这是我们最新精选的有关信号处理的全球英文出版物和专利，涉及许多科学在线期刊，分类并侧重于信号处理、量化、去噪、数字信号、模拟信号和卷积。
Learning-based segmentation of diffusion-weighted MR images with arbitrary q -space samplings
Published on 2026-06-02 by @MIT
Abstract: AbstractSegmenting anatomical regions is a crucial step in many diffusion-weighted MRI (dMRI) workflows, such as region-of-interest analysis or anatomically-constrained tractography, which enable in vivo studies of brain microstructure and connectivity. However, convolutional neural networks (CNNs)—the foundation of most state-of-the-art segmentation models—require structured inputs with a fixed number of channels. This makes them ill-suited for dMRI, where acquisition protocols vary widely [...]
Our summary: This work presents a novel method for segmenting diffusion-weighted MRI data using geometric deep learning. It directly maps unstructured dMRI data to anatomical segmentations without requiring diffusion model fits. The proposed approach achieves robust generalization and superior performance compared to existing methods.
segmentation, diffusion-weighted MRI, geometric deep learning, convolutional neural networks
Publication
Processor-implemented method and system for generating a high definition (hd) map for high precision position estimation and map maintenance
Patent published on the 2026-05-28 in WO under Ref WO2026107580 by MICRO ENG TECH INC [CA] (Zekry Ahmed Mohamed Ayman Riad Ahmed [ca], Elhabiby Mohamed Mamdouh [ca], El-sheimy Naser Mahmoud [ca])
Abstract: A processor-implemented method and system for generating a high-definition (HD) map for high-precision position estimation and map maintenance is provided. The method includes mapping an input red-green-blue (RGB) data into a latent space vector representing an image location and an orientation in a three-dimensional (3D) space. The method further includes mapping the image location and the orientation of the input RGB data to a high-dimensional synthesized RGB image. The method further includes[...]
Our summary: The method generates a high-definition map for precise position estimation and maintenance. It utilizes RGB data to create a latent space vector and synthesize high-dimensional images. The system employs a CNN to classify images for model retraining and updates the HD map as needed.
HD Map, Position Estimation, Generative Adversarial Networks, Convolutional Neural Network
Patent
Denoising of medical image data
Patent published on the 2026-05-28 in WO under Ref WO2026109285 by KONINKLIJKE PHILIPS N V [NL] (Brown Kevin Martin [nl], SchnellbÄcher Nikolas David [nl])
Abstract: Medical image data is denoised by applying a first machine-learning algorithm to process the medical image data to produce first denoised image data and first noise data representing the noise in the medical image data. The first noise data is inverted to produce second noise data. The second noise data is combined with the first denoised image data, and the resulting combined image data is processed using a second machine-learning algorithm to produce second denoised image data.[...]
Our summary: Medical image data is denoised using a first machine-learning algorithm. The first noise data is inverted and combined with the first denoised image data. A second machine-learning algorithm processes the combined data to produce second denoised image data.
Denoising, Medical Imaging, Machine Learning, Image Processing
Patent
Rotated constellations for frequency diversity
Patent published on the 2026-05-28 in US under Ref US20260150107 by QUALCOMM INCORPORATED [US] (Sen Pinar [us], Taherzadeh Boroujeni Mahmoud [us], Yang Wei [us], Ivanov Kirill [us])
Abstract: Methods, systems, and devices for wireless communications are described. A wireless communications system may implement unitary rotations of constellations to introduce frequency diversity for signaling, such as control signaling, in accordance with a coupling matrix or a rotation angle. The digital signal may be decoded to obtain the original data. As such, frequency diversity may be introduced to signals by performing one or more unitary rotations and transmitted rotated versions of the signal[...]
Our summary: This content describes methods for introducing frequency diversity in wireless communications through unitary rotations of constellations. It outlines how a wireless system can utilize a coupling matrix or rotation angle to enhance signaling robustness. User equipment can receive configurations to apply these rotations, improving communication reliability.
frequency diversity, unitary rotations, wireless communications, coupling matrix
Patent
Sparse 1-bit quantization-based over-the-air federated learning method and system
Patent published on the 2026-05-28 in US under Ref US20260148092 by CHUNG ANG UNIV INDUSTRY ACADEMIC COOPERATION FOUNDATION [KR] (Cho Sung Rae [kr], Lee Dong Hyun [kr], Oh Jun Suk [kr])
Abstract: Sparse 1-bit quantization-based over-the-air federated learning method and system are disclosed. The A sparse 1-bit quantization-based over-the-air federated learning method, comprising: receiving a global parameter vector from a server at each communication round t and updating a local parameter vector using the global gradient vector; deriving a local gradient vector by applying a first-order approximation algorithm using a local dataset, compressing the local gradient vector through 1-bit qua[...]
Our summary: The method involves receiving a global parameter vector and updating a local parameter vector. It compresses the local gradient vector using 1-bit quantization and sparsifies it. The server aggregates the transmitted sparsified vectors to calculate and broadcast the global gradient vector.
federated learning, 1-bit quantization, sparse representation, global parameter vector
Patent
Dual in-line package defect image post-processing system combining bilayer occlusion perception and soft non-maximum suppression and method thereof
Patent published on the 2026-05-28 in US under Ref US20260148361 by INVENTEC CORP [TW] (Zhuang Zi-yue [cn])
Abstract: [0000] A dual in-line package defect image post-processing system combining bilayer occlusion perception and soft non-maximum suppression and a method thereof are disclosed. In the system, a bilayer convolutional network model models a defect region of a dual in-line package image as a top layer and a bottom layer overlapped with each other, detects an occlusion part and infers an occluded part to output masks and detection boxes. Each of the masks and the detection boxes is filtered based on an[...]
Our summary: The system utilizes a bilayer convolutional network to model and detect defects in dual in-line package images. It incorporates occlusion perception and soft non-maximum suppression to enhance detection accuracy. The output includes simplified polygonal masks and high-confidence detection boxes for effective defect inspection.
bilayer convolutional network, occlusion perception, soft non-maximum suppression, defect inspection
Patent
Method and system for anomaly detection, health index classification, and remaining useful life (rul) prediction of circuit breakers
Patent published on the 2026-05-28 in US under Ref US20260148133 by THE HONG KONG POLYTECHNIC UNIV [HK] (Bu Siqi [hk], Lin Pui Yu [hk], Yeung Him Ching [hk], Lu Genghong [hk], Fu Yicheng [hk], Wang Bingsen [hk], Dan Yihua [hk], Wong Chak Nam [hk], Lam Kin Man [hk])
Abstract: [0000] Disclosed is a method for anomaly detection, health index classification, and remaining useful life (RUL) prediction of circuit breakers, including the steps of: capturing and collecting real-time trip coil waveform signals from each of circuit breakers; transmitting the collected trip coil waveform signals to a circuit breaker health monitoring system; processing the collected signals data; analyzing the processed data using artificial intelligence and signal processing algorithms for an[...]
Our summary: The method captures real-time trip coil waveform signals from circuit breakers. It processes and analyzes the signals using AI and signal processing for anomaly detection and health index classification. The system predicts the remaining useful life (RUL) of the circuit breakers based on the analyzed data.
anomaly detection, health index classification, remaining useful life, signal processing
Patent
Impact of vascular geometry, magnetic field calculations, and water diffusion models
Published on 2026-05-26 by @MIT
Abstract: AbstractBiophysical simulations have guided the development of blood oxygenation level-dependent (BOLD) functional MRI (fMRI) acquisitions and signal models that relate the BOLD signal to the underlying physiology, such as calibrated BOLD and vascular fingerprinting. Numerous simulation techniques have been developed, however, few of them have been directly compared, thus limiting the assessment of the accuracy and interchangeability of these methods as well as the accuracy of the quantitative t[...]
Our summary: This study compares eight simulation approaches for BOLD fMRI, focusing on vascular geometry, magnetic field calculations, and water diffusion models. Most simulations showed good agreement with the reference approach, but biases were noted in two-dimensional geometries. The findings support the use of the reference method for exploratory simulations, enhancing the reliability of BOLD signal interpretations.
vascular geometry, magnetic field calculations, water diffusion models, BOLD fMRI
Publication
