233results about How to "Eliminate noise" patented technology

The invention provides a three-dimensional terrain model real-time smooth drawing method with combination of a GPU technology, and belongs to the technical field of image processing. The objective of the invention is to provide the three-dimensional terrain model real-time smooth drawing method with combination of the GPU technology so that cache reuse in multiple times of drawing can be realized based on the current popular programmable GPU technology with a global digital elevation model acting as a data source, and load of computation space is effectively reduced. The method comprises the steps of construction of a multi-resolution pyramid model, elimination of image noise points, filtering of images, partitioning of planar projection of the earth according to equal latitude and longitude, and construction of different hierarchical levels of pyramid layers according to a mode from the top to the bottom. Acceleration and enhancement of terrain rendering are realized based on the programmable GPU technology, i.e. all phases of a graphical drawing pipeline are controlled by using shader languages, two and textures are respectively generated by vertex information and index information of elevation data to be stored in video memory for scheduling of whole terrain drawing; and vertex interpolation and migration are performed in the geometric phase by utilizing a curved surface subdivision and fractal technology so that procedural details are generated and the phenomenon of edges and corners of the terrain mesh when resolution is insufficient can be compensated.

A speech encoder comprises an encoding element for encoding a noise reduced speech signal, and a noise suppression element that takes a noisy speech signal and generates the noise reduced speech signal by maximizing the signal to noise ratio (SNR) of the noisy speech signal without suppressing the voiced speech components of the noisy speech signal. The noise suppression element may use harmonic modeling techniques that maximize the SNR in each sub-band of the noisy speech signal by reconstructing the voiced speech components of the noisy voiced speech signal emphasizing harmonic frequencies within each sub-band. The SNR is further maximized by eliminating noise components between signal peaks at the harmonic frequencies, and eliminating noise at signal peaks at the harmonic frequencies by smoothing harmonic parameters generated by the reconstruction of the voiced speech components of the noisy speech signal.

Disclosed is a pop-up type portable terminal. The pop-up type portable terminal includes a main housing, a sub-housing slidably coupled to the main housing such that the sub-housing is slidably accommodated in or withdrawn from the main housing, a main board installed in the main housing, a guide beam fixed to one side of the main board and extending lengthwise along the main board, a guide rail installed at a rear surface of the sub-housing and extending lengthwise along the sub-housing so as to engage with the guide beam for guiding a sliding movement of the sub-housing, and a torsion coil spring installed in the main housing and having a first end portion fixedly coupled to the main housing and a second end portion fixedly coupled to the sub-housing in order to provide bias force in a direction urging the sub-housing to be withdrawn from the main housing. Noise caused by friction between the guide beam and the guide rail is thereby significantly reduced.

A lithographic projection apparatus projects a pattern from a patterning device onto a substrate using a projection system. The apparatus has a liquid supply system to supply a liquid to a space between the projection system and the substrate. The apparatus also has a fluid removal system including a chamber to hold liquid and having an open end adjacent to a volume in which fluid will be present. The open end removes, through a pressure differential across-the open end when liquid is present in the chamber, substantially only liquid from the volume when liquid in the volume is adjacent to the open end but not gas from the volume when gas in the volume is adjacent to the open end.

The invention relates to a depth generating method based on a background difference sectional drawing and a sparse optical flow method. The depth generating method comprises the following steps: selecting suitable colour space and establishing a colour model; establishing a background model fast and effectively so as to exactly describe the background of a video sequence; classifying each pixel point according to the background model; optimizing the classified result of the pixel; evaluating an optical flow value of a feature point by the sparse optical flow method; evaluating the optical flow values of all the points by interpolation; and obtaining a depth value by the optical values. The invention has the advantages as follows: the sheltering problem which cannot be solved by singly using the optical flow method can be avoided; and the generation of the depth of the video sequence can be automatically realized.

The invention discloses an image depth estimation system and method based on cavity convolution and semi-supervised learning, and solves the problem of estimating scene depth from a single image. According to the invention, a conventional encoder is adopted; the network structure model of the decoder is improved, and a cavity convolution module is added between an encoder module and a decoder module. The image depth estimation method specifically comprises the steps of obtaining a training set and a test set; using a semi-supervised learning strategy to train a model; testing the precision ofthe model by using a test set; estimating image depth using a model. According to the method, the spatial perception capability of the network is improved by using hole convolution, a semi-supervisedlearning strategy is adopted, and the output depth map is optimized through the depth map smoothing error. The method has the characteristics of small parameter model, high prediction precision and complete detail information. The method is applied to the fields of image three-dimensional reconstruction and automatic driving.

The invention discloses a spatial neighborhood information weighted hyper-spectral remote sensing image classification scheme aiming at a problem that noising exists in an image classification result existing in the conventional hyper-spectral remote sensing image classification technology. The scheme comprises the following steps: performing partial principal component analysis on preprocessed image data to acquire a converted characteristic matrix; performing quick clustering on the characteristic matrix by using a spatial neighborhood information weighted FCM (Fuzzy C-Means) algorithm to obtain a classification result of a hyper-spectral image. According to the hyper-spectral remote sensing image classification scheme, the effect of principal component analysis feature dimension reduction and the rich spatial neighborhood information on the hyper-spectral image are fully combined; the classification result of the hyper-spectral image is improved while the efficiency of the algorithm is guaranteed; compared with the conventional method, the scheme has the advantages that the amount of calculation can be reduced, the classification result can be improved, the phenomenon of noising caused by a same object with different spectrums and noise can be overcome and an excellent classification result can be obtained.

The invention discloses a gluing detection system and method based on deep learning, and the method comprises the following steps: constructing an initial deep residual error network model which is ofa five-layer structure and comprises an input layer, a convolution layer, a residual error layer, a full connection layer and a joint loss function layer; inputting the training sample into the deepresidual network model for training to obtain a gluing detection model; obtaining an original gluing image, and carrying out the preprocessing to obtain a to-be-detected gluing image; and inputting the to-be-detected gluing image into the gluing detection model, obtaining a score of gluing image detection, and judging whether the to-be-detected gluing image is qualified or not according to the score. According to the invention, the gluing of the vehicle window in the gluing workshop is automatically detected by constructing the deep residual network model, the gluing quality can be quickly andaccurately judged, the detection result is uploaded in real time, the timely processing is facilitated, and the automation degree and the production efficiency of the gluing workshop are improved.

The invention relates to an unsupervised image style migration method based on dual learning, and belongs to the field of computer vision. The method comprises the following steps: firstly, preprocessing training data, and then designing network structures of a generator and a discriminator; designing a loss function and training the generator and the discriminator by using the training data and the loss function to obtain an unsupervised image style migration network ST; introducing an aesthetic scoring model to maximize the aesthetic quality score of the generated image; taking basic pixel features and advanced semantic features of the images as dual consistency constraints of unsupervised training, and dynamically adjusting the weights of the two features; adaptively adjusting convergence rates of the model in different style migration directions by using a style balance technology; and finally, performing style migration on the input image by applying ST. Compared with an existingmethod, the method has the advantages that the target image with higher quality can be generated, good universality is achieved, meanwhile, the training process of the model is more stable, and selection and design of the network structure are more flexible.

A color image processing method and apparatus thereof are provided. The color image processing method includes the steps of (a) sorting image pixels according to the color distance between the image pixels and a central pixel, (b) grouping the sorted pixels into groups in which the difference in the intragroup color distance is minimum and the difference in the intergroup color difference is maximum, and (c) performing filtering by replacing a central pixel value with a predetermined pixel value determined by pixel values of pixels in the groups. The color image processing method can reduce the generation of edge blurring of the image in removing impulse noise from an image and filtering the same.

A one-way clutch includes an outer member with inner teeth; an inner member rotatably retained in the outer member concentrically with the same and having a depression formed in an outer circumference thereof; at least one planet gear loosely and rotatably retained in the depression and engaging the inner teeth; and a removed portion formed in the outer circumference of the inner member adjacent the depression. When the inner member rotates in one direction with respect to the outer member, the planet gear freely rotates inside the depression. When the inner member rotates in an opposite direction with respect to the outer member, a part of teeth of the planet gear engages the removed portion. The planet gear is retained in the depression via a moving-range regulating member for regulating a moving range of the planet gear in radial and circumferential directions.

The invention belongs to the field of optical measurement, and specifically discloses a three-dimensional point cloud planeness computing method based on local optimization. The method comprises the following steps: selecting a measurement plane, obtaining the point cloud of the measurement plane, and performing grid division; performing planeness fitting on the point cloud in a single grid to obtain a fitting plane, and then computing the distance between each point in the grid to the fitting plane, performing descending order arrangement on absolute values of the distances, and finally eliminating the front points as noise points; and repeating the computing until all grids are traversed, and performing planeness fitting on the remaining point cloud to obtain the planeness of the measurement plane. According to the three-dimensional point cloud planeness computing method disclosed by the invention, in view of the characteristic that the noise points are generated easily in the optical measurement and the original data size is large, the grid division is performed on the point cloud of the measurement plane, the points with relatively large distances with the fitting plane in eachgrid are removed, and then the remaining points are fitted to obtain the planeness of the measurement plane, thereby avoiding the problem of uneven local distribution of the point cloud due to the noise points and making the computing be more accurate.

A hands-free digital push-to-talk device (102) includes a digital background noise suppressor (302), a digital voice activity detector (304), an audio buffer (306), as well as a decision handler (308), embedded inside the device's (102) digital signal processor (222). Audio is buffered until the decision handler (308) determines that speech is present on an audio stream fed to the voice activity detector (304). The decision handler (308) makes the decision by assigning weighted values to each voice activity detector (304) determination, the weighted value varying depending on the state of the device (102) and temporal distance from the present time.

The present invention discloses a method for automatic contour extraction and path generation of a bitmap. The method comprises the following steps of (1) reading the bitmap file, and storing the read data; (2) converting the bitmap file into a grey-scale map; (3) selecting a binarization algorithm to perform binarization processing; (4) performing contour extraction preprocessing; (5) performing contour extraction; and (6) obtaining cutting path generation. According to the method, the cutting path can be generated by the bitmap file without participation of professional designers or with participation of only a few professional designers, complexity of cutting path generation is reduced, generation efficiency and accuracy of the cutting path can be largely improved, cutting precision and automation degree are increased, and the method has good prospect of promotion and application.

The invention discloses a construction method of a noise type asphalt pavement. The construction method is characterized by including the steps: firstly, after construction of a roadbed and a base on the roadbed is completed, paving a lower asphalt layer on the base, paving an upper asphalt layer 4cm in thickness on the surface of the lower asphalt layer, and forming an initial pressing face on the upper asphalt layer by means of rolling of a tire roller, wherein the upper asphalt layer is made of stone matrix asphalt composed of lignocelluloses, asphalt, broken stones and mineral powder; secondly, uniformly spreading coarse-grain mixtures on the surface of the initial pressing face; thirdly, re-rolling the surface of the initial pressing face by the tire roller until set compactness and voidage are reached, and performing final rolling and polish for the pavement by the aid of a steel-wheel roller so as to form an asphalt surface; and finally, washing the asphalt surface with high-pressure water to achieve structural depth, and forming the rough pavement uniform in structure after descumming. By the aid of the method, pavement construction quality is greatly improved, and forming standardization of the noise type asphalt pavement is effectively guaranteed.

The invention relates to electromagnetic valves with wastewater throttling devices in the water treatment field, in particular to an electromagnetic valve provided with a wastewater throttling device and having anti-blocking and silencing functions. According to the technical scheme, the electromagnetic valve provided with the wastewater throttling device and having the anti-blocking and silencing functions is characterized in that one side of a valve body is provided with a containing cavity, a wastewater rod is inserted in the containing cavity, and a buffer cavity is formed between the middle of the wastewater rod and the inner wall of the containing cavity; and the valve body is provided with anti-blocking pin holes penetrating through a water inlet channel and the buffer cavity, the wastewater rod is provided with a wastewater hole penetrating through the buffer cavity and a water outlet channel, the aperture of the anti-blocking pin holes is small than that of the wastewater hole, and water in the water inlet channel can pass through the anti-blocking pin holes, the buffer cavity and the wastewater hole and then flows out from the water outlet channel. By the adoption of the technical scheme, the problems existing in the prior art are solved, and the novel electromagnetic valve provided with the wastewater throttling device can well prevent the wastewater hole from being blocked and effectively reduce noise.

A first transmission line for transmitting a first signal and a second transmission line for transmitting a second signal, which has the reverse phase of the first signal, are connected in series with a common mode choke coil. A third transmission line and fourth transmission line are each connected in series with the common mode choke coil, and transmit the first and second signals. A semiconductor device is connected in series with the third and fourth transmission lines, so as to transmit and receive the first and second signals. One end of a first terminator is connected in parallel with the first transmission line, and the other end is connected to the common mode choke coil. One end of a second terminator is connected in parallel with the second transmission line, and the other end is connected to the common mode choke coil. The noise eliminating capability of the common mode choke coil is increase by means of this structure.

Method for predicting properties of a cellulose-fibre-based product, a suspension of cellulose fibre being analysed during manufacture of the product by means of spectroscopic measurements (13, 15) in a selected spectrum in the wavelength range 200-25000 nm. According to the invention, each sample quantity is diluted (3, 4) prior to said analysis, following which one partial flow (6) is dewatered and dried (8) and used for said spectroscopic measurements (13, 15), while a second partial flow (5) of the diluted sample quantity is used for analysis (7, 14) of physical fibre data, by means of image analysis, each sample quantity analysed generating at least 100 data points from the spectroscopic measurement, and at least 50 data points with regard to said physical fibre data, in the form of one or more physical fibre property distributions, which data points are combined in multivariate data processing (16), for said prediction.

The invention discloses an agricultural park scene lane departure early warning control method based on an end-to-end convolutional neural network. The method comprises the following steps: collectinga real-time road image, transmitting the real-time road image to an efficient semantic segmentation convolutional neural network, and performing pixel-level marking; performing region-of-interest selection on the image marked by the neural network; extracting an edge image of the road image in the region-of-interest selection by adopting differential edge detection; obtaining a coordinate point set corresponding to edge pixels in the edge image in an image coordinate system; detecting a road edge fitting straight line from the coordinate point set corresponding to the edge pixel by adopting aHough transform straight line detection algorithm; obtaining an optimal fitting straight line through screening and fusion; and according to the optimal fitting straight line result, extracting the relative distance between the vehicle body and the left and right edges of the road in the real-time road image and the characteristics of the focus of the vehicle-mounted camera to distinguish the vehicle pose state, and carrying out corresponding vehicle body adjustment to realize centered driving of the vehicle.

The embodiment of the invention, which is applied to the technical field of information processing, discloses a parameter adjusting method of terminal equipment and terminal equipment. When a aluetooth module of the terminal equipment sends played audio data to a bluetooth headset or bluetooth vehicle-mounted unit by bluetooth connection, the terminal equipment determines whether a time interval for sending audio data by the bluetooth host module to a bluetooth control module in the bluetooth format is larger than a preset time value; if so, a certain functional parameter being a scanning parameter of another wireless communication module sharing a channel with the bluetooth module is adjusted, thereby eliminating a factor causing interruption, noises or breaking of audio data received from the bluetooth headset or bluetooth vehicle-mounted unit.

The invention discloses a fruit branch and leaf region obtaining method of red apple images in a garden. The method comprises steps of 1: acquiring apple RGB images; 2: obtaining fruit images: extracting R-G chromatic aberration images from the RGB images, successively carrying out corrosion operation, hole filling, small region removing and open operation on the R-G chromatic aberration images and finally carrying out threshold segmentation; 3: obtaining leaf images: subtracting the fruit images from the RGB images, extracting 2G-R-B chromatic aberration images from the subtracted images and carrying out small region removing and threshold segmentation; and 4: obtaining branch images: carrying out dynamic threshold segmentation on the R-G chromatic aberration images, adding the segmented images into the obtained fruit images to obtain fruit addition images, then subtracting the fruit addition images and the leaf images from the RGB images, extracting R-G chromatic aberration images from the subtracted images and then carrying out open operation, small region removing and threshold segmentation. According to the invention, the method lays a foundation for precise fruit identification and positioning, effective obstacle avoidance and successful picking for following picking robots.

A through substrate which comprises a silicon substrate (10) having a through hole (19) penetrating a front surface (11) and a back surface (12), a oxidized silicon film (13) being provided along the inner wall surface of the through hole (19), layers (14, 15) comprising Zn and Cu, respectively, being formed on the inner wall surface of the oxidized silicon film (13), and a Cu plating layer (18) which has been grown from a Cu seed layer (17) along the inner wall surface of layers (14, 15) comprising Zn and Cu, respectively, via an insulating layer (16) between them. The above through substrate can provide a through electrode capable of avoiding the noise due to the cross talk.

A two-way soft closing device is mounted in a sliding door and has a sliding door track, two soft closing activation triggers, two roller carriers and a soft closing device. The sliding door track is an elongated member. The soft closing activation triggers are mounted movably in two ends of the sliding door track respectively. The roller carriers are mounted slidably in the sliding door track and are spaced apart at an interval. The soft closing device is mounted in the sliding door track, is located between the roller carriers and is mounted firmly in one of the roller carriers. Therefore, the soft closing device absorbs a collision between a sliding door plate and a doorframe when the sliding door is opened or is closed. Meanwhile, the two-way soft closing device eliminates noise coming from the collision and prevents an elder or a little child from hit by the sliding doorplate.

The invention discloses a lifting mechanism. The lifting mechanism comprises an installation frame, a driving motor, a screw rod and a nut, wherein the driving motor is arranged on the installation frame; the screw rod is driven by the driving motor to rotate; the nut is arranged on the screw rod; a load of the lifting mechanism is installed on the nut and does synchronous lifting motion along with the nut. The lifting mechanism is characterized by also comprising a one-way damping device, wherein under the falling state of the lifting mechanism, the one-way damping device can generate a damping force to the screw rod, and the damping force is used for offsetting and eliminating the influence of the load on the screw rod. The invention also discloses a range hood adopting the lifting mechanism. The lifting mechanism and the range hood disclosed by the invention have the advantages that the lifting mechanism adopts a screw-rod and nut mechanism, the load is installed on the nut and does synchronous lifting motion along with the nut, the one-way damping device can generate a damping force to the screw rod, and the damping force is used for offsetting and eliminating the influence of the load on the screw rod, so that the jittering and abnormal noise in the motion process of the lifting mechanism can be eliminated, and the unstable phenomenon of weightlessness caused by self weight of the load in the falling process of the mechanism is avoided.

There is provided an operating method of an image sensor including: storing a first charge from an optoelectronic element to a pixel buffer circuit within a first exposure period; transferring the first charge from the pixel buffer circuit to a first storage circuit within a second exposure period; storing a second charge from the optoelectronic element to the pixel buffer circuit within the second exposure period after the first charge is transferred; transferring the second charge from the pixel buffer circuit to a second storage circuit; and comparing charges stored in the first storage circuit and in the second storage circuit to output an analog image signal.

The invention relates to a gain control method and device. The method includes the steps of collecting an output signal of an amplifier; when it is determined that the output signal is at a zero crossing point or is located in an adjacent section of the zero crossing point, controlling an AGC control circuit to output a corresponding gain signal, and controlling gains of the amplifier through a gain control circuit. By means of the scheme, the music playing quality can be improved.

The invention discloses an audio processing method and apparatus, a sound box system and a storage medium. The audio processing method specifically comprises the following steps: dividing collected environmental sound according to preset bands, and marking the strength values of the environmental sound of the bands, after a recording instruction is received, dividing audio signals picked up by a microphone according to the preset bands, marking the strength values of the audio signals of the bands, then comparing calibration values of the bands of the environmental sound and the calibration values of the bands of the audio signals with a preset reference sound calibration value, judging audio main sound, all environmental noise and recorded environmental noise, finally performing gain andencoding processing on the audio main sound, and performing attenuation and encoding processing on the all environmental noise and recorded environmental noise, outputting audio digital signals to a back end of the sound box system for playing. The problem that a back end power amplifier cannot reduce the noise, resulting in worse playing effect, is solved, and the playing sound effect of the sound box system is improved.

The invention relates to a logarithmic cylinder roller, whose generatrix of support surface is a logarithmic curve. The middle of said generatrix is protruded and two ends of generatrix are connected to two chamfering arc. The invention can eliminate the edge effect in the contact of cylinder roller to uniform distribute the contact pressure along the axes and improve the bearing ability and rigidity of roller contact couple, and the end part of cylinder roller can form a lubricate film to reduce the friction and abrasion in rolling. The better lubricate property can reduce the vibration and noise, improve the stability, improve the bearing ability and the application for worse working condition. The invention can apply the rolling needle too and other linear contact mechanics elements.

The invention discloses a method and device for extracting a magneto-rheological polishing removal function. The method comprises the following steps of acquiring the surface shape detection data of a substrate element before and after spot collection, wherein the edge of the substrate element is provided with a position mark; according to the surface shape detection data before and after speckle collection and the position mark of the substrate element, calculating to obtain the residual data; carrying out median filtering processing on the residual data; for the residual error data after filtering processing, adopting an elliptic arc to approach the contour of a removal function, and generating a plurality of removal functions containing the same parameters; and executing the data offset on the plurality of removal functions containing the same parameters, averaging, and extracting the removal functions. According to the method, the magneto-rheological polishing removal function can be accurately extracted, and the accuracy of the removal function and the magneto-rheological machining precision are improved.

The invention discloses a high-throughput functional gene screening method and system for cell phenotypic image quantitative analysis, and belongs to the technical field of gene screening. The methodcomprises the steps of using a full-automatic fluorescence microscope to shoot a cell image of the phenotype to be screened and a cell image without the phenotype to be screened; respectively converting the images into a to-be-screened phenotype and a black-white binary image without the to-be-screened phenotype; then segmenting into an image containing a single phenotypic cell to be screened andan image containing a single phenotypic cell without being screened; taking the corresponding part of the image containing the single to-be-screened phenotypic cell in the cell image of the to-be-screened phenotypic cell as a positive training set, and taking the corresponding part of the image containing the single to-be-screened phenotypic cell in the cell image of the to-be-screened phenotypiccell as a negative training set to obtain a final model capable of identifying the cell phenotypic; and carrying out final model identification on the image of the knockout or overexpression cell of the to-be-screened gene to obtain the relevance between the to-be-screened gene and the phenotype. The method improves the screening efficiency and accuracy.