91results about How to "Improve noise reduction efficiency" patented technology

The invention relates to a shift register. A shift register unit comprises a drop-down driving unit, a first drop-down unit, at least one second drop-down unit and at least one drop-down signal input end, wherein the drop-down driving unit is used for inputting a first drop-down signal to a drop-down node; the first drop-down unit is connected to the drop-down node, a pull-up node and a signal output end, and is used for denoising the pull-up node and / or the signal output end according to the first drop-down signal; the at least one drop-down signal input end is in one-to-one correspondence connection with the at least one second drop-down unit, and each second drop-down unit is connected to the pull-up node and the signal output end, and is used for denoising the pull-up node and / or the signal output end according to a second drop-down signal input by the correspondingly connected drop-down signal input end. By means of the technical scheme of the shift register, after the pull-up node realizes the pull-up function and the signal output end outputs the signal and before a first signal input end of the shift register unit receives a next high-level signal, the pull-up node at the output end can be continuously denoised, and accordingly, the denoisinig efficiency is improved.

Embodiments of the application disclose a method and a device for shooting an image, a storage medium, electronic equipment and a method for training a model, wherein the electronic equipment is capable of receiving an input shooting command; then shooting a to-be-shot scene according to the shooting command to obtain a shot image; and performing de-noising processing on the shot image through animage de-noising model, and using a de-noised image obtained through the de-noising processing as a result image of the shooting command, wherein the image de-noising model is obtained by model training in which multiple image sets shot in advance are used as training inputs, and de-noised images which are obtained by synthetic de-noising respectively performed on the multiple image sets are usedas target outputs. Through adoption of the above technical scheme, as long as the training of the image de-noising model is completed, when subsequent shooting is performed, the image de-noising modelobtained by training can be used to perform de-noising processing on shot images, thereby saving the time of shooting multiple images and performing image synthesis, and achieving the purpose of improving an image de-noising efficiency.

The invention discloses a shift register unit, a driving method, a gate drive circuit and a display device, which belong to the technical field of display. The shift register unit comprises an input module, an output module, a pull-down control module, a pull-down module, a first reset control module and a reset module, wherein the input module is used for controlling potential of a pull-up node;the output module is used for inputting first clock signals from a first clock signal end to the output end; the pull-down control module is used for controlling potential of a pull-down node; the pull-down module is used for controlling potential of the pull-up node and the output end under control of the pull-down node and a second power signal; the first reset control module is used for controlling potential of the second control node under control of the first control node and reset signals from a reset signal end; and the reset module is used for controlling potential of the pull-up nodeunder control of the second control node and third power signals from a third power end. The efficiency of noise reduction on the output end is improved.

The invention relates to the field of earphone noise reduction technologies and provides a method and device for controlling noise reduction of an earphone. The method for controlling noise reduction of the earphone includes the steps that a noise signal of the environment is collected through a terminal microphone; the collected noise signal is processed to generate a judgment result; according to the judgment result, the connected earphone is controlled to switch on or switch off a noise reduction function. The method and device for controlling noise reduction of the earphone solve the problems that for an existing earphone noise reduction method, a hardware switch adjusting noise reduction needs to be arranged on the earphone, highly integration of the earphone is not facilitated, and automatic noise reduction can not be achieved. The method and device for controlling noise reduction of the earphone have the advantages that on one hand, a manufacturer does not need to arrange hardware for adjusting noise reduction; on the other hand, automatic noise reduction is achieved, on the premise that the hardware switch on the earphone is not needed, the noise reduction function of the earphone can be automatically switched on or switched off, so the integration degree of the earphone is improved, and the efficiency of noise reduction of the earphone is also improved.

The invention relates to the navigation and guidance technologies and provides a technical scheme of a geomagnetic aided inertial navigation method. According to the scheme, at least more than one magnetic measuring apparatus is used for simultaneously measuring real-time geomagnetic information of transporters; and a geomagnetic matching navigation mode is adopted to conduct rough matching, noise reduction and precise matching on the measured real-time geomagnetic information sequentially. The magnetic measuring manner reduces the error of the apparatus and ensures the reliability of the measured value. By adopting the matching mode, the non-matching points with the reference geomagnetic map are eliminated and the range of the noise reduction and the precise matching is reduced, thereby improving the noise reduction efficiency of the real-time geomagnetic information and saving the precise matching time of the non-matching points; and meanwhile, the speed of the precise matching is improved and the precision of the navigation is ensured, thereby further promoting the quality of the geomagnetic aided inertial navigation.

Disclosed are a method and a terminal used for image noise reduction, wherein the method comprises: obtaining image data of an image; performing wavelet decomposition on at least one component of three components of luminance component y and chrominance components u and v in the image data to obtain a high frequency wavelet coefficient and a low frequency wavelet coefficient of each component; performing recursive noise reduction on the low frequency wavelet coefficient of the each component in at least one direction to obtain a low frequency wavelet coefficient after the noise reduction of the each component; performing wavelet reconstruction based on the high frequency wavelet coefficient of the each component and the low frequency wavelet coefficient after the noise reduction of the each component to obtain at least one component after the noise reduction; when the at least one component after the noise reduction comprises three components, combining the three components after the noise reduction to obtain image data after the noise reduction; and when the at least one component after the noise reduction comprises one or two components, combining the at least one component after the noise reduction with other components in the three components to obtain the image data after the noise reduction.

The invention discloses a video image noise reduction method. The method comprises the steps of obtaining a target video, and performing image block matching on frame images of the target video; extracting image blocks and pixel points of the corresponding matched image blocks at the same position, and calculating the Euclidean distance between the pixel points; when the Euclidean distance is larger than a preset threshold value, obtaining the structure tensor of the image blocks where the pixel points are located, and calculating the structure weight of the image blocks; by performing self-adaption weighted filtering on the image blocks and the corresponding matched image blocks through the structure weight, adopting the pixel points in the image blocks and the corresponding matched imageblocks for performing noise reduction. The invention further discloses a video image noise reduction device and a computer readable storage medium. By adding the self-adaption weighted filtering noise reduction operation, the images can accurately estimate actual image signals during noise reduction, and the advantage of improving the video image conversion efficiency is achieved.

A method for reducing image noise includes calculating a mean pixel-correlation between a central pixel and neighboring pixels within a working window in an image. The method also includes calculating a weighted horizontal similarity between the central pixel and horizontal neighboring pixels and a weighted vertical similarity between the central pixel and vertical neighboring pixels within the working window. Then, the method determines if the central pixel is noise. If so, a noise-reduction process is executed.

The invention discloses a video denoising method and apparatus. The method comprises the following steps: obtaining a video sequence using a noise-containing reference frame as the center from a to-be-denoised video according to a preset time radius; executing the following denoising operations on each reference block, which is pre-divided and is not overlapped with each other in the noise-containing reference frame: determining a search domain of the reference block on each video frame of the video sequence; matching a similar image block of the reference block in the search domain on each video frame; and performing multi-frame average calculation on the similar image blocks of a plurality of reference blocks matched by the reference block in the video sequence frames to obtain a denoising result of the reference block. By application of the scheme provided by the invention, the algorithm complexity can be reduced, the denoising efficiency can be improved, and meanwhile, the denoising effect of noise inhibition and edge protection can be basically guaranteed.

The invention provides a dual-mode silencer for an internal combustion engine exhausting system. The dual-mode silencer comprises a housing, a first end cover, a second end cover, an input pipe, multiple output pipes and a control device, wherein the first end cover and the second end cover are respectively located at two ends of the housing, the input pipe outward extends from the first end cover so as to collect exhausted gas from an internal combustion engine, the multiple output pipes outward extend from the second end cover, and the multiple output pipes, the input pipe and the housing of the silencer form multiple exhausting paths. The control device is arranged on one of the multiple output pipes and used for operably switching the silencer to be in a first operating mode or a second operating mode according to one or more conditions of the internal combustion engine. The corresponding exhausting path is defined in each operating mode, and at least most of the exhausted gas is transferred through the exhausting paths. The exhausting path corresponding to the first operating path is different from the exhausting path corresponding to the second operating path.

The invention provides a silencing component of a compressor and a compressor with the same. The silencing component comprises a fixed part and a silencer. The fixed part comprises an air inlet and a cylinder connection. The silencer is connected with the fixed part and has a silencing chamber. The silencing chamber has an opening facing the fixed part. The opening is blocked by the fixed part. The silencer further comprises a division plate for dividing the silencing chamber into a first chamber having a first resonant frequency and a second chamber having a second resonant frequency. An airflow passage allowing a gas to pass through is formed between the division plate and the fixed part. An air outlet is formed on the fixed part and / or silencer. The set position of an air inlet is corresponding to that of the first chamber. The set position of the air outlet is corresponding to that of the first chamber. The silencing component of the invention improves the silencing efficiency and improves the silencing effect.

The invention discloses a method for measuring the geometric dimension of a refractory brick. The method includes the steps of carrying out filtering and noise reduction on an original refractory brick color image to obtain a noise-reduced image; seperating a refractory brick region from the noise-reduced image to obtain a refractory brick image; carrying out morphological processing on the refractory brick image, so that the edges of the refractory brick image are in smooth transition, and removing the isolated point noise at the edges of the refractory brick image to obtain a corrected refractory brick image; using a Canny operator for carrying out edge detection on the corrected refractory brick image; extracting the contour linear line of the refractory brick by using a Hough transformation method and a K-means method, and forming the outer contour of the refractory brick. The method for measuring the geometric dimension of a refractory brick improves efficiency and precision.

The invention discloses a voice noise reduction method and relates to the field of machine learning, and the method is used for solving the problems of large calculation amount and large resource occupation of the existing voice noise reduction. The method comprises the following steps of obtaining voice data; preprocessing the voice data, and extracting multi-dimensional features of the preprocessed voice data; inputting the multi-dimensional features into a preset voice noise reduction model to obtain a frequency band gain coefficient; dividing the voice data into a plurality of frequency bands, and filtering noise data in the frequency bands according to the frequency band gain coefficient; and recovering the filtered voice data into a voice data stream and outputting the voice data stream. The invention further discloses a voice noise reduction device, electronic equipment and a computer storage medium. According to the invention, the frequency band gain coefficient is calculated to realize voice noise reduction.

The invention relates to a generator set shock absorption and noise reduction method based on noise source recognition. The method comprises the steps that a preset noise source is collected, a presetmatrix is established, and a noise reduction plan is preset; a noise detection module is arranged, and a noise monitor is connected with the corresponding position of the generator set; operating thegenerator set, detecting noise and establishing a detection matrix; comparing the detection matrix with a preset matrix; and selecting a corresponding noise reduction plan according to the detectionresult. According to the invention, different types of standard noise sources are pre-recorded; taking the noise source as a judgment reference; according to the method, the preset matrix is set to store the specified parameters in the noise into the sound source library, the types of the detected noise can be accurately judged by sequentially comparing the detected noise with the preset noise, the specific noise reduction method is used in a targeted mode according to the types of the noise, and the shock absorption efficiency of the method on the generator set can be effectively improved.

The invention relates to an image noise reduction method and a system thereof. The method comprises the following steps of identifying a face area in an image whose noise needs to be reduced and filtering a color value of a non-skin color pixel point in the face area; identifying a dark portion area in the image whose noise needs to be reduced and carrying out gray processing on the dark portion area; and carrying out white noise reduction processing on a current image, wherein the current image is the image whose noise needs to be reduced after color value filtering of the non-skin color pixel point and the gray processing. By using a scheme in the invention, a noise reduction effect of the image can be increased.

The invention discloses a multi-channel three-cavity silencer which comprises a silencer barrel body, an air inlet tube and an exhaust tube and further comprises a first cavity partition plate and a second cavity partition plate, wherein the first and second cavity partition plates divide silencer barrel body into a first cavity, a second cavity and a third cavity, the air inlet tube extends into the first cavity, the exhaust tube extends into the third cavity, a second cavity central tube crossing the cavities is arranged in the second cavity in the middle, the cavities are provided with multiple partition plates, the partition plates respectively divide the cavities into cavity groups formed by sub-cavities, and the inside of the second cavity central tube is a resistive silencing portion made of a silencing material. A fin structure is added for the multi-channel three-cavity silencer, the multi-channel three-cavity silencer is provided with the sub-cavities and the resistive silencing portion, and a combined silencing structure is formed to improve the cavity utilization rate and noise reduction effect. In addition, a metal layer of the fin structure effectively helps a silencing layer absorbing sound wave energy and enables the barrel body to effectively radiate heat.

The invention is suitable for the field of noise reduction earphones, and provides an environment noise collection method and apparatus by a noise reduction earphone. The method includes the steps: collecting environment noise by a noise reduction earphone microphone, and generating a corresponding simulated noise signal; sampling the simulated noise signal according to a pre-configured sampling frequency and generating noise sampling values; quantifying the multiple generated noise sampling values on the basis of a pre-configured quantification level, and generating an optimal noise sampling value; generating a noise value according to the optimal noise sampling value through a preset conversion formula; and matching the generated noise value with a pre-stored noise value in a pre-stored corresponding relation to obtain the noise level of the generated noise value. The beneficial effects of the invention reside in that firstly, the noise level of environment noise can be obtained without arrangement of hardware for detecting the noise level on a noise reduction earphone; and secondly, the noise reduction function can be triggered in time to achieve noise reduction, so the noise reduction efficiency is improved.

The invention is applicable to the field of earphone noise reduction and provides an earphone noise reduction method and device. The earphone noise reduction method includes: acquiring current noise frequency; judging whether the frequency is in the preconfigured frequency range, wherein the frequency range comprises the frequency heard by a user; on yes judgment, calling the noise reduction mode and setting the noise reduction level that the frequency corresponds to according to the frequency. The earphone noise reduction method and device solves the problem in the existing earphone noise reduction mode that preferable noise reduction mode selection and different noise reduction level setting cannot be achieved according to the sound frequency heard by the user, and noise cannot be eliminated effectively. Under the condition that an earphone is used normally, the current noise reduction mode can be switched to the preferable noise reduction mode quickly, the electric quantity of the earphone is saved, and the noise reduction efficiency is improved.

The invention discloses a method and a device for removing noise by adopting multiple microphones. The method comprises the steps of receiving first sound signals from a first direction; receiving second sound signals from a second direction which is different from the first direction; comparing sound volumes in the first sound signals and the second sound signals; carrying out volume noise reduction processing for sound signals with the same volumes in both the first sound signals and the second sound signals; and storing a sound frequency of a person when speaking, comparing the stored sound frequency with a sound frequency in the first sound signals and the second sound signals, and carrying out frequency noise reduction processing for the sound with the frequency outside the scope of the stored sound frequency of the first sound signals and the second sound signals. As a first microphone and a second microphone are adopted in the method, and a first noise reduction module and a second noise reduction module respectively carry out volume noise reduction processing and frequency noise reduction processing for the first sound signals and the second sound signals, the method and device for removing noise by adopting multiple microphones have the advantages of simple structure, low production cost, high noise reduction efficiency, unchanged tone quality and the like.

The invention discloses a hyperspectral image noise reduction method based on a three-dimensional quasi-recurrent neural network, and belongs to the field of image processing. The hyperspectral imagenoise reduction method based on the three-dimensional quasi-recurrent neural network comprises the steps of constructing the three-dimensional quasi-recurrent neural network taking three-dimensional convolution, quasi-recurrent pooling and an alternating direction structure as a core according to space-spectrum correlation and spectrum global correlation; training a three-dimensional quasi-recurrent neural network through the noise reduction training data set; and removing complex noise in the noisy hyperspectral image by using the trained three-dimensional quasi-recurrent neural network. According to the method, the noise reduction recovery of the hyperspectral image can be completed with high quality, the recovery result can be ensured to have space and spectrum fidelity, the recovery efficiency of the hyperspectral image is greatly improved, the method can be used for the hyperspectral image acquired by an imaging system with any spectral resolution, and the application range of thehyperspectral image is expanded. The method can be applied to the fields of remote sensing imaging, geological exploration, agricultural production and biomedicine.

The noise reduction method for non-local mean filtering comprises the following steps: acquiring one pixel in an original image as a current noise reduction pixel, and selecting a first search windowby taking the current noise reduction pixel as a center; Calculating the similarity between each pixel in the first search window and the current noise reduction pixel; Selecting the pixel with highersimilarity as the center to determine a second search window, calculating the similarity between the second search window and the current noise reduction pixel, and selecting the pixel with higher similarity as the center to determine a third search window until the boundary of the original image is searched; And selecting the pixel with the highest similarity with the current noise reduction pixel according to the calculated similarity result to perform noise reduction on the current noise reduction pixel. By calculating the similarity between the pixels in the similar window of the pixel neighborhood and the current noise reduction pixels and determining the search direction according to the pixels with the high similarity, the number of times of similarity calculation of the number oftimes of search can be greatly reduced, the complexity of noise reduction processing can be reduced, and the noise reduction efficiency can be improved.

The invention relates to a bent type noise reduction device for a cooling tower. The bent type noise reduction device comprises multiple noise elimination flow deflector sets. Each noise elimination flow deflector set comprises multiple noise elimination flow deflectors. An air inlet in the bottom of the cooling tower and an air outlet in the top of the cooling tower are provided with the multiple noise elimination flow deflector sets correspondingly. All the noise elimination flow deflectors in each noise elimination flow deflector set are mutually parallel and horizontally installed at equal intervals. The noise elimination flow deflector sets on the air inlet are distributed in the vertical direction, and the noise elimination flow deflector sets on the air outlet are horizontally installed. The noise elimination flow deflectors in the same noise elimination flow deflector set are mutually parallel. The bent type noise reduction device is characterized in that the noise elimination flow deflectors are in a bent shape, and the center lines of the noise elimination flow deflectors are of a left-right symmetrical structure. The invention further relates to a stepped noise reduction device for the cooling tower and a corrugated type noise reduction device for the cooling tower.

The invention discloses a central air conditioner and a denoising method and device thereof, relates to the technical field of denoising and is intended to reduce denoising cost and enhance denoisingeffect. The method herein includes: allowing a denoising controller to receive a reference noise signal acquired and transmitted by a reference noise sensor in a target denoising module, and receive an error noise signal acquired and transmitted by an error noise sensor in the target denoising module, wherein the target denoising module is nay of at least two denoising modules; the denoising controller processes the reference noise signal of the target denoising module and the corresponding error noise signal so as to acquire a corresponding target denoising signal of the target denoising module; the denoising controller controls a speaker in the corresponding denoising module according to the target denoising signal to output a denoised sound, so that the denoised sound counteracts noiseto be eliminated.

The invention provides an audio processing method and an intelligent sound box. The method comprises the following steps: obtaining to-be-played audio data, and obtaining current operation informationwhich comprises an operation voltage, an operation current and an operation vibration amplitude; acquiring magnetic field interference information of the current environment, and performing noise reduction on the to-be-played audio according to the operation voltage, the operation current and the magnetic field interference information; performing voice endpoint detection on the to-be-played audio data, and performing audio segmentation on the to-be-played audio data according to a voice endpoint detection result to obtain effective audio; and performing audio noise reduction on the effectiveaudio in the to-be-played audio according to the operation vibration amplitude, and playing the to-be-played audio data after audio noise reduction. According to the invention, through obtaining themagnetic field interference information based on the current environment, noise reduction is carried out on noise data generated in different magnetic field interference environments, and the audio noise reduction efficiency is improved.

The invention relates to artificial intelligence, and provides an audio noise reduction method and device, equipment and a storage medium. The method comprises the following steps: pre-processing noise frequency to obtain frequency spectrum information, processing the frequency spectrum information based on a frequency domain signal processing network to obtain frequency spectrum mask features, acquiring time-frequency features according to the frequency spectrum information and the frequency spectrum mask features, processing the time-frequency features based on a time domain signal processing network to obtain time-frequency mask features, generating a predicted audio according to the time-frequency features and the time-frequency mask features, adjusting network parameters of a preset learner based on the predicted audio and the pure audio to obtain a noise reduction model, acquiring a request audio, and performing noise reduction processing on the request audio based on the noise reduction model to obtain a target audio. According to the method, the noise reduction accuracy and real-time performance of the request audio can be improved. In addition, the invention also relates to a block chain technology, and the target audio can be stored in a block chain.

The invention provides a method and a system for reducing color noise of an image. The method for reducing the color noise of the image comprises the following steps of: converting RGB parameters of each pixel point into YUV parameters of each pixel point; obtaining a Y channel parameter of the current pixel point, and carrying out filtering processing to obtain a UV channel first control parameter; obtaining a UV channel second control parameter corresponding to each pixel point of the to-be-denoised image; calculating a UV channel control parameter of each pixel point; carrying out pixel domain division on each pixel point, respectively calculating the sum of absolute difference values of a U channel and a V channel of a current pixel domain and an adjacent pixel domain, and respectively carrying out smoothing processing to obtain a corresponding U channel weighted weight and a V channel weighted weight; filtering to obtain a corresponding U channel noise reduction result and a V channel noise reduction result; and obtaining filtered YUV parameters according to the U channel noise reduction result and the V channel noise reduction result of each pixel point, and converting the filtered YUV parameters into RGB parameters.

The invention discloses an automobile exhaust manifold silencer which comprises a main body, an air inlet pipe and an end cover. A silencer core body is fixed in the main body, and both ends of the silencer core body run through the end part of the main body and the middle part of the end cover. An annular groove is arranged in the open end face of the main body. An extension part which matches the annular groove is arranged in the middle part of the seal end of the end cover. An inclined face A is arranged at the upper part of the seal end face of the end cover. An inclined face B which shares the same inclination direction with the inclined face A is arranged on the end face of the main body. The inclined face A, the outer wall of the extension part and the inclined face B form an annular filling groove. The end part of the silencer core body is in butt joint with an intake pipe. A helical diversion trench is arranged in the inner wall of the intake pipe in the direction, which points to the main body, of the axis. On the seal end face of the end cover, an L-shaped contact face is formed on the outer wall between the extension end and a projection. Through a metal adhesive in the filling groove, the outer wall is completely separated from the inside of the main body to ensure that the airtightness of the silencer.

The invention belongs to the technical field of silencer manufacturing and particularly relates to novel automotive silencing equipment. The novel automotive silencing equipment comprises a shell, a front end cover, a rear end cover, an air inlet pipe, a buffer pipe, an air outlet pipe, a first partition plate, a second partition plate, a third partition plate, a partition plate support, a first cavity, a second cavity, a third cavity, a fourth cavity, silencing cotton and glass fiber cotton. The front end cover and the rear end cover are arranged at the two ends of the shell respectively. The first partition plate, the second partition plate and the third partition plate are sequentially arranged inside the shell. The second partition plate is connected with the third partition plate through the partition plate support. The shell is divided into the first cavity, the second cavity, the third cavity and the fourth cavity through the first partition plate, the second partition plate and the third partition plate. The novel automotive silencing equipment is good in silencing effect and high in silencing speed.

The invention provides a charging pile noise reduction control method, device and system based on clustering analysis, and the method comprises the steps: collecting a noise signal, and taking the noise signal as a feedforward signal; performing signal fusion on the feedforward signal to obtain a primary signal; processing the primary signal by using an FxLMS algorithm, and outputting a secondary sound field signal; offsetting the noise signal and the secondary sound field signal to obtain a residual noise signal; collecting residual noise signals, and calculating a decibel mean value; acquiring a noise threshold interval corresponding to the current time period; judging whether the decibel mean value converges to the lower limit of the noise threshold interval; judging whether the decibel mean value exceeds the upper limit of the noise threshold interval or not; according to the invention, two sensors are adopted to collect noise signals, and a signal fusion technology is utilized to obtain accurate primary signals. And meanwhile, different requirements on noise in different environments and time periods where the charging pile is located are considered, a dynamic threshold interval is set, and the noise reduction efficiency is improved.