Abstract: Some embodiments are directed to a method of estimating a velocity of a contrast agent. The method includes receiving a plurality of video frames that were produced using a dynamic contrast enhanced imaging process, each video frame including a plurality of pixels/voxels. Information from the video frames is used to estimate velocity vectors indicating the velocity and direction of the agent with the vascular networks. The estimated velocity can be used to diagnose cancer, such as prostate cancer. Instead of velocity vectors, agent trajectories can be determined also used for the same purpose.

Abstract: Architecture that enables the identification of entities such as people and content in live broadcasts (e.g., streaming content (e.g., video) of live events) and non-live presentations (e.g., movies), in realtime, using recognition processes. This can be accomplished by extracting live data related to a live event. With respect to people entities, filtering can be performed to identify the named (people) entities from the extracted live data, and trending topics discovered as relate to the named entities, as associated with the live event. Multiple images of the named entities that capture the named entities under different conditions are captured for the named entities. The images are then processed to extract and learn facial features (train one or more models), and facial recognition is then performed on faces in the video using the trained model(s).

Abstract: A method for preventing a person from taking a dangerous selfie is executed by a mobile electronic apparatus including a multi-axis sensing device and an image acquisition device disposed on a surface of the mobile electronic apparatus. The method includes the following steps: using the image acquisition device operating in a shooting mode to capture an image of the person in a background; using the multi-axis sensing device to detect orientation information of the mobile electronic apparatus; determining whether the person is in a potentially dangerous environment according to the orientation information and the image; and enabling the image acquisition device to suspend or exit the shooting mode when the person is in the potentially dangerous environment.

Abstract: A method for estimating a sub-pixel position of an extreme point in an image uses a parametric function. The parametric function is fitted locally to a group of neighboring pixels in the image, and a spatial position of an extreme point of the parametric function is identified. If the extreme point of the parametric function is of a different type than the extreme point in the image, or the position of the extreme point of the parametric function is located outside an area defined by the pixel positions in the group of neighboring pixels, a pixel is removed from the group of neighboring pixels and the fitting of the parametric function is repeated. In the end, the sub-pixel position is estimated as the position of the extreme point of the parametric function. A level of uncertainty of the estimated sub-pixel position is estimated as the number of repetitions needed before arriving at the estimate of the sub-pixel position.

Abstract: An image coding apparatus comprising, a generating unit configured to generate a plurality of pieces of sub-band data from RAW data, a quantization parameter generating unit configured to divide each piece of sub-band data into equal numbers of segments, and generate a first quantization parameter set individually for each segment and common for the plurality of pieces of sub-band data in the same segment, a quantizing unit configured to quantize the segments of the plurality of pieces of sub-band data on the basis of the quantization parameter generated by the quantization parameter generating unit, and a coding unit configured to code a quantization result, obtained from the quantizing, for each piece of sub-band data.

Abstract: Provided are a printed matter inspection device and a printed matter inspection method for appropriately determining visibility of a streaky concentration defect, regardless of the size of a variation of pixel values of an image. A printed matter inspection device determines visibility of a streak from a basic streak intensity that is an intensity of the streak detected from a printed matter, color information and frequency feature information on a streak near-field area that is continuous from an area of the streak and has a variation of image feature amounts that is within a predetermined first range, and frequency feature information on an outer peripheral area that is in contact with the streak near-field area, and determines whether the quality of the printed matter is good or not on the basis of at least the visibility.

Abstract: An AR based fault detection and maintenance system analyzes real-time video streams from a remote user device to identify a specific context level at which a user is to handle an equipment and provides instructions corresponding to the specific context level. The instructions enable generating AR simulations that guide the user in executing specific operations including repairs on faulty components of the equipment. The received video stream is initially analyzed to identify a particular equipment which is to be handled by the user. Fault prediction procedures are executed to identify faults associated with the equipment. The instructions to handle the faults are transmitted to the user device as AR simulations that provide step-by-step simulations that enable the user to execute operations as directed by the instructions.

Abstract: Provided herein are methods for analysis of target cells on a population or individual basis, including before and after contact with a stimulus in order to determine the effect of such stimulus on the target cells. Also provided are devices for performing such methods. The analysis methods involve identifying and measuring or tracking morphological changes that occur in target cells over a period of time. Tracking is accomplished using imaging systems capable of imaging target cells individually over a period of time either continuously or at discrete intervals of time.

Abstract: A convolutional neural network (CNN) processing method and apparatus. The apparatus may select, based on at least one of a characteristic of at least one kernel of a convolution layer or a characteristic of an input of the convolution layer, one operation mode from a first operation mode reusing a kernel, of the at least one kernel, and a second operation mode reusing the input, and perform a convolution operation based on the selected operation mode.

Abstract: A medical scan comparison system is operable to receive a medical scan via a network and to generate similar scan data. The similar scan data includes a subset of medical scans from a medical scan database and is generated by performing an abnormality similarity function to determine that a set of abnormalities included in the subset of medical scans compare favorably to an abnormality identified in the medical scan. At least one cross-sectional image is selected from each medical scan of the subset of medical scans for display on a display device associated with a user of the medical scan comparison system in conjunction with the medical scan.

Abstract: Systems and methods for surveillance are described, including an image capture device configured to mounted to an autonomous vehicle, the image capture device including an image sensor. A storage device is included in communication with the processing system, the storage device including a pruned convolutional neural network (CNN) being trained to recognize obstacles in a road according to images captured by the image sensor by training a CNN with a dataset and removing filters from layers of the CNN that are below a significance threshold for image recognition to produce the pruned CNN. A processing device is configured to recognize the obstacles by analyzing the images captured by the image sensor with the pruned CNN and to predict movement of the obstacles such that the autonomous vehicle automatically and proactively avoids the obstacle according to the recognized obstacle and predicted movement.

Abstract: A monitoring system is disclosed. The system comprises a range detection system, configured for scanning a scene to provide range data arranged gridwise over a plurality of picture elements, and a data processing system configured for processing the range data to monitor hygienic activity, preferably hand hygienic activity, of an individual in the scene. Preferably, the picture elements are devoid of any color data and any grayscale data. In some embodiments of the present invention the data processing system identifies one or more of the moments for hand hygiene as defined by the World Health Organization, and monitors the hygienic activity in relation to these five moments.

Abstract: A medical scan image analysis system is operable to receive a plurality of medical scans that represent a three-dimensional anatomical region and include a plurality of cross-sectional image slices. A plurality of three-dimensional subregions corresponding to each of the plurality of medical scans are generated by selecting a proper subset of the plurality of cross-sectional image slices from each medical scan, and by further selecting a two-dimensional subregion from each proper subset of cross-sectional image slices. A learning algorithm is performed on the plurality of three-dimensional subregions to generate a fully convolutional neural network. Inference data corresponding to a new medical scan received via the network is generated by performing an inference algorithm on the new medical scan by utilizing the fully convolutional neural network. An inferred abnormality is identified in the new medical scan based on the inference data.

Abstract: A line segment detection method includes acquiring an image, extracting from the image a straight line that includes a line segment, acquiring a luminance distribution of the straight line, generating a virtual function model that contains a distribution of a foreground and a background of the image by using the luminance distribution, and determining whether luminance values near end points (start point and end point) of the line segment contained in the virtual function model probabilistically belong to the foreground or the background, and estimating end points (start point and end point) in the virtual function model.

Abstract: An image may include depictions of different parts of a scene, such as the ground, the sky, and objects between the ground and the sky. The image may be divided into multiple bands, and the image may be scored based on colors of pixels within different bands of the image.

Abstract: Described in detail herein are methods and systems for performing a physical object assessment. The system entails an automated system to receive, unpack and assess a physical object based on automated determined attributes. The system can determine an element based on the assessment and generate a webpage including the attributes of the physical object and the determined element.

Abstract: The invention relates to a car park monitoring system for detecting free and occupied parking spaces in a car park area, comprising a detection unit and a camera system, the camera system being designed to supply image evidence from within the car park area, and the detection unit being designed to assign, depending on said image evidence, an occupation status to the parking spaces which indicates whether a parking pace is free or occupied. The car park monitoring system is characterized in that the detection unit is further designed to store the image evidence for at least one corresponding parking space as a first snapshot when the occupation status of the parking space in question changes from free to occupied, and additionally to store the image evidence as a second snapshot when a limiting condition is met.

Abstract: A magnetic ink character reader includes a transport unit configured to move a medium along a transport path, a reader configured to magnetically read magnetic ink characters printed on the medium, while the medium is moved along the transport path and to output a signal waveform corresponding to the magnetic ink characters, and a processor configured to compare a transportation direction of the medium through the reader to a forward reading direction of the reader, and to invert, in horizontal and vertical directions, the signal waveform output from the reader when the transportation direction is not the forward reading direction. The processor performs magnetic character recognition using the signal waveform from the reader when the transportation direction is the forward reading direction and the inverted signal waveform when the transportation direction is a reverse reading direction opposite the forward reading direction.

Abstract: Systems and methods for detecting document contours in a digital image. An example method may involve: receiving an image comprising one or more documents and detecting a set of lines in the image; identifying a plurality of intersection points corresponding to the set of lines; determining a vanishing point based on the plurality of intersection points; generating a candidate polygon based on the set of lines; evaluating the candidate polygon based on a parameter associated with the vanishing point; and indicating the candidate polygon represents a contour of one of the one or more documents.

Abstract: System and methods for monitoring the growth of an aquatic plant culture and detecting real-time characteristics associated with the aquatic plant culture aquatic plants. The systems and methods may include a control unit configured to perform an analysis of at least one image of an aquatic plant culture. The analysis may include processing at least one collected image to determine at least one physical characteristic or state of an aquatic plant culture. Systems and methods for distributing aquatic plant cultures are also provided. The distribution systems and methods may track and control the distribution of an aquatic plant culture based on information received from various sources. Systems and methods for growing and harvesting aquatic plants in a controlled and compact environment are also provided. The systems may include a bioreactor having a plurality of vertically stacked modules designed to contain the aquatic plants and a liquid growth medium.