Abstract: Apparatus, methods, and articles of manufacture for implementing crowdsourcing pipelines that generate training examples for machine learning expression classifiers. Crowdsourcing providers actively generate images with expressions, according to cues or goals. The cues or goals may be to mimic an expression or appear in a certain way, or to “break” an existing expression recognizer. The images are collected and rated by same or different crowdsourcing providers, and the images that meet a first quality criterion are then vetted by expert(s). The vetted images are then used as positive or negative examples in training machine learning expression classifiers.

Abstract: Techniques using small form factor infrared imaging modules are disclosed. An imaging system may include visible spectrum imaging modules, infrared imaging modules, illumination modules, and other modules to interface with a user and/or a monitoring system. Visible spectrum imaging modules and infrared imaging modules may be positioned in proximity to a scene that will be monitored while visible spectrum-only images of the scene are either not available or less desirable than infrared images of the scene. Imaging modules may be configured to capture images of the scene at different times. Image analytics and processing may be used to generate combined images with infrared imaging features and increased detail and contrast. Selectable aspects of non-uniformity correction processing, true color processing, and high contrast processing, may be performed on the captured images. Control signals based on the combined images may be presented to a user and/or a monitoring system.

Abstract: A rectangular region group storage unit stores a group of rectangular regions indicating portions to be recognized for a crowd state on an image. A crowd state recognition dictionary storage unit stores a dictionary of a discriminator acquired by machine learning by use of a plurality of pairs of crowd state image as an image which expresses a crowd state at a predetermined size and includes a person whose reference site is expressed as large as the size of the reference site of a person defined for the predetermined size, and training label for the crowd state image. A crowd state recognition unit extracts regions indicated in the group of rectangular regions stored in the rectangular region group storage unit from a given image, and recognizes states of the crowds shot in the extracted images based on the dictionary.

Abstract: The present invention includes a method and an apparatus for determining the viscosity of a fluid. The apparatus comprising that includes a microchannel connected to a glass capillary in fluid communication with the microchannel, a digital camera positioned with respect to the glass capillary to capture two or more images of a fluidic slug as a fluid travels within the glass capillary, and a processor communicably coupled to the digital camera that determines a viscosity of the fluid based on the two or more digital images.

Abstract: An image processing device, a method for processing an image, and a recording medium storing a program for causing a computer to execute the method. The image processing device and the method includes converting captured images in which image capturing ranges partially overlap into images in a coordinate system different from a coordinate system, based on a first conversion data, detecting an amount of misalignment in an overlapping area, calculating correction amounts for the first conversion data related to generation of the second conversion image for each coordinate value of the second conversion image, based on the detected amount of misalignment, detecting a subject area in which a subject exists from the second conversion image, and correcting the first conversion data related to generation of the second conversion image, using one of the calculated correction amounts of a subject area in which a dynamic subject exists.

Abstract: A fingerprint of an image identified within a received message is generated following analysis of the message. A spam detection engine identifies an image within a message and converts the image into a grey scale image. The spam detection engine analyzes the grey scale image and assigns a score. A fingerprint of the grey scale image is generated based on the score. The fingerprint may also be based on other factors such as the message sender's status (e.g. blacklisted or whitelisted) and other scores and reports generated by the spam detection engine. The fingerprint is then used to filter future incoming messages.

Abstract: Disclosed are techniques for more accurately estimating the pose of a camera used to capture a three-dimensional scene. Accuracy is enhanced by leveraging three-dimensional object priors extracted from a large-scale three-dimensional shape database. This allows existing feature matching techniques to be augmented by generic three-dimensional object priors, thereby providing robust information about object orientations across multiple images or frames. More specifically, the three-dimensional object priors provide a unit that is easier and more reliably tracked between images than a single feature point. By adding object pose estimates across images, drift is reduced and the resulting visual odometry techniques are more robust and accurate. This eliminates the need for three-dimensional object templates that are specifically generated for the imaged object, training data obtained for a specific environment, and other tedious preprocessing steps.

Abstract: An image processing method includes the following steps. An original image is read, and the original image includes a red channel layer, a green channel layer and a blue channel layer. A processing is performed on the red channel layer, the green channel layer and the blue channel layer, respectively, to get a derived red channel layer, a derived green channel layer and a derived blue channel layer. And the derived red channel layer, the derived green channel layer and the derived blue channel layer are combined to form a new image. The processing includes the following steps. Gradient-variations of a plurality of directions of a region are calculated. An average of the gradient-variations is calculated. A calculating procedure is provided to decide diffusion-coefficient equations of the directions, and each of the diffusion-coefficient equations is substituted into an anisotropic diffusion equation.

Abstract: An image acquisition part acquires image data from an image sensor provided in a room. A storage stores information indicating a transition model of a human body posture and test information indicating a test area of the image data in each posture of the transition model. A posture estimator calculates a statistic in the test area of the image data by referring to the information indicating the transition model and the test information, and estimates a current human body posture from a human body posture before one point of time based on the statistic.

Abstract: An image processing apparatus includes: a detecting unit that detects regions of interest that are estimated as an object to be detected, from a group of a series of images acquired by sequentially imaging a lumen of a living body, and to extract images of interest including the regions of interest; a neighborhood range setting unit that sets, as a time-series neighborhood range, a neighborhood range of the images of interest in the group of the series of images arranged in time series so as to be wider than an interval between images that are continuous in time series in the group of the series of images; an image-of-interest group extracting unit that extracts an image-of-interest group including identical regions of interest from the extracted images of interest, based on the time-series neighborhood range; and a representative-image extracting unit that extracts a representative image from the image-of-interest group.

Abstract: In embodiments of the present invention improved capabilities are described for producing background separated product images for print and on-line display. An image formation system provides lighting of a product to facilitate acquiring images that can be automatically processed to generate high resolution item-only images free of quality defects and imaging artifacts. Image processing programs accurately detect an outline of an item in a set of digital images taken using the image formation system and automatically store processed images in an image library. The images in the library may be repurposed for print, sales display, transmission to a user, on-line customer support, and the like. A user display configured with an adaptable user interface facilitates user interaction with images in the library.

Abstract: A system and method for generating compact iris representations based on a database of iris images includes providing full-length iris codes for iris images in a database, where the full-length iris code includes a plurality of portions corresponding to circumferential rings in an associated iris image. Genuine and imposter score distributions are computed for the full-length iris codes, and code portions are identified that have a contribution that provides separation between imposter and genuine distributions relative to a threshold. A correlation between remaining code portions is measured. A subset of code portions having low correlations within the subset is generated to produce a compact iris representation.

Abstract: The embodiments relate to a method, an apparatus, and a computer program for evaluating an x-ray image of a breast produced during a mammography. In order to simplify the evaluation of such an x-ray image in respect of the breast density, a method is proposed to automatically determine the masking risk caused by the mammographically dense tissue and to use this for categorizing, describing, and/or representing the breast density.

Abstract: Provided are methods and computing devices wherein in one embodiment, user input on a touch screen of a computing device within which a text document is displayed is associated with both: a) a user chosen command, such as a selection or an insertion of text characters, to be applied within the document, and b) a positional location in the document, representing a user chosen document location. A region associated with the positional location in the document is defined, and data comprises representation of the user chosen command is processed to determine the user chosen document location, being within the region and in proximity to the positional location in the document, based on the document location being capable of being either starting location or ending location within user chosen locations in the document at which to apply the user chosen command.

Abstract: An ultrasound system and method are described for acquiring a sequence of ultrasound images of the carotid artery during the delivery of a contrast agent. Plaque in the images is identified and a time-intensity curve is calculated for pixels in the images. The intensity values before and after the arrival of contrast are compared to identify pixels or groups of pixels having perfusion. An anatomical image may be formed showing areas in an image of the plaque of the intensity and presence of perfusion, or the perfusion may be quantified by determining the percentage of pixels in the plaque image which exhibit perfusion. The extent and degree of perfusion is an indicator of the risk of plaque particulates in the blood stream which may lead to stroke-related symptoms.

Abstract: A system for detecting and identifying foliage includes a tracking component, a tracking parameters component, and a classification component. The tracking component is configured to detect and track one or more features within range data from one or more sensors. The tracking parameters component is configured to determine tracking parameters for each of the one or more features. The tracking parameters include a tracking age and one or more of a detection consistency and a position variability. The classification component is configured to classify a feature of the one or more features as corresponding to foliage based on the tracking parameters.

Abstract: Systems and methods that provide for an interactive media guidance application implemented in a user device, e.g., an app on a mobile phone. The interactive media guidance application uses the geographical location, e.g., longitude and latitude coordinates, and geospatial orientation, e.g., gyroscopic angles, of the user device to allow the user to capture images of the location in the same manner as the location was depicted in his or her favorite film or television program. The interactive media guidance application can deliver to the user the desired immersive experience where he or she sees and has the same viewpoint as the director or another person associated with the film or television program.

Abstract: A character recognition device includes circuitry that is configured to remove duplicate successive points of a plurality of points in a handwritten stroke to form an enhanced handwritten stroke; space the plurality of points a uniform distance apart; detect primary strokes and secondary strokes of the enhanced handwritten stroke; merge the primary strokes; generate a primary merged stroke; extract raw point-based features from local features of the primary merged stroke; extract statistical features from computed statistics associated with the raw point-based features to form primary merged stroke features; train and classify data from the primary merged stroke features and secondary stroke features to form stroke models; determine a plurality of primary merged stroke model candidates from the stroke models; compute a likelihood value for a combined set of primary stroke candidates and a set of secondary stroke candidates; and determine the handwritten stroke from the computing.

Abstract: A character recognition device includes circuitry that is configured to remove duplicate successive points of a plurality of points in a handwritten stroke to form an enhanced handwritten stroke; space the plurality of points a uniform distance apart; detect primary strokes and secondary strokes of the enhanced handwritten stroke; merge the primary strokes; generate a primary merged stroke; extract raw point-based features from local features of the primary merged stroke; extract statistical features from computed statistics associated with the raw point-based features to form primary merged stroke features; train and classify data from the primary merged stroke features and secondary stroke features to form stroke models; determine a plurality of primary merged stroke model candidates from the stroke models; compute a likelihood value for a combined set of primary stroke candidates and a set of secondary stroke candidates; and determine the handwritten stroke from the computing.

Abstract: Apparatus and methods are described including, using a computer processor (28), automatically identifying whether a given pixel (111) within an image corresponds to a portion of an object. A set of concentric circles (132a-c) that are disposed around the pixel are sampled, and a first function is applied to each of the circles such that the circles are defined by a first set of rotationally invariant descriptors. A second function is applied to the set of circles to generate a second set of descriptors, each of which represents a difference between respective pairs of the circles. A third function is applied such that the second set of descriptors becomes rotationally invariant. The processor identifies whether the given pixel corresponds to the portion of the object, based upon the first and second sets of rotationally invariant descriptors. Other applications are also described.