Abstract: Disclosed herein are systems, methods, and instrumentalities associated with reconstructing magnetic resonance (MR) images based on under-sampled MR data. The MR data include 2D or 3D information, and may encompass multiple contrasts and multiple coils. The MR images are reconstructed using deep learning (DL) methods, which may accelerate the scan and/or image generation process. Challenges imposed by the large quantity of the MR data and hardware limitations are overcome by separately reconstructing MR images based on respective subsets of contrasts, coils, and/or readout segments, and then combining the reconstructed MR images to obtain desired multi-contrast results.

Abstract: An image of a portion of a semiconductor die is obtained that shows one or more structures in a first process layer and one or more structures in a second process layer. Using machine learning, a first region is defined on the image that at least partially includes the one or more structures in the first process layer. Also using machine learning, a second region is defined on the image that at least partially includes the one or more structures in the second process layer. An overlay offset between the one or more structures in the first process layer and the one or more structures in the second process layer is calculated using the first region and the second region.

Abstract: A system and method for object monitoring, detection, and segmentation in electro-optical (EO) satellite imagery data comprising an image processing engine configured to prepare EO data for analysis by an inference engine which utilizes one or more trained models to perform object detection or image segmentation. The workflow begins with ingesting satellite data, followed by de-hazing to remove atmospheric interference. Image enhancement improves resolution and geo-registration ensures precise spatial alignment. The processed image is then fed into a machine learning-based object detection or image segmentation network, trained to identify specific objects of interest. This integrated approach leverages advanced technologies to extract actionable insights from satellite data, enabling efficient and precise object monitoring, detection, and segmentation.

Abstract: An image processing method, an image processing device, a training method and a computer-readable storage medium. The image processing method includes: extracting a characteristic vector in an image to be recognized; based on the characteristic vector of the image to be recognized, acquiring a predicted score value of the image to be recognized; and based on the predicted score value, determining a category of an image information of the image to be recognized; wherein the image to be recognized is a face image, and the image information is a facial expression.

Abstract: Disclosed is a stereophotogrammetric method based on binocular vision, including the following steps: image acquisition, image correction and stereo matching are performed; cost matching and cost aggregation are performed on images of different sizes after correction; image segmentation is performed on the corrected image to determine edge pixel points of the object to be measured; and a pixel distance at an edge of the object to be measured is calculated to measure the size of the object. The method of the present invention enhances the matching accuracy of contour pixels of the object to be measured and improves the measurement accuracy.

Abstract: Methods, systems and computer program products (“software”) enable a virtual three-dimensional visual experience (referred to herein as “V3D”) in videoconferencing and other applications, and capturing, processing and displaying of images and image streams.

Abstract: An intensity spectrum designing unit of a data generating device includes an initial value setting unit that sets a plurality of objects of a first generation of an intensity spectrum function A(?) and a phase spectrum function ?(?), an evaluation value calculating unit that calculates an evaluation value for each of a plurality of objects of an n-th generation, an object selecting unit that selects two or more objects used for generating a plurality of objects of an (n+1)-th generation among objects of the n-th generation on the basis of superiority of the evaluation value, and a next-generation generating unit that generates a plurality of objects of the (n+1)-th generation on the basis of the selected two or more objects. The evaluation value calculating unit, the object selecting unit, and the next-generation generating unit repeat processes while 1 is added to n until a predetermined condition is satisfied.

Abstract: Door open monitoring for an intelligent device is disclosed. In a method for monitoring a door of an intelligent device according to an exemplary embodiment of the present disclosure, it is decided whether to send a message relating to the ventilation of the inside of a drum by analyzing an image generated through a camera placed in the door. A door monitoring system of the present disclosure may be associated with an artificial intelligent module, an unmanned aerial vehicle (UAV), a robot, an augmented reality (AR) device, a virtual reality (VR) device, a 5G service-related device, etc.

Abstract: Enhanced eye-tracking techniques for augmented or virtual reality display systems. An example method includes obtaining an image of an eye of a user of a wearable system, the image depicting glints on the eye caused by respective light emitters, wherein the image is a low dynamic range (LDR) image; generating a high dynamic range (HDR) image via computation of a forward pass of a machine learning model using the image; determining location information associated with the glints as depicted in the HDR image, wherein the location information is usable to inform an eye pose of the eye.

Abstract: In some embodiments, apparatuses and methods are provided herein useful to train a machine learning algorithm to detect text of interest. In some embodiments, there is provided a system to detect vertically oriented text of interest including a first data set comprising a plurality of captured digital images each depicting an object of interest and a second data set comprising a plurality of augmented digital images each depicting a captured digital image augmented with a synthetic text image; a first control circuit configured to cause the machine learning algorithm to output a machine learning model trained to automatically detect occurrences of vertically oriented text of interest based on the first data set and the second data set; at least one camera; and a second control circuit configured to execute the machine learning model to automatically detect vertically oriented text of interest on the object of interest.

Abstract: The invention relates to a real-time object detection and 3D localization method based on a single frame image. Comprising following steps: S1: inputting a 2D RGB image; S2: performing feature extraction on the 2D RGB image, extracting features of a deep network and a shallow network respectively; S3: carrying out 2D object detection and applying to subsequent modules; S4: estimating vertices, instance-level depth and center point of a 3D-box respectively; S5: adding a regularization term for maintaining horizontal locality into prediction of center point of a 3D-box to constrain and optimize the prediction of center point of the 3D-box; and S6: outputting a 2D RGB image with a 3D-box tag in combination with predictions of all modules. The invention increases the speed of model training convergence and the accuracy of 3D object detection and localization, and meets the accuracy requirements of an Advanced Driver Assistant System (ADAS) with a low hardware cost.

Abstract: A method for assessing cardiothoracic ratio (CTR) includes following steps. A testing X-ray image database of a subject is provided. A first image data classifying step is performed, wherein the testing X-ray image database is classified by a first deep learning neural network classifier to obtain a testing chest X-ray image data. A second image data classifying step is performed, wherein the testing chest X-ray image data is classified by a second deep learning neural network classifier to obtain a target chest X-ray image data. A feature extracting step is performed, wherein a diameter of thoracic cavity and a diameter of cardiac silhouette of the target chest X-ray image data are captured automatically and then trained to achieve a convergence by a third deep learning neural network classifier. An assessing step is performed, wherein an assessing result of CTR is obtained according to a feature of CTR.

Abstract: Artificial intelligence (AI) recognition of echocardiogram (echo) images by a mobile ultrasound device comprises receiving a plurality of the echo images captured by the ultrasound device, the ultrasound device including a display and a user interface (UI) that displays the echo images to a user, the echo images comprising 2D images and Doppler modality images of a heart. One or more neural networks process the echo images to automatically classify the echo images by view type. The view type of the echo images is simultaneously displayed in the UI of the ultrasound device along with the echo images. A report is generated showing the calculated measurements of features in the echo images. The report showing the calculated measurements is displayed on a display device.

Abstract: Provided is a weakly supervised pathological image tissue segmentation method based on an online noise suppression strategy, including: acquiring a hematoxylin-eosin (H&E) stained graph, processing the H&E stained graph to obtain a data set, dividing the data set, training a classification network based on a divided data set, and generating a pseudo-label; suppressing a noise existing in the pseudo-label based on the online noise suppression strategy, and training a semantic segmentation network through the pseudo-label after noise suppression and a training set corresponding to the pseudo-label to obtain a prediction result of the semantic segmentation network after the training, and taking the prediction result as a final segmentation result.

Abstract: A vehicle wheel alignment system has a plurality of cameras, each camera for viewing a respective target disposed at a respective wheel of the vehicle and capturing image data of the target as the wheel and target are continuously rotated a number of degrees of rotation without a pause. The image data is used to calculate a minimum number of poses of the target of at least one pose for every five degrees of rotation as the wheel and target are continuously rotated the number of degrees of rotation without a pause. At least one of the cameras comprises a data processor for performing the steps of preprocessing the image data, and calculating an alignment parameter for the vehicle based on the preprocessed image data.

Abstract: Mechanisms are provided to improve an output of a trained machine learning (ML) computer model based on label co-occurrence statistics. For a corpus, label vector representations of the knowledge data structures are generated. Co-occurrence scores for each pairing of labels, across the label vector representations, are generated. A vector output of the ML computer model is received and a knowledge driven reasoning (KDR) computer model is configured with threshold(s) and delta value(s) specifying condition(s) of a co-occurrence of a first label in the output with a second label in the plurality of labels which, if present, causes the delta value(s) to be applied to modify a probability value associated with the second label in the output of the ML computer model. The KDR computer model is executed on the output of the ML computer model to modify probability value(s) in the output.

Abstract: An In-Building Communications system is disclosed which permits communication in tunnels, underground parking garages, tall buildings such as skyscrapers, buildings having thick walls of concrete or metal, and/or any building which has communication dead zones due to electromagnetic shielding. The invention includes a portable bi-directional amplifier (BDA) system, an outdoor antenna system attached to the building or independently mountable, an indoor antenna system attached to the building or independently mountable inside the building, and a standardized, In-Building Communications (IBC) interface box affixed preferably to the exterior of the building. The interface box communicates with antenna systems attached to the building.

Abstract: Aspects of the disclosure are directed towards artificial intelligence-based modeling of target objects, such as aircraft parts. In an example, a system initially trains a machine learning (ML) model based on synthetic images generated based on multi-dimensional representation of target objects. The same system or a different system subsequently further trains the ML model based on actual images generated by cameras positioned by robots relative to target objects. The ML model can be used to process an image generated by a camera positioned by a robot relative to a target object based on a multi-dimensional representation of the target object. The output of the ML model can indicate, for a detected target, position data, a target type, and/or a visual inspection property. This output can then be used to update the multi-dimensional representation, which is then used to perform robotics operations on the target object.

Abstract: An image processing method is provided that automatically calculates Body Surface Area (BSA) score using machine learning techniques. A Felzenszwalb image segmentation algorithm is used to define proposed regions in each of a plurality of training set images. The training set images are oversegmented, and then each of the proposed regions in each of the plurality of oversegmented training set images are manually classified as being a lesion or a non-lesion. A Convolutional Neural Network (CNN) is then trained using the manually classified proposed regions in each of the plurality of training set images. The trained CNN is then used on test images to calculate BSA scores.

Abstract: A method for the computing and memory resource-conserving semantic segmentation of image data of an imaging sensor with the aid of an artificial neural network, in particular, a convolutional neural network, the artificial neural network including an encoder path and a decoder path. The method includes: dividing an input tensor into at least one first slice tensor and at least one second slice tensor as a function of a division function, the input tensor being dependent on the image data; outputting the at least one first slice tensor to the decoder path of the neural network; connecting the at least one first slice tensor to the at least one second slice tensor as a function of a connecting function to obtain an output tensor; and outputting the output tensor to the encoder path of the artificial neural network.