公开(公告)号：US20240013035A1

公开(公告)日：2024-01-11

申请号：US18238459

申请日：2023-08-26

Applicant: ZHEJIANG LAB ,  ZHEJIANG UNIVERSITY

Inventor： Mengxiao ZHANG ,  Huajin TANG ,  Chaofei HONG ,  Xiao WANG ,  Mengwen YUAN ,  Yujing LU ,  Wenyi ZHAO ,  Gang PAN

IPC: G06N3/049 ,  G06N3/10 ,  G06N3/08

CPC classification number: G06N3/049 ,  G06N3/10 ,  G06N3/08

Abstract: A computing platform (10), a method, and an apparatus (20) for spiking neural network (SNN) learning and simulation are provided. The computing platform (10) includes a neuron dynamics simulation module (11), a neuron conversion module (12), an SNN construction and weight learning module (13), and a neural network level parameter and weight access module (14). The neuron dynamics simulation module (11) simulates changing features of neurons. The neuron conversion module (12) performs operations on a calculation graph. The SNN construction and weight learning module (13) updates and iterates connection weights. The neural network level parameter and weight access module (14) stores overall network detail parameters.

公开(公告)号：US11868869B1

公开(公告)日：2024-01-09

申请号：US18215784

申请日：2023-06-28

Applicant: ZHEJIANG LAB

Inventor： Gang Huang ,  Wei Hua ,  Yongfu Li

IPC: G06N3/04 ,  G06N3/049

CPC classification number: G06N3/049

Abstract: The present invention relates to the field of smart grids, and provides a non-intrusive load monitoring method and device based on temporal attention mechanism. The method comprises the following steps: obtaining a total load data, an equipment load data, and corresponding sampling time of a building during a certain period of time; integrating the total load data and the equipment load data with the corresponding sampling time to obtain an enhanced total load data and an enhanced equipment load data; using a sliding window method to segment the enhanced total load data and the enhanced equipment load data, and constructing a deep learning training dataset; constructing a neural network model based on a deep learning training framework and training the model using the training dataset. The present invention can effectively extract the working time mode of the load and its inherent dependencies, thereby improving the accuracy of load monitoring.

公开(公告)号：US20240006372A1

公开(公告)日：2024-01-04

申请号：US18328797

申请日：2023-06-05

Applicant: ZHEJIANG LAB

Inventor： Weihao WANG ,  Shunbin LI ,  Guandong LIU ,  Ruyun ZHANG ,  Qinrang LIU ,  Zhiquan WAN ,  Jianliang SHEN

IPC: H01L23/00

CPC classification number: H01L24/81 ,  H01L24/11 ,  H01L24/14 ,  H01L2224/1403 ,  H01L2224/145 ,  H01L2224/14131 ,  H01L2224/11462

Abstract: A system-on-wafer structure and a fabrication method. The structure includes a wafer substrate, an integrated chiplet, a system configuration board and a thermal module. The wafer substrate and the integrated chiplet are bonded through a wafer micro bump array and a chiplet micro bump array. The wafer substrate and the system configuration board are bonded through a copper pillar array on wafer substrate topside and a pad on system configuration board backside. A molding layer is provided between the wafer substrate and the system configuration board, and is configured to mold the wafer substrate, the integrated chiplet and the copper pillar array. Integrated chiplet are electrically connected to each other through a re-distributed layer in wafer substrate. The integrated chiplet is electrically connected to the system configuration board through the re-distributed layer and the copper pillar array. The thermal module is attached to the backside of the wafer substrate.

公开(公告)号：US11861505B2

公开(公告)日：2024-01-02

申请号：US17833088

申请日：2022-06-06

Applicant: ZHEJIANG LAB

Inventor： Hongsheng Wang ,  Hujun Bao ,  Guang Chen

IPC: G06N3/10 ,  G06N7/01 ,  G06N3/04

CPC classification number: G06N3/10 ,  G06N3/04 ,  G06N7/01

Abstract: The disclosure discloses a method of executing dynamic graph for neural network computation and the apparatus thereof. The method of executing dynamic graph includes the following steps: S1: constructing and distributing an operator and a tensor; S2: deducing an operator executing process by an operator interpreter; S3: constructing an instruction of a virtual machine at runtime by the operator interpreter; S4: sending the instruction to the virtual machine at runtime by the operator interpreter; S5: scheduling the instruction by the virtual machine; and S6: releasing an executed instruction by the virtual machine. According to the method of executing dynamic graph for neural network computation and the apparatus thereof provided by the disclosure, runtime is abstracted to be the virtual machine, and the virtual machine acquires a sub-graph of each step constructed by a user in real time through the interpreter and schedules, the virtual machines issues, and executes each sub-graph.

公开(公告)号：US20230421500A1

公开(公告)日：2023-12-28

申请号：US18092939

申请日：2023-01-04

Applicant: ZHEJIANG LAB

Inventor： Congqi SHEN ,  Xingchang GUO ,  Hanguang LUO ,  Qi XU ,  Tao ZOU ,  Ruyun ZHANG

IPC: H04L45/7453 ,  H04L67/63 ,  H04N21/232

CPC classification number: H04L45/7453 ,  H04N21/232 ,  H04L67/63

Abstract: A content store-and-forward method, apparatus, an electronic apparatus and a storage medium, the method comprising the following steps: receiving an interest packet in a named data network; forwarding the interest packet to a storage node, so that the storage node looks up the corresponding content, and packages the content into a data packet containing the hash value of the name identifier and the corresponding content; receiving the data packet forwarded by the storage node; forwarding the data packet to an interest packet port, wherein the interest packet port is a port that has once received the name identifier corresponding to the content in the data packet; forwarding another data packet to the storage node, so that the storage node parses the hash value and the content of the name identifier in the another data packet, and stores the hash value of the name identifier and the corresponding content.

公开(公告)号：US20230419639A1

公开(公告)日：2023-12-28

申请号：US18464249

申请日：2023-09-10

Applicant: ZHEJIANG LAB

Inventor： Zhende LIU ,  Wuyue ZHANG ,  Peng LIU

IPC: G06V10/762 ,  G06V10/26 ,  G06T7/62

CPC classification number: G06V10/762 ,  G06V10/26 ,  G06T7/62 ,  G06V2201/07 ,  G06V20/13

Abstract: The present application discloses a target detection method and device for a high-resolution remote sensing image, which comprises: acquiring an original high-resolution remote sensing image from a sensor; acquiring target information of an area and an expanded area of the area expanding around by a predetermined distance; adaptively partitioning the original high-resolution remote sensing image to obtain different cluster areas to be detected, and obtaining an area required to be detected and an area not required to be detected; selecting a model of the area required to be detected and generating a target detection scheme; executing the target detection scheme to obtain a detection result; determining whether a computing platform has extra computing resources to detect the area not required to be detected, if so, performing dynamic partition detection for the area to obtain a detection result, and merging the detection results into a target detection result.

公开(公告)号：US20230418836A1

公开(公告)日：2023-12-28

申请号：US17981368

申请日：2022-11-04

Applicant: ZHEJIANG LAB

Inventor： Peilei WANG ,  Ruyun ZHANG ,  Tao ZOU ,  Shunbin LI ,  Peilong HUANG

IPC: G06F16/27 ,  H04L67/56 ,  G06F9/54 ,  G06F16/25

CPC classification number: G06F16/27 ,  H04L67/56 ,  G06F9/54 ,  G06F16/256

Abstract: Disclosed are an input/output proxy method and apparatus for a mimic Redis database. Through a pseudo server module, it is ensured that the interface of the Redis database is consistent with the external interface of the native Redis, so that it is convenient to implant the Redis database into arbitrary Redis application scenarios; the isolation of the modules inside is realized by independent processes, thus facilitating independent development, maintenance and expansion; and the synchronization function is integrated into the input/output proxy to achieve resource reuse; for the synchronization function, the random credit attenuation mechanism is cleverly utilized to ensure the synchronization function while taking into account the saving of resources.

公开(公告)号：US20230410560A1

公开(公告)日：2023-12-21

申请号：US17950033

申请日：2022-09-21

Applicant: ZHEJIANG LAB

Inventor： Hongsheng WANG ,  Guang CHEN ,  Hujun BAO

IPC: G06V40/20 ,  G06V20/64 ,  G06V10/56 ,  G06V10/82

CPC classification number: G06V40/25 ,  G06V20/64 ,  G06V10/56 ,  G06V10/82

Abstract: Disclosed are a method and apparatus for constructing a three-dimensional data set of a pedestrian re-identification based on a neural radiation field. The method includes the following steps: S1: capturing images of pedestrians to be entered by a group of cameras at different viewing angles; S2: generating a three-dimensional spatial position point set by sampling through camera rays in the scenario, and converting observation directions of the cameras corresponding to the three-dimensional spatial position point set into three-dimensional Cartesian unit vectors; and S3: inputting, into a multi-layer sensor, the three-dimensional spatial position point set and the observation directions converted into the three-dimensional Cartesian unit vectors, to output corresponding densities and colors. The method and apparatus of the present disclosure gives a brand-new method for constructing a pedestrian re-identification data set, and provides a new idea of data set construction.

公开(公告)号：US20230376355A1

公开(公告)日：2023-11-23

申请号：US18095381

申请日：2023-01-10

Applicant: ZHEJIANG LAB

Inventor： Mengmeng LIU ,  Qiongqian YANG ,  Dandan HUANG ,  Chen XU ,  Yanlin LIU ,  Zhenting LI

IPC: G06F9/50

CPC classification number: G06F9/5055

Abstract: The disclosure discloses a computing task allocation method, an updating method for computing task allocation, a terminal and a network device. When a computing task of a terminal is generated, computing task allocation is performed using at least one of a centralized mode, a distributed mode, or a hybrid mode; the computing task allocation includes communication resource allocation, computing resource allocation, and a task offloading decision; the above computing task allocation method subjected to dynamically updating according to a terminal state, a network state or a task state. Therefore, the compromise problem between overall system performance optimization and device fairness in a cloud-edge collaborative IoT system is solved.

公开(公告)号：US11823384B2

公开(公告)日：2023-11-21

申请号：US17766204

申请日：2021-01-23

Applicant: ZHEJIANG LAB ,  MINFOUND MEDICAL SYSTEMS CO., LTD

Inventor： Fan Rao ,  Wentao Zhu ,  Bao Yang ,  Ling Chen ,  Hongwei Ye

IPC: G06T7/00

CPC classification number: G06T7/0012 ,  G06T2207/10081 ,  G06T2207/10104 ,  G06T2207/20081

Abstract: Disclosed is a CT image generation method for attenuation correction of PET images. According to the method, a CT image and a PET image at T1 and a PET image at T2 are acquired and input into a trained deep learning network to obtain a CT image at T2; the CT image can be applied to the attenuation correction of the PET image, thereby obtaining more an accurate PET AC (Attenuation Correction) image. According to the CT image generation method for attenuation correction of PET images, the dosage of X-rays received by a patient in the whole image acquisition stage can be reduced, and physiological and psychological pressure of the patient is relieved. In addition, the later image acquisition only needs a PET imaging device, without the need of PET/CT device, cost of imaging resource distribution can be reduced, and the imaging expense of the whole stage is reduced.