Abstract: Imaging data (20) are acquired by a PET scanner (6) or other imaging device. Iterative image reconstruction of the imaging data is performed to generate a reconstructed image (22). The iterative image reconstruction includes performing an update step (24) that includes an edge preserving prior (28) having a spatially varying edge preservation threshold (30) whose value at each image voxel depends on a noise metric (32) in a local neighborhood of the image voxel. The noise metric may be computed as an aggregation of the intensities of neighborhood image voxels of the reconstructed image in the local neighborhood of the image voxel. The edge preserving prior may be a Relative Difference Prior (RDP). For further noise suppression, during the iterative image reconstruction image values of image features of the reconstructed image that have spatial extent smaller than a threshold (38) may be reduced.

Abstract: A computer-implemented method or protecting a machine-learning model against training data attacks is disclosed. The method comprises performing an initial training of a machine-learning system with controlled training data, thereby building a trained initial machine-learning model and identifying high-impact training data from a larger training data set than in the controlled training data, wherein the identified individual training data have an impact on a training cycle of the training of machine-learning model, wherein the impact is larger than a predefined impact threshold value. The method also comprises building an artificial pseudo-malicious training data set from the identified high-impact training data and retraining the machine-learning system comprising the trained initial machine-learning model using the artificial pseudo-malicious training data set.

Abstract: Disclosed herein is a power semiconductor device including a semiconductor body, a first load terminal, a second load terminal, an active region, an edge termination region, and a thin film layer that includes a bulk material and a laminar filler compound. Furthermore, a method of producing such a power semiconductor device is described herein, the method including: providing a thin film including a mixture of a bulk material component and a laminar filler compound onto a surface of at least parts of the edge termination region and/or over at least parts of the first load terminal; and curing the obtained mixture of the bulk material and laminar filler compound to generate a thin-film layer that includes a bulk material and a laminar filler compound.

Abstract: Bevel saws and methods of adjusting a fence of a bevel saw are disclosed herein. The bevel saws include a workpiece support that defines a workpiece support surface. The bevel saws also include a user-actuated assembly and an assembly attachment structure that operatively attaches the user-actuated assembly to the workpiece support. The assembly attachment structure includes a blade bevel angle adjustment mechanism configured to adjust a blade bevel angle of the bevel saw blade. The bevel saws also include a fence that defines a fence surface that extends away from the workpiece support surface. The fence includes a first fence region and a second fence region. The first fence region and the second fence region define a blade-receiving gap therebetween, and the fence further includes a gap adjustment mechanism configured to automatically adjust the blade-receiving gap based, at least in part, on the blade bevel angle.

Abstract: Provided herein are compositions and methods useful for the detection of MTB. In particular, provided herein are kits, reagents, reaction mixtures, and methods involving such for nucleic acid amplification and detection procedures, which specifically and sensitively detect MTB in samples.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a prediction of an audio signal. One of the methods includes receiving a request to generate an audio signal conditioned on an input; processing the input using an embedding neural network to map the input to one or more embedding tokens; generating a semantic representation of the audio signal; generating, using one or more generative neural networks and conditioned on at least the semantic representation and the embedding tokens, an acoustic representation of the audio signal; and processing at least the acoustic representation using a decoder neural network to generate the prediction of the audio signal.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a prediction of an audio signal. One of the methods includes receiving a request to generate an audio signal conditioned on an input; processing the input using an embedding neural network to map the input to one or more embedding tokens; generating a semantic representation of the audio signal; generating, using one or more generative neural networks and conditioned on at least the semantic representation and the embedding tokens, an acoustic representation of the audio signal; and processing at least the acoustic representation using a decoder neural network to generate the prediction of the audio signal.

Abstract: A multi-purpose detection circuit for object detection and vehicle position determination is described. For example, the circuit is configurable for detecting foreign metallic objects, living objects, and a vehicle or type of vehicle above an inductive wireless power transmitter. The circuit is also configurable for determining the vehicle's position relative to the inductive wireless power transmitter. An example apparatus includes a measurement circuit including a multiplexer, electrically connected to a plurality of inductive and capacitive sense circuits, for measuring one or more electrical characteristics in each of the inductive and capacitive sense circuits according to a predetermined time multiplexing scheme.

Abstract: A secure guest of a computing environment requests confidential data. The confidential data is included in metadata of the secure guest, which is stored in a trusted execution environment of the computing environment. Based on the request, the confidential data is obtained from the metadata of the secure guest that is stored in the trusted execution environment.

Abstract: A method for coating objects starts with dispensing, by means of a first painting system, a first coating layer on a first object. A property of the first coating layer is measured to thereby obtain a first property value, and a coating variable of a second painting system is adjusted based on the first property value. A second coating layer is dispensed on the first coating layer by means of the second painting system which is different from the first painting system. By providing a measurement between the two painting systems, and by adjusting the coating parameters of the downstream painting system based on the measurement results, the overall coating can be corrected by compensating by means of the downstream painting system for a defective coating dispensed by the upstream painting system.

Abstract: At least one request to store diagnostic state of a virtual machine is obtained. Based on obtaining the at least one request, a store of diagnostic state of the virtual machine is performed to provide stored diagnostic state of the virtual machine. The performing the store includes encrypting the diagnostic state of the virtual machine that is unencrypted and being stored to prevent a reading of the diagnostic state of the virtual machine by an untrusted entity prior to encrypting the diagnostic state of the virtual machine that is unencrypted and being stored.

Abstract: A virtual machine is dispatched and based on the dispatch, a determination is made as to whether a select area of memory expected to be accessible to the virtual machine and used in communication between the virtual machine and an operating system is accessible to the virtual machine. Based on determining that the select area of memory is inaccessible to the virtual machine, virtual machine execution is exited with a select interception code.

Abstract: Deferred reclaiming of secure guest resources within a computing environment is provided, which includes initiating, by a host of the computing environment, removal of a secure guest from the computing environment, while leaving one or more resources of the secure guest to be reclaimed asynchronous to the removal of the secure guest. The deferring also includes reclaiming the one or more secure guest resources asynchronous to the removal of the secure guest, where the one or more secure guest resources are available for reuse as the one or more secure guest resources are reclaimed asynchronous to the removal of the secure guest.

Abstract: A secure guest of a computing environment requests confidential data. The confidential data is included in metadata of the secure guest, which is stored in a trusted execution environment of the computing environment. Based on the request, the confidential data is obtained from the metadata of the secure guest that is stored in the trusted execution environment.

Abstract: A method for inspection of a target object, the method including irradiating a reference surface having a non-flat reference profile with radiation; determining reference response data based on detected radiation having interacted with the reference surface; irradiating a target object with radiation, the target object including a target surface having a non-flat target profile corresponding to the reference profile; determining inspection response data based on detected radiation having interacted with the target object; and determining at least one parameter of the target object based on the reference response data and the inspection response data. An alternative method; a control system for controlling an emitter system and a detector system; and an inspection system including a control system, an emitter system and a detector system, are also provided.

Abstract: An example loudspeaker arrangement includes a seat configured to support a listener sitting in the seat so that a head of the listener is in a listening position; and a loudspeaker array secured to the seat and disposed in a position in front of a backrest of the seat and to the side of the head when the head is in the listening position. The loudspeaker array includes at least one loudspeaker and has a main broadcasting axis representative of a main broadcasting direction, the main broadcasting direction of the loudspeaker array pointing to the head.

Abstract: A method for training an operator of a spray drying plant, the spray drying plant including a plurality of plant elements including pre-processing elements, a spray drying element, post-processing elements, powder recovery elements, and a programmable logic controller, PLC. The method including the steps of: obtaining a transient model of the spray drying plant, wherein the transient model includes transient sub-models of the plurality of plant elements; calculating repeatedly simulated sensor data based on the obtained transient model, using a processing unit; displaying on a display a training human machine interface, tHMI, configured to communicate with the transient model of the spray drying plant, and to display the simulated sensor data and control data for the transient model; and updating the transient model based on operator input on the tHMI, for controlling the transient model of the spray drying plant.

Abstract: A computer-implemented method for creating a secure software container. The method comprises providing a first layered software container image, transforming all files, except corresponding metadata, of each layer of the first layered software container image into a volume, the volume comprises a set of blocks, wherein each layer comprises an incremental difference to a next lower layer, encrypting each block of the set of blocks of a portion of the layers, and storing each encrypted set of the blocks as a layer of an encrypted container image along with unencrypted metadata for rebuilding an order of the set of blocks equal to an order of the first layered software container image, so that a secure encrypted software container is created.

Abstract: The invention relates to optical surveying such as in building construction, road construction, landscaping and similar. A first image sensor obtains in a first wavelength range a first image of a scene within a field of view captured by an optical arrangement such as a telescope. A light emitter emits light in a second wavelength range and a second image sensor obtains a second image of the field of view in the second wavelength range. A target position of a reflecting target is found in the first image by detecting a known image pattern of the reflecting target in the first image. A region of interest in the second image is then a defined based on the identified target position in the first image, for detecting a reflector position of a reflector of the reflecting target in the region of interest. With the invention it becomes possible to improve the identification of a reflective target at reduced processing time, even if reflections from other objects than the reflective target are present.

Abstract: A method for coating objects starts with dispensing, by means of a first painting system, a first coating layer on a first object. A property of the first coating layer is measured to thereby obtain a first property value, and a coating variable of a second painting system is adjusted based on the first property value. A second coating layer is dispensed on the first coating layer by means of the second painting system which is different from the first painting system. By providing a measurement between the two painting systems, and by adjusting the coating parameters of the downstream painting system based on the measurement results, the overall coating can be corrected by compensating by means of the downstream painting system for a defective coating dispensed by the upstream painting system.