公开(公告)号：US20220407102A1

公开(公告)日：2022-12-22

申请号：US17821882

申请日：2022-08-24

Applicant: KOREA ELECTRONICS TECHNOLOGY INSTITUTE

Inventor： Je-nam LEE ,  Youngkwon KIM ,  Eunbyul DO

IPC: H01M8/18 ,  H01M8/02

Abstract: A zinc-bromine flow battery is proposed. The battery may include a conductive interlayer that can reduce the amount of inactive zinc in the form of dendrites on a negative electrode, thereby improving the zinc desorption process and improving the capacity and lifespan characteristics of the battery. The battery may include a membrane, a first electrode stacked on one side of the membrane, and a second electrode stacked on other side of the membrane. The battery may also include a conductive interlayer interposed between a negative electrode from among the first and second electrodes and the membrane and having a log value of hydrogen generation exchange current density of −4 or less in an acid-based electrolyte.

公开(公告)号：US20220383104A1

公开(公告)日：2022-12-01

申请号：US17512463

申请日：2021-10-27

Applicant: Korea Electronics Technology Institute

Inventor： Choong Sang CHO ,  Young Han LEE

IPC: G06N3/08 ,  G06N3/04 ,  G06T7/11 ,  G06K9/62

Abstract: There are provided a method and a system for image segmentation utilizing a GAN architecture. A method for training an image segmentation network according to an embodiment includes: inputting an image to a first network which is trained to output a region segmentation result regarding an input image, and generating a region segmentation result; and inputting the region segmentation result generated at the generation step and a ground truth (GT) to a second network, and acquiring a discrimination result, the second network being trained to discriminate inputted region segmentation results as a result generated by the first network and a GT, respectively; and training the first network and the second network by using the discrimination result. Accordingly, region segmentation performance of a semantic segmentation network regarding various images can be enhanced, and a very small image region can be exactly segmented.

公开(公告)号：US20220373207A1

公开(公告)日：2022-11-24

申请号：US17622480

申请日：2021-05-07

Applicant: Korea Electronics Technology Institute

Inventor： Ki Woong KWON ,  Sang Hun KIM

IPC: F24F11/46 ,  F24F11/64 ,  G05B19/042 ,  G06K9/62 ,  G06N20/00

Abstract: Provided are a system and a method for predicting an occupancy time of a user in a city building based on big data for auto control of heating or cooling for energy saving. The user occupancy time prediction system according to an embodiment includes: a sensor configured to collect data regarding whether a user occupies a predetermined space in a building; a database configured to store the collected data; a data pre-processing unit configured to process the stored data into a format suitable for machine learning; and a prediction unit configured to input the processed data into a machine learning model, and to predict an expected unoccupancy time of the user regarding the predetermined space in the building. Accordingly, a user occupancy/unoccupancy time may be predicted by analyzing big data which uses previous occupancy data of a user in a city building, and energy may be saved by adjusting a temperature of a heating or cooling device before the unoccupancy time. In addition, when the temperature of the heating or cooling device is adjusted before unoccupancy is predicted, the temperature is only changed to the extent that the user does not recognize inconvenience, and comfortability may be maintained or improved.

公开(公告)号：US20220369955A1

公开(公告)日：2022-11-24

申请号：US17772692

申请日：2019-11-11

Applicant: KOREA ELECTRONICS TECHNOLOGY INSTITUTE

Inventor： Yeonkug MOON ,  Minjoon KIM ,  Donghyun LEE ,  Sunghun CHAE

IPC: A61B5/11 ,  A61B5/22

Abstract: A play experience system disclosed in the present invention comprises an experience device including a plurality of experience rooms and a plurality of experience instruments provided in each of the plurality of experience rooms for the user to experience the play-type; a sensor device comprising at least one or more sensors being disposed in a predetermined area of each of the plurality of experience instruments; and a processing device configured to receive body composition information and heart rate information of the user and comprehensive measurement information on physical development items of the user from the at least one or more sensors, and to obtain result information on a current level of the physical development items of the user by comparing the received comprehensive measurement information on the physical development items with pre-stored standard information on the physical development items.

公开(公告)号：US11482134B2

公开(公告)日：2022-10-25

申请号：US16536151

申请日：2019-08-08

Applicant: KOREA ELECTRONICS TECHNOLOGY INSTITUTE

Inventor： Hye Dong Jung ,  Sang Ki Ko ,  Han Mu Park ,  Chang Jo Kim

IPC: G09B21/00 ,  G06T11/00 ,  G10L15/22 ,  G10L15/24 ,  G10L25/63 ,  G10L25/90 ,  G06V40/20 ,  G06V40/16

Abstract: Disclosed is a method of providing a sign language video reflecting an appearance of a conversation partner. The method includes recognizing a speech language sentence from speech information, and recognizing an appearance image and a background image from video information. The method further comprises acquiring multiple pieces of word-joint information corresponding to the speech language sentence from joint information database, sequentially inputting the word-joint information to a deep learning neural network to generate sentence-joint information, generating a motion model on the basis of the sentence-joint information, and generating a sign language video in which the background image and the appearance image are synthesized with the motion model. The method provides a natural communication environment between a sign language user and a speech language user.

公开(公告)号：US11477206B2

公开(公告)日：2022-10-18

申请号：US16820508

申请日：2020-03-16

Applicant: SAMSUNG ELECTRONICS CO., LTD. ,  KOREA ELECTRONICS TECHNOLOGY INSTITUTE

Inventor： Ji-min Chung ,  Seung-woo Kum ,  Young-sun Ryu ,  Tae-beom Lim

IPC: H04L9/40 ,  H04W4/70 ,  H04W4/08 ,  H04L67/12

Abstract: An example terminal includes a communication circuitry configured to communicate with a server; and a data processor configured to request the server to include a second user in a relationship group of a first user and to extend, to the relationship group, a range of authorization for an Internet of Things (IoT) apparatus registered as an apparatus of the first user.

公开(公告)号：US20220326600A1

公开(公告)日：2022-10-13

申请号：US17698370

申请日：2022-03-18

Applicant: KOREA ELECTRONICS TECHNOLOGY INSTITUTE

Inventor： Hyeong Keun KIM ,  Seul Gi KIM ,  Hyun Mi KIM ,  Jin Woo CHO ,  Hye Young KIM

IPC: G03F1/64 ,  G03F7/20

Abstract: This application relates to a pellicle for extreme ultraviolet lithography containing amorphous carbon and a manufacturing method thereof. In one aspect, the pellicle includes a substrate having an opening formed in a central portion, a support layer formed on the substrate to cover the opening, and a pellicle layer formed on the support layer and containing amorphous carbon. The pellicle layer may include a core layer formed on the support layer, and a capping layer formed on the core layer and may further include a buffer layer. At least one of the core layer, the capping layer, or the buffer layer may be an amorphous carbon layer.

公开(公告)号：US11469475B2

公开(公告)日：2022-10-11

申请号：US16835464

申请日：2020-03-31

Applicant: Korea Electronics Technology Institute

Inventor： Sanggil Woo ,  Jisang Yu ,  Goojin Jeong ,  Jenam Lee

IPC: H01M50/46 ,  H01M4/134 ,  H01M4/36 ,  H01M4/62 ,  H01M4/38 ,  H01M10/0525 ,  H01M10/0562 ,  H01M4/02

Abstract: The present disclosure provides a separator, a lithium metal negative electrode, and a lithium metal secondary battery which include a solid superacid coating layer. The solid superacid coating layer suppresses a growth of lithium dendrites in a lithium metal secondary battery employing lithium metal as a negative electrode by improving a mobility and a reaction uniformity of lithium at an interface of the lithium metal negative electrode and an electrolyte solution. In the lithium metal secondary battery, the solid superacid coating layer comprising solid superacid material having a porous structure is formed on at least one of the lithium metal negative electrode and the separator.

公开(公告)号：US20220317973A1

公开(公告)日：2022-10-06

申请号：US17254254

申请日：2020-11-30

Applicant: KOREA ELECTRONICS TECHNOLOGY INSTITUTE

Inventor： Dong Yeob SHIN ,  Tae Beom LIM ,  Yong Seok LIM

IPC: G06F7/544 ,  G06F7/523 ,  G06F7/50 ,  G06F5/01 ,  G06F1/03 ,  G06N3/04

Abstract: The present invention relates to a pop count-based deep learning neural network computation method, a multiply accumulator, and a device thereof. The computation method according to an exemplary embodiment of the present invention is a computation method for a deep learning neural network, including a step of generating one-hot encoding codes according to the type of first multiplication result values for a multiplication (first multiplication) of weights (W) and input values (A); a step of performing a pop-count for each generated code; and a step of accumulating result values for a constant multiplication (second multiplication) between each type of the first multiplication result value and each count value of the pop-count which are different constant values.

公开(公告)号：US11449738B2

公开(公告)日：2022-09-20

申请号：US16537293

申请日：2019-08-09

Applicant: Korea Electronics Technology Institute

Inventor： Sang Yub Lee ,  Jae Jin Ko ,  In Pyo Cho ,  Jae Kyu Lee

IPC: G06N3/063 ,  G06F12/02

Abstract: Provided are a system and method for managing a neuromorphic memory. The system includes a memory having a multi-dimensional lattice structure, a prescience estimator (PSE) configured to separately estimate memory use of two or more data sets and determine a sequence of inputting the two or more data sets to the memory, and a prescience manager (PSM) configured to allocate memory regions in which the two or more data sets will be stored according to estimation results of the PSE, store the data sets, and perform learning and inference.