point, identifying, for each of the plurality of feature points in the second image, a position of a center of gravity of the set of points, and determining an amount of movement of the image-capture target by performing weighting processing, the weighting processing being performed on each of amounts of displacements and a number of points in the set of points acquired for each of the plurality of feature points, each of the amounts of displacements being a displacement from a position of each of the plurality of feature points in the first image to the position of the center of gravity.

Abstract: Described is a system for multi-view embedding for object recognition. During operation, the system receives an input image and transforms raw data of objects in the image with corresponding labels into low-level features and high-level semantic representations of the labels. A trained object recognition model is generated by embedding the low-level features with multiple high-level semantic representations. The system then receives data of an unknown object and assigns a label to the unknown object using the trained object recognition model. Finally, a device can be controlled based on the label.

Abstract: A method for operating a robotic system includes obtaining and processing first data representative of an object at a start location; obtaining additional data as the object is transferred to a task location, wherein the additional data includes information regarding one or more edges and/or one or more surfaces separate from portions of the object captured in the first data; and creating registration data representative of the object based on the additional data.

Abstract: A medical image display apparatus for displaying a medical image generated from first medical image data obtained from an object by a probe, including at least one processor configured to obtain information about a position at which tissue from the object is extracted by a needle, based on a signal received from a sensor attached to at least one from among the probe and the needle, and locate a tissue extraction position in the first medical image data based on the information, and a display configured to display a first medical image on which the tissue extraction position is marked, the first medical image being generated from the first medical image data.

Abstract: A computer implemented method comprises receiving one or more images of a computing environment comprising a plurality of interconnected components, analyzing the or each received image to identify each component shown in the image(s) and the connection(s) of each identified component, by identifying a set of attributes for each component from the image(s) and matching the identified attributes to attributes of known components stored in a database, obtaining a specification for each identified component, and generating a document comprising each identified component, its respective specification and the connection(s) of each identified component.

Abstract: An exemplary system, method and computer accessible medium for generating an image(s) of a portion(s) of a patient can be provided, which can include, for example, receiving first imaging information related to the portion(s), receiving second information related to modelling information of a further portion(s) of a further patient(s), where the modelling information includes (i) an under sampling procedure, and/or (ii) a learning-based procedure, and generating the image(s) using the first information and the second information. The modelling information can include artifacts present in a further image of the further portion(s). The image(s) can be generated by reducing or minimizing the artifacts. The second information can be generated, for example using a variational network(s).

Abstract: The invention relates to the field of intelligent electronics, and more particularly, to a rapid recognition method and a household intelligent robot. The method, applicable to the household intelligent robot, comprises the steps of: pre-setting a plurality of personal files corresponding to different users; collecting identification information associated with features of the user, and establishing an association between the identification information and the personal file corresponding to the user; the household intelligent robot collecting the features of the user and matching the user features with stored identification information, so as to identify the user; if the user is successfully identified, executing a retrieving step, otherwise, exiting; and the retrieving step comprising retrieving the corresponding personal file according to the identified user, and working according to the personal file.

Abstract: A content saliency network is a machine-learned neural network that predicts the saliency of elements of a content item. The content saliency network may be used in a method that includes determining a set of elements in a UI and computing a first context vector for the content. The method may also include, for each element in the set of elements, computing a vector of simple features for the element, the simple features being computed from attributes of the element, computing a second context vector for the element, computing a third context vector for an intermediate context of the element, and providing the vectors to the content saliency network. The content saliency network provides a saliency score for the element. The method further includes generating an element-level saliency map of the content using the respective saliency scores for the set of elements and providing the saliency map to a requestor.

Abstract: Provided is a method and system for determining a complex situation according to the time flow of events occurring in each device. In a method for determining a complex situation according to an embodiment of the present invention, a current situation is determined by detecting events occurring in devices and referring to a complex situation determination rule. The complex situation determination rule is a rule in which the current situation is mapped according to the events, which have occurred in the devices, and time intervals between the events. The present invention enables a more reliable determination of the current situation by complexly considering not only the events occurring in the devices but also time intervals between the events.

Abstract: An image processing apparatus allows strength of color moire reduction processing to be changed according to a parameter at the time of development in the processing for reducing color moire. A processor of this image processing apparatus performs development processing on an image, acquires a singular value with respect to a pixel value contained in a partial region in the image on which the development processing has been performed, corrects the singular value according to a value of a parameter in the development processing, and calculates a corrected pixel value in the partial region with use of the corrected singular value.

Abstract: Method for generating a test-set for inspection of a design being printed by a printing-press, each color-unit in the printing-press prints a respective color. The design is composed of original-layers. Inspection includes determining the origin of at least one defect in the printed-design. The method includes the procedures of generating defective-layer or layers of the design, by introducing at least one selected defect to at least one selected original-layer, in a selected location or locations and combining layers using a trained-synthesis-neural-network. The layers include the defective-layer or layers and remaining ones of the original-layers. The trained-synthesis-neural-network provides a plurality of features respective of each pixel. The method also includes the procedure of generating the test-set from the output of the synthesis-neural-network. The test-set includes at least one synthesized-test-image. The synthesized-test-image includes at least one synthesized-defect at the selected location.