232 results about "Accuracy improvement" patented technology

Improved cassette-based medical fluid therapy systems, apparatuses and methods are provided. In one aspect, various peristaltic pump tubing materials are provided. Cassette and membrane materials, configurations and manufacturing improvements are provided. Various aspects of the invention include: a head height sensing and compensating method and apparatus; an admixing method and apparatus; a pH measuring method and apparatus; an air detecting and removal methods and apparatuses; an active priming apparatus and method; and a volumetric accuracy improvement apparatus and method.

A direct digital synthesizer (DDS) drives a receive sampling gate at a frequency that is offset from a transmit pulse frequency to produce an expanded time sampled echo signal. The frequency offset generates a smoothly slipping phase between realtime received echoes and the sampling gate that stroboscopically expands the apparent time of the sampled echoes with an exemplary factor of 1-million and a range accuracy of 1-centimeter. The flexibility and repeatability of the digitally synthesized timing system is a quantum leap over analog prior art. The rock solid stability of the DDS allows further accuracy improvement via an error correction table.

The present invention is related to position calculation and navigation systems, and more particularly, to a method and apparatus for making accuracy improvements to a GPS receiver's navigation solutions. According to a first aspect, selective cartography information from a map database accessed by a GPS receiver from a location-aiding device is integrated into the position calculations performed by the GPS receiver. According to another aspect, selective cartography information includes indices representing associated attributes of geographical objects, such as, a width-indicator index for a road segment.

In a radio-communication system, when wanting to perform high-accuracy scheduling in a base station, a large number of information items, such as channel quality information, desired precoding matrix information and so on, are fed back from a mobile station with a high degree of accuracy. The base station monitors the feedback information and, upon acquisition of it, transmits to the mobile station a certain kind of feedback information which is judged to contribute to scheduling accuracy improvement while preventing the other kinds of feedback information from being sent to the mobile station.

The invention discloses a road surface traffic sign recognition method based on a convolution neural network, and the method comprises the following steps: image collection and preprocessing; and convolution neural network structure design and training. The method employs a V-parallax method to obtain a road surface area from an original image, can reduce the impact caused by non-road-surface interference, and improves the extraction precision of a road surface area. The invention employs a vertical view to reconstruct the road surface area, enables unparallel lines, presented in a visual image because of a view angle, to be reconstructed into approximately parallel lines, facilitates the recognition of a road surface traffic sign, and improves the capability of adapting to view angle inclination. The method the deep learning method (convolution neural network), can extract recessive characters reflecting the data essence from a large number of training samples. Compared with a shallow-layer learning classifier, the method is higher in learning efficiency and recognition precision.

The invention discloses a vehicle target detection method and aims to solve the technical problems that a traditional method is complicated in detection process, large in calculated quantity and inaccurate in detection. According to the method, a VOC 2007 vehicle dataset is made according to an ImageNet dataset; a training model is configured by the adoption of a caffe deep learning framework andan improved Faster R-CNN algorithm, and an Inception network structure is introduced to perform feature extraction on an image; a sliding window with the area of 642 is additionally arranged and usedfor detecting a small target; a target detection problem in a scene is converted into a target dichotomy problem, and a vehicle target is detected and recognized; an RPN loss function is used to perform optimization; and a SoftMax algorithm is utilized to classify vehicle image features, and therefore a final result is obtained. The method has the advantages that the calculated quantity of the image features is reduced, the detection process is simplified, the image feature extraction capability is enhanced, network recognition precision is improved, and the detection result is more comprehensive.

A method and system for dose correction of a particle beam writer is disclosed. The method and system includes reading a file of writing objects that includes dose intensity, calculating a rate of dose intensity change between adjacent writing objects, selecting a writing object that may need accuracy improvement of dose correction based on the rate of dose intensity change, and improving accuracy of the dose correction of the writing object that is selected and its adjacent objects.

The present disclosure relates generally to techniques for enhancing recurrent neural networks (RNNs) implemented on an integrated circuit. In particular, approximations of activation functions used in an RNN, such as sigmoid and hyperbolic tangent, may be implemented in an integrated circuit, which may result in increased efficiencies, reduced latency, increased accuracy, and reduced resource consumption involved with implementing machine learning.

A novel Parametric Systematic Error Correction (ParSEC) system is disclosed which provides improved system accuracy for image navigation and registration (INR). This system may be employed in any suitable imaging system and, more specifically, to all imaging systems that exhibit systematic distortion. The ParSEC system may be employed to any such system regardless of sensing type (remote or in situ) or imaging media (photon or charged particle) and is further applicable to corrected imaging of any celestial body currently detectable to remove distortion and systematic error from the imaging system employed. The ParSEC system of the instant invention comprises a software algorithm that generates at least about 12 correction coefficients for each of the INR system measurements such as stars, visible landmarks, infrared (IR) landmarks and earth edges. An iterative estimation algorithm such as, for example, least squares or Kalman filters may be employed to determine the at least about 12 correction coefficients from each set of measurement residuals. The improved image products provide more accurate weather forecasting such as wind velocity and temperature.

A system and method for sensing and compensating for unintended joint movement of a robotic arm caused by application of a load. The system may have a plurality of external encoders each in intimate contact with an external edge portion of one of a plurality of robotic arm joints to sense joint movement caused by application of the load, and a computing device configured for calculating a compensation amount based on the sensed joint movement and sending the calculated compensation amount to a corresponding robot motor encoder to correct the position of the joint by the compensation amount. The method may comprise applying the load one portion at a time, such that a portion of the load is applied, the compensation amount is calculated, the position of the joint is corrected, and then the process repeats, with another portion of the load applied to the robotic arm.

The present invention is related to location positioning systems, and more particularly, to a method and apparatus for making accuracy improvements to a GPS receiver's navigation solutions. According to a first aspect, cartography information from a map database embedded within the GPS receiver is integrated into the position calculations performed by the GPS receiver. According to another aspect, the map database embedded within the GPS receiver is optimized for the purpose of improving the accuracy of the GPS receiver's position calculations.

The invention provides a video image classification method based on a time-space co-occurrence double-flow network. The main content of the video image classification method comprises data input, a time-space double-flow network, fusion and an SVM classifier. The process comprises the steps of firstly inputting image and light stream information, performing early fusion by being combined with a time network and a space network, enabling fusion output to act as a feature vector, inputting the feature vector into the SVM classifier, and acquiring a final classification result. According to the invention, a method that the early fusion double-flow network is combined with time information and space information (time-space co-occurrence) is adopted, a video data set of the monkey is utilized, and more frames, namely, more space data, are utilized from each piece of video so as to generate significant improvement in precision; the space information and the time information are combined and form mutual complementation, and the precision reaches 65.8%. A smaller number of separated clusters are formed through using the co-occurrence method, the separated clusters generally stay together more closely, and the time information is better utilized.

A container inspection system with CT tomographic scanning function, which is related to a large container inspection system for Customs, is disclosed, comprising: a radiation source; an annular rotatable rack with the radiation source provided at the outside thereof and rotated with it; an annular rack body for supporting the annular rotatable rack in a vertical plane; a driving device for rotating the annular rotatable rack; a detector array provided in the inner side of said annular rotatable rack and opposed to a side where the radiation source is provided; a transmission device passing through the annular rotatable rack and annular rack body, and transmitting a container truck to be detected in linear movement; a remote control device for controlling the operations of the radiation source, the driving device and the transmission device, and receiving / displaying the image signal obtained by the detector array. Not only can general inspection of the objects to be detected be undertaken, but also does repeated tomographic inspection of suspicious locations within the inspected objects, thus achieves effective accuracy improvement of the contrabands detection.

The invention relates to a single-phase ground line selection method with the function of accuracy improvement for a small-current grounding system. The method comprises: (1), whether a fault occurs is monitored; if so, a fault line section algorithm is started; (2), Fourier transform is carried out on sampling data of a zero-sequence voltage and a zero-sequence current in a fifth cycle after fault occurrence and amplitudes and phase angles of the zero-sequence voltage and zero-sequence current fundamental waves of all paths of outgoing lines are obtained; and if a neutral point passes through an arc suppression ground system, amplitudes and phase angles of harmonic waves are calculated for five times; (3), amplitude-comparing-method fault measurement is constructed; (4), a phase-comparing-method fault measurement is constructed; (5), initial-half-wave-method fault measurement is constructed; (6), wavelet-analytical-method fault measurement is constructed; and (7), a serial number of a line with the largest fault measurement is found out under each criterion; if the serial numbers are identical, the line is a fault line and a line selection result is outputted, and the line selection is completed; and if the serial numbers are different, a multi-criterion fusion method based on a D-S evidence theory is started. With the method, accuracy of the fault line selection algorithm is improved effectively.

The invention discloses a satellite-based augmentation system positioning accuracy improvement method based on regional correction. The satellite-based augmentation system positioning accuracy improvement method comprises the steps of: dividing a region to be observed into a plurality of observation regions, and setting a plurality of observation stations in each observation region; acquiring residual data of each observation station through processing observation data; acquiring clock difference mean value data of each observation station according to the observation residual data; acquiring ambiguity reduction value data of each observation region according to the residual data; acquiring regional correction value data of each observation region according to the residual data and the ambiguity reduction value data; and providing a user with a regional correction value data calling service through a network or a satellite link. According to the satellite-based augmentation system positioning accuracy improvement method based on regional correction, the wide-range regional design for the observation stations is carried out, the regional correction values used for network and satellite-based broadcast are calculated on the basis of the observation stations, the error correction in user positioning is refined, and the user positioning accuracy can be improved; and the satellite-based augmentation system positioning accuracy improvement method has the advantages of high accuracy and wide application range.

A method of and an apparatus for determining tire shapes are provided which enable both an apparatus weight reduction and a determination accuracy improvement. The apparatus of the present invention includes: a portable base member; a non-contact displacement gage held on the base member so as to be freely movable along a predetermined scanning direction; a weight supported on the base member so as to be movable along the scanning direction; and means for symmetrically driving the displacement gage and weight while synchronizing them with each other. Using this apparatus, the method includes: setting the base member in proximity to a tire tread portion so that the displacement gage will scan the tread portion; allowing the displacement gage and weight to symmetrically move along the scanning direction while synchronizing them with each other; and determining the shape of the tread portion, based on an output from the displacement gage.

A system and method for sensing and compensating for unintended joint movement of a robotic arm caused by application of a load. The system may have a plurality of external encoders each in intimate contact with an external edge portion of one of a plurality of robotic arm joints to sense joint movement caused by application of the load, and a computing device configured for calculating a compensation amount based on the sensed joint movement and sending the calculated compensation amount to a corresponding robot motor encoder to correct the position of the joint by the compensation amount. The method may comprise applying the load one portion at a time, such that a portion of the load is applied, the compensation amount is calculated, the position of the joint is corrected, and then the process repeats, with another portion of the load applied to the robotic arm.

Channel access delays and reception uncertainty are modeled as protocol-independent generic processes that are optimized for improved simulation performance. The generic process components are designed such that each different protocol can be modeled using an arrangement of these components that is specific to the protocol. In this way, speed and / or accuracy improvements to the generic process components are reflected in each of such protocol models. If an accurate analytic model is not available for the generic process component, a prediction engine, such as a neural network, is preferably used. The prediction engine is trained using the existing detailed models of network devices. Once trained, the prediction engine is used to model the generic process, and the protocol model that includes the generic component is used in lieu of the detailed models, thereby saving substantial processing time.

The invention belongs to the technical field of laser remote sensing and particularly relates to a laser altimeter atmospheric delay range finding error correction method for a land object. The invention first deduces the theoretical relationship between the atmospheric delay error and the surface atmospheric pressure, and thus simplifies the high pressure equation. According to the altitude and atmospheric pressure of the weather station measured by the ground meteorological station, the atmospheric pressure attenuation factor of each weather station is calculated, and the attenuation model of the atmospheric pressure with the height change is established. According to the latitude and longitude of the laser foot, nearest k weather stations are The results of cross validation show that the accuracy of the atmospheric pressure space-time interpolation of the method is higher than that of the NECP-based reanalysis data calculation method, and the method provided by the invention is of great value in the measurement accuracy improvement of the laser altimeter satellite.

The invention discloses a terminal control operation method, a terminal and a computer readable storage medium. The method is applied to operation accuracy improvement of a terminal, and includes thesteps that an operation gesture issued by a user for the terminal is detected, and the operation gesture is matched with preset gestures; when at least two preset gestures exist and are successfully matched with the operation gesture, one preset gesture with the highest matching priority is selected as a target gesture of the operation gesture according to the matching priorities of all the presetgestures, and controlling over the terminal is achieved according to the target gesture. The invention also discloses the terminal and the computer readable storage medium; a procedure capable of achieving the functions is stored in the computer readable storage medium, and the technical scheme is applied through a processor; the matching accuracy of the operation gesture issued by the user and the preset gestures is increased, controlling over the terminal is achieved according to the target gesture, time is saved, and better user experiences are brought to the user.

The invention discloses a hot-spot video push method and apparatus. With the method and apparatus, hot-spot videos can be pushed according to watching favorites of users recently, thereby achieving objectives of push content accuracy improvement and push content hot-spot enhancement. The method comprises: according to a video subscription theme set by a client in advance, a video meeting the video subscription theme is searched regularly; the searched video serving as a pushed video is pushed to the client; click information, recorded by the client, of a plurality of pushed videos by the user is obtained; when all videos, clicked by the user by Nth times in succession, among the multiple pushed videos match a hot-spot event base successfully, a video meeting the video subscription theme is searched regularly in the hot-spot event base according to the video subscription theme set by the client in advance by beginning with the next-time video push of successful video push to the client by Nth times in succession. With the method, the accuracy of the pushed content and the pushed content hot-spot enhancement can be effectively improved.

The present invention discloses a primer system for concurrently detecting 10 gene polymorphism sites related to genetic deafness. Based on the product prepared by the primer system, 10 gene polymorphism sites related to genetic deafness can be concurrently detected. Use of the product, genotypes of 10 gene polymorphism sites of a subject are detected, and the detection results can be used for assisted clinical diagnosis, and can further be used for epidemiological investigation, and can also be used for prenatal screening, neonate screening and other fields. With the present invention, 10 gene polymorphism sites on different genes can be concurrently detected in a reaction system, such that advantages of low cost, convenient operation, accuracy improvement and sensitivity improvement are provided compared with sequencing, real-time fluorescence quantitative PCR and other technologies.

A novel Parametric Systematic Error Correction (ParSEC) system is disclosed which provides improved system accuracy for image navigation and registration (INR). This system may be employed in any suitable imaging system and, more specifically, to all imaging systems that exhibit systematic distortion. The ParSEC system may be employed to any such system regardless of sensing type (remote or in situ) or imaging media (photon or charged particle) and is further applicable to corrected imaging of any celestial body currently detectable to remove distortion and systematic error from the imaging system employed. The ParSEC system of the instant invention comprises a software algorithm that generates at least about 12 correction coefficients for each of the INR system measurements such as stars, visible landmarks, infrared (IR) landmarks and earth edges. An iterative estimation algorithm such as, for example, least squares or Kalman filters may be employed to determine the at least about 12 correction coefficients from each set of measurement residuals. The improved image products provide more accurate weather forecasting such as wind velocity and temperature.

The invention provides a method for improving the identification precision of kinetic parameters of a modularized robot joint, and the method comprises the steps: carrying out the kinetic modeling through a Newton-Euler method, and enabling an improved Fourier series motivation track to be applied to a device for improving the identification precision of the kinetic parameters of the modularized robot joint; collecting an angle value, an angular speed value and a moment value through a sensor unit; carrying out the differential operation through the angular speed measured by a tachometer, obtaining an angular acceleration value, and carrying out the preliminary processing of information; carrying out the parameter estimation, and substituting parameters into a kinetic equation for estimating the kinetic parameters. The method employs the multi-sensor fusion technology, achieves the direct real-time measurement of the angles of a joint end and a motor end through two absolute coders, and improves the safety of a device. The mode of direct measurement reduces the calculation burden of a control unit, thereby facilitating the implementation of the real-time control, solving problems of interference and error increase caused by the indirect measurement, and improves the parameter recognition precision.

A portable-device position determination system has a portable device that is configured to be possessed by a user, and an in-vehicle device that is mounted on a vehicle and conducts wireless communication with the portable device. The in-vehicle device transmits a request signal to the portable device. The portable device detects a received signal intensity of the request signal to send back the received signal intensity to the in-vehicle device. The in-vehicle device compares the received signal intensity of the request signal detected by the portable device to a threshold to determine a position of the portable device. The in-vehicle device includes a transmission controller that generates the request signal, supplies an electric power to a transmitting antenna, and transmits the request signal from the transmitting antenna, and a current detector that detects a current passed through the transmitting antenna when the request signal is transmitted.

The invention provides a face recognizing and aligning method based on a shared convolutional characteristic. The method comprises four steps of firstly performing multilayer convolution and pooling operation on an input image, and extracting a convolutional characteristic; performing face recognition according to the extracted convolutional characteristic, and outputting face position information in the image; then according to the face position information which is input in the previous step, extracting the convolutional characteristic which corresponds with the face position, performing face characteristic point positioning on the face, and outputting face characteristic point positions such as eyebrows, canthuses, nose and mouth corners in the face image; and performing rotating and scaling processing on the face according to a face image and corresponding face characteristic point points which are output in the face characteristic point detecting step, and outputting the aligned face image. The face recognizing and aligning method can realize automatic detection and automatic alignment for the face in the image and has advantages of high speed and high accuracy. Furthermore the method facilitates accuracy improvement of face recognition and face identification technology.

The invention provides a positioning accuracy improvement method for a mobile terminal. The method comprises the following steps: identifying NLOS (none line of sight) propagation of BS-MS (based station-mobile server) links by acquiring AOA / TOA (angle of arrival / time of arrival) measuring parameters of each link in a time interval; if all BS-MS links have no LOS (line of sight) transmission, acquiring a final position estimation of a target mobile terminal by using a Co-HTAP (hybrid positioning) algorithm; otherwise, judging the number of the mobile terminals having the LOS propagation links, and acquiring the final position estimation of the target mobile terminal by using a corresponding algorithm according to the number of the mobile terminals having the LOS propagation links.

The invention discloses a spatial distribution forecasting method for heavy metal of cadmium in flat terrain area soil. The method comprises: firstly, utilizing a radial basis function neural network model to create the non-linear mapping relationships of various influencing factors on the cadmium content in soil; based on the change of correlation of the various influencing factors in different scopes from an analyzing region, dividing the analyzing region into a region within a 10km scope and a region beyond a 10km scope; constructing neural network models that reveal the non-linear mapping relationships of the various influencing factors on the cadmium content in soil respectively; and using the HASM model to simulate the residuals of the forecasting result by the neural network models so as to obtain the forecasting result of the spatial distribution of cadmium in soil in the analyzing region. According to the invention, the forecasting accuracy of the forecasting result to the verification point increases by 5.56% to 17.65%. And at the same time, the method could well reflect the detailed information of the spatial change of cadmium content in soil in an analyzing region.

The invention provides an electromagnetic wave propagation characteristic-based indoor distance measurement positioning method. The electromagnetic wave propagation characteristic-based indoor distance measurement positioning method is realized based on a positioning system. The method includes the following steps that: a control point information table and an electromagnetic wave propagation model table are built at a positioning calculation server; a plurality of positioning labels are preset in a target area which requires positioning accuracy improvement so as to be adopted as control points, and the coordinates of the control points are measured, and real distances between the control points and each anchor point are calculated, and the coordinates of the control points and the real distances between the control points and the anchor points are written into the control point information table; the differences of the real distances between each control point and each anchor point, and observation distances are calculated, and are written into the electromagnetic wave propagation model table as distance correction values; and when indoor positioning is performed on user labels, control points adjacent to the positions of users are selected according to the electromagnetic wave propagation model table or the control point information table, and distance correction values of the positions of the users are calculated according to the distance correction values, and the distance values of the user labels and each anchor point are corrected, and the accurate coordinates of the users are re-calculated.