公开(公告)号：US11915470B2

公开(公告)日：2024-02-27

申请号：US17746697

申请日：2022-05-17

申请人： Quanzhou Equipment Manufacturing Research Institute

发明人： Xian Wei ,  Jielong Guo ,  Chao Li ,  Hai Lan ,  Dongheng Shao ,  Xiaoliang Tang ,  Xuan Tang ,  Zhiyuan Feng

IPC分类号： G06V10/80 ,  G06V10/25 ,  G06V20/58 ,  G06V10/77 ,  G06V10/82 ,  G06V10/10 ,  G01S13/86 ,  G01S13/931

CPC分类号： G06V10/806 ,  G01S13/865 ,  G01S13/867 ,  G06V10/16 ,  G06V10/25 ,  G06V10/7715 ,  G06V10/82 ,  G06V20/58 ,  G01S13/931 ,  G06V2201/07

摘要： A target detection method based on fusion of vision, lidar and millimeter wave radar comprises: obtaining original data detected by a camera, a millimeter wave radar, and a lidar, and synchronizing the millimeter wave radar, the lidar, and the camera in time and space; performing a calculation on the original data detected by the millimeter wave radar according to a radar protocol; generating a region of interest by using a position, a speed, and a radar reflection area obtained from the calculation; extracting feature maps of a point cloud bird's-eye view and the original data detected by the camera; projecting the region of interest onto the feature maps of the point cloud bird's-eye view and the original data detected by the camera; fusing the feature maps of the point cloud bird's-eye view and the original data detected by the camera, and processing a fused image through a fully connected layer.

公开(公告)号：US11783037B1

公开(公告)日：2023-10-10

申请号：US18317512

申请日：2023-05-15

申请人： Quanzhou Equipment Manufacturing Research Institute

发明人： Jielong Guo ,  Xian Wei ,  Xuan Tang ,  Hui Yu ,  Dongheng Shao ,  Jianfeng Zhang ,  Jie Li ,  Yanhui Huang

IPC分类号： G06F21/56 ,  G06N3/08 ,  G06F21/55

CPC分类号： G06F21/566 ,  G06F21/552 ,  G06N3/08 ,  G06F2221/034

摘要： Disclosed are a defense method and a model of deep learning model aiming at adversarial attacks in the technical field of image recognition, which makes full use of the internal relationship between the adversarial samples and the initial samples, and transforms the adversarial samples into common samples by constructing a filter layer in front of the input layer of the deep learning model; the parameters of the filter layer are trained by using the adversarial attack samples, so as to improve the ability of the model to resist adversarial attack; then the trained filter layer is combined with the learning model after the adversarial training, and a deep learning model with strong robustness and high classification accuracy is obtained, which ensures that the recognition ability of the initial sample is not reduced while resisting the adversarial attacks.

公开(公告)号：US11875576B2

公开(公告)日：2024-01-16

申请号：US18340090

申请日：2023-06-23

申请人： QUANZHOU EQUIPMENT MANUFACTURING RESEARCH INSTITUTE

发明人： Jielong Guo ,  Xian Wei ,  Xuan Tang ,  Hui Yu ,  Jianfeng Zhang ,  Dongheng Shao ,  Xiaodi Yang ,  Yufang Xie

IPC分类号： G06V20/58 ,  G06V10/77 ,  G06V10/32 ,  G06V10/82 ,  G06V10/26 ,  G06V10/774 ,  G06V10/776 ,  G06V10/764

CPC分类号： G06V20/582 ,  G06V10/26 ,  G06V10/32 ,  G06V10/764 ,  G06V10/774 ,  G06V10/776 ,  G06V10/7715 ,  G06V10/82

摘要： Provided is a traffic sign recognition method based on a lightweight neural network, which including: a lightweight neural network model is constructed for training and pruning to obtain a lightweight neural network model; the lightweight neural network model comprises a convolution feature extraction part and a classifier part; the convolution feature extraction part includes one layer of conventional 3×3 convolution and 16 layers of separable asymmetric convolution. The classifier part includes three layers of separable full connection modules.

公开(公告)号：US20220277557A1

公开(公告)日：2022-09-01

申请号：US17746697

申请日：2022-05-17

申请人： Quanzhou Equipment Manufacturing Research Institute

发明人： Xian WEI ,  Jielong Guo ,  Chao Li ,  Hai Lan ,  Dongheng Shao ,  Xiaoliang Tang ,  Xuan Tang ,  Zhiyuan Feng

IPC分类号： G06V10/80 ,  G06V10/77 ,  G06V10/82 ,  G06V10/25 ,  G06V10/10 ,  G06V20/58 ,  G01S13/86

摘要： A target detection method based on fusion of vision, lidar and millimeter wave radar comprises: obtaining original data detected by a camera, a millimeter wave radar, and a lidar, and synchronizing the millimeter wave radar, the lidar, and the camera in time and space; performing a calculation on the original data detected by the millimeter wave radar according to a radar protocol; generating a region of interest by using a position, a speed, and a radar reflection area obtained from the calculation; extracting feature maps of a point cloud bird's-eye view and the original data detected by the camera; projecting the region of interest onto the feature maps of the point cloud bird's-eye view and the original data detected by the camera; fusing the feature maps of the point cloud bird's-eye view and the original data detected by the camera, and processing a fused image through a fully connected layer.