Patents by Inventor Yudong Liang
Yudong Liang has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11954842Abstract: The invention discloses a powder leakage monitoring device and a powder leakage monitoring method. The powder leakage monitoring device comprises light field camera, 3D PTZ (3-Dimensional Pan/Tilt/Zoom) and computer. Wherein, the light field camera records original light field images of monitored area; the 3D PTZ under the light field camera adjusts the shooting angle of the light field camera when it rotates according to the set direction; and the computer respectively connects to the light field camera and the 3D PTZ, which generates refocused images corresponding to the original light field images, and determines the spatial coordinates of the powder leakage point and the hazard range of the powder leakage in the monitored area according to the refocused images and the shooting angle. Therefore, the range and accuracy of powder leakage monitoring are both increased by using this invention.Type: GrantFiled: October 31, 2022Date of Patent: April 9, 2024Assignee: HUANENG NANJING JINLING POWER GENERATION CO., LTD.Inventors: Yudong Liu, Lei Wang, Kai Shi, Yong Xia, Xudong Si, Dongfang Li, Xuedong Liang
-
Publication number: 20240046105Abstract: A system for image quality assessment of non-aligned images includes a first deep path portion of a convolutional neural network having a set of parameters and a second deep path portion of the convolutional neural network sharing a set of parameters with the first deep path convolutional neural network. Weights are shared between the first and second deep path convolutional neural networks to support extraction of a same set of features in each neural network pathway. Non-aligned reference and distorted images are respectively provided to the first and second deep paths of the convolutional neural network for processing. A concatenation layer is connected to both the first and second deep paths convolutional neural network, and a fully connected layer is connected to the concatenation layer to receive input from both the first and second deep paths of the convolutional neural network, generating an image quality assessment as a linear regressor and outputting an image quality score.Type: ApplicationFiled: October 16, 2023Publication date: February 8, 2024Inventors: Jinjun Wang, Yudong Liang
-
Patent number: 11816576Abstract: A system for image quality assessment of non-aligned images includes a first deep path portion of a convolutional neural network having a set of parameters and a second deep path portion of the convolutional neural network sharing a set of parameters with the first deep path convolutional neural network. Weights are shared between the first and second deep path convolutional neural networks to support extraction of a same set of features in each neural network pathway. Non-aligned reference and distorted images are respectively provided to the first and second deep paths of the convolutional neural network for processing. A concatenation layer is connected to both the first and second deep paths convolutional neural network, and a fully connected layer is connected to the concatenation layer to receive input from both the first and second deep paths of the convolutional neural network, generating an image quality assessment as a linear regressor and outputting an image quality score.Type: GrantFiled: July 20, 2021Date of Patent: November 14, 2023Assignee: DEEP NORTH, INC.Inventors: Jinjun Wang, Yudong Liang
-
Publication number: 20210350243Abstract: A system for image quality assessment of non-aligned images includes a first deep path portion of a convolutional neural network having a set of parameters and a second deep path portion of the convolutional neural network sharing a set of parameters with the first deep path convolutional neural network. Weights are shared between the first and second deep path convolutional neural networks to support extraction of a same set of features in each neural network pathway. Non-aligned reference and distorted images are respectively provided to the first and second deep paths of the convolutional neural network for processing. A concatenation layer is connected to both the first and second deep paths convolutional neural network, and a fully connected layer is connected to the concatenation layer to receive input from both the first and second deep paths of the convolutional neural network, generating an image quality assessment as a linear regressor and outputting an image quality score.Type: ApplicationFiled: July 20, 2021Publication date: November 11, 2021Inventors: Jinjun Wang, Yudong Liang
-
Patent number: 11100402Abstract: A system for image quality assessment of non-aligned images includes a first deep path portion of a convolutional neural network having a set of parameters and a second deep path portion of the convolutional neural network sharing a set of parameters with the first deep path convolutional neural network. Weights are shared between the first and second deep path convolutional neural networks to support extraction of a same set of features in each neural network pathway. Non-aligned reference and distorted images are respectively provided to the first and second deep paths of the convolutional neural network for processing. A concatenation layer is connected to both the first and second deep paths convolutional neural network, and a fully connected layer is connected to the concatenation layer to receive input from both the first and second deep paths of the convolutional neural network, generating an image quality assessment as a linear regressor and outputting an image quality score.Type: GrantFiled: January 16, 2020Date of Patent: August 24, 2021Assignee: DEEP NORTH, INC.Inventors: Jinjun Wang, Yudong Liang
-
Publication number: 20200280115Abstract: An integrated filter system, including: a dielectric slab assembly; and a filter circuit arranged on the dielectric slab assembly, where the filter circuit includes a low-pass filter and a band-pass filter connected to one another in series. The present invention further discloses an antenna system. By means of the foregoing implementations, the filter system and the antenna system are integrated and miniaturized and have smaller weights, and the filter system and the antenna system have a simple structure and lower costs, and stable and reliable whole performance.Type: ApplicationFiled: January 20, 2017Publication date: September 3, 2020Applicants: TONGYU COMMUNICATION INC., TONGYU COMMUNICATION INC.Inventors: Hui CAI, Pengbo WANG, Qi ZHOU, Jianzhao YU, Yudong LIANG, Qilue FU, Gang JU
-
Publication number: 20200234141Abstract: A system for image quality assessment of non-aligned images includes a first deep path portion of a convolutional neural network having a set of parameters and a second deep path portion of the convolutional neural network sharing a set of parameters with the first deep path convolutional neural network. Weights are shared between the first and second deep path convolutional neural networks to support extraction of a same set of features in each neural network pathway. Non-aligned reference and distorted images are respectively provided to the first and second deep paths of the convolutional neural network for processing. A concatenation layer is connected to both the first and second deep paths convolutional neural network, and a fully connected layer is connected to the concatenation layer to receive input from both the first and second deep paths of the convolutional neural network, generating an image quality assessment as a linear regressor and outputting an image quality score.Type: ApplicationFiled: January 16, 2020Publication date: July 23, 2020Inventors: Jinjun Wang, Yudong Liang
-
Patent number: 10540589Abstract: A system for image quality assessment of non-aligned images includes a first deep path portion of a convolutional neural network having a set of parameters and a second deep path portion of the convolutional neural network sharing a set of parameters with the first deep path convolutional neural network. Weights are shared between the first and second deep path convolutional neural networks to support extraction of a same set of features in each neural network pathway. Non-aligned reference and distorted images are respectively provided to the first and second deep paths of the convolutional neural network for processing. A concatenation layer is connected to both the first and second deep paths convolutional neural network, and a fully connected layer is connected to the concatenation layer to receive input from both the first and second deep paths of the convolutional neural network, generating an image quality assessment as a linear regressor and outputting an image quality score.Type: GrantFiled: October 24, 2017Date of Patent: January 21, 2020Assignee: DEEP NORTH, INC.Inventors: Jinjun Wang, Yudong Liang
-
Publication number: 20190122115Abstract: A system for image quality assessment of non-aligned images includes a first deep path portion of a convolutional neural network having a set of parameters and a second deep path portion of the convolutional neural network sharing a set of parameters with the first deep path convolutional neural network. Weights are shared between the first and second deep path convolutional neural networks to support extraction of a same set of features in each neural network pathway. Non-aligned reference and distorted images are respectively provided to the first and second deep paths of the convolutional neural network for processing. A concatenation layer is connected to both the first and second deep paths convolutional neural network, and a fully connected layer is connected to the concatenation layer to receive input from both the first and second deep paths of the convolutional neural network, generating an image quality assessment as a linear regressor and outputting an image quality score.Type: ApplicationFiled: October 24, 2017Publication date: April 25, 2019Inventors: Jinjun Wang, Yudong Liang
-
Patent number: D734879Type: GrantFiled: March 7, 2014Date of Patent: July 21, 2015Assignee: Zhejiang Shengui Lighting Co, Ltd.Inventors: Liang Chen, Yudong Liang, Yinxiao Zhu
-
Patent number: D982623Type: GrantFiled: October 21, 2021Date of Patent: April 4, 2023Assignee: Guangzhou Jingwushi Information Technology Co., Ltd.Inventors: Yudong Liang, De Zhang