Abstract: A method and apparatus are disclosed herein for generating and sending priority alert notifications based on medical information, such as, for example, medical information obtained from analyzing medical images. In one embodiment, the method comprises: determining, using an image analysis engine, whether one or more features in a medical image of a patient meet predefined criteria, the predefined criteria being indicative of a medical condition; determining, using the image analysis engine, whether an alert notification is to be sent regarding results of determining whether the one or more features in the medical image meet the predefined criteria; and sending the alert notification with indicia indicative of a priority level if the one or more features in the medical image meet the predefined criteria, including sending medical information that prompted the image analysis engine to send the notification at the priority level.

Abstract: According to one aspect, scene classification may be provided. An image capture device may capture a series of image frames of an environment from a moving vehicle. A temporal classifier may classify image frames with temporal predictions and generate a series of image frames associated with respective temporal predictions based on a scene classification model. The temporal classifier may perform classification of image frames based on a convolutional neural network (CNN), a long short-term memory (LSTM) network, and a fully connected layer. The scene classifier may classify image frames based on a CNN, global average pooling, and a fully connected layer and generate an associated scene prediction based on the scene classification model and respective temporal predictions. A controller of a vehicle may activate or deactivate vehicle sensors or vehicle systems of the vehicle based on the scene prediction.

Abstract: An environment monitoring method and an electronic device are provided, the method divides the satellite image into a plurality of first divided images with overlapping areas, a first multi-dimensional feature map is obtained by inputting the plurality of first divided images into an environment monitoring model, the environmental monitoring model fully combines the correlation between the environmental information of different dimensions, the environmental features of a plurality of different dimensions are correlated through an association network. By utilizing the environment monitoring method, a large area of the environment monitoring effectively is realized, and accuracy of environmental detection is improved.

Abstract: Embodiments herein describe a system for providing visual feedback to an associate when stowing packages into containers. As an associate carries a package towards a container, a camera disposed over the container captures an image of the package that a computer vision system uses to identify a barcode on the package. Using the barcode, the system can determine the desired destination of the package and whether the container the associate is approaching is the correct destination of the package. If so, the feedback system can activate a directional light on above the container which outputs a first color (e.g., green), thereby informing the associate that the container she is approaching is the correct container. However, if container is not the correct destination of the package, the directional light outputs a different color light (e.g., red) which informs the associate the container is the wrong destination for the package.

Abstract: Methods and systems for training a machine learning model using synthetic defect images are provided. One system includes one or more components executed by one or more computer subsystems. The one or more components include a graphical user interface (GUI) configured for displaying one or more images for a specimen and image editing tools to a user and for receiving input from the user that includes one or more alterations to at least one of the images using one or more of the image editing tools. The component(s) also include an image processing module configured for applying the alteration(s) to the at least one image thereby generating at least one modified image and storing the at least one modified image in a training set. The computer subsystem(s) are configured for training a machine learning model with the training set in which the at least one modified image is stored.

Abstract: The present disclosure provides a computer-implemented method, computer system and computer program product for object detection. According to the computer-implemented method, a first object can be classified to be a first category based on detection data acquired in a detection area. Then, a confidence score for the first category can be calculated based on historical detection data of the detection area, wherein the confidence score presents possibility degree of at least one object of the first category existing in the detection area. Whether classification of the first object is abnormal can be determined based on the confidence score.

Abstract: The present invention relates to an object region detection method, an object region detection apparatus, and a non-transitory computer-readable medium thereof, and more particularly, to an object region detection method, an object region detection apparatus, and a non-transitory computer-readable medium thereof, capable of further accelerating analysis of object recognition and tracking by preliminarily detecting an object region based on a parameter value obtained from an image decoding process and referring to the detected object region for image analysis.

Abstract: The present disclosure is directed toward systems and methods for tracking objects in videos. For example, one or more embodiments described herein utilize various tracking methods in combination with an image search index made up of still video frames indexed from a video. One or more embodiments described herein utilize a backward and forward tracking method that is anchored by one or more key frames in order to accurately track an object through the frames of a video, even when the video is long and may include challenging conditions.

Abstract: Provided is creating an augmented reality video using a set-top box. The set-top box may include a communication circuit and a processor. The communication circuit may be configured to receive a first video stream including a target object image in real time from a server and to receive a second video stream captured by a user device from the user device. The processor may be configured to: extract the target object image from each frame of the first video stream using WebGL, correct a color of a boundary line of the target object image between the target object and the background, and composite the extracted target object image into an each frame of the second video stream.

Abstract: A system for inspecting a customized dental device associated with a dental application for manufacturing defects is disclosed. The system obtains images of the customized device and identifies an identifier of the customized device. The system determines a digital file associated with the customized device based on the identifier, the digital file including a first digital model of the customized device and/or a second digital model of a mold used during manufacture of the customized device. The system determines an intended property for the customized device based on at least one of the first digital model or the second digital model, determines an actual property of the customized device from the images, determines whether there is a manufacturing defect in the customized device by comparing the intended property for the customized device with the actual property of the customized device, and outputs an output associated with the determination.

Abstract: Provided are mechanisms and processes for visual feature tagging in multi-view interactive digital media representations (MIDMRs). In one example, a process includes receiving a visual feature tagging request that includes an MIDMR of an object to be searched, where the MIDMR includes spatial information, scale information, and different viewpoint images of the object. A visual feature in the MIDMR is identified, and visual feature correspondence information is created that links information identifying the visual feature with locations in the viewpoint images. At least one image associated with the MIDMR is transmitted in response to the feature tagging request.

Abstract: The present invention provides an image processing circuit including a region type determination circuit and a filtering circuit. In the operations of the image processing circuit, the region type determination circuit receives pixel values of a plurality of pixels within a region of an image frame, and uses a high-pass filter to filter the pixel values many times to generate a plurality of filtered pixel values, and determines if the region is an edge region, a non-edge region or a mosquito noise region to generate a determination result. The filtering circuit generates a plurality of weights according to the determination result, and uses the plurality of weights to filter a center pixel of the plurality of pixels within the region to generate an adjusted pixel value.

Abstract: A measurement processing device includes an imaging unit, an image acquisition unit, a display unit, a measurement point display control unit, a corresponding-point calculation unit, a corresponding-point image generation unit, and a corresponding-point image display control unit. The measurement point display control unit causes a measurement point indicating a measurement position designated by a user to be displayed on a first image. The corresponding-point image generation unit generates a corresponding-point image that includes the corresponding point calculated by the corresponding-point calculation unit and a region in the vicinity of the corresponding point in the second image and is constituted of all or a part of the second image. The corresponding-point image display control unit causes the corresponding-point image to be displayed so that a straight line passing through the measurement point and the corresponding point is orthogonal to the parallax direction on a display screen.

Abstract: When a device is successfully unlock using a facial recognition authentication process, feature vectors generated from images obtained during the facial recognition authentication process may be stored as temporary templates on the device. After a period of time, one of the temporary templates may be selected to be used as a “virtual” template for the device. For example, a median temporary template in the temporary templates may be selected as the virtual template. The performance of the virtual template may then be assessed over time and compared to the performance of templates generated from an enrollment process to determine if and how the virtual template is implemented on the device.

Abstract: In various embodiments, a defective pixel detection application automatically detects defective pixels in video content. In operation, the defective pixel detection application computes a first set of pixel intensity gradients based on a first frame of video content and a first neighborhood of pixels associated with a first pixel. The defective pixel detection application also computes a second set of pixel intensity gradients based on the first frame and a second neighborhood of pixels associated with the first pixel. Subsequently, the defective pixel detection application computes a statistical distance between the first set of pixel intensity gradients and the second set of pixel intensity gradients. The defective pixel detection application then determines that the first pixel is defective based on the statistical distance.

Abstract: A system and method for generating simulated post-contrast T1-weighted magnetic resonance (MR) images without the use of exogenous contrast material based upon patient-specific non-contrast MR images using machine learning/artificial intelligence techniques to train the system to generate post-contrast T1-weighted magnetic resonance images based upon retrospectively collected non-contrast MR images of various sequence types including T1-weighted, T2-weighted, FLAIR (Fluid-Attenuated Inversion Recovery), and/or DWI (Diffusion-Weighted Imaging).

Abstract: Computer-implemented methods and systems are provided for quantitative hemodynamic flow analysis, which involves retrieving patient specific image data. A 3D reconstruction of a vessel of interest can be created from the patient specific image data. Geometric information can be extracted from the 3D reconstruction. A lesion position can be determined. Patient specific data can be obtained. Hemodynamic results can be calculated based on the geometric information, the lesion position and the patient specific data.

Abstract: Disclosed are systems and methods related to facial recognition. An image of a subject can be captured via a camera on a mobile device. The image can be classified according to a device type, whether the image is captured indoors or outdoors, and a standoff distance. Facial features can be extracted from the image based on the image category. The facial features can be compared with a predefined set of facial features in a database. An identification of the subject can be made in response to the comparison.

Abstract: The present disclosure provides a picture hook identification method as well as an apparatus and a terminal device using the same. The method includes; determining geometric parameter(s) of an identification object based on image(s) collected b a camera and internal parameter(s) of the camera; comparing the geometric parameters of the identification object with geometric parameter(s) of a target picture book; and determining the identification object as the target picture book, if a difference between the geometric parameters of the identification object and the geometric parameters of the target picture book is within a preset range. In this manner, the target picture book is further filtered by matching the geometric parameters, which can reduce misidentification to improve the accuracy of identifying the picture book.

Abstract: A method and a system are described for counting objects in an image. The method includes receiving at least one image comprising one or more objects. The method includes determining contours of the one or more objects using one or more morphological operations on the at least one image. The method further includes identifying shapes of the one or more objects based on counting a number of contours associated with each of the one or more objects. The method includes comparing the shapes of the one or more objects with one or more predefined training images to identify one or more objects of interest. The method includes counting the one or more objects of interest based on the shapes of the one or more objects of interest.