NEURAL NETWORK COMPUTING SYSTEM AND METHOD OF EXECUTING NEURAL NETWORK MODEL

    公开(公告)号:US20240256834A1

    公开(公告)日:2024-08-01

    申请号:US18543628

    申请日:2023-12-18

    Inventor: Jungho KIM

    CPC classification number: G06N3/045

    Abstract: Provided is a neural network computing system that includes: a processor comprising a plurality of computing devices, a memory storing at least one instruction related to execution of a neural network model, a memory controller; and a system bus. The processor is configured to execute the at least one instruction to: determine a normalized target performance of the neural network model, determine a normalized target performance of each of the plurality of computing devices, determine a normalized target performance of the memory controller and the system bus, determine an operating frequency for each of a plurality of hardware devices based on the determined normalized target performance(s), and execute the neural network model by operating the plurality of hardware devices based on the determined operating frequency for each of the plurality of hardware devices.

    MAGNETIC SHIELD STRUCTURE FOR SPEAKER AND ELECTRONIC DEVICE INCLUDING THE SAME

    公开(公告)号:US20200169816A1

    公开(公告)日:2020-05-28

    申请号:US16684661

    申请日:2019-11-15

    Abstract: Provided is an electronic device. The electronic device may include: a housing including a front plate, a rear plate facing away from the front plate, and a side member surrounding a first space between the front plate and the rear plate; a display panel exposed to an exterior of the electronic device through the front plate and configured to detect a pen input using a magnetic field; a speaker structure disposed between the display panel and the rear plate, and including a first surface facing the display panel, a second surface facing in a direction opposite the display panel, a side surface surrounding a second space between the first surface and the second surface, and a yoke facing the first surface; and a first shield structure made of a ferromagnetic material and disposed on the first surface.

    ULTRASONIC IMAGING APPARATUS AND METHOD FOR CONTROLLING ULTRASONIC IMAGING APPARATUS
    5.
    发明申请
    ULTRASONIC IMAGING APPARATUS AND METHOD FOR CONTROLLING ULTRASONIC IMAGING APPARATUS 审中-公开
    超声波成像装置及超声波成像装置的控制方法

    公开(公告)号:US20150289846A1

    公开(公告)日:2015-10-15

    申请号:US14684770

    申请日:2015-04-13

    CPC classification number: A61B8/5207 G01S7/52047 G01S15/8952

    Abstract: An ultrasonic imaging apparatus includes: an ultrasound probe configured to transmit ultrasonic waves to a target region of an object in a plurality of directions, and to receive vibration waves generated from the object; and an image processor configured to generate image signals in the plurality of directions based on the vibration waves generated according to transmission of the ultrasonic waves in the plurality of directions, and to combine the image signals in the plurality of directions, wherein the ultrasound probe includes ultrasound elements configured to respectively generate ultrasonic waves of different frequencies, the ultrasonic waves intersecting each other in the target region of the object.

    Abstract translation: 超声波成像装置包括:超声波探头,被配置为在多个方向上将物体的目标区域发送超声波,并且接收从物体产生的振动波; 以及图像处理器,其被配置为基于根据所述多个方向上的超声波的发送而产生的振动波来生成所述多个方向上的图像信号,并且将所述多个方向的图像信号组合,其中,所述超声波探头包括 超声波元件被配置为分别产生不同频率的超声波,超声波在物体的目标区域中彼此相交。

    NEURAL NETWORK COMPUTING SYSTEM AND METHOD OF EXECUTING NEURAL NETWORK MODEL

    公开(公告)号:US20240004774A1

    公开(公告)日:2024-01-04

    申请号:US18112769

    申请日:2023-02-22

    CPC classification number: G06F11/3442 G06N3/10

    Abstract: A neural network computing system includes a processor including heterogeneous computing devices configured to execute a neural network model; a memory configured to buffer input data and output data of the neural network model; a memory controller configured to control data input and data output of the memory; and a system bus configured to support communication between the processor and the memory controller. The processor determines a target execution time for each node included in the neural network model based on a target end-to-end execution time of the neural network model; controls operating frequencies of hardware devices including the heterogeneous computing devices, the memory controller, and the system bus based on a target computing device for execution of each node, an amount of work for each node, and the target execution time for each node, and executes the neural network model by operating at the operating frequencies.

    ULTRASOUND IMAGING APPARATUS AND METHOD FOR CONTROLLING THE SAME
    9.
    发明申请
    ULTRASOUND IMAGING APPARATUS AND METHOD FOR CONTROLLING THE SAME 审中-公开
    超声成像装置及其控制方法

    公开(公告)号:US20150289847A1

    公开(公告)日:2015-10-15

    申请号:US14683304

    申请日:2015-04-10

    Abstract: The present disclosure provides an ultrasound imaging apparatus. The ultrasound imaging apparatus includes an ultrasonic probe for transmitting an ultrasound signal toward an object, receiving an echo ultrasound signal reflected from the object, and transforming the echo ultrasound signal to an electric signal; a beamformer for beamforming and outputting the electric signal; an image reconstruction unit for generating a reconstructed image by applying a Point Spread Function (PSF) to an ultrasonic image corresponding to the output signal; and an image post-processor for performing filtering to emphasize a peak point of the reconstructed image. The use of a filter to emphasize a peak point may facilitate an accurate setting of a phase parameter to be used in estimation of a PSF. Accordingly, an almost ideal PSF may be estimated, image reconstruction is performed using the estimated PSF, and thus a high resolution image may be obtained.

    Abstract translation: 本公开提供了一种超声成像装置。 超声波成像装置包括:超声波探头,用于向物体发送超声波信号;接收从物体反射的回波超声信号,将回波超声信号变换为电信号; 用于波束形成和输出电信号的波束形成器; 图像重建单元,用于通过对与输出信号相对应的超声波图像应用点扩展函数(PSF)来产生重构图像; 以及用于执行滤波以强调重构图像的峰值点的图像后处理器。 使用滤波器来强调峰值点可以有助于在PSF的估计中使用的相位参数的精确设置。 因此,可以估计几乎理想的PSF,使用估计的PSF执行图像重建,因此可以获得高分辨率图像。

Patent Agency Ranking