Abstract: A system is disclosed for auditing waste retrieved by a service vehicle. The system may have an optical sensor mountable onboard the service vehicle and configured to capture image data associated with the waste as the waste falls into the service vehicle during completion of a waste service activity. The system may also have at least one controller in communication with the optical sensor and being configured to generate at least one of an alert and a recommendation regarding a mix of the waste based on the image data.

Abstract: A vascular assessment apparatus is disclosed. The apparatus is configured to receive a stenotic model having measurements of a coronary vessel tree of a subject, the stenotic model created from a set of medical images recorded before the subject underwent a stent procedure. The apparatus is also configured to receive a post-medical image of the coronary vessel tree of the subject after the stent has been placed into the coronary vessel tree and identify vascular features within the post medical image that correspond to vascular features that are provided among the set of medical images. The apparatus is further configured to modify geometrical information for the corresponding vascular features in the stenotic model to create an updated stenotic model from the identified vascular features of the post medical image and calculate an index indicative of vascular function, based, at least in part, on the updated stenotic model.

Abstract: In an X-ray CT apparatus of an embodiment, an X-ray tube emits X-rays. A detector detects the X-rays emitted from the X-ray tube and having passed through a subject. Processing circuitry collects projection data based on data detected by the detector. The Processing circuitry generates a reconstructed image from the projection data. A display displays a display image based on the reconstructed image. A input circuitry receives an operation to rotate a first display image based on a first reconstructed image generated by the processing circuitry on a screen of the display, and specify a certain region on a second display image whose axis is in a direction different from a slice direction. The processing circuitry generates a second reconstructed image from the projection data, having higher resolution than that of the first display image, for the certain region.

Abstract: There is provided a computerized method and system of controlling a quality measure in a compression quality evaluation system, the method comprising: calculating a value indicative of a characteristic of an input image; and configuring the quality measure upon a criterion being met by the value, the quality measure being indicative of perceptual quality of a compressed image compressed from the input image. The value can be a smoothness value indicative of an extent of smoothness of an input image, and the configuration of the quality measure can be performed upon a smoothness criterion being met by the smoothness value. Alternatively, the value can be a gradient value indicative of an extent of gradual spatial changes of an input image, and the configuration of the quality measure can be performed upon a gradient criterion being met by the gradient value.

Abstract: A method and an apparatus for recognizing a specific object inside an image in an electronic device are provided. The method includes displaying at least one image; detecting at least one gesture; selecting a recognition function related to at least one object existing in the at least one image according to the detected at least one gesture; and recognizing the at least one object using the selected recognition function.

Abstract: A license plate number input device comprises: a characteristic information extraction unit for extracting characteristic information on a license plate from a vehicle image; a keypad provision unit for providing at least one input keypad on the basis of the extracted characteristic information; and an out unit for displaying input information input by a user through the provided input keypad.

Abstract: A system for registering a three dimensional map of an environment includes a data collection device, such as a robotic device, one or more sensors installable on the device, such as a camera, a LiDAR sensor, an inertial measurement unit (IMU), and a global positioning system receiver. The system may be configured to use the sensor data to perform visual odometry, and/or LiDAR odometry. The system may use IMU measurements to determine an initial estimate, and use a modified generalized iterative closest point algorithm by examining only a portion of scan lines for each frame or combining multiple feature points across multiple frames. While performing the visual and LiDAR odometries, the system may simultaneously perform map registration through a global registration framework and optimize the registration over multiple frames.

Abstract: An image processing apparatus includes an association unit, a storage unit, a first estimation unit, and a second estimation unit. The association unit associates detection results of a first object and a second object. The storage unit stores a positional relationship between respective fields of view of a first image capturing unit and a second image capturing unit. The first estimation unit estimates a three-dimensional position of the first object based on the positional relationship and the first object detected in respective images captured by the first image capturing unit and the second image capturing unit. The second estimation unit estimates a three-dimensional position of the second object based on the three-dimensional position of the first object associated by the association unit.

Abstract: Embodiments are directed toward analyzing images of cables and electronic devices to augment those images with information relating to the installation or troubleshooting of such cables and electronic devices. The images are analyzed to determine non-text characteristics of a connector of the cable and non-text characteristics of at least one port on the electronic device. These non-text characteristics can be compared to each other to determine if the connector is compatible with one of the ports on the electronic device. Similarly, these non-text characteristics can be compared with non-text characteristics of known connectors and ports to determine a type of the connector and a type of the ports on the electronic device, which is used to determine their compatibility. The images are then modified or overlaid with information identifying the type of connector or port, their compatibility or lack thereof, or instructions for connecting the compatible connector and port.

Abstract: A system and method for tracking and annotating objects in a 3D model is disclosed. The method includes receiving a signal feed from a group of sensors including at least one depth sensor, determining a reference signal feed from the signal feed, determining a volumetric representation of at least one object not present in the reference signal feed, assigning a unique volumetric profile to at least one object, and/or storing the unique volumetric profile in an object database. The signal feed may include at least one 3D area of interest. Also, the unique volumetric profile may include a unique volumetric id obtained from a group of attributes. The unique volumetric profile may include at least a unique volumetric position and a unique volumetric centroid. The group of sensors may further include video cameras, thermal, and chemical sensors.

Abstract: A method includes determining a registration transform between first three dimensional pre-scan image data and second three dimensional pre-scan image data based on a predetermined registration algorithm. The method further includes registering first volumetric scan image data and second volumetric scan image data based on the registration transform. The method further includes generating registered image data. A system (100) includes a pre-scan registerer (122) that determines a registration transform between first three dimensional pre-scan image data and second three dimensional pre-scan image data based on a predetermined registration algorithm. The system further includes a volume registerer (126) that registers first volumetric scan image data and second volumetric scan image data based on the registration transform, generating registered image data.

Abstract: In one embodiment, at least one of number of times a frame is paused by a plurality of users (users) and attention-activity of the users for the frame is tracked for each frame of an asset (video etc.). An interactive version of at least one frame is pre-generated based on the tracking. The interactive version is stored to enable playing of the interactive version of the at least one frame. In another embodiment, pausing of a currently playing frame (frame) of the asset is determined. A determination to replace the frame is made based on at least one of attention-activity of a user in the frame, and detecting metadata, of the frame, specifying that the frame is to be replaced. An interactive version of the frame is generated, based on at least one of the attention of the user and the metadata, to replace the frame with the interactive version.

Abstract: A device for use in identifying an explosive includes a processor and an induction heat source in communication with the processor. The induction heat source is configured to emit radiation to heat a metallic component of an explosive device by way of induction. The device also includes a temperature sensor in communication with the processor that is configured to detect heat emitted from the metallic component of the explosive device. The processor is configured to identify a location of the metallic component of the explosive device based on the detected heat. A gas sensor is used to detect one or more gases emitted from a non-metallic explosive device. The device further includes an action arm configured to conduct a detonation attempt at the location of the metallic component of the explosive device.

Abstract: For image quality scoring of an image from a medical scanner, a generative model of an expected good quality image may be created using deep machine-learning. The deviation of an input image from the generative model is used as an input feature vector for a discriminative model. The discriminative model may also operate on another input feature vector derived from the input image. Based on these input feature vectors, the discriminative model outputs an image quality score.

Abstract: A decision support method for use by an operator surrounded by adverse entities in a battlefield environment comprises generating a layered representation of the physical environment surrounding the operator from sensor information by mapping the spherical physical environment of the operator into a geometrical representation suitable for display on a screen, the representation being segmented into a plurality of layers having respective sizes, each layer being associated with a respective category of tactical actions. The representation further comprises visual elements representing adverse entities in the surrounding physical environment of the operator, each visual element being represented so as to be superposed with a given layer.

Abstract: Systems and methods track one or more points between images. A point for tracking may be selected, at least in part, on a determination of how discriminable the point is relative to other points in a region containing the point. A point of an image being tracked may be located in another image by matching a patch containing the point with another patch of the other image. A search for a matching patch may be focused in a region that is determined based at least in part on an estimate of movement of the point between images. Points may be tracked across multiple images. If an ability to track one or more points is lost, information about the points being tracked may be used to relocate the points in another image.

Abstract: A device for use in a security system includes a sensor, an image capture device, an action arm, and a processor in communication with the sensor, the image capture device, and the action arm. The processor is configured to receive data from the sensor and the image capture device and identify a target based on an analysis of the received data. The processor is also configured to determine, based on the analysis of the received data, that the target is a drone or other mechanical device. The processor is also configured to determine an action to take on the identified target. The processor is further configured to cause the action arm to perform the determined action on the target.

Abstract: An objective of the present invention is to obtain a vehicle-mounted recognition device capable of recognizing an object to be recognized in a shorter processing time than previously possible.

Abstract: An image upsampling system, a training method thereof and an image upsampling method are provided, the feature images of an image are obtained by using the convolutional network, upsampling processing is performed on the images with the muxer layer to synthesize every n×n feature images in the input signal into a feature image with the resolution amplified by n×n times, in the upsampling procedure with the muxer layer, information of respective feature images in the input signal is recorded in the generated feature image(s) without loss; and thus, every time when the image passes through a muxer layer with an upsampling multiple of n, the image resolution can be increased by n×n times.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying an object from a video. One of the methods includes obtaining multiple frames from a video, where each frame of the multiple frames depicts an object to be recognized, and processing, using an object recognition model, the multiple frames to generate data that represents a classification of the object to be recognized.