-
公开(公告)号:US20180239956A1
公开(公告)日:2018-08-23
申请号:US15875227
申请日:2018-01-19
申请人: MindMaze Holding SA
发明人: Tej TADI , Robert LEEB , Nicolas BOURDAUD , Gangadhar GARIPELLI , Skander MENSI , Nicolas MERLINI , Yann LEBRUN
CPC分类号: G06K9/00302 , G06F3/015 , G06K9/00315 , G06K9/0051 , G06K9/00523 , G06K9/6234 , G06K9/6278 , G06K9/6285 , G06K9/6287 , G10L15/04 , G10L2015/025 , G10L2021/105
摘要: A system, method and apparatus for detecting facial expressions according to EMG signals.
-
公开(公告)号:US20180096218A1
公开(公告)日:2018-04-05
申请号:US15711295
申请日:2017-09-21
CPC分类号: G06K9/46 , G06K9/00442 , G06K9/0051 , G06K9/00986 , G06K9/036 , G06K9/6212 , G06T5/002 , G06T7/0002 , G06T2207/10004 , G06T2207/20076 , H04N1/409 , H04N1/504
摘要: A signal processing method includes addition of noise obtained by multiplying generated random number by K to the input pixel signal I(x), a binarization process of comparing the result of the addition with two thresholds, and a process of calculating a probability. The binarization process includes a first nonlinear process and a second nonlinear process. The first nonlinear process outputs “P” in a case where I(x) after the addition of the noise is greater than the threshold T1 and less than the second threshold T2. The second nonlinear determines “1” or “0” for a processing target pixel, in which the result of the first nonlinear process is “P,” based on input pixel signals of pixels around the processing target pixel. The process of calculating a probability calculates a probability J(x) that the result of the first nonlinear process is “1,” or the result of the first nonlinear process is “P” and the result of the second nonlinear process is “1”.
-
公开(公告)号:US20170356936A1
公开(公告)日:2017-12-14
申请号:US15499973
申请日:2017-04-28
申请人: MCMASTER UNIVERSITY
发明人: Mahmoud ISMAIL , Saeid HABIBI , Samir ZIADA
CPC分类号: G01R19/0053 , G01M13/04 , G01M13/045 , G01M99/005 , G01M99/008 , G06K9/0051 , G06K9/00523 , G06K9/6247
摘要: Systems and methods relating to fault detection and diagnosis. Signals received from sensors are first filtered to remove noise and are then analyzed using wavelet packet transform (WPT) based PCA. The results of the PCA analysis are then automatically classified to thereby quickly and easily determine what issues there may be in a finished product or in a machine being monitored.
-
公开(公告)号:US20170325753A1
公开(公告)日:2017-11-16
申请号:US15668730
申请日:2017-08-04
发明人: Uli TESSEL , Harald GREINER
CPC分类号: A61B5/7425 , G06K9/0051 , G06T1/00 , G06T1/0021 , G06T7/231 , G06T7/285 , G06T2207/20068 , G09G5/377
摘要: The display of correct and reliable operating image data which may change over the time is correctly and reliably displayed. A secured image is generated based on a source image by generating at least one secured image part. The secured image part is generated by combining a source image part of the source image with a corresponding secure pattern. The displaying of the secured image is monitored by detecting the corresponding secure pattern of the secured image part, updating the detected secure pattern, checking the currentness or correctness of the detected secure pattern
-
公开(公告)号:US09734424B2
公开(公告)日:2017-08-15
申请号:US14252638
申请日:2014-04-14
发明人: Sean Ryan Francesco Fanello , Cem Keskin , Pushmeet Kohli , Shahram Izadi , Jamie Daniel Joseph Shotton , Antonio Criminisi
CPC分类号: G06K9/52 , G06K9/0051 , G06K9/40 , G06K9/6268 , G06T5/002 , G06T5/20 , G06T2207/10028 , G06T2207/20081 , H04N5/217
摘要: Filtering sensor data is described, for example, where filters conditioned on a local appearance of the signal are predicted by a machine learning system, and used to filter the sensor data. In various examples the sensor data is a stream of noisy video image data and the filtering process denoises the video stream. In various examples the sensor data is a depth image and the filtering process refines the depth image which may then be used for gesture recognition or other purposes. In various examples the sensor data is one dimensional measurement data from an electric motor and the filtering process denoises the measurements. In examples the machine learning system comprises a random decision forest where trees of the forest store filters at their leaves. In examples, the random decision forest is trained using a training objective with a data dependent regularization term.
-
公开(公告)号:US20170014083A1
公开(公告)日:2017-01-19
申请号:US15076314
申请日:2016-03-21
申请人: Masimo Corporation
发明人: Mohamed K. Diab , Rex J. McCarthy
IPC分类号: A61B5/00 , A61B5/1455 , A61B5/024
CPC分类号: A61B5/7257 , A61B5/0205 , A61B5/02416 , A61B5/02433 , A61B5/1455 , A61B5/14551 , A61B5/14552 , A61B5/6826 , A61B5/7203 , A61B5/7214 , A61B5/7225 , A61B5/726 , A61B5/7278 , A61B5/742 , G06K9/0051
摘要: A method and an apparatus to analyze two measured signals that are modeled as containing desired and undesired portions such as noise, FM and AM modulation. Coefficients relate the two signals according to a model defined in accordance with the present invention. In one embodiment, a transformation is used to evaluate a ratio of the two measured signals in order to find appropriate coefficients. The measured signals are then fed into a signal scrubber which uses the coefficients to remove the unwanted portions. The signal scrubbing is performed in either the time domain or in the frequency domain. The method and apparatus are particularly advantageous to blood oximetry and pulserate measurements. In another embodiment, an estimate of the pulserate is obtained by applying a set of rules to a spectral transform of the scrubbed signal. In another embodiment, an estimate of the pulserate is obtained by transforming the scrubbed signal from a first spectral domain into a second spectral domain. The pulserate is found by identifying the largest spectral peak in the second spectral domain.
摘要翻译: 一种用于分析被建模为包含期望和不期望的部分(例如噪声,FM和AM调制)的两个测量信号的方法和装置。 系数根据根据本发明定义的模型关联两个信号。 在一个实施例中,使用变换来评估两个测量信号的比率以找到适当的系数。 然后将测量的信号馈送到信号洗涤器中,该信号洗涤器使用系数去除不需要的部分。 信号擦除在时域或频域中进行。 该方法和装置对于血氧饱和度测量和脉动测量是特别有利的。 在另一个实施例中,通过将一组规则应用于经擦除的信号的频谱变换来获得脉冲发生器的估计。 在另一个实施例中,通过将擦除的信号从第一光谱域变换成第二光谱域来获得脉冲发生器的估计。 通过识别第二个光谱域中最大的光谱峰值来发现脉冲发生器。
-
公开(公告)号:US09492096B2
公开(公告)日:2016-11-15
申请号:US14467957
申请日:2014-08-25
申请人: VivaQuant LLC
发明人: Brian Brockway , Marina Brockway
IPC分类号: A61B5/04 , A61B5/0456 , A61B5/00 , A61B7/00 , A61B5/0245 , A61B5/0408 , A61B5/0452 , G06F17/14 , G06K9/00 , G06F19/00
CPC分类号: A61B5/04012 , A61B5/0006 , A61B5/0245 , A61B5/04017 , A61B5/04087 , A61B5/0452 , A61B5/0456 , A61B5/7253 , A61B7/00 , G06F17/14 , G06K9/0051 , G06K9/0053 , G16H50/30
摘要: Aspects of the present disclosure are directed to processing ECG signals from a subject. As may be implemented in accordance with one or more embodiments, respective electrodes sense ECG signals from a subject, and the ECG signals are digitized and processed, such as to remove noise, detect a QRS complex, evaluate quality, detect arrhythmia and/or to store the signals. In response to an input from the subject, one or more of the ECG signals is recorded along with sound from the user, such as to concurrently record the user's voice for describing conditions in connection with the recording of the ECG signals. This approach can be carried out in an enclosed housing, operated adjacent the subject's thorax. The processed digitized ECG signals and the audio signals are then communicated for receipt by an external device.
摘要翻译: 本公开的方面旨在处理来自对象的ECG信号。 如可以根据一个或多个实施例来实现的,各个电极感测来自对象的ECG信号,并且ECG信号被数字化和处理,诸如去除噪声,检测QRS复合物,评估质量,检测心律失常和/或 存储信号。 响应于来自主体的输入,一个或多个ECG信号与来自用户的声音一起被记录,例如同时记录用户的语音以便描述与ECG信号的记录有关的条件。 这种方法可以在封闭的房屋中进行,在受试者的胸部附近操作。 处理的数字化ECG信号和音频信号然后被传送以供外部设备接收。
-
公开(公告)号:US09438204B2
公开(公告)日:2016-09-06
申请号:US14010562
申请日:2013-08-27
申请人: MEDIATEK INC.
发明人: Chien-Hua Hsu
IPC分类号: H03B1/00 , H03K5/00 , H04B1/10 , H03H21/00 , A61B5/00 , A61B5/04 , A61B5/0428 , A61B5/0452 , G06K9/00
CPC分类号: H03H21/0025 , A61B5/04017 , A61B5/0428 , A61B5/0452 , A61B5/7203 , A61B5/725 , G06K9/0051 , G06K9/0055 , H03H21/0021 , H03H2021/0081 , H03H2021/0094 , H03H2210/033
摘要: Embodiments of the invention disclose a signal processing device and a signal processing method and a device and a method for signal processing. The signal processing device includes a sampling module, a first segmentation module, a second segmentation module, and a detection module. The sampling module samples an input signal to generate a sample signal. The first segmentation module calculates a first segment value according to the sample signal during a first time interval. The second segmentation module calculates a second segment value according to the sample signal during a second time interval different in length from the first time interval. The detection module generates a detection signal according to the determination of whether the first segment value lies out of a first range, and whether the second segment value lies out of a second range.
摘要翻译: 本发明的实施例公开了信号处理装置和信号处理方法以及用于信号处理的装置和方法。 信号处理装置包括采样模块,第一分段模块,第二分割模块和检测模块。 采样模块对输入信号进行采样以产生采样信号。 第一分割模块在第一时间间隔期间根据采样信号计算第一分段值。 第二分段模块在与第一时间间隔不同的第二时间间隔内根据采样信号计算第二分段值。 所述检测模块根据所述第一段值是否在第一范围内的判定以及所述第二段值是否处于第二范围之外,生成检测信号。
-
公开(公告)号:US09398999B2
公开(公告)日:2016-07-26
申请号:US14629868
申请日:2015-02-24
IPC分类号: A61H31/00 , A61B5/04 , A61B5/0402 , A61B5/113 , A61B5/00 , A61M16/10 , G06K9/00 , A61B5/0205 , A61B5/11 , A61M16/00 , A61N1/39
CPC分类号: A61H31/005 , A61B5/0205 , A61B5/04017 , A61B5/0402 , A61B5/11 , A61B5/1135 , A61B5/4836 , A61B5/6825 , A61B5/6831 , A61B5/721 , A61B5/725 , A61B5/7257 , A61B5/742 , A61B2562/0219 , A61H31/00 , A61H31/006 , A61H31/007 , A61H2201/5007 , A61H2201/5048 , A61H2201/5058 , A61H2201/5064 , A61H2201/5071 , A61H2201/5084 , A61H2230/04 , A61H2230/06 , A61H2230/855 , A61M16/00 , A61M16/10 , A61M2230/06 , A61N1/39 , G06F19/00 , G06K9/0051 , G09B23/288
摘要: A chest compressions monitor for measuring the depth of chest compressions achieved during CPR. A displacement detector produces a displacement indicative signal indicative of the displacement of the CPR recipient's chest toward the recipient's spine. A signaling mechanism provides chest compression indication signals directing a chest compression force being applied to the chest and a frequency of such compressions.
-
公开(公告)号:US09292909B2
公开(公告)日:2016-03-22
申请号:US14138055
申请日:2013-12-21
申请人: FLIR Systems, Inc.
发明人: Nicholas Högasten , Theodore R. Hoelter , Katrin Strandemar , Malin Ingerhed , Mark Nussmeier , Eric A. Kurth , Pierre Boulanger , Barbara Sharp
IPC分类号: G06K9/40 , G06T5/00 , H04N5/33 , G06T5/20 , G06K9/00 , H04N5/232 , G06K9/60 , H04N5/225 , B60R11/04
CPC分类号: G06T5/003 , B60R11/04 , G06K9/0051 , G06K9/00791 , G06K9/605 , G06T5/002 , G06T5/20 , G06T2207/10016 , G06T2207/10048 , G06T2207/20201 , G06T2207/30252 , H04N5/2253 , H04N5/2257 , H04N5/23258 , H04N5/23267 , H04N5/33 , H04N5/332 , H04N5/3651
摘要: Techniques are disclosed for systems and methods using small form factor infrared imaging devices to image scenes in proximity to a vehicle. An imaging system may include one or more infrared imaging devices, a processor, a memory, a display, a communication module, and modules to interface with a user, sensors, and/or a vehicle. Infrared imaging devices may be positioned in proximity to, mounted on, installed in, or otherwise fixed relative to a vehicle. Infrared imaging devices may be configured to capture infrared images of scenes in proximity to a vehicle. Various infrared image analytics and processing may be performed on captured infrared images to correct and/or calibrate the infrared images. Monitoring information, notifications, and/or control signals may be generated based on the corrected infrared images and then presented to a user and/or a monitoring and notification system, and/or used to control aspects of the vehicle.
摘要翻译: 公开了使用小尺寸红外成像装置对车辆附近的场景进行成像的系统和方法的技术。 成像系统可以包括一个或多个红外成像设备,处理器,存储器,显示器,通信模块以及与用户,传感器和/或车辆接口的模块。 红外成像装置可以被定位在车辆附近,安装在,安装在或相对于车辆固定的位置。 红外成像装置可以被配置为捕获靠近车辆的场景的红外图像。 可以对所捕获的红外图像执行各种红外图像分析和处理,以校正和/或校准红外图像。 监视信息,通知和/或控制信号可以基于经校正的红外图像生成,然后呈现给用户和/或监视和通知系统,和/或用于控制车辆的各个方面。
-
-
-
-
-
-
-
-
-