104results about How to "Eliminate individual differences" patented technology

A substrate-supporting device for CVD having a substrate-supporting region includes: a substrate-supporting surface which is a continuous surface defining a reference plane on which a substrate is placed; and multiple dimples having bottom surfaces lower than the reference plane. The respective dimples are isolated from each other by a portion of the substrate-supporting surface.

A substrate-supporting device for CVD having a substrate-supporting region includes: a substrate-supporting surface which is a continuous surface defining a reference plane on which a substrate is placed; and multiple dimples having bottom surfaces lower than the reference plane. The respective dimples are isolated from each other by a portion of the substrate-supporting surface.

The invention discloses a rehabilitation training evaluation method and system based on the interaction between exercise information and myoelectricity. Exoskeletons are arranged on front half parts of rehabilitation information gloves, bending sensors and pressure sensors are arranged on the portions, corresponding to joints of the upper limbs and hands, of the rehabilitation information gloves,multiple myoelectric electrodes are arranged on muscle parts corresponding to joint movement, and the bending sensors, the pressure sensors and the myoelectric electrodes are separately connected witha computer; the computer uses the position, the velocity, the acceleration and the pressure as input data for neural network training, myoelectric signal features are used as guide signals for performing neural network training, and the computer displays visual feedback refined to the color change of each exercise muscle through a display unit so that healthy people or patients can conduct activetraining in line with continuous movement intention according to the movement and multiple muscle color changes. Real-time synchronization of the intention and complex and fine movement is achieved in a real sense, and evaluation of the targeted quantification of the single muscle of the patients with different movement disorder conditions is achieved.

There is provided a method of driving an image sensor which can remove FPN resulting from inter-chip variations without requiring any dark correction. A semiconductor photosensor chip has a plurality of sensor modules mounted on a mounting substrate, and a semiconductor device in which at least an N signal input buffer circuit for receiving N signals, an S signal input buffer circuit for receiving S signals, a differential circuit for calculating any difference between the outputs from the N and S signal input buffer circuits, and a voltage clamping circuit for clamping the output from the differential circuit are formed on a single semiconductor substrate, and the voltage clamping circuit clamps the reset state of S and N signal common output line.

The invention belongs to the technical field of human body physiological parameter measurement, and relates to a reflective pulse blood oxygen detecting method based on diffusion theory, which includes the steps: (1) establishing the relationship between organizer absorption coefficient and SaO2 (oxygen saturation); (2) establishing the relationship between reflection light intensity and organizer absorption coefficient; (3) establishing the relationship between an R value and the reflection light intensity; (4) acquiring pulse wave signals of a red light wavelength and a near-infrared light wavelength; (5) detecting features of pulse wave signals of the infrared light wavelength to extract peak values which are the maximum value and the minimal value of the reflection light intensity; (6) calculating blood oxygen saturation coefficient R; and (7) plugging the R value into a calibration curve to obtain needed blood oxygen saturation coefficient in real time. The reflective pulse blood oxygen detecting method based on the diffusion theory can be used for realizing non-invasive real-time measurement, and is higher in precision and wider in application range.

The embodiment of the invention provides a lease item scheduling method and system, and belongs to the technical field of sharing. The method comprises the steps: determining the residual amount of lease items in a target region; obtaining application environment parameters of the target region for the lease items; determining the demand quantity of the target region for the lease items based on aprediction model and the application environment parameters; and determining the number of lease items to be scheduled to the target region according to the residual amount and the demand quantity. Therefore, the method can achieve the real-time prediction of the lack of vehicles to be scheduled to the target region based on the model, eliminates the personalized difference and error caused by the dependence on the manual estimation and prediction, achieves the intelligent scheduling of lease items, and guarantees the maximization of the utilization rate of lease items.

To provide, via a simple method, a piezoelectric vibrator wherein nonuniformities in the adjusted amounts of the characteristic frequencies are small, and the characteristic frequencies can be adjusted with little decrease in the vibration characteristics, whereby individual differences pertaining to the characteristic frequency can be eliminated. In a vibrator 70, adjustment electrodes 72 are formed in advance on part of an electrode 60, and drive electrode 61 and adjustment electrodes 72 that have initially been made conductive relative to each other are electrically cut off and insulated from each other by cutting conductive parts 73, or the mutually insulated drive electrode 61 and adjustment electrodes 72 are made conductive (shorted) using solder, a wire, or another electrically conductive member, whereby the characteristic frequencies are adjusted simply and reliably. Individual differences in the vibrators 70 related to the resonance frequency (characteristic frequency) can thereby be eliminated, and reliability can be greatly improved.

First, normal calibration is executed by introducing a standards sample to an MS portion, thereafter, an adjusting sample is introduced to a GC portion, and a mass spectrum is acquired by executing mass spectrometry by the MS portion with regard to an object substance separated by a column. A data processing portion compares a measured mass spectrum and a standard spectrum and modifies a standard spectrum constituting a target in executing calibration by using a standard sample from the result. The standard spectrum modified in this way is reflected with an error factor of a dispersion in a purity of the standard sample or the like and therefore, when recalibration of the MS portion using the standard sample is executed by constituting a target by the modified standard spectrum, a difference among apparatus is resolved.

A pressure control system is used for eliminating individual differences of a plurality of pressure controllers used for controlling pressures of a plurality of pressure-controlled sections. The pressure control system includes a plurality of pressure controllers for supplying a pressurized fluid to a plurality of pressure-controlled sections, a master pressure controller for supplying a pressurized fluid having a reference pressure, a plurality of calibration chambers corresponding to the pressure controllers. The pressure control system further includes differential-pressure detecting devices provided in the calibration chambers to detect a differential pressure between the pressurized fluid supplied from the master pressure controller and the pressurized fluid supplied from the pressure controller, and an arithmetic device configured to receive a signal from the differential-pressure detecting device and adjust an output of the pressure controller so that the above differential pressure becomes zero or approximately zero.

The invention provides an identification method for blood serum proteome of broiler chickens. The identification method comprises the following steps: (1) collecting a blood serum sample; (2) carrying out SDS-PAGE (Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis); (3) carrying out in-gel enzymolysis on protein gel; (4) analyzing protein by nano-HPLC-MS / MS (nano-High Performance Liquid Chromatography-Mass Spectrometry / Mass Spectrometry); searching a protein data base through Protein Pilot software to identify obtained data. By adopting the method provided by the invention, the quantity and types of blood serum protein expression of the broiler chickens can be rapidly found at high throughput, and a method and a concept are provided for development of research on nutrients of the broiler chickens and immune functions of the broiler chickens.

A defrosting method for a heat pump comprises following steps: 1, arranging more than one temperature sensor for detecting the temperature of an evaporator on the evaporator of the heat pump; 2, defrosting the heat pump for the first time by a timing defrosting method during running, and carrying out defrosting after the detected temperature of the evaporator keeps lower than 0 DEG C for 40 minutes; and 3, determining a temperature maximum Tmax of the evaporator in a defrosting period, beginning to time after the first defrosting is finished, recording the temperature maximum Tmax of the evaporator after a first time interval T1, and recording the temperature maximum Tmax by a master control chip of the heat pump. On the basis of the traditional intelligent defrosting, and the step 3 is added to automatically judge a reference value which is the temperature maximum Tmax. The defrosting method for the heat pump has the advantages that operation is flexible, defrosting is effective, frostless defrosting can be avoided to the greatest extent, and an application range is wide.

The invention provides a difference depth feature-based facial expression recognition method and system and belongs to the pattern recognition technical field. The method includes the following stepsthat: face depth features are extracted from a sample face region frame sequence, and neutral expression frames and peak expression frames are extracted from the sample face region frame sequence according to the face depth features; and the face depth features of the neutral expression frames are subtracted from the face depth features of the peak expression frames of the sample image frame sequence, so that difference depth features can be obtained; and the difference depth features of the sample face region frame sequence and corresponding facial expressions are adopted as training input, so that a facial expression classifier can be obtained through training, and the facial expression classifier is adopted to achieve facial expression classification. According to the difference depth feature-based facial expression recognition method and system of the invention, the concept of difference is introduced into the depth features, and the difference depth features are used to representfacial expressions, and therefore, facial expression information can be preserved as much as possible, and at the same time, individual differences and environmental noises can be eliminated; and thedistinguishing capability of used features for the facial expressions is strong, so that the robustness of facial expression recognition can be improved.

The embodiment of the invention discloses a control method and device of a proportional solenoid and engineering machinery using the device. The method comprises the step of judging the current mode, and if the current mode is a self-learning mode, outputting the given working current to the proportional solenoid, and acquiring the corresponding relation according to the given working current; and if the current mode is a working mode, according to the given signal and the corresponding relation, outputting the pulse width to the proportional solenoid to modulate the PWM value. The invention improves the accuracy and the reliability of the system action.

First, normal calibration is executed by introducing a standards sample to an MS portion, thereafter, an adjusting sample is introduced to a GC portion, and a mass spectrum is acquired by executing mass spectrometry by the MS portion with regard to an object substance separated by a column. A data processing portion compares a measured mass spectrum and a standard spectrum and modifies a standard spectrum constituting a target in executing calibration by using a standard sample from the result. The standard spectrum modified in this way is reflected with an error factor of a dispersion in a purity of the standard sample or the like and therefore, when recalibration of the MS portion using the standard sample is executed by constituting a target by the modified standard spectrum, a difference among apparatus is resolved.

The invention discloses wearable equipment capable of automatically adjusting body temperature, and a method for adjusting body temperature. The equipment comprises a wearable component, wherein the wearable component comprises at least one separated component; the at least one separated component comprises a flexible wearable outer shell adapted to the shape of a corresponding part of a human body, and at least one stimulating electrode, a power supply management module, a control module and a wireless module which are arranged on the flexible wearable outer shell; the flexible wearable outershell is used for being attached to the surface of the human body; the stimulating electrode is used for being in contact with the skin of the human body; an upper computer terminal controls to generate and transmit a stimulating instruction to the wireless module in wireless manner; and the control module controls the power supply management module to output a stimulating electric signal used for stimulating the corresponding part of the human body to the stimulating electrode according to the stimulating instruction, and the adjustment of the body temperature of a wearer is realized throughstimulating the increase / decrease in heat generated by the human body by means of the stimulating electrode. The wearable equipment and the method solve the problem that the body temperature of a human body can only be quickly adjusted according to external environmental factors under the traditional cold or hot weather conditions.

The invention provides a signal processing system for a four-quadrant photoelectric detector, relates to the field of signal detecting and processing of photoelectric detectors, and aims at solving the problems of distortion of pulse signal acquiring, reduction of system precision and influence from environmental temperature of the exiting processing system for the four-quadrant photoelectric detector. The system is that a signal outputted by the four-quadrant photoelectric detector flows through an amplitude limiting protecting circuit and then is outputted to a high-pass filter; the filtered signal is converted into a differential signal through a single-end converting differential driver; the differential signal is adjusted by a variable gain adjusting module to be zoomed to reach a proper amplitude value and then inputted into a high speed differential analog-to-digital converter; the generated digital signal is inputted to a signal processing unit; the signal processing unit sends a control signal to a detector gain control module according to the intensity of the received digital signal; the detector gain control module controls the output gain of the four-quadrant photoelectric detector; the signal processing unit processes the signal to obtain the spot center position and then sends to a host control system. With the adoption of the system, the spot center position calculation precision can be improved.

The invention discloses a brain MR image-based multi-dimensional feature classification algorithm. The algorithm comprises the steps of performing regional division on brain MR images and extracting a plurality of ROI characteristics; selecting a mark characteristic, establishing a correlation among the ROI characteristics about the mark characteristic, and forming related characteristics; extracting the ROI characteristics and the related characteristics from the brain MR images to form an ROI characteristic set and a related characteristic set; selecting an optimal ROI characteristic subset and an optimal related characteristic subset from the ROI characteristic set and the related characteristic set through a mixed characteristic algorithm; and setting a weight factor of a function proportion of the ROI characteristics in a classifier, and integrating the optimal ROI characteristic subset and the optimal related characteristic subset through the weight factor and a multi-kernel SVM model to form a multi-kernel classifier. According to the algorithm, high-dimensional characteristics are obtained; local change and function connection change caused by related diseases can be analyzed at the same time; the classification accuracy is high; and the algorithm can assist in diagnosis of different diseases.

The invention provides an air conditioner outdoor heat exchanger self-cleaning control method, an air conditioner and a computer readable storage medium. The method comprises the following steps thatafter a heating mode works for a first preset time, an initial defrosting mode is entered, after the work of the initial defrosting mode is finished, whether the number of times that the temperature of an outdoor unit coil is not higher than a preset filth blockage temperature threshold value is larger than the preset number of times or not is judged, if yes, the situation that filth blockage exists in the heat exchanger is determined, and an preset self-cleaning operation is executed. The invention further provides the air conditioner applying the method and a computer readable storage mediumfor achieving the method. According to the air conditioner outdoor heat exchanger self-cleaning control method, the air conditioner and the computer readable storage medium, the filth blockage of theoutdoor heat exchanger of the air conditioner can be accurately judged, calculation of the self-cleaning execution time and duration of the outdoor heat exchanger of the air conditioner is more accurate, the better self-cleaning effect of the outdoor heat exchanger is achieved, and the heat exchange effect of the outdoor heat exchanger is improved.

The invention discloses a multi-earphone volume control circuit and a control method. The multi-earphone volume control circuit comprises a coder-decoder for performing digital-to-analogue conversion on audio signals, and a single chip; the audio output end of the coder-decoder is simultaneously connected with the audio input ends of a plurality of earphone amplifiers, respectively; the audio output end of each earphone amplifier is correspondingly connected with an earphone; the single chip comprises a plurality of volume regulating command output ends which are correspondingly connected with the volume regulating command input ends of the earphone amplifiers one to one; one volume regulating command output end of the single chip is connected with the control signal input end of the coder-decoder. The multi-earphone volume control circuit and the control method have the advantages that the plurality of earphone amplifier are correspondingly connected with the corresponding earphones, respectively, and used for processing the audio signals of the earphone connected with the earphone amplifiers, and therefore, the volume of each earphone is controlled separately, individual difference can be eliminated and the design is more practical.

The invention relates to a facial expression analysis method and a device based on expression intensity label distribution, and a computer readable medium. The method comprises the following steps: automatically encoding a face image sequence to generate an expression intensity label; and selecting neutral expressions and non-neutral expressions from the expression sequence as input, training theinput data by adopting a convolutional neural network to obtain an expression recognition model, and predicting the category of the expression image and the intensity of the category according to thejoint expression intensity distribution output by the expression recognition model. According to the method, the training data of expression intensity labeling can be obtained, so that the correctnessand robustness of expression recognition are improved.

The invention discloses a method for generating dynamic data of a dynamic seat based on video processing. The method comprises the following steps of identifying an image of a film video, and selecting a plurality of feature points of any feature objects on the image as required; tracking the position change of all the feature points through a video processing algorithm by utilizing a computer; calculating and obtaining a movement parameter and a movement trail of a virtual camera through video processing algorithm according to the position change of the feature points; calculating and obtaining the practical movement trails and movement parameters of the feature objects through the video processing algorithm according to the movement parameter and the move trail of the virtual camera; generating the dynamic data of the dynamic seat according to the practical movement trails and the movement parameters of the feature objects. By means of the method for generating dynamic data of the dynamic seat based on video processing, automatic generation of the dynamic data is almost achieved, generation efficiency of the dynamic data is greatly improved, accuracy of the dynamic data obtained through the method is obviously improved, and the method can eliminate the hysteretic property of the dynamic data.

The invention discloses a washing machine and a wet cloth weighing method and device thereof. The method comprises the following steps: after the washing machine enters a dewatering program, controlling the rotating speed of a dewatering motor of the washing machine to be increased from zero speed; in a stage of increasing the rotating speed of the dewatering motor, acquiring first total power required for the dewatering motor which drives an inner barrel of the washing machine to rotate by a first number of turns in a process of increasing a first rotating speed to a second rotating speed; in a second stage of stable running at the second rotating speed of the dewatering motor of the washing machine, acquiring a second total power required for the dewatering motor which drives the inner barrel of the washing machine to rotate by a second number of turns in a process of stable running at the second rotating speed, and acquiring an average torque ripple value of the dewatering motor which drives the inner barrel to rotate in the process of stable running at the second rotating speed; and calculating weight of clothes in the inner barrel according to the first total power, the second total power and the average torque ripple value. According to the method of the invention, wet cloth weight of the clothes can be conveniently and accurately acquired in a dewatering process.

The invention provides a shadow shielding identification method and device for a photovoltaic string. The method comprises the following steps: obtaining the direct current power of each photovoltaicstring corresponding to a target inverter in an effective acquisition period at each acquisition moment; performing data standardization processing on the direct current power of each photovoltaic string at each acquisition moment to obtain the power generation efficiency of each photovoltaic string at each acquisition moment; respectively calculating power generation efficiency discrete rates ofall the photovoltaic strings at each acquisition moment; at each acquisition moment, clustering the power generation efficiency of all the photovoltaic strings to obtain two clustering clusters, and calculating a relative error between central values of the two clustering clusters; and at each acquisition moment, when the power generation efficiency dispersion rate is greater than a dispersion rate threshold value and the relative error is greater than an error threshold value, judging that the photovoltaic string in the clustering cluster with the smaller central value has shadow shielding atthe acquisition moment. According to the invention, accurate identification of shadow shielding of the photovoltaic string is realized on the basis of not needing meteorological data.

Disclosed herein is a disc playback apparatus with a built-in disc drive which performs hand-close process. The control unit 60 has the A / D input port P which is given the voltage of the terminal of the loading motor 78, after voltage reduction by the dividing resistors R1 and R2. It acquires the voltage applied to the A / D input port P at intervals of 20 ms if the disc tray 10 is in its completely ejected state and compares the acquired voltage with a predefined threshold value. If the acquired voltage is lower than the predefined threshold value, it clears the count, and if the acquired voltage is higher than the predefined value, it increments the count. If the count over 5, the control unit 60 delivers control signals to the driver circuit 88, thereby driving the loading motor to retract the disc tray 10.

The invention provides a control method for the fuel injection amount of a methanol engine, an electric control unit and a vehicle. The control method includes the steps that the target injection amount epsilon of fuel injectors is determined according to received input signals; calculation is carried out to obtain the injection pulse width alphaj of the fuel injectors under the injection pressuredifference Xi of the current fuel injector by using the following formula; according to the alphaj obtained through calculation, the injection pulse width of the fuel injectors is controlled; f(alphaj) is generated by fitting the relation of the injection pulse width alphaj of the multiple fuel injectors in the same model and corresponding injection pulse width correction coefficients; and g(Xi)is generated by fitting the relationship between the injection pressure difference Xi of the multiple fuel injectors in the same model and corresponding injection pressure difference correction coefficients. The control precision of the injection quantity is improved, the possibility of knocking of an air cylinder is reduced, and better play is given to the optimal performance of the methanol engine.

The invention discloses a calorimeter performance degradation evaluation method, a storage medium and equipment; and the method comprises the steps: carrying out the abnormal point removal and trend term removal preprocessing of the monitoring data of a calorimeter, constructing a calorimeter performance degradation feature set, and selecting an optimal feature through employing multiple evaluation indexes; obtaining a normal state reference by using an improved DBSCAN algorithm, and constructing a performance degradation index fused with Mahalanobis distance and cosine similarity; secondly, providing an adaptive degradation stage division method based on the pauta criterion, completing threshold determination of four degradation stages in combination with a calorimeter performance degradation curve, and determining the stage by calculating degradation characteristic state points of the calorimeter to be detected and performance degradation index values of all the references; and providing an evaluation standard based on judgment accuracy to realize accurate evaluation of the performance degradation of the heat meter. The method can improve the operation efficiency and result accuracy of a calorimeter degradation evaluation algorithm to a certain extent, and achieve the performance degradation stage mode recognition of the calorimeter.

The invention discloses a two-way joint bending measurement device in the technical field of measurement devices, and solves the problem that the two-way joint bending degree cannot be measured. The two-way joint bending measurement device is characterized by comprising an amperometric two-way bending sensor, an elastic base, a driver and a data collection processor, wherein the driver comprises asensor interface, an information distribution interface and an original data processing module; the data collection processor comprises an information collection interchange interface and a data processing and visualization module. The amperometric two-way bending sensor is fixedly installed on the elastic base, and the amperometric two-way bending sensor comprises a light emission assembly, a light receiving assembly and a light guide element. The luminous fluxes of the light guide element correspond to the two-way bending degrees of the light guide element one to one, according to output current, the two-way bending degree of a to-be-measured object is judged, and the function of automatically monitoring two-way bending of the joint of the object in real time on line at a large time interval can be achieved.

The invention discloses a face expression analysis method and system based on a sorting convolutional neural network, and belongs to the field of image processing and pattern recognition. According tothe invention, the interested expressions are spliced with other types of expressions; Any two frames of images are selected from the spliced facial expression sequence to serve as input, the expression intensity sorting model is trained through the sorting convolutional neural network, intensity estimation of interested expressions can be achieved through the single expression intensity sortingmodel, and estimation of expression categories can be achieved through combination of multiple expression intensity sorting models. According to the method, the category and intensity of expressions can be estimated at the same time, individual difference and environmental noise are eliminated while facial expression information is reserved to the maximum extent, and therefore the correctness androbustness of expression analysis are improved, and the method has extremely high practical application prospects.

The invention discloses an animal autonomous respiration device used for hyperpolarization gas MRI. A respiration connector is respectively connected with an existing hyperpolarization gas transmission module and an MRI compatible respirator. The respiration connector is directly connected with a respiration cavity through a respiration pipeline, a pressurization cavity is directly connected with the respiration cavity through a perforated diaphragm, the pressurization cavity is connected with the outer diameter of the respiration pipeline, a normal pressure baffle is provided with vent holes, the normal pressure baffle is connected with the tail end of the respiration cavity, an anesthetic gas recovery box is connected with the outer diameter of the pressurization cavity and the normal pressure baffle, an anesthetic gas inlet pipeline is sequentially connected with a flow meter and a pressure meter, and the pressure meter is connected with the pressurization cavity through a pipeline. A branch is formed between the flow meter and the pressure meter through a three-way valve, and the branch is directly connected with the MRI compatible respirator. The animal autonomous respiration device used for hyperpolarization gas MRI is simple in structure, is convenient to use, achieves hyperpolarization gas tests, provides an autonomous respiration mode, avoids damage to the lungs of animals caused by a passive mechanical respiration mode, and guarantees authenticity and reliability of measuring results.

Most chilled-mirror dew-point hygrometers measure dew-point temperatures by using platinum resistors; however, since platinum resistors have the defect of temperature drift, the measured values have certain difference from actual values. Therefore, correction with standard samples is required frequently. The invention discloses a precision chilled-mirror dew-point hygrometer with a temperature self-correction function, and a method for eliminating drift of measured temperature values. Before dew point measurement begins, two precision resistors R1 and R2 with resistances equal to platinum resistor resistances equivalent to upper limit and lower limit temperatures of the sample to be measured are used to substitute the platinum resistors, and are respectively and successively connected to a measuring circuit, and the measuring circuit is corrected. After the dew point is reached, the output voltage of the measuring circuit is recorded; based on the correction result, the drifted measured dew point value is converted into a standard value.