中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

2350 records (took 0.038 seconds)

    PROGRAMMABLE SWITCH-BASED SERVICE FUNCTION CHAINING SYSTEM AND METHOD
    192.
    发明申请
    PROGRAMMABLE SWITCH-BASED SERVICE FUNCTION CHAINING SYSTEM AND METHOD 有权

    公开(公告)号：US20220103658A1

    公开(公告)日：2022-03-31

    申请号：US17296352

    申请日：2019-08-13

    Applicant: KOREA UNIVERSITY RESEARCH AND BUSINESS FOUNDATION

    Inventor： Sangheon PACK ,  Jaewook LEE ,  Haneul KO ,  Hochan LEE

    IPC: H04L69/22 ,  H04L49/00 ,  H04L47/2483

    Abstract: A service function chaining system is provided. The service function chaining system includes a PDP management module implementing SFC entities in consideration of an order of a service function chain (SFC) on a programmable data plane (PDP) and an SFC management module that transmits a first SFC entity information, which is information on SFC entities to be implemented on the PDP, to the PDP management module or transmits an SFC rule to a node included on the PDP, and the PDP management module includes a receiving unit for receiving the first SFC entity information, a processing unit for generating a second SFC entity information by processing the first SFC entity information to enable the SFC entities to be implemented on the PDP and to be processed according to an SFC order, and implementing unit for implementing the SFC entities on the PDP using the second SFC entity information.

    METHOD FOR CELL IMAGE SEGMENTATION USING SCRIBBLE LABELS, RECORDING MEDIUM AND DEVICE FOR PERFORMING THE METHOD
    193.
    发明申请
    METHOD FOR CELL IMAGE SEGMENTATION USING SCRIBBLE LABELS, RECORDING MEDIUM AND DEVICE FOR PERFORMING THE METHOD 有权

    公开(公告)号：US20220101522A1

    公开(公告)日：2022-03-31

    申请号：US17477827

    申请日：2021-09-17

    Applicant: Korea University Research and Business Foundation

    Inventor： Won-Ki JEONG ,  Hyunsoo LEE

    IPC: G06T7/00 ,  G06T7/11 ,  G06T7/194 ,  G06T5/00 ,  G06N20/20

    Abstract: A cell image segmentation method using scribble labels includes iteratively pre-training via an image segmentation network (U-Net) using a cell image and scribble labels indicating a cell region and a background region as training data, calculating an exponential moving average (EMA) of image segmentation prediction probabilities at a predetermined interval during the pre-training, self-training by assigning the cell region and the background region for which the EMA of image segmentation prediction probabilities is over a preset threshold to be a pseudo-label, and iteratively refining the image segmentation prediction probability based on a scribbled loss (Lsp) obtained through a result of the training and an unscribbled loss (Lup). Accordingly, it is possible to achieve cell image segmentation with high reliability using only scribble labels.

    METHOD, SYSTEM AND COMPUTER PROGRAM FOR COLORING BASED ON DEEP LEARNING
    194.
    发明申请
    METHOD, SYSTEM AND COMPUTER PROGRAM FOR COLORING BASED ON DEEP LEARNING 有权

    公开(公告)号：US20220092106A1

    公开(公告)日：2022-03-24

    申请号：US17449710

    申请日：2021-10-01

    Applicant: NAVER WEBTOON LTD. ,  KOREA UNIVERSITY RESEARCH AND BUSINESS FOUNDATION

    Inventor： Jae Hyuk CHANG ,  Jae Gul CHOO ,  Seung Joo YOO ,  Sung Hyo CHUNG ,  Ga Young LEE ,  Hyo Jin BAHNG

    IPC: G06F16/535 ,  G06N3/04

    Abstract: A deep learning-based coloring system includes a memory network configured to provide a color feature in response to a specific query and a coloring network configured to perform coloring, based on the color feature generated by the memory network. The memory network includes: a query generation unit configured to generate a query; a neighbor calculation unit configured to calculate k-nearest neighbors, based on similarities between the query and key memory values; a color feature determination unit configured to generate color features for indicating color information stored in the key memory; a threshold triplet loss calculation unit configured to calculate a threshold triplet loss, based on a comparison between a threshold and a distance between the color features; and a memory update unit configured to update a memory, based on whether a distance between a top value and a value of a newly input query is within the threshold.

    APPARATUS FOR DETECTING POSITION OF RADIATION SOURCE
    198.
    发明申请
    APPARATUS FOR DETECTING POSITION OF RADIATION SOURCE 有权

    公开(公告)号：US20220026589A1

    公开(公告)日：2022-01-27

    申请号：US17496975

    申请日：2021-10-08

    Applicant: Korea University Research and Business Foundation ,  Korea Atomic Energy Research Institute

    Inventor： Jung-Yeol YEOM ,  Seop HUR ,  Inyong KWON ,  Chanho KIM

    IPC: G01T1/169 ,  G01T1/20 ,  G01T7/00

    Abstract: The present invention relates to an apparatus for determining the location of a radiation source. The apparatus for determining the location of a radiation source according to the present invention comprises: a collimator part for selectively passing radiation therethrough according to the direction in which the radiation is incident; a scintillator part for converting the radiation incident from the collimator part into a light ray; a first optical sensor for converting the light ray incident from one end of the scintillator part into a first optical signal; a second optical sensor for converting the light ray incident from the other end of the scintillator part into a second optical signal; and a location information acquisition part for acquiring information on the location where the light ray is generated in the scintillator part, by using the second optical signal and the second optical signal.

Patent Agency Ranking