公开(公告)号：US20210074366A1

公开(公告)日：2021-03-11

申请号：US16802477

申请日：2020-02-26

Applicant: KIOXIA CORPORATION

Inventor： Koji HORISAKI ,  Kazuhisa HORIUCHI ,  Ryo YAMAKI ,  Gibeom PARK ,  Youyang NG

IPC: G11C16/26 ,  G11C16/12 ,  G11C16/34 ,  G06F13/16

Abstract: A memory controller performs a reference read on a plurality of memory cells using reference read voltages, generates a histogram indicating the number of memory cells in different threshold voltage bins based on results of the reference read, estimates actual read voltages based on the histogram and a first estimation function, and reads data using the actual read voltages. When reading of the data with the actual read voltages estimated using the first estimation function fails, the memory controller estimates actual read voltages using a second estimation function different from the first estimation function and reads the data with the actual read voltages estimated using the second estimation function.

公开(公告)号：US20210295942A1

公开(公告)日：2021-09-23

申请号：US17184120

申请日：2021-02-24

Applicant: KIOXIA CORPORATION

Inventor： Ryo YAMAKI ,  Youyang NG ,  Koji HORISAKI ,  Kazuhisa HORIUCHI ,  Gibeom PARK

IPC: G11C29/42 ,  G11C29/44 ,  G11C29/18 ,  G11C29/50

Abstract: A memory system includes a non-volatile memory having a plurality of memory cells and a memory controller. The memory controller is configured to generate a histogram indicating, with respect to each of a plurality of threshold voltage levels for multi-level cell (MLC) reading, a number of memory cells at the threshold voltage level, based on data read from the plurality of memory cells using a plurality of reference read voltages, estimate a plurality of read voltages for MLC reading of the plurality of memory cells as estimation values by inputting the histogram into a read voltage estimation model, determine, through MLC reading of the plurality of memory cells using a plurality of sets of read voltages, a set of read voltages for MLC reading as observation values, and update one or more parameters of the read voltage estimation model based on the estimation values and the observation values.

公开(公告)号：US20240311997A1

公开(公告)日：2024-09-19

申请号：US18589658

申请日：2024-02-28

Applicant: Kioxia Corporation

Inventor： Shuhei IIJIMA ,  Osamu YAMANE ,  Youyang NG ,  Yuchieh LIN ,  Takuji OHASHI ,  Takeshi FUJIWARA

IPC: G06T7/00 ,  G06T5/50 ,  G06T7/194 ,  G06V10/44 ,  G06V10/764 ,  G06V20/70

CPC classification number: G06T7/0008 ,  G06T5/50 ,  G06T7/194 ,  G06V10/44 ,  G06V10/764 ,  G06V20/70 ,  G06T2207/20224 ,  G06T2207/30148

Abstract: A semiconductor image processing apparatus including a processing circuitry, the processing circuitry configured to identify a label corresponding to a feature amount included in an input image by using an identifier, learn a model for inferring the feature amount included in the input image and learns the identifier, and perform additional learning of the model based on the input image and the learned identifier.

公开(公告)号：US20230290125A1

公开(公告)日：2023-09-14

申请号：US17942815

申请日：2022-09-12

Applicant: Kioxia Corporation

Inventor： Bo WANG ,  Youyang NG ,  Yuchieh LIN ,  Kengo NAKATA ,  Takeshi FUJIWARA

IPC: G06V10/774 ,  G06V10/40 ,  G06V10/82 ,  G06V10/764 ,  G06V20/70 ,  G06V10/776

CPC classification number: G06V10/7747 ,  G06V10/40 ,  G06V10/82 ,  G06V10/764 ,  G06V20/70 ,  G06V10/776

Abstract: An image processing apparatus has a first image acquisitor that acquires a source image, a second image acquisitor that acquires a first target image, a label acquisitor that acquires a label, a feature extractor including a first neural network that extracts a feature of the source image and a feature of the first target image, a class classifier including a second neural network that performs a class classification of the source image and the first target image, a domain classifier including a third neural network that performs a domain classification of the source image and the first target image, a processor that assigns a pseudo label to the first target image, a self-learner that performs a self-learning of the first neural network, the second neural network, and the third neural network, and a learner that learns the first, second and third neural networks, by performing a back propagation process.

公开(公告)号：US20220076398A1

公开(公告)日：2022-03-10

申请号：US17191063

申请日：2021-03-03

Applicant: Kioxia Corporation

Inventor： Youyang NG ,  Bo WANG ,  Takuji OHASHI ,  Osamu YAMANE ,  Takeshi FUJIWARA

IPC: G06T7/00 ,  G06K9/62

Abstract: An information processing apparatus has an acquisitor configured to acquire an entire area image obtained by capturing an entire area of a processing surface of a wafer including at least one defect, a training image selector configured to select, as a training image, a partial image including at least one defect from the entire area image, a model constructor configured to construct a calculation model of generating a label image obtained by extracting and binarizing the defect included in the partial image, and a learner configured to update a parameter of the calculation model based on a difference between the label image generated by inputting the training image to the calculation model and a reference label image obtained by extracting and binarizing the defect of the training image.

公开(公告)号：US20210027457A1

公开(公告)日：2021-01-28

申请号：US16933441

申请日：2020-07-20

Applicant: Kioxia Corporation

Inventor： Atsushi NAKAJIMA ,  Youyang NG ,  Yuko Kono ,  Takuji Ohashi ,  Chihiro Ida

IPC: G06T7/00 ,  G06F30/27 ,  G06F30/3308

Abstract: A semiconductor image processing apparatus has an image input unit inputs a first semiconductor image, an exposure condition input unit configured to input exposure conditions, a generator performs a process of extracting a feature amount in consideration of the exposure conditions while reducing resolution of the first semiconductor image and thereafter use the extracted feature amount to increase the resolution to generate a second semiconductor image, and a discriminator configured to discriminate whether the input image is the second semiconductor image or a third semiconductor image provided in advance. The generator performs learning so that the discriminator erroneously discriminates the second semiconductor image as the third semiconductor image based on a result discriminated by the discriminator. The discriminator performs learning so as not to erroneously discriminate the second semiconductor image as the third semiconductor image, and not to erroneously discriminate the third semiconductor image as the second semiconductor image.