727results about How to "Reduce noise interference" patented technology

The invention discloses a speech emotion recognition method based on a long short time memory network and a convolutional neural network. According to the method, a speech emotion recognition system based on LSTM (long short time memory network) and CNN (convolutional neural network) is constructed; with a speech sequence as input of the system, the LSTM and the CNN are trained by virtue of a back propagation algorithm and parameters of the networks are optimized, so that an optimized network model is obtained; and by virtue of the network model which is trained, the newly inputted speech sequence is emotionally classified into six emotions, namely being sad, being happy, being disgusted, being fearful, being scared and being neutral. The method, which takes two network models, namely the LSTM and the CNN, into comprehensive consideration, avoids cumbersome artificial selection and characteristic extraction, and the accuracy of emotion recognition is enhanced.

The invention discloses a shift register and a grid line driving device, which relate to the technical field of liquid crystal display. In order to reduce noise jamming of the shift register at the standby unattended time, the shift register and the grid line driving device are invented. The shift register comprises a first thin film transistor, a second thin film transistor, a third thin film transistor, a fourth thin film transistor, a capacitor and a drop-down module, wherein the drop-down module connected among a clock signal end, a first node and a signal output end is connected with a low level signal end and used for maintaining the first node and the signal output end to have a low level at the standby unattended time of the shift register. The grid line driving device comprises a plurality of the shift registers which are connected in series. The grid line driving device can be used for driving a grid line.

The invention discloses a method for detecting a random access leader. After a random access channel (RACH) signal is treated with the simplified time domain-related operation, noise mean estimation is carried out to each root sequence; an absolute threshold is then detected according to a signal obtained by the noise estimation and signals in a search window are detected; the signals which exceed the detection threshold are screened and then a timing position can be determined in a fast and accurate way so as to provide reliable uplink timing adjustment information to a mobile terminal, guarantee the accuracy of the uplink synchronization and effectively eliminate the timing error due to the over-sampling and reduce the false alarm in the detection.

The invention relates to a cruise image based imaging method of a water-surface aerial view, which comprises the steps of: firstly, establishing an imaging system which comprises a server image processing system, a client interface display and interactive system, and a video collecting system of a cruise boat; installing a video camera and a GPS (Global Position System) device on the boat; collecting water-surface video data; processing the video data by the server image processing system to obtain a water aerial view; and inputting the water-surface aerial view into the client interface display and interactive system. A water-surface video is collected by the cruise boat, and is converted into the water aerial view. The method has the advantages of small distortion, short period, high aging performance, low investment and good compatibility, and is disturbed by less noise.

The invention belongs to the detection field, and relates to an ultrasonic wave flow rate detecting system and a detecting method, in particular to a high-precision fluid flow rate ultrasonic wave detecting system and a detecting method. The ultrasonic wave flow rate detecting system includes a system control part, a circuit part and an auxiliary device. The ultrasonic wave flow rate detecting method is characterized in that an improved time difference method is adopted for calculating the speed of a fluid; the delay window reception technique is adopted, and the influence of noise on the detection precision during detection is reduced through pulse width detection; and as the resolution of measuring time for the system is further improved through the adoption of the interpolation correlation method and the correlation coefficient judgment method, and reaches 1.25ns, the accuracy and the precision of the ultrasonic flow-meter are remarkably improved. The detecting system provided by the invention has the characteristics of small size, low power consumption, stable and reliable measurement and the like; the precision reaches 0.5; the sensitivity attains 0.3mm/s; and the system is suitable for measuring the flow rates and the flow velocities of various liquids in plastic or metal pipes with diameters being 25mm-8m.

The invention relates to a binocular vision positioning method for target grabbing of an underwater robot, and belongs to the field of computer vision. The method is mainly used for accurately acquiring three-dimensional information of a grabbed target when an underwater robot works. The method comprises the following steps: double-target positioning: calculating internal and external parameters of left and right cameras; target detection: positioning a target object detection frame; binocular image correction: carrying out distortion correction and stereo correction, and determining a right image target area; binocular image stereo matching: extracting image feature points, describing the feature points, performing stereo matching, and removing mismatching; and calculating the three-dimensional information of the target in the image under the left camera coordinate. According to the method, the accurate parallax value is obtained by extracting feature points, removing unstable featurepoints through non-maximum suppression, constructing a binary descriptor, matching the feature points and removing mismatching. Through the scheme, the binocular stereo matching robustness can be improved, and meanwhile, the three-dimensional information of the detection target can be accurately obtained, so that the real-time positioning requirement on the target when the underwater robot grabsthe target is met.

The invention discloses a method for real-time monitoring gas heating furnace flame on the basis of ROI average image analysis which belongs to the field of industrial heating furnace flame video monitoring. The method comprises following steps: 1) setting a flame region of interest(ROI) through a human-computer interaction method; 2) extracting a ROI image from a flame monitoring video flow and calculating an average image; 3) pre-treating the ROI average image, including noise smoothing, threshold segmenting, morphological processing, to determine the most probable flame pixel region; 4) extracting flame features from the pre-treated image, comprising flame region area, average brightness, mass center and flame color principal components; 5) synthetically analyzing flame feature, establishing judgment rules, automatically identifying flame burning status and outputting; 6) establishing off-line self-study model by use of historical data, optimizing system parameters, increasing method robustness. The method can be applied for real-time monitoring combustion conditions of gas heating furnace with single-burner or multi-burner and guarantees safety of industrial production.

The invention discloses an illegal parking detection method based on image texture. The method sequentially includes the following steps of obtaining the background, segmenting targets and analyzing the texture. The background obtaining is mainly achieved through the regional division of a determined detection area and a detection area, the establishment of a detection area model and the background renewing. The target segmenting is mainly achieved through the dynamic threshold value selection, the target segmentation and the morphology filtering. By means of the target segmentation method where dynamic threshold values are adopted, moving targets can be more accurately obtained through segmentation; meanwhile, the noise interference generated in the segmentation process can be effectively eliminated through the opening operation in the morphology filtering; the texture analyzing is mainly achieved through the static target detection, the running and parked vehicle recognition and the illegal parking event conforming; whether static objects appear or not is judged by analyzing the process during which the image gray value reaches the steady state; meanwhile, the false alarm existing in the process is eliminated through the entropy of an image gray level co-occurrence matrix.

The invention belongs to the technical field of fine measurement, and in particular relates to a small module arc gear measurement method. According to the technical scheme, the small module arc gear non-contact optical fine measurement method comprises the steps that a microscopic image of a small module arc gear is collected; the image edge of the small module arc gear is accurately identified; by using a least square circle method, the geometric center of the small module arc gear is fit; by using a maximum inscribed circle method, the center hole assembly center of the small module arc gear is fit; by using a polar coordinate system, the tooth pitch deviation of the small module arc gear is measured; by using a segmentation measurement method, the tooth profile deviation of the small module arc gear is measured; and the maximum tooth thickness deviation of the small module arc gear is measured. According to the method, the problem that the basic parameters, the geometry parameters, the tooth pitch deviation, the tooth profile deviation and the maximum tooth thickness deviation of the arc gear are difficultly measured is solved, wherein the module of the arc gear is below 0.2mm, and the tip diameter is below 2mm.

The invention discloses a method and a device for real-time detection of moving electrocardiosignal QRS (magnetic resonance angiography) waves. The method includes steps: performing four-point smoothing to electrocardiosignal data, subtracting a baseline by the smoothed data, and then performing square processing to the data in order to obtain a data sequence S; respectively performing first difference and second difference operations of the data sequence S to obtain data sequences D and E respectively; picking a certain amount of data from the data sequences D and E for calculating initial threshold values respectively, and updating the threshold values regularly; calculating and marking a second difference minimum value point meeting conditions of the threshold values as an R-type wave point; detecting a Q-type wave and an S-type wave near the R-type wave point, and excluding false detected QRS waves according to waveform features of the QRS waves; and finding a real R wave point in the electrocardiosignal data. By the aid of the method and the device, the QRS waves can be detected on a mobile device quickly and efficiently in real time.

The invention discloses a short-impending rainfall prediction method based on ConvLSTM and 3D-CNN, and belongs to the technical field of weather forecast. The method comprises the following steps: firstly, inputting a historical radar echo map, a gridding temperature and total rainfall at a moment t, and performing data cleaning and denoising on the historical radar echo map and the gridding temperature; performing statistical analysis on the rainfall data imbalance problem, and establishing a new loss function using different weights at different rainfall rate levels; secondly, standardizingthe gridding temperature and the total rainfall by using a meteorological data mapping method based on power and logarithm transformation; and finally, fusing the t-moment input data subjected to theprevious step into a data block, carrying out model building and testing based on a convolutional long-term and short-term memory neural network and a three-dimensional convolutional neural network, and outputting a short-term and temporary rainfall prediction result. According to the method, the rainstorm prediction precision can be improved, the meteorological data is reasonably visualized and standardized, the image features of various meteorological data are fused, and the noise interference is reduced.

The invention provides a superconductive quantum bit measuring system, comprising a signal source which current signal output is connected with an object Josephson node with two ends respectively connected with two inputs of a low-noise amplifier, while the output of the low-noise amplifier is connected with the positive end of the amplifier, and the negative end of the amplifier is grounded, the output of the amplifier is connected with a stop end of a counter which start end is connected with a synchronous signal output of the signal source, the current signal output of the signal source and the object Josephson node are serially connected with a copper powder filter between, the current signal output of the signal source and the copper powder fiber are serially connected with a RC filter between. The inventive system can effectively reduce noise and meet the demand of superconductive quantum bit measurement.

The invention discloses a microwave differential gain-based coherent Doppler wind lidar. According to the technical scheme of the invention, a balanced detector is connected with a bandpass filter, so that invalid signals, away from the Doppler frequency shift, are filtered and eliminated. A pre-amplifier is adopted to amplify an electrical signal outputted by the detector. Meanwhile, a power divider is adopted to divide amplified electrical signal into a monitoring channel and a detection signal. The detection signal is filtered by a pre-emphasis filter. The frequency-domain change of the detection signal is converted to the intensity change of the signal. Through the comparison with the monitoring channel and the transmittance curve of the pre-emphasis filter, the Doppler shift information is obtained. The interference of invalid echo signals is reduced. By adopting the pre-emphasis filter, echo signals are subjected to frequency discrimination, so that the computation amount of signal processing is reduced. The sensitivity and the signal-to-noise ratio of the system are improved. The coherent Doppler wind lidar is higher in resolution distance.

The invention relates to a transient state power quality disturbance classification recognition method based on BUD spectrum kurtosis. The method adopts the BUD spectrum kurtosis method to calculate kurtosis of two disturbing signals of transient state pulse and transient oscillation, selecting maximum value, minimum value and average value of the kurtosis as characteristic quantity and inputting SVM of PSO optimizing parameters to conduct practice tests. Simulation data are obtained through PSCAD/EMTDC, and the method is used for conducting analysis. Results show that the method based on the BUD kurtosis is capable of effectively extracting disturbance characteristic quantity and has good noiseproof performance. An SVM classifier can be used for recognizing two disturbances when small samples are overlapped with other disturbances, and recognition rate is high.

The invention discloses a shift register, a grid line integrated drive circuit, an array substrate and a display. Two electronic switch modules are additionally arranged in the conventional shift register and are arranged between a drop-down node, a low level signal end and a pull-up node, and between the drop-down node, the low level signal and a signal output end respectively; the two electronic switch modules are started when the shift register is in the non-work time and the drop-down node is in low level and are used for reducing the noise of discharging the pull-up node and the signal output end respectively, so that the noise interference of the shift register in the non-working time is effectively reduced.

The invention relates to an answer extraction method for a community question and answer system. The method comprises the following steps: firstly, carrying out word segmentation and stop word removalpreprocessing on a question and answer data set; selecting a word meaning similarity scheme, and setting a threshold value to execute normalization of synonymous words and synonymous words; calculating smooth inverse frequency similarity based on question classification and public component removal, sorting the smooth inverse frequency similarity, and selecting k candidate questions with closestsemantics; calculating the similarity of the k candidate questions by considering the character level characteristics and the dependency pyramid characteristics, namely, based on the character vectorcalculation similarity, sorting the candidate questions, and sorting the candidate questions according to the comprehensive dependency similarity; and finally, selecting an optimal answer by comprehensively balancing the ranking and numerical values. A problem classification strategy is adopted, and the range of subsequent calculation is narrowed; public components of the data set are removed at the sentence level, and the smooth inverse frequency similarity is adopted as a sorting reference to accurately screen the first k candidate problems; the questions with the closest semantics are moreefficiently and accurately determined in the community question and answer data, and the efficiency and accuracy of answer extraction are improved.

The invention relates to an image processing method and device, electronic equipment and a storage medium. The method comprises the following steps: screening a to-be-processed image sequence, and determining a target image sequence; Performing feature extraction on the target image sequence to obtain a feature map of the target image sequence; Determining a to-be-segmented image according to thefeature map, the target image sequence and the to-be-processed image sequence; And performing region segmentation processing on the to-be-segmented image, and determining a target region of the to-be-segmented image. According to the image processing method provided by the embodiment of the invention, the target image sequence can be screened out from the to-be-processed image sequence, the noiseinterference of time information is reduced, and the calculation amount of feature extraction processing is reduced. Moreover, due to the fact that noise interference is reduced, the recognition precision of the target area can be improved, namely the processing precision of segmentation processing on the segmented image is improved.

The invention discloses an environment-friendly vehicle orientation loudspeaker system. The environment-friendly vehicle orientation loudspeaker system mainly comprises a dedicated short range communication (DSRC) module, a data processing controller, an ultrasonic testing orientation loudspeaker module, a loudspeaker, a control key 1, a control key 2 and a power source, wherein the data processing controller leads out data lines to be connected with the DSRC module, the ultrasonic testing orientation loudspeaker module, the control key 1, the control key 2 and the power source respectively, the loudspeaker is connected with an output interface of the data processing controller, the DSRC module comprises a microwave transceiver antenna, a modem and a coder decoder, the ultrasonic testing orientation loudspeaker module comprises 16 ultrasonic detectors and 16 orientation loudspeakers, and the data processing controller is an advanced RISC machines (ARM) processor. By means of the environment-friendly vehicle orientation loudspeaker system, noise interference that is brought to other pedestrians around can be avoided to the maximum degree, and living environments of people can be further improved.

The invention relates to a method and apparatus for detecting infrared dim small targets under a complex background, belonging to the technical field of infrared target detection. At first, the invention adopts a bilateral filtering algorithm to preprocess the single-frame infrared images in an image sequence successively, so as to reduce noise interference in the image and enhance the signal-to-noise ratio at the target; then, a single-frame infrared target detection algorithm based on block extremum is used to detect the potential candidate target sets in each preprocessed image frame by frame; based on the spatio-temporal continuity of small targets in multi-frame images, a pipeline filtering algorithm is used to identify the possible real targets in the candidate target set; finally the false alarm removal strategy is used to eliminate the false target and finally confirm the true target. Through the process, the invention overcomes the shortcomings of weakening target intensity, blurring edge, enlarging contour and the like brought by the prior infrared image processing technology, and can better retain and enhance the dim small target information, improve the target detectionrate, and reduce the missed detection rate and false alarm rate of the target.