Abstract: An apparatus for controlling docking with a spacecraft includes at least one camera for generating at least one image pixel stream of the spacecraft. A field programmable gate array (FPGA) receives the at least one image pixel stream from the at least one camera. The FPGA texture processes the at least one pixel stream to generate at least one texture map for the at least one image pixel stream. A processor receives the at least one texture map from the FPGA and calculates texture map statistics responsive to the generated at least one texture map, generates thresholding results for the at least one pixel image stream responsive to the generated at least one texture map, determines a bus centroid of the spacecraft responsive to the generated thresholding results and outputs the determined bus centroid of the spacecraft. A docking controller controls docking of with the spacecraft responsive to the determined bus centroid of the spacecraft.

Abstract: Devices, systems and processes for the detection of unsafe cabin conditions that provides a safer passenger experience in autonomous vehicles are described. One example method for enhancing passenger safety includes capturing at least a set of images of one or more passengers in the vehicle, determining, based on the set of images, the occurrence of an unsafe activity in an interior of the vehicle, performing, using a neural network, a classification of the unsafe activity, and performing, based on the classification, one or more responsive actions.

Abstract: Detection of pathological abnormalities in tissue samples and/or pluralities of cells is a highly specialized and time-consuming effort, usually performed by a select group of clinicians and technical personnel. Described herein are methods for more automatable, consistent and comprehensive cell sample analysis to deliver a rapid, reliable and detailed classification, e.g., diagnosis, of the status of cells present in a sample, particularly, but not limited to cancer diagnosis.

Abstract: Provided is a system and method of generating training data for training a Deep Neural Network usable for examination of a semiconductor specimen. The method includes: obtaining a first training image and first labels respectively associated with a group of pixels selected in each segment, extract a set of features characterizing the first training image, train a machine learning (ML) model using the first labels, values of the group of pixels, and the feature values of each of the set of features corresponding to the group of pixels, process the first training image using the trained ML model to obtain a first segmentation map, and determine to include the first training image and the first segmentation map into the DNN training data upon a criterion being met, and to repeat the extracting of the second features, the training and the processing upon the criterion not being met.

Abstract: Provided herein is a system and method that acquires data and determines a driving action based on the data. The system comprises a processor configured to acquire data of nonuniform resolution over a field of view of the sensor, and a controller configured to determine a driving action of a vehicle based on the data, and perform the driving action.

Abstract: A computing system includes one or more processors and a memory storing instructions that, when executed by the one or more processors, cause the system to perform operations. The operations include obtaining sensor data from a sensor of a vehicle, the sensor data including point cloud frames at different positions, orientations, and times, the sensor data used to generate a map, determining a position and an orientation of the sensor corresponding to a capture of each of the point cloud frames according to a simultaneous localization and mapping (SLAM) algorithm, and depicting, on an interface, a graphical illustration of the determined positions at which the point cloud frames were captured.

Abstract: A lane-line recognizing apparatus for a vehicle includes an edge-point detector and an approximate-line calculation processor. The edge-point detector is configured to detect edge points on the basis of brightness variation within a detection region for a lane line. The approximate-line calculation processor is configured to calculate an approximate line of a point group including the edge points. The lane-line recognizing apparatus has: a first mode in which the lane-line recognizing apparatus is configured to mainly search for standard edge candidate points having brightness relatively high; and a second mode in which the lane-line recognizing apparatus is configured to mainly search for opposite edge candidate points having brightness relatively low. The lane-line recognizing apparatus is configured to selectively perform switching between the first and the second modes in accordance with the number of the detected opposite edge candidate points.

Abstract: Provided is a method for locating a human target using superficial vein characteristics, which includes acquiring a model displaying both of the human target and superficial veins; projecting the model on a body surface where the superficial veins are located, so that a ratio of a projection on the body surface to actual dimensions of a human body is (0.9-1.1):1; and adjusting a position of the projection of the model, observing the superficial veins, and making at least two of the superficial veins coincide with veins in the projection of the model, and determining that a position of the human target in the projection of the model is a projection of the human target on the actual body surface.

Abstract: Positron emission tomography (PET) with partially known attenuation accounts for the missing attenuation. Since a computed tomography (CT) scan may provide attenuation for less than all the locations used in PET reconstruction, artificial intelligence corrects for the missing attenuation. For example, the unknown attenuation or attenuation correction factors are estimated by the artificial intelligence. The known and estimated attenuations or correction factors are used in the PET reconstruction, providing more uniform PET sensitivity and better accounting for scatter. As another example, the artificial intelligence alters intensity of the activity in some locations to account for reconstruction with missing attenuation information, correcting for sensitivity variation and/or lack of scatter information for some locations.

Abstract: A method involves use of multiple convolutional neural networks and multiple segmentation masks to programmatically generate a stool rating for a digital image of a diaper with stool. The method includes generating, by a first convolutional neural network, a first mask representing an identification of an area of the digital image that corresponds to stool, and a second mask representing an identification of an area of the digital image that corresponds to a diaper. The method further includes generating a third mask representing an intersection of the first and second masks, and generating a modified digital image utilizing the third mask. The method further includes determining, by a second convolutional neural network, a stool rating for the digital image of the diaper with stool by utilizing the modified digital image as input for the second convolutional neural network.

Abstract: An operation method with fingerprint event recognition, an apparatus, and a mobile terminal relate to the field of communications technologies. The method includes receiving a fingerprint event from a user at a preset position on a mobile terminal, and displaying on a display screen of the mobile terminal at least one shortcut when the fingerprint matches a stored fingerprint event. A sliding gesture input from the user is detected, the sliding gesture starting at the preset position and extending from the preset position to select a first shortcut from the at least one shortcut, thereby improving operation efficiency.

Abstract: A method of processing functional OCT image data, acquired by an OCT scanner scanning a retina that is being repeatedly stimulated by a light stimulus, to obtain a response of the retina to the light stimulus, comprising: receiving OCT image data generated by the OCT scanner repeatedly scanning the retina over a time period, and a sequence of stimulus indicators each indicative of a stimulation of the retina by the light stimulus in a respective time interval of a sequence of time intervals spanning the time period; calculating, for each stimulus indicator, a product of the stimulus indicator and a respective windowed portion of the sequence of B-scans comprising a B-scan based on a portion of the OCT image data generated while the retina was being stimulated in accordance with the stimulus indicator; and combining the calculated products to generate the indication of the response.

Abstract: The technology provides a sign detection and classification methodology. A unified pipeline approach incorporates generic sign detection with a robust parallel classification strategy. Sensor information such as camera imagery and lidar depth, intensity and height (elevation) information are applied to a sign detector module. This enables the system to detect the presence of a sign in a vehicle's externa environment. A modular classification approach is applied to the detected sign. This includes selective application of one or more trained machine learning classifiers, as well as a text and symbol detector. Annotations help to tie the classification information together and to address any conflicts with different the outputs from different classifiers. Identification of where the sign is in the vehicle's surrounding environment can provide contextual details.

Abstract: Provided herein are devices, systems, and methods for a three-dimensional registering, tracking, and/or guiding an object of interest, such a body part, a surgical tool, or an implant, during a surgical procedure. Such devices, systems, and methods may offer minimally invasive, high precision registering, tracking, and/or guiding of the object of interest using a patterned light beam and data processing using artificial intelligence. The methods, devices, and systems disclosed herein may be compatible with a simple marker or markers placed by a minimally invasive method on the object of interest.

Abstract: A method, apparatus, device and storage medium for recognizing a bill image may include: performing text detection on a bill image, and determining an attribute information set and a relationship information set of each text box of at least two text boxes in the bill image; determining a type of the text box and an associated text box that has a structural relationship with the text box based on the attribute information set and the relationship information set of the text box; and extracting structured bill data of the bill image, based on the type of the text box and the associated text box that has the structural relationship with the text box.

Abstract: A method and system for activity classification. A pressure sensor receives input data resulting from physical activity of a subject performing an activity. The input data includes pressure data from at least one pressure sensor, and may include other data acquired through other types of sensors. A deep learning neural network is applied to the input data for identifying the activity. The neural network is trained with reference to training data from a training database. The training data may include empirical data from a database of previous data of corresponding activities, synthesized data prepared from the empirical data or simulated data. The training data may include data from physical activity of the subject being monitored by the system. Different aspects of the neural network may be trained with reference to the training data, and some aspects may be locked or opened depending on the application and the circumstances.

Abstract: An example operation may include one or more of receiving, by an accident processing node, an accident report generated by a transport, retrieving, by the accident processing node, a time and a location of the accident from the accident report, querying, by the accident processing node, a plurality of transport profiles on a storage based on the time and the location of the accident, and retrieving video data associated with the time and the location of the accident from the plurality of the transport profiles.

Abstract: A system uses a machine learning based model to determine attributes describing states of mind and behavior of traffic entities in video frames captured by an autonomous vehicle. The system classifies video frames according to traffic scenarios depicted, where each scenario is associated with a filter based on vehicle attributes, traffic attributes, and road attributes. The system identifies a set of video frames associated with ground truth scenarios for validating the accuracy of the machine learning based model and predicts attributes of traffic entities in the video frames. The system analyzes video frames captured after the set of video frames to determine actual attributes of the traffic entities. Based on a comparison of the predicted attributes and actual attributes, the system determines a likelihood of the machine learning based model making accurate predictions and uses the likelihood to generate a navigation action table for controlling the autonomous vehicle.

Abstract: Provided is an information processing device that includes a movement estimation unit that estimates position and posture movement by using corresponding-point information obtained based on image capturing, and a corresponding-point information selection unit that selects the corresponding-point information to be used by the movement estimation unit from among a plurality of pieces of corresponding-point information. The plurality of pieces of corresponding-point information includes first corresponding-point information obtained based on image capturing by a first camera set including a first camera and a second camera, and second corresponding-point information obtained based on image capturing by a second camera set including a third camera and a fourth camera. The first camera set has a wider view angle than that of the second camera set. The first camera set and the second camera set have image capturing ranges that are at least partially identical to each other.

Abstract: A method of a mobile device configured to photograph an object includes controlling a display panel to display an image, a touch panel to receive a user input, and a network interface to wirelessly communicate with an external network, unlocking the mobile device when the mobile device is in an area condition, a wireless communication condition, and a time condition, and unlocking the mobile device according to a password when the mobile device is not in the time condition.