1664results about How to "Avoid inaccuracies" patented technology

A transceiver (10) includes an RF transmitter (12) and an RF receiver (14) coupled together through a duplexer (30) or non-filtering multiport device (30′). Either device may leak significant portions (56, 58) of the transmit signal (20) into the receive signal (44), and may significantly distort the transmit signal (20). Distortion is compensated in the transmitter (12) through the use of a linear predistorter (68) that is adjusted in response to an RF feedback signal obtained from the antenna-side of the device. Transmit signal leakage is compensated in the receiver (14) by producing an RF cancellation signal (106) that, when combined with the receive signal (44) at RF at least partially cancels the transmit signal portions (56, 58) leaked into the receive signal (44). Residual leakage signal and intermodulation products thereof may be cancelled digitally.

The invention discloses an ReLU convolutional neutral network-based image denoising method. The method comprises the following steps of building an ReLU convolutional neutral network model, wherein the ReLU convolutional neutral network model comprises a plurality of convolutional layers and active layers after the convolutional layers, wherein the active layers are ReLU functions; selecting a training set, and setting training parameters of the ReLU convolutional neutral network model; training the ReLU convolutional neutral network model by taking a minimized loss function as a target according to the ReLU convolutional neutral network model and the training parameters of the ReLU convolutional neutral network model to form an image denoising neural network model; and inputting a to-be-processed image to the image denoising neural network model, and outputting a denoised image. According to the ReLU convolutional neutral network-based image denoising method disclosed by the invention, the learning ability of the neural network is greatly enhanced, accurate mapping from noisy images to clean images is established, and real-time denoising can be realized.

The disclosure is directed towards the assessments used to facilitate management of members in a given workforce. More particularly, elements comprising a system, method, and apparatus are operative to quantify individual worker engagement with resources made available in a digitally implemented office network. Such elements may be further operative to compare each individual worker's engagement with such resources against his own historical engagement, each other worker's historical and present engagement, as well as any predetermined engagement thresholds in order to gauge worker performance while avoiding traditional limitations imposed by remote-work arrangements.

A method for time synchronization that avoids time synchronization inaccuracies caused by asymmetric delays internal to a device. Time synchronization according to the present teachings includes determining an asymmetry between an internal delay of an inbound timing packet in a device and an internal delay of an outbound timing packet in the device and correcting a time synchronization in response to the asymmetry.

The invention discloses a polyphone model training method for speech synthesis, and a speech synthesis method and device. The method comprises the following steps of processing a voice data set and a text set so as to generate a training corpus set, wherein the text set corresponds to the voice data set, and the training corpus set comprises texts and Pinyin sequences corresponding to the texts; extracting feature information of the texts; and training polyphone models according to the feature information and the Pinyin sequence. According to the polyphone model training method for speech synthesis, in a polyphone model training process, manual labeling on Pinyin of the texts is not required, a training period of the polyphone models is greatly shortened, meanwhile, the circumstance that the trained polyphone models are inaccurate due to wrong manual labeling is avoided, and accuracy of the trained polyphone models is improved.

The application discloses a method and a system for updating statistical data. The method comprises the following steps: a service system receives a service request from a user, conducts data processing for the current service and saves the service data processed into a service database and the message queue in a calculation server; the calculation server acquires the current service data processed from the message queue, conducts increment statistical computing to acquire increment statistical data of the current service processed and updates the statistical data into the statistical table database; a task server periodically traverses the service data in the service database, executes the statistical computation based on the service database to acquire the statistical data for the whole service database and applies the statistical data for the whole service database to overwrite all of the data in the statistical table database and update the statistical table database.

The invention relates to a load analysis method for a four-wheel six-component road spectrum of a finished automobile. The method comprises the following steps: 1 testing a four-wheel six-component road spectrum of a finished automobile; 2 collecting the design data of the finished automobile; 3 building and adjusting a dynamic model of the finished automobile; 4 carrying out virtual iteration; 5 obtaining Z-direction displacement of a wheel center; 6 decomposing the load of the finished automobile; 7 processing the load result. The six components of four wheel centers of the finished automobile are directly measured as external road excitation input; the actual stress condition of each wheel center under the typical characteristic road excitation of test site reliability is reflected; the technical effect of obtaining the load of the finished automobile system and spare parts is realized by a virtual iteration technique and a dynamics simulation technique; the anti-fatigue performance of the finished automobile structure can be accurately predicted and optimized at the earlier stage; problem rectification of the sample automobile stage is reduced; the research and development cycle is shortened; the research and development cost is also reduced.

The invention relates to an on-line large power grid recovery assistant decision-making system. The system comprises two work modes, namely a real-time assistant mode and a research and analysis mode, a fracture surface operation analysis module, a calculation configuration module, a calculation function module and a calculation management module, wherein the fracture surface operation analysis module is used for fracture surface reading and fracture surface adjustment, the calculation configuration module is used for setting calculation parameters and selecting a black-start power supply and a started power supply, the calculation function module is used for system partitioning, load flow calculation, electromagnetic transient calculation, electromechanical transient calculation and path search and assessment, and the calculation management module is used for displaying system partitioning schemes, voltage waveforms, ranking of available paths and a black-start generating scheme. When the on-line large power grid recovery assistant decision-making system is used for developing an on-line electric power system recovery assistant decision-making supporting system, existing data resources in a dispatching center can be sufficiently utilized, the advantages of the method of pre-arranged scheme recovery and intelligent dispatching are combined, a model does not need to be rebuilt, the construction time and cost of the system are reduced, the real-time performance and the accuracy of suggestions to decision-making are improved, and a powerful support is provided for rapid and reliable recovery of a power grid after power outages.

An extracorporeal blood chamber for an optical blood monitoring system includes an opaque chamber body in order to prevent inaccuracies when measuring oxygen saturation levels due to light ducting, which can occur at low oxygen saturation levels and low hematocrit levels. In one embodiment, the blood chamber need not include a moat as is present in conventional blood chambers.

Search time is reduced with a search engine that includes a bi-directional inverted index facility which can be accessed with a keyword search in one of a number of languages and provide a listing of documents contained in all of those languages. The keywords in all supported languages are preferably stored in an inverted index lookup table cross referenced to documents in those language containing the keywords. Keywords with the same meaning in different languages are accessible together when that keyword in one of the languages is queried. The search engine containing the table can identify pertinent documents either in a selected language, a second language or in all supported languages, as determined by the user. Information about each document can include not only the identity of the document but also information used in ranking the documents such as the number of times that a keyword appears in that document, and the keywords proximity to other keywords. The use of the inverted index table therefore reduces search time by eliminating the need for translation of keywords, their identification in documents and accumulating of ranking information at search runtime and avoids inaccuracies which may result from full text translations of documents.

Embodiments of the invention relate to a facial image series-based attribute identification method and device. The method comprises the following steps of: extracting facial attribute features in a facial image series through a convolutional neural network and a recurrent neural network; selecting a target facial attribute feature in the facial attribute features through a machine learning model; and when a plurality of primary attribute identification results obtained through identifying the target facial attribute feature by a plurality of preset classifiers are obtained, fusing the plurality of primary identification results to generate an advanced identification result of the facial image attribute identification. According to the method and device, the features between different facial images in the facial image series are correlated, and the complementary relationship between different images in the facial image series is fully mined, so that the problem of low correctness caused by single-facial image attribute identification in the prior art can be well avoided.

A method and apparatus for avoiding pressing inaccuracies on a touch panel. The touch panel includes a panel, a control module having an analog-to-digital converter (ADC) and a delay buffer, and a processor, the delay buffer able to buffer a trigger signal for interrupting the processor. The method includes sequentially receiving a plurality of analog signals generated by the panel when the panel is touched with the control module, transforming the plurality of analog signals into a plurality of corresponding digital signals and transmitting the plurality of digital signals to the processor with the analog-to-digital converter, neglecting a first m digital signals of the plurality of digital signals, and neglecting a last n digital signals of the plurality of digital signals, where m and n are integers with values equal to or larger than 1.

The invention discloses an on-line estimation early warning method for an SOH (State Of Health) of an electric vehicle battery pack. The method comprises the following steps: S1, measuring initial maximum capacity of each single battery; S2, acquiring real-time charge and discharge current of each single battery by using a battery management system and estimating an SOC value before and after charge and discharge of each single battery; S3, calculating the SOH value after each charge and discharge of each single battery; S4, judging whether the single battery of which the SOH value is smaller than a first preset threshold value exists in the battery pack, executing a step S6 if the single battery exists, and otherwise, executing a step S5; S5, calculating an average value of the SOH values of all single batteries in the battery pack, judging whether the single battery of which the SOH value differs from the average value for more than a second preset threshold value exists in the battery pack, executing a step S6 if the single battery exists, and otherwise, returning to execute the step S1; S6, triggering the battery management system to enter a protection state and emitting alarm. The method has the advantages of high real-time property, high accuracy, simple implementation mode and low cost and can be widely applied to the monitoring field of the battery pack.

A metrology/inspection system moves the imaging and/or measuring equipment of the system relative to a wafer. Accordingly, measurement or inspection of the wafer does not require that the wafer be mounted on a precision stage. This allows the wafer to be at rest on any structure native in a processing apparatus when the system measures or inspects the wafer. Accordingly, measurement does not require removing the wafer from the processing apparatus and does not delay processing since the wafer can be measured, for example, during a required cool down period of device fabrication process. Alignment of an optical system includes pre-alignment base on edge detection using the optical system and more precise alignment using image recognition. An R-θ stage can position the optical system at inspection areas on the wafer. Image rotation can provide a fixed orientation for all images at the various inspection areas and can maintain the fixed orientation when moving from one inspection area to the next.

A method and an apparatus are provided for post-layout optimization of an integrated circuit. In one instance, only local transformations accomplished by incremental changes to placement and routing are provided, so as to avoid the costly design iteration loop that requires re-synthesis, re-place and re-route. Optimization can be provided in multiple optimization phases each accomplishing a specified set of transformations. Static timing analysis is performed at the end of each set of local transformations to determine if further optimization steps are required. In one instance, the physical design is first scanned for mismatch between drivers and loads. Then, in a second optimization phase, “hot spots” in the physical design are identified for local transformation using a “bidirectional combinational total negative slack” (BCTNS) algorithm. In subsequent phases, optimization based on meeting setup times and hold times in a critical path are performed.

An operation control method of a synchronization variable speed constant frequency wind generating set includes the strategies of AYC, synchronization control, low wind speed ceiling capacity follow control and high wind speed constant power control which includes the AYC, the synchronization control, the low wind speed ceiling capacity follow control and the high wind speed constant power control. An anemoscope and a wind vane measure the wind speed and wind direction information to generate a yaw deviation signal; a yaw motor is driven to execute left and right yaw according to the yaw direction and angle and limit the yaw velocity and meanwhile, the steering turns of the engine room are measured with the help of cables winding around a sensor, so as to execute safe untwist action, thus ensuring that the wind generating set realizes quick, accurate and safe effect of the wind wheel against wind. All the control strategies cooperate in work from the start to synchronization power generation according to the requirement of the set and external conditions, so as to ensure the safe and efficient operation of the wind generating set.

A method for controlling an operation of a vapor compression system (VCS) determines values of measured outputs of the operation of the VCS that include performance and constrained outputs. The method determines states of the VCS using an estimator model of the VCS defining a relationship between the states of the VCS, control inputs and controlled outputs, such that a difference between outputs predicted using the estimator model and the measured outputs asymptotically approaches zero. The states of the VCS include a main state representing the operation of the VCS and an auxiliary state representing the effect of unknown disturbances on each measured output of the VCS. The control inputs for controlling the operation of the VCS are determined using a prediction model, such that the constrained output satisfies the constraints, and a difference between the performance output and the value of the setpoint asymptotically approaches zero.

A method for detecting breast cancer can include the step of positioning a transmitting antenna and a receiving antenna about a breast so that the transmitting antenna is positioned to transmit microwave energy into the breast and the receiving antenna is positioned to receive the transmitted energy after the energy has passed into the breast. Microwave energy can be transmitted from the transmitting antenna. The microwave energy can be received by the receiving antenna and used to determine a presence of tumors within the breast. A relative position of the transmitting antenna and the receiving antenna can be adjusted about the breast. After each position adjustment, the transmitting and the receiving steps can be repeated.

A system and method for controlling call admission in a wireless communications network recursively estimates reverse link load levels as a function of changes in base station receive power and/or the number of mobiles served in the cell/sector. In one implementation, a call admission controller utilizes multiple load level estimating methods, whereby a first method estimates load as a function of changes in base station receive power and changes in the number of served mobiles. The call admission controller uses a second method to recursively update load level estimates as proportionally changing with number of users. The call admission controller may utilize a third estimate method as an accuracy check based on a changes in base station receive power measurements. The call admission controller outputs a call blocking command when load level estimates exceed a threshold to avoid cell/sector overload conditions and network instability.

The invention provides a method for detecting soldering points of electronic product complete machines and performing tin soldering on the electronic product complete machines. The method is implemented by the aid of an automatic tin soldering machine with a CPU (central processing unit). The automatic tin soldering machine structurally comprises a workbench, an X-axis, a Y-axis, a Z-axis and a four-axis numerical control system, wherein the four-axis numerical control system is formed around a rotating shaft of the Z-axis. A soldering head is fixed on the Z-axis which is fixedly provided with an industrial camera. The method includes that standard image information of various kinds of qualified electronic product complete machines is stored in a data storage unit of the CPU, the industrial camera acquires an image of a to-be-detected electronic product complete machine, the image is stored in the data storage unit and is transmitted to a data processor, the data processor compares to-be-detected image information with the standard information in the data storage unit, if the image information is identical to the standard information, tin soldering points of the to-be-detected complete machine are qualified, or otherwise, the tin soldering points are unqualified, the data storage unit displays differences of the to-be-detected image on a display screen in a partial enlarged manner, repair soldering is performed, and if the soldering points are still unqualified, manual detection is performed.

The invention discloses a pertinent bill content identification system and method based on table positioning and depth learning. The invention is directed to a particular bill and automatically locates the cell position of the table in the ticket, a text block is extracted, and character segmentation of the text block is required in comparison with a general ocr recognition method. The invention eliminates the modification step, and can directly recognize indefinite-length multi-character text block by constructing a sample pertinently and training a depth learning model, and can effectively deal with background texture and text blur.

The invention discloses a portable fluorescence quantitative PCR detector, and belongs to the fields of biology and medical detection. The detector comprises an optical excitation unit, an optical detection unit, and a temperature control unit. The optical excitation unit comprises a plurality of light sources, an exciting light optical filter, and a holophote. The temperature control unit comprises a reaction tank, a heat cover, a heating/refrigerating sheet, a heat dissipating part, and a temperature control circuit. The optical detection unit comprises a plurality of light splitting sheets, an optical filter, an imaging sensor, and a detection circuit. The light given off by the light source goes through the exciting light optical filter and then is converted into exciting light with a specific wavelength; then the exciting light is reflected by the holophote and is projected into a PCR tube to excite the fluorescence substances in the reagent, thus the fluorescence substances will give off light, part of the light goes though the hole in the side surface of the reaction tank and enters the optical detection unit, the light is split and filtered in the optical detection unit, and finally the light form an image on the corresponding image sensor.