-
公开(公告)号：US20230097391A1
公开(公告)日：2023-03-30
申请号：US18071106
申请日：2022-11-29
Inventor： Luyan LIU , Kai MA , Yefeng ZHENG
IPC: G06V10/776 , G06V10/764 , G06V20/70 , G06V10/762
Abstract: An image processing method can reduce costs related to manual labeling, improve training efficiency, and increase a quantity of training samples, thereby improving the accuracy of an image classification model. First images and second images are processed using an image classification model to obtain predicted classification results. The first images include a classification label and the second images include a pseudo classification label. A first loss value indicating accuracy is acquired based on the predicted classification results, the corresponding classification labels, and the corresponding pseudo classification labels. A second loss value indicating accuracy is acquired based on the predicted classification results and the corresponding pseudo classification labels. A model parameter of the image classification model is updated based on the first loss value and the second loss value. Classification processing and acquisition is performed until a target image classification model is obtained.
-
公开(公告)号：US20230080533A1
公开(公告)日：2023-03-16
申请号：US17992759
申请日：2022-11-22
Inventor： Luyan LIU , Kai MA , Yefeng ZHENG
Abstract: An electroencephalogram signal classification method includes: obtaining a first electroencephalogram signal; processing the first electroencephalogram signal using at least two electroencephalogram signal classification models to obtain respective motor imagery probability distributions from the at least two electroencephalogram signal classification models; and determining a motor imagery type of the first electroencephalogram signal based on the motor imagery probability distributions. A plurality of electroencephalogram signal classification models is respectively trained using an augmented data set obtained through augmentation. During prediction, by combining the plurality of electroencephalogram signal classification models, the accuracy of classifying an electroencephalogram signal to determine a motor imagery type may be improved, when using a model trained with a relatively small number of training samples.
-
公开(公告)号：US20230101539A1
公开(公告)日：2023-03-30
申请号：US18076098
申请日：2022-12-06
Inventor： Luyan LIU , Kai MA , Yefeng ZHENG
IPC: A61B5/00 , A61B5/117 , G06F18/2413
Abstract: A physiological electric signal classification processing method includes: performing data alignment on an initial physiological electric signal corresponding to a target user identity based on target signal spatial information corresponding to the target user identify to obtain a target physiological electric signal; performing spatial feature extraction on the target physiological electric signal based on a target spatial filtering matrix to obtain a target spatial feature, the target spatial filtering matrix being generated based on target training physiological electric signals corresponding to a plurality of training user identities respectively and training labels corresponding to the target training physiological electric signals, the target training physiological electric signals being obtained by performing data alignment on initial training physiological electric signals based on training signal spatial information corresponding to the training user identities; and obtaining a classification result corresponding to the initial physiological electric signal based on the target spatial feature.
-
Patent Agency Ranking
4.发明申请公开(公告)号：US20220215558A1
公开(公告)日：2022-07-07
申请号：US17703829
申请日：2022-03-24
Inventor： Luyan LIU , Kai MA , Yefeng ZHENG
Abstract: A method for three-dimensional edge detection includes obtaining, for each of plural two-dimensional slices of a three-dimensional image, a two-dimensional object detection result and a two-dimensional edge detection result, stacking the two-dimensional object detection results into a three-dimensional object detection result, and stacking the two-dimensional edge detection results into a three-dimensional edge detection result. The method also includes performing encoding according to a feature map of the three-dimensional image, the three-dimensional object detection result, and the three-dimensional edge detection result, to obtain an encoding result, and performing decoding according to the encoding result, the three-dimensional object detection result, and the three-dimensional edge detection result, to obtain an optimized three-dimensional edge detection result of the three-dimensional image.
-
公开(公告)号：US20250005888A1
公开(公告)日：2025-01-02
申请号：US18808070
申请日：2024-08-18
Inventor： Ruifen ZHANG , Luyan LIU , Hong WANG , Yawen HUANG , Yefeng ZHENG
IPC: G06V10/42 , G06T5/10 , G06V10/771
Abstract: An image processing method includes: obtaining, through a plurality of radio frequency coils, a plurality of pieces of corresponding undersampled frequency-domain data respectively; and performing, by using a plurality of image processing networks that are cascaded, an information supplement operation respectively on the plurality of pieces of frequency-domain data to obtain a plurality of corresponding target restored images, and determining a target reconstructed image based on the plurality of target restored images, a piece of frequency-domain data being configured for obtaining one target restored image, and an image processing network including an image restoring network, a frequency-domain complement network, and a susceptibility estimation network.
-
公开(公告)号：US20230077726A1
公开(公告)日：2023-03-16
申请号：US17992565
申请日：2022-11-22
Inventor： Luyan LIU , Kai MA , Yefeng ZHENG
IPC: A61B5/00
Abstract: A method for classification processing of an electrophysiological signal, including acquiring an electrophysiological signal collected by an acquisition device, and acquiring a channel association feature corresponding to the acquisition device. The channel association feature indicates spatial locations of multiple acquisition channels of the acquisition device, each of the multiple acquisition channels collecting the electrophysiological signal at a respective spatial location. The method further includes extracting a time feature corresponding to the electrophysiological signal, and generating an embedded feature based on the channel association feature and the time feature, and extracting a spatial feature corresponding to the embedded feature, and obtaining a classification result corresponding to the electrophysiological signal based on the spatial feature.
-
公开(公告)号：US20220148191A1
公开(公告)日：2022-05-12
申请号：US17587825
申请日：2022-01-28
Inventor： Luyan LIU , Kai MA , Yefeng ZHENG
Abstract: An image segmentation method includes obtaining target domain images and source domain images, and segmenting the source domain images and the target domain images by using a generative network in a first generative adversarial network. The method further includes segmenting the source domain images and the target domain images by using a generative network in a second generative adversarial network, and determining a first source domain image and a second source domain image according to source domain segmentation losses, and determining a first target domain image and a second target domain image according to target domain segmentation losses. The method also includes performing cross training on the first generative adversarial network and the second generative adversarial network to obtain a trained first generative adversarial network; and segmenting a to-be-segmented image based on the generative network in the trained first generative adversarial network.
-
8.发明申请公开(公告)号：US20210407086A1
公开(公告)日：2021-12-30
申请号：US17470433
申请日：2021-09-09
Inventor： Luyan LIU
Abstract: This application provides a method and apparatus for training an image segmentation model, a device, and a storage medium. The method includes: training an initial image segmentation model by using source domain samples, to obtain a pre-trained image segmentation model; extracting a predicted segmentation result of a source domain image and a predicted segmentation result of a target domain image by using the pre-trained image segmentation model; training a first discriminator by using the predicted segmentation result of the source domain image and the predicted segmentation result of the target domain image; training a second discriminator by using the predicted segmentation result of the source domain image and a standard segmentation result of the source domain image; and iteratively training the pre-trained image segmentation model according to a loss function of the pre-trained image segmentation model, an adversarial loss function of the first discriminator, and an adversarial loss function of the second discriminator, until convergence, to obtain a trained image segmentation model.
-
-
-
-
-
-
-
Search Results
8
records
(took 0.013 seconds)
