109 results about "Filter window" patented technology

An intra frame prediction method and apparatus based on interpolation filtering, a video encoding method and apparatus, a video decoding method and apparatus, and a system. The intra prediction method based on interpolation filtering comprises: determining, based on a determined filtering window and a reconstructed value of a sample in a reconstruction area neighboring to a current block, a filtering coefficient used to perform interpolation filtering on the current block; and performing interpolation filtering based on a sample value of a reference sample neighboring to a current sample and the filtering coefficient, to determine a predicted value of the current sample.

The invention relates to a three-waveband medium-long wave infrared optical filter and a preparation method, the three-waveband medium-long wave infrared optical filter comprises a front film system, a substrate and a back film system, and the substrate is located between the front film system and the back film system; the front film system is formed by alternately overlapping germanium layers and zinc selenide film layers; and the back film system is formed by alternately overlapping germanium layers and zinc selenide film layers. According to the invention, the light energy of three wave bands of 3780-4000 nm, 4500-4900 nm and 7850-9500 nm is improved, the luminous flux of other wave bands is well inhibited, the technical problem of infrared filtering windows in some special infrared spectrum detection systems is solved, and a new scheme is provided for the application of the infrared spectrum detection systems.

The invention relates to the technical field of underwater facility measurement, and provides an underwater facility defect geometric feature measurement method and system, a medium and equipment, and the method comprises the steps: calculating the actual sound velocity based on the water temperature, water pressure and salinity, correcting a sonar point cloud, calculating the sound velocity change rate, and adjusting the size of a filtering window, performing smooth filtering on the corrected sonar point cloud; for the surface image, calculating an image dark channel, after smoothing the dark channel image through scattering filtering, calculating transmissivity, fusing a scattering coefficient, removing surface image scattered light, and correcting the scattering coefficient through turbidity; based on the salinity and the water temperature, the underwater refraction angle of the laser is calculated, and the three-dimensional scanning point cloud is corrected; registering and fusing the point clouds to generate a facility surface point cloud model; based on the surface image, texture information of the facility surface is extracted, after the texture information is mapped to the facility surface point cloud model, a defect area is segmented, and geometric feature parameter extraction is carried out. And the detection reliability in a complex water area environment is improved.

The invention relates to a dual-waveband medium-long wave infrared optical filter based on a germanium / zinc sulfide asymmetric film system and a preparation method of the dual-waveband medium-long wave infrared optical filter. The dual-waveband medium-long wave infrared optical filter comprises a front film system, a substrate and a back film system, the front film system and the back film system are formed by alternately overlapping germanium layers and zinc sulfide film layers; the average transmittance of the infrared optical filter in the wave band of 3800 nm to 4800 nm and the wave band of 8000 nm to 10000 nm is not lower than 91%. According to the two-waveband medium-long wave infrared optical filter disclosed by the invention, 3800 to 4800 nm amp is realized; according to the infrared filtering window, the light energy of medium waves and long waves of 8000-10000 nm is improved, the luminous flux of other wavebands is well inhibited, the technical problem of the infrared filtering window in some special infrared spectrum detection systems is solved, and a new scheme is provided for infrared application.

The invention discloses an element stripping method and device based on multispectral data, equipment and a medium, and relates to the field of remote sensing image processing, and the method comprises the steps: collecting a multispectral image of a target region; a self-adaptive window filtering algorithm is adopted to pre-process the multispectral image to obtain a multispectral enhanced image, and the self-adaptive window filtering algorithm calculates a window adjustment coefficient based on noise density, gray variance and gradient magnitude; spectral features and texture features are extracted from the multispectral enhanced image; fusing the spectral features and the texture features, and constructing fusion features of each pixel in the multispectral enhanced image; determining a ground feature category of each pixel based on the fusion features to strip a plurality of vectorized layers containing different elements; according to the method, the three indexes of the noise density, the gray variance and the gradient magnitude can be fused, the window adjustment coefficient is obtained, dynamic self-adaptive change of the filtering window size is achieved, and higher-precision data support is provided for subsequent element extraction.

The invention discloses a multi-frame reconstruction multispectral imaging system based on dynamic space-time joint coding. The system comprises a filtering window module used for limiting the spectral range of imaging; the spatial code modulation module is used for providing spectral codes in a spatial dimension; the time coding modulation module is used for carrying out inter-frame modulation on the spectrum codes provided by the space coding modulation module; the optical imaging module is used for imaging the light field modulated by the time coding modulation module on the color acquisition module; the color acquisition module is used for carrying out further spectral coding on the light field to obtain a measurement image; a calculation reconstruction module; the method is used for recovering a high-dimensional multispectral image data cube by using a measurement image. The method has the advantages of being light in weight, low in cost, high in underqualitative relieving capacity, efficient and stable in reconstruction method, high in adaptability and practicability and the like.

The invention relates to the technical field of data processing, in particular to an in-transit monitoring management method and system for logistics transportation, and the method comprises the steps: taking a mean value of adjacent data of any data in a vehicle speed data sequence as an ideal value of the data, calculating the noise intensity of the data, and correcting the noise intensity, the noise intensity represents the difference between the data and the ideal value, screening the corrected data with relatively low noise degree from the adjacent data of the preset filtering window of the data, adding the data into the filtering window to obtain the filtering window updated for the first time, and repeating the updating process on the basis of the updated filtering window, and calculating the optimization degree of the filtering window updated each time, performing mean filtering on the data by using the filtering window of which the optimization degree meets a preset screening condition, and monitoring the transportation process based on the filtered data. According to the invention, the accuracy of the filtered vehicle speed data is ensured, and accurate monitoring of the logistics transportation process can be realized.

The invention discloses an insect body echo self-adaptive filtering method based on micro-motion phase self-correlation matching, belongs to the technical field of radars, and is used for solving the problems that an existing filtering mode of the main Doppler frequency of an insect target has certain limitation, full-scene adaptation is difficult to realize by a fixed window length, and the filtering efficiency is low. And low-frequency filtering causes that the insect target in low-frequency vibration flight cannot be effectively extracted and identified. The method comprises the following steps: carrying out instantaneous Doppler frequency model modeling processing on an original insect echo signal to obtain a target instantaneous Doppler frequency model; performing smoothing processing based on different sliding window lengths on the target echo phase to obtain a micro-motion phase after the main Doppler component is filtered; carrying out self-correlation function calculation under a related self-adaptive window length on the micro-motion phase after the main Doppler component is filtered out; determining an optimal filtering window length corresponding to the current insect phase signal; and performing slow-time signal echo compensation processing on the current insect phase signal to obtain a micro-motion signal of the current insect.

The invention relates to the technical field of navigation and positioning, in particular to an unmanned ship attitude-combined positioning mapping method, which comprises the following steps of: acquiring laser point cloud data and ship attitude information; constructing a dynamic height filtering window, and eliminating water surface reflection noise points; density clustering parameters are adjusted, and spatial isolated point clusters are eliminated; positioning and mapping are realized through tight coupling optimization; in the prior art, a fixed threshold filtering scheme is adopted, and inshore structure reservation and long-distance noise elimination are difficult to consider, so that wharfs, reefs and other characteristics in lake surface mapping are mistakenly deleted or a large number of water surface false points are remained; according to the scheme, a ship attitude adaptive dynamic height filtering window is innovatively provided, the filtering boundary is intelligently scaled in a wave disturbance environment through fusion of radar real-time height, point cloud statistics and a horizontal distance segmentation adjustment mechanism, specular reflection false points and wave false point clusters are accurately eliminated, and nearshore topographic features are completely reserved; and the structural integrity and the space authenticity of the water surface map are obviously improved.

A deep learning segmentation method for asphalt mixture CT images relates to the technical field of aggregate image segmentation. CT images of asphalt mixtures are collected and voids are removed. Image equalization is performed using a gamma beam hardening correction algorithm. A filter window is selected, and an adaptive bilateral filtering denoising algorithm based on local image information is used to denoise the image. The U-Net model is improved, with the Inception convolution module replacing the standard convolution operation and residual connections replacing skip-layer connections. A spatial attention mechanism is introduced, and a joint loss function is used. Image samples are selected, aggregates are labeled, and data augmentation is used to increase the sample size to form a training set for model training. After training, this set is used to segment other images. By equalizing and denoising CT images and using the improved U-Net model, the method can achieve accurate segmentation of aggregate-mortar boundary information, avoiding aggregate adhesion issues.

One variation of a respirator device includes a shroud center portion, a filter assembly, and a gasket. The shroud central portion defines: a transparent central region configured to align with a user's nose and mouth; and a filter window laterally offset from the transparent central region. A filter assembly includes a filter media and a seal that: surrounds a perimeter of the filter media; the filter assembly is positioned within the filter window and defines a central groove configured to mate with an edge of the filter window to instantaneously retain the filter assembly within the filter window. The pad is disposed on the back side of the central portion of the lens and is configured to: seal against the face of the user; and positioning the lens center portion and the filter assembly offset from the user's face to form an airflow channel to direct airflow through the user's smooth cheek region.

The embodiment of the application provides an interferometric phase denoising method and device, relates to data processing technology, and the interferometric phase denoising method comprises the following steps: performing multi-scale frequency decomposition on an original interferometric phase map based on a preset frequency domain filtering window to obtain hierarchical phase signals containing different frequency components; calculating residual components of adjacent levels according to the difference between principal component signals of adjacent levels to obtain residual domain signals corresponding to each level respectively; performing denoising processing on the residual domain signals by using a pre-trained denoising model to obtain denoised residual domain signals corresponding to each level respectively; and performing phase reconstruction on the denoised residual domain signals corresponding to each level respectively to obtain a target interferometric phase map.The application can effectively maintain the authenticity and integrity of the phase structure while suppressing noise, effectively reduce the generation of artifacts and improve the convergence stability, and ensure the reliability of phase unwrapping and subsequent deformation inversion.

The invention relates to a shooting method and device, electronic equipment and a storage medium. The shooting method comprises the following steps: in response to shooting of shooting equipment in a motion process, obtaining an image frame sequence collected by the shooting equipment; performing motion estimation on feature points of adjacent frames in the image frame sequence to obtain inter-frame motion vectors of the adjacent frames, and determining the length of a filtering window based on the inter-frame motion vectors; the inter-frame motion vector of the adjacent frames comprises an inter-frame horizontal motion vector and an inter-frame vertical motion vector, and the filtering window length comprises a horizontal direction filtering window length and a vertical direction filtering window length; based on the horizontal direction filtering window length and the vertical direction filtering window length, determining a target filtering window, and based on the target filtering window, performing smooth filtering on the displacement change vector of the shooting equipment between the adjacent frames; and carrying out image stabilization processing on the adjacent frame images based on the displacement change vector after smooth filtering, and outputting the image after image stabilization processing as a shot image.

This invention belongs to the field of nasal alar pulse wave signal processing technology, and provides a method, apparatus, medium, and wearable device for nasal alar pulse wave signal processing. The method includes: extracting the respiratory cycle based on the respiratory signal; determining the length of the mean sliding filter window to filter out the DC component of the original nasal alar pulse wave signal, obtaining a filtered nasal alar pulse wave signal; extracting the amplitude of the AC component of the filtered nasal alar pulse wave signal; adjusting the current of the light source element corresponding to the pulse wave signal sensing element to make the amplitude of the AC component of the filtered nasal alar pulse wave signal greater than or equal to a preset amplitude threshold; further filtering the nasal alar pulse wave signal using an adaptive weighted moving average filtering algorithm; and determining whether a nasal alar pulse wave signal meeting preset quality requirements is obtained based on a comparison between the noise figure of the further filtered nasal alar pulse wave signal and a preset noise figure threshold.

The utility model discloses a wafer transportation box, which is characterized in that an upper cover and a bottom shell are respectively provided with a filter window which is integrally formed, and a filter membrane is arranged in an inner interlayer of each filter window, so that air flow in the wafer transportation box is realized, the humidity of the air is ensured to be consistent with that of external air, and the cleanliness in the wafer transportation box is ensured.

The invention relates to the technical field of data denoising, in particular to a data processing system for a cosmetic medical dressing preparation process, which comprises a processor and a memory, the processor executes a computer program stored in the memory to implement the following steps: according to an outlier degree index value of the process monitoring data and a noise degree index value of each process monitoring data in the filtering window of the process monitoring data, performing weight distribution on the process monitoring data in the filtering window of the process monitoring data, obtaining a target weight value of each piece of process monitoring data in a filtering window of the process monitoring data; and according to the data in the filtering window of the process monitoring data and the target weight value of each process monitoring data in the filtering window, performing weighted mean filtering on the process monitoring data sequence to obtain filtering data. Noise can be completely filtered out as much as possible, and meanwhile distortion of non-noise data or key data can be avoided as much as possible.

The invention discloses an ADC50Hz or 60Hz noise filtering method based on a software algorithm, and the method comprises the steps: carrying out the initialization configuration, and collecting original data; preprocessing and acquiring detection data; an improved Goertzel algorithm is used to acquire frequency spectrum parameters; calculating a three-level extraction factor, and sampling the data output by the ADC to obtain sample data; after forward and reverse compensation is carried out on the sample data, pulse noise suppression of a two-dimensional window is carried out on pulse noise; calculating a transmission quality index, and adjusting the size of a filtering window according to the transmission quality index; and carrying out amplitude calibration on the filtering result by using an improved comb algorithm, and outputting a standard signal after noise reduction processing. On one hand, the noise reduction processing scheme does not need to additionally add a filter circuit, so that the circuit size and the hardware cost are reduced; and on the other hand, the parameters in the method can be adaptively adjusted according to the frequency spectrum parameters, so that the adaptability is improved.

The application discloses a kind of low-illumination scene-oriented adaptive image noise reduction processing method, it is related to digital image processing technical field.The steps include: step S100: low-illumination image is collected and preprocessed, preprocessing includes dark current correction, bad point repair and lens distortion correction;By using back-illuminated CMOS sensor to collect image and execute dark current correction, bad point repair, the preprocessing operation of lens distortion correction, in combination with average brightness and gray standard deviation division low-illumination grade, according to noise intensity and distribution characteristics identify noise type, again based on preset parameter mapping table adaptive matching filter window size, noise reduction intensity weight and detail retention threshold technical means, it is difficult to eliminate in the process of low-illumination image acquisition fixed noise and system error, traditional noise reduction algorithm parameter fixed cannot adapt to different low-illumination grade and noise type, the problem of over-reduction or noise reduction deficiency, image edge detail loss is serious in noise reduction process.

The invention relates to the technical field of performance detection, in particular to an intelligent measurement switch performance detection method, device and system, and the method specifically comprises the steps: obtaining various types of electrical parameter data of an intelligent measurement switch under each working condition, calculating the difference between the electrical parameter data and a corresponding rated value, recording the difference as an instantaneous deviation value, constructing a first characteristic value of each electrical parameter under each working condition based on the difference of the instantaneous deviation values under different working conditions, and analyzing the overall deviation under each working condition; the interference influence characteristic coefficient of each electrical parameter is constructed through the difference between the instantaneous deviation and the overall deviation of each electrical parameter, and the filtering window size of each electrical parameter under each working condition is constructed by combining the data quantity between adjacent zero point moments in a sequence formed by the instantaneous deviation values of all moments so as to carry out filtering. The performance test of the intelligent measurement switch is carried out through the filtered data; the interference influence of different interference sources under different working conditions is reduced, and the accuracy of intelligent measurement switch performance test analysis is improved.

The application relates to the technical field of inductance alignment control, in particular to a super-mini inductance high-precision alignment device and a control method thereof. The method comprises the following steps: images of a substrate and a mask plate are collected from preset angles respectively, and are recorded as collected images; for any collected image and a standard template image thereof, feature coefficients in horizontal and vertical directions of the collected image and the standard template image thereof are obtained, and the collected image is divided into small blocks; adjustment coefficients of filter processing in the small blocks of the collected image are obtained, the size of a filter window in the small blocks of the collected image is obtained, and filter processing is performed on the collected image; the alignment of the substrate and the mask plate is measured, if the alignment does not meet the standard, the relative positions of the substrate and the mask plate are adjusted until the alignment meets the standard. The application aims to improve the accuracy of subsequent alignment mark position matching by accurately preprocessing the substrate and the mask plate, and to improve the high-precision alignment control precision of the super-mini inductance.

The invention provides a depth map processing method and system and a three-dimensional image reconstruction method. The depth map processing method comprises the following steps: traversing each pixel point of a depth map, and initializing a mask; respectively setting filtering window radiuses in the row direction and the column direction based on the resolution of the depth map; adopting expansion convolution to determine sampling points of a filtering window; judging the effectiveness of each sampling point in the filtering window, gradually reducing the calculation range of the adjacent points, determining pixel points corresponding to the pseudo edge point set according to the judgment of the adjacent points, and resetting the mask; and in combination with morphological operation, the depth map is processed, and a final image with accurate three-dimensional data is obtained. According to the error information of the depth map data output by the structured light camera and the line spectrum confocal sensor, the depth map processing method is provided, so that the three-dimensional surface reconstructed by the output final image can output accurate three-dimensional data.

The invention relates to the technical field of vehicles, in particular to a vehicle window anti-pinch method and device, electronic equipment, a storage medium and a program product.The method comprises the steps that the motor rotating speed when a vehicle window ascends is collected; determining a target rotating speed according to the motor rotating speed, and matching an anti-pinch monitoring data window and current filtering time according to the target rotating speed; and based on a preset anti-pinch force, performing anti-pinch control according to the target rotating speed, the anti-pinch monitoring data window and the current filtering time. Therefore, by adding the car window self-adaptive learning algorithm, correcting the current filtering window in real time based on the rotating speed of the motor and dynamically adjusting the anti-pinch monitoring response time, the problems of deviation between the actual anti-pinch force and the target, wrong anti-pinch or non-anti-pinch and the like caused by aging of a mechanical structure in the related technology are solved, and the safety and reliability of the car window anti-pinch system are improved.

The invention provides a trailing point cloud filtering processing method, device and equipment and a readable storage medium. The method comprises the following steps: acquiring a laser point cloud generated in the working process of a rotating mirror type laser radar; edge area detection of two detection objects is carried out on the laser point cloud through a sliding filtering window, and whether a trailing point cloud exists in the sliding filtering window is judged; if the trailing point cloud exists, segmenting the trailing point cloud based on a distance threshold, and generating a to-be-processed trailing point cloud with a close-shot target distance and a long-shot target distance; whether the to-be-processed trailing point cloud is directly deleted or not is judged, if yes, the to-be-processed trailing point cloud is directly deleted, and if not, restored filtering processing is carried out on the to-be-processed trailing point cloud, and then the filtered point cloud is output. According to the method, the object contour edge is extracted by sliding the filtering window, the recognition accuracy and processing efficiency of the trailing point cloud are improved, the point cloud information of the object edge is reserved by restoring the trailing point cloud, and the point cloud quality is improved.

The present application relates to a kind of three-waveband middle-long wave infrared filter and preparation method, including front film system, substrate and back film system, the substrate is located between the front film system and back film system;The front film system is composed of germanium layer and zinc selenide film layer alternately superimposed;The back film system is composed of germanium layer and zinc selenide film layer alternately superimposed.The present application realizes the light energy promotion of 3780-4000nm, 4500-4900nm, 7850-9500nm middle-long wave three wavebands, and very good inhibits the light flux of other wavebands, solves the technical problem of infrared filter window in some special infrared spectrum detection system, provides new scheme for the application of infrared spectrum detection system.

The application relates to the technical field of InSAR, in particular to a landslide disaster deformation monitoring method, system and equipment based on InSAR. The method comprises the following steps: collecting SAR images of a target area in real time, taking the first collected SAR image and any one of the remaining SAR images except the first one as master and slave images respectively, and generating multiple interference fringe images; analyzing interference phase values in the interference fringe images to adjust the window size when filtering the interference fringe images; and the filtering result is used for subsequent acquisition of a landslide disaster deformation monitoring result of the target area. The application aims to effectively avoid the real ground deformation information of landslide disaster deformation in the target area from being excessively smoothed by adaptively adjusting the filtering window size, and improve the reliability of the landslide disaster deformation monitoring result of the target area.

The application provides a local adaptive tone mapping method, comprising the following steps: S1, inputting an image and extracting image brightness; S2, performing logarithmic transformation on the image brightness; S3, calculating the image variance D of logY, and the window size is M*M; S4, adjusting the filtering window size N and the filtering coefficient of the bilateral filter according to the variance D; S5, taking the output of the bilateral filtering as the base layer logYBase, and the difference between logY and the base layer is the detail layer logYDetail, logYDetail=logY-logYBase; S6, compressing the base layer, enhancing the detail layer, and adding the base layer and the detail layer to obtain the enhanced brightness graph: logYEn=f(logYBase)+alpha*logYDetail, wherein f(·) represents a mapping curve, which is usually similar to a gamma shape, and a user can adjust it according to requirements, alpha represents an enhancement coefficient, and the greater the value, the stronger the image details; S7, converting the logarithmic domain brightness back to the linear domain: YEn=10 logYEn ^logYEn; S8, restoring the image color according to the RGB three-channel values and Y of the original image and the enhanced YEn to obtain a tone-mapped image.

This invention relates to a dual-band mid-to-long-wave infrared filter based on a germanium / zinc sulfide asymmetric film system and its fabrication method, comprising a front film system, a substrate, and a back film system; both the front and back film systems are composed of alternating layers of germanium and zinc sulfide; the infrared filter exhibits an average transmittance of not less than 91% in the 3800nm-4800nm ​​and 8000nm-10000nm wavelength bands. The dual-band mid-to-long-wave infrared filter of this invention achieves enhanced light energy in both the 3800-4800nm ​​and 8000-10000nm mid-to-long-wave bands, while effectively suppressing light flux in other bands. It solves the technical problems of infrared filter windows in some special infrared spectral detection systems, providing a new solution for infrared applications.

The invention discloses a method and a system for searching and correcting dead pixels, electronic equipment and a storage medium, which are used for efficiently processing the dead pixels. The method for searching and correcting the dead pixels comprises the following steps: acquiring a target dead pixel coordinate information list, and creating a row index table and a column coordinate ordered queue according to the target dead pixel coordinate information list; performing line caching on the target image according to the size of a preset filtering window to obtain line pixel data of the target image; searching in a row index table according to row coordinate information of the row pixel data to obtain the number of dead pixels of each row of pixels in the target image; according to the number of the defective pixels of the indexed current row pixel and the corresponding row coordinate information, reading column coordinate information of the defective pixels of the row pixel which is the same as the corresponding row coordinate information in the target image from the column coordinate ordered queue, and generating defective pixel coordinates of the target image in combination with the corresponding row coordinate information; and performing correction according to the dead pixel coordinates of the target image and a preset filtering window.

The embodiment of the invention provides a de-noising result determination method and a chroma de-noising system. The method comprises the steps that cached image data are acquired from N line caches through a first-stage filter, first-stage chromaticity denoising is conducted on each pixel point of an image to be denoised according to the image data, a first denoising result is obtained, N is an odd number larger than 1 and smaller than a first threshold value, and the image to be denoised comprises the image data; according to the noise level of the image to be denoised, post-filtering is carried out on the first denoising result through a post-filter, and the post-filter comprises a second-stage filter cascaded with the first-stage filter. Through the embodiment of the invention, the problem of high hardware resource consumption caused by the fact that a filtering window is adopted for filtering processing in a denoising method in the related technology can be solved, and the effect of realizing chroma denoising under low cache is achieved.

The embodiment of the invention provides an interferometric phase denoising method and device, and relates to the data processing technology, and the method comprises the steps: carrying out the multi-scale frequency decomposition of an original interferometric phase diagram based on a preset frequency domain filtering window, so as to obtain a hierarchical phase signal containing different frequency components; according to the difference between the principal component signals of the adjacent hierarchies, residual components of the adjacent hierarchies are calculated, and residual domain signals corresponding to all the hierarchies are obtained; de-noising processing is carried out on the residual domain signals by using a pre-trained de-noising model to obtain de-noised residual domain signals corresponding to each level; and performing phase reconstruction on the denoised residual domain signal corresponding to each level to obtain a target interferometric phase diagram. According to the invention, the authenticity and integrity of the phase structure can be effectively maintained while the noise is suppressed, the generation of artifacts is effectively reduced, the convergence stability is improved, and the reliability of phase unwrapping and subsequent deformation inversion is ensured.