Abstract: Described are methods, systems, apparatus, and computer program products for determining the presence of an object on a target surface. A machine vision system includes a first image capture device configured to image a first portion of a target surface from a first viewpoint and a second image capture device configured to image a second portion of the target surface from a second viewpoint. The machine vision system is configured to acquire a first image from the first image capture device, a second image from the second image capture device, rectify the first image and second image, retrieve a disparity field, generate difference data by comparing, based on the mappings of the disparity field, image elements in the first rectified image and a second image elements in the second rectified image; and determine whether the difference data is indicative of an object on the target surface.

Abstract: Embodiments are described for detecting optical discrepancies associated with image capture analyzing pixels in multiple images corresponding to common points of reference in a physical environment. In an embodiment, photometric error values are averaged over time to compute the mean error at each pixel. Once the estimate of the mean error has a sufficient number of updates above a specified value, the estimate is thresholded to provide a mask of any optical discrepancies occurring in the stereo pair of images. Applications include detecting optical discrepancies in images captured for use by a visual navigation system in guiding an autonomous vehicle (e.g., an unmanned aerial vehicle).

Abstract: Disclosed herein are methods and systems for determining liveness of a user, comprising analyzing visual content of a screen of a client device used by a user to access an online service, adjusting one or more visual objects displayed on the screen of the client device according to dynamically changing patterns, capturing a sequence of images depicting one or more reflecting surfaces associated with the user viewing the screen while the visual objects are displayed, analyzing the images to identify a reflection of the displayed visual objects in the reflecting surfaces and verifying liveness of the user based on one or more of a plurality of reflection attributes of the identified reflection.

Abstract: A reduction in concentration due to a change in an emotion is inhibited. A change in an emotion of the human is suitably reduced. Part (in particular, an eye or an eye and its vicinity) or the whole of a user's face is detected, a feature of the user's face is extracted from data on the detected part or whole of the face, and an emotion of the user is estimated from the extracted feature of the face. In the case where the estimated emotion is an emotion that might reduce concentration, for example, a stimulus is applied to the sense of sight, the sense of hearing, the sense of touch, the sense of smell, or the like of the user to recover the concentration of the user.

Abstract: Systems and techniques are provided for modeling three-dimensional (3D) meshes using images. An example method can include receiving, via a neural network system, an image of a target and metadata associated with the image and/or a device that captured the image; determining, based on the image and metadata, first 3D mesh parameters of a first 3D mesh of the target, the first 3D mesh parameters and first 3D mesh corresponding to a first reference frame associated with the image and/or the device; and determining, based on the first 3D mesh parameters, second 3D mesh parameters for a second 3D mesh of the target, the second 3D mesh parameters and second 3D mesh corresponding to a second reference frame, the second reference frame including a 3D coordinate system of a real-world scene where the target is located.

Abstract: Disclosed are techniques for automated analysis and assessments of contrast medium absorption phases in contrast medium based medical images. A target image set includes a plurality of medical images acquired to image a plurality of contrast medium absorption phases. For the images of the target image set, a set of contrast medium absorption phase probabilities are determined corresponding to likelihoods that a given image corresponds to a given contrast medium absorption phase. The determined sets of contrast medium absorption phases are compared against a reference set of contrast medium absorption phases to determine differences to determine a set of matching scores indicative of how closely the contrast medium absorption phases of the target image set align with the plurality of contrast medium absorption phases as compared to the reference set of contrast medium absorption phases.

Abstract: A system for plant parameter detection, including: a plant morphology sensor having a first field of view and configured to record a morphology measurement of a plant portion and an ambient environment adjacent the plant, a plant physiology sensor having a second field of view and configured to record a plant physiology parameter measurement of a plant portion and an ambient environment adjacent the plant, wherein the second field of view overlaps with the first field of view; a support statically coupling the plant morphology sensor to the physiology sensor, and a computing system configured to: identify a plant set of pixels within the physiology measurement based on the morphology measurement; determine physiology values for each pixel of the plant set of pixels; and extract a growth parameter based on the physiology values.

Abstract: A method of detecting and diagnosing cancers characterized by the presence of at least one nodule/neoplasm from an imaging scan is presented. To detect nodules in an imaging scan, a 3D CNN using a single feed forward pass of a single network is used. After detection, risk stratification is performed using a supervised or an unsupervised deep learning method to assist in characterizing the detected nodule/neoplasm as benign or malignant. The supervised learning method relies on a 3D CNN used with transfer learning and a graph regularized sparse MTL to determine malignancy. The unsupervised learning method uses clustering to generate labels after which label proportions are used with a novel algorithm to classify malignancy. The method assists radiologists in improving detection rates of lung nodules to facilitate early detection and minimizing errors in diagnosis.

Abstract: An example aquatic imaging system comprises a light source, a first platform coupled with a image capture device and a second platform that is parallel to the platform, the image capture device having a first field of view, and, the second platform being coupled to a organism tank, first organism tank having an inner wall, outer wall and a base that defines a well capable of retaining water, the base being parallel to the second platform, the organism tank configured to receive a light beam originating from the light source and configured to project at least a portion of the light beam through the well and in a directional plane that is parallel to the base, the image capture device configured to direct the first field of view from the first platform through the well in the organism tank.

Abstract: An intelligent recognition method of hyperspectral image of parasites in raw fish relates to optical detection technology, and includes step 1: obtaining a hyperspectral image of the raw fish in a wavelength range from 300 to 1100 nm; step 2: extracting a grayscale image of the hyperspectral image at a wavelength value of 437 nm, and obtaining a position range of fish meat in the grayscale image by performing a median filtering process and a binarization process on the grayscale image; step 3: extracting spectral signals of pixel points in the position range of the hyperspectral image, performing a first-order derivative process on the spectral signals, and import the spectral signals after the first-order derivative process into a preset first model, a second model, and a third model for analysis. The method can accurately distinguish the parasite body in the raw fish.

Abstract: Systems and methods for detecting, characterizing, and addressing surgical smoke are disclosed. In some embodiments, a surgical system receives image data representing an image of a surgical environment and generates sets of values based on the received image. The sets of values may include representations of atmospheric light in the image and representations of contrast values in the image. The system may determine from the sets of values whether predetermined smoke severity criteria are met, and may responsively automatically engage and/or disengage one or more surgical devices, such as a surgical smoke evacuation system and/or an image processing system (e.g., a video enhancement feature) configured to improve quality of smoky surgical images.

Abstract: Disclosed are methods related to guiding robotic surgery using optical coherence tomography (OCT) and computer-readable media and computer systems executing the methods. They may include receiving a series of cross-sectional slices of 3D space obtained from an OCT probe over biological tissue and processing and filtering the series of slices. The processing and filtering may include spatially smoothing the intensity values of each slice, thresholding each slice after it has been blurred, performing a connected-component analysis to identify blobs on the thresholded slice, filtering the blobs, performing edge detection, and invoking a selective median filter. The processing and filtering can be used to construct a depth map from the received series of cross-sectional slices in order to guide a robotic end effector, based on the depth map.

Abstract: A system and a method of calculating coordinates of a pupil center point are provided. The system for acquiring the coordinates of the pupil center point includes a first camera, a second camera, a storage and a processor. The first camera is configured to capture a first image including a face and output the first image to the processor, the second camera is configured to capture a second image including a pupil and output the second image to the processor, a resolution of the first camera is smaller than a resolution of the second camera, and the storage is configured to store processing data, and the processor is configured to: acquire the first image and the second image; extract a first eye region corresponding to an eye from the first image; convert the first eye region into the second image, to acquire a second eye region corresponding to the eye in the second image; and detect a pupil in the second eye region and acquire the coordinates of the pupil center point.

Abstract: This disclosure provides a system and method for producing ultrasound images based on Full Waveform Inversion (FWI). The system captures acoustic/(an)elastic waves transmitted through and reflected and/or diffracted from a medium. The system performs an FWI process in a time domain in conjunction with an accurate wave propagation solver. The system produces 3D maps of physical parameters that control wave propagation, such as shear and compressional wavespeeds, mass density, attenuation, Poisson's ratio, bulk and shear moduli, impedance, and even the fourth-order elastic tensor containing up to 21 independent parameters, which are of significant diagnostic value, e.g., for medical imaging and non-destructive testing.

Abstract: The present disclosure discloses a method and apparatus for detecting an obstacle, and relates to the technical field of intelligent transportation. A specific implementation plan is: acquiring a current image acquired by a camera; inputting the current image into a pre-trained detection model to obtain a position of a detection frame of an obstacle and determine a first pixel coordinate of a grounding point in the current image; determining an offset between the current image and a template image; converting the first pixel coordinate into a world coordinate of the grounding point based on the offset; and outputting the world coordinate of the grounding point as a position of the obstacle in a world coordinate system. This embodiment solves the problem of camera jitter from an image perspective, greatly improves the robustness of the roadside perception system, and saves computing resources.

Abstract: A security system (10) includes an image information acquisition unit (11) that acquires input image information on an image taken of a person in a store, a tracking unit (12) that tracks an action of a hand of the person based on the input image information, and a suspicious action detection unit (13) that detects a suspicious action of the person based on the tracked action of the hand. A security system, a security method, and a security program capable of accurately detecting a suspicious action are thereby provided.

Abstract: Disclosed is a system and method for segmentation of selected data. In various embodiments, automatic segmentation of fiber tracts in an image data may be performed. The automatic segmentation may allow for identification of specific fiber tracts in an image.

Abstract: A scanner acquires a set of images of a hand of a user to facilitate identification. These images may vary, due to changes in relative position, pose, lighting, obscuring objects such as a sleeve, and so forth. A first neural network determines output data comprising a spatial mask and a feature map for individual images in the set. The output data for two or more images is combined to provide aggregate data that is representative of the two or more images. The aggregate data may then be processed using a second neural network, such as convolutional neural network, to determine an embedding vector. The embedding vector may be stored and associated with a user account. At a later time, images acquired from the scanner may be processed to produce an embedding vector that is compared to the stored embedding vector to identify a user at the scanner.

Abstract: This invention can readily, reliably perform identity verification and input of user information when opening an account from a terminal. An information processing apparatus includes a first feature extractor extracting a first feature from a face image of a user included in a moving image, a second feature extractor extracting a second feature from a face image of an identity verification document for authenticating the user included in the moving image, a collator collating whether the first feature and the second feature match each other, a determiner instructing the user to perform a predetermined action, and determining whether an action of the user included in the moving image corresponds to the instructed predetermined action, and a verifier verifying that the user is the person himself or herself, when the first feature and the second feature match each other and the action of the user corresponds to the instructed predetermined action.

Abstract: Described is a system for counting stacked items using image analysis. In one implementation, an image of an inventory location with stacked items is obtained and processed to determine the number of items stacked at the inventory location. In some instances, the item closest to the camera that obtains the image may be the only item viewable in the image. Using image analysis, such as depth mapping or Histogram of Oriented Gradients (HOG) algorithms, the distance of the item from the camera and the shelf of the inventory location can be determined. Using this information, and known dimension information for the item, a count of the number of items stacked at an inventory location may be determined.