72results about How to "Smooth border" patented technology

The invention discloses a video image skin-tone detection method based on multi-color gamut selective morphology processing and is characterized in that: the mathematical morphology processing technology is introduced, video image data is input and passes through two parallel skin-tone detection modules based on RGB color gamut and YUV color gamut, pixels in the image are analyzed, selected and roughly judged whether the pixels are skin-tone pixels of the human body respectively in an RGB color space and a YUV color space under the control of a register operation management module MCU, the detection result is made a comprehensive judgment and analysis by a comprehensive judgment module for skin-tone detection, then a binary image obtained through the judgment is made a selective morphology filtering based on the content information in a selective morphology filtering module, finally the detected skin-tone is protected or processed under the control of the register operation management module MCU according to the need of the user to obtain the image which meets the visual habit of human eyes.

The invention discloses a deformable context coding network model and a liver and liver tumor segmentation method, which can accurately determine the contour positions of a liver and a liver tumor, realize more accurate liver and liver tumor segmentation, have a wide application prospect, and are suitable for popularization and application. According to the network model of the invention, deformable convolution is utilized to enhance the feature representation capability of a traditional encoder, help the traditional encoder to learn a convolution kernel with adaptive spatial structure information, and eliminate interference of different sizes of liver tumor positions; according to the method, the global feature information in the image is coded by using the Ladder spatial pyramid poolingmodule for extracting the multi-scale context information, so that the contour positions of the liver and the liver tumor are determined more accurately, more accurate liver and liver tumor segmentation is realized, and the method has a wide application prospect.

The invention relates to an image-based digital oil painting canvas automatic generation method. With the image-based digital oil painting canvas automatic generation method adopted, any images can be automatically converted to digital oil painting canvases, and the movement direction and size of a brush can be reasonably prompted. The method comprises the following steps that: firstly, a statistical filtering method is provided so as to remove boundary sawteeth generated by Mean-Shift image segmentation, so that smooth boundary lines can be obtained; secondly, representative colors are selected through adopting K means clustering, and an adaptive digital display adjustment method is provided, and therefore, the clarity and visibility of the display of corresponding digits can be ensured, and the digits do not overlap with boundaries; the direction field of an image space is calculated, so that streamlines for prompting the movement direction and size of the brush can be generated; and finally, the above boundary lines, color digits and brush streamlines are synthesized, and a digital oil painting canvas consistent with inputted image content can be generated.

The invention relates to a pixel array, electro optical device, an electric apparatus and a method of driving the pixel array. In the pixel array, pixels are two-dimensionally arranged, each of the pixels including a subpixel of the first color having a highest luminosity factor, a subpixel of the second color and a subpixel of the third color having a lowest luminosity factor. A circuit element in each of subpixels of the first color to the third color in each of the pixels is arranged in a row direction. A light emitting region of a subpixel of the first color and a light emitting region of a subpixel of the second color are arranged in the first direction inclined to the row direction. A light emitting region of a subpixel of the third color is arranged in the second direction orthogonal to the first direction, with respect to the light emitting region of the subpixel of the first color and the light emitting region of the subpixel of the second color. The data line extends and is connected to circuit elements in the sub pixels. Data lines connected with the circuit elements in the first color and the light emitting region of the subpixel of the second color, data lines connected with the circuit elements in the second color and the light emitting region of the subpixel of the third color and data lines connected with the circuit elements in the third color and the light emitting region of the subpixel of the first color are arranged repeatedly.

The invention discloses a remote sensing sub-pixel mapping method based on the spatial distribution characteristics of features, comprising the following steps: first, classifying remote sensing images into an area pattern, a line pattern and a point pattern according to the spatial geometric characteristics of features; then, under the hypothesis of spatial dependence, using a sub-pixel mapping method based on vector boundary to process area features, using a line feature template sub-pixel mapping method to process line features, and using a sub-pixel mapping method based on space pattern consistency matching to process point features; and finally, embedding the sub-pixel mapping results of the three spatial patterns to get a sub-pixel map of the images. According to the invention, a sub-pixel mapping theoretical model capable of processing point, line and area features at the same time and based on the spatial distribution characteristics of features is built theoretically, and the simulation precision is high.

Disclosed are an Al-Mg-Zn-Mn aluminum alloy welding wire and a preparation method thereof. The welding wire and the method belong to the technical field of aluminum alloys. The welding wire comprises,by mass, 0.5-3.0% of Zn, 4.5-5.5% of Mg, 0.4-1.0% of Mn, 0-0.15% of Ti, 0-0.15% of Cr, 0-0.25% of Fe, 0-0.25% of Si, 0-0.15% of the total amount of other impurity elements, 0-0.05% of single impurityelement and the balance Al. The preparation method comprises the steps of smelting, casting, preprocessing after heating to 200-400 DEG C and performing heat preserving for 2-6 hours, extruding afterperforming heat preserving under the temperature of 440-470 DEG C for 2-4 hours, rolling after heating to 480-550 DEG C and performing heat preserving for 2-4 hours, in-process annealing and bright-cold drawing. According to the method, high-temperature and long-duration uniform processing is not needed, energy is saved, processing procedures of welding wire are shortened, wire braking during drawing is reduced, yield of drawing is improved, production cost is reduced, and metal microstructure can be refined, so that weld metal has natural aging strengthening capability to improve the performance of weld joints.

The invention discloses a method for realizing cerebrovascular image identification by using fast boundary tracking. The method comprises the following steps: obtaining an original cerebrovascular image needed to be identified and performing pretreatment on the original cerebrovascular image; performing gradient treatment on the original cerebrovascular image; using a pixel point with the biggest gradient value in the image after the gradient treatment as a starting pixel point to perform tracing judgment on boundary points; and repainting all the obtained boundary points into an image which is the needed processed cerebrovascular image. The method is characterized in that three adjacent points of the prior point which is far from the last point are selected as candidate nodes, and the next adjacent point is determined by distributing different weights according to the importance differences of position information of the nodes so as to obtain a boundary tracing image of the cerebrovascular image. The boundary image extracted by the method has distinct outline, low noise, high precision, and smooth boundary, improves the quality of boundary extraction, and reduces the computation time greatly; besides, the method is applied to a system for computer-aided diagnosis of cerebrovascular diseases, and improves the accuracy of a computer in judging cerebrovascular pathological change parts and types.

The invention discloses a suppression method for partial discharge at a flange of a GIS/GIL supporting insulator. According to the invention, reduction of electric field intensity in a flange-side local air gap area is taken as an optimization target, and optimal space distribution of flange-side insulation internal dielectric parameters of supporting insulator is solved by using a variable density or level set algorithm; and extracting the geometric contour of a high-dielectric region by using an image segmentation algorithm, and acquiring a CAD drawing of the geometric shape of the high-dielectric region through a parameter modeling mode. According to the invention, dielectric functional gradient material distribution is introduced in consideration of the mechanical properties and interface bonding strength of a cast workpiece, the structural design of a high-dielectric area workpiece is improved; a high-dielectric composite material is prepared in a high-dielectric filler/polymer blending mode, and manufacturing of the workpiece is completed through 3D printing; and finally, the high-dielectric workpiece is put into a traditional epoxy pouring metal mold, and thermocuring epoxy resin is poured to complete manufacturing of the supporting insulator.

The invention provides a preparation method of a 5356 aluminum alloy welding wire, and belongs to the technical field of aluminum alloy welding wires. The preparation method comprises the following steps that a 5356 aluminum alloy cast ingot is prepared through direct water cooling semi-continuous casting, then is heated to 460-520 DEG C, and keeps warm for 2-4 hours, and hot extrusion is carriedout to obtain an 5356 aluminum alloy hot extrusion line rod; the 5356 aluminum alloy hot extrusion line rod is heated to 480-550 DEG C, heat preservation is carried out for 3-6 hours, rolling is carried out, the obtained 5356 aluminum alloy wire blank is subjected to middle annealing, the wire blank of the annealing state welding wire is peeled off, and the wire blank is pulled in a plurality oftimes, so that the 5356 aluminum alloy welding wire with the diameter of 1.2 mm is obtained. According to the preparation method, the production process of the 5356 aluminum alloy welding wire is remarkably shortened, the energy is saved, the production efficiency is improved, the production cost of the 5356 aluminum alloy welding wire can be greatly reduced, the 5356 aluminum alloy welding wire with fine alloy phase particles uniformly distributed on an alpha-A1 matrix can be obtained, the quality of the welding wire is improved, the stability of the welding process is guaranteed, and the comprehensive performance of a welded joint is improved.

The invention discloses an automatic extraction method of human chest organ tissue, belongs to the technical field of medical image processing, and aims to solve the problems that boundaries of organ tissue are unsmooth and observation by researchers and clinicians is affected due to the fact that an existing image segmentation technology is not clear enough. The automatic extraction method is characterized by comprising steps as follows: reading three-dimensional tomography images; performing fitting and primary segmentation processing on each image in a sequence with a least square ellipse fitting algorithm, parts outside sternums are removed, and parts inside the sternums are reserved; image noise of the parts inside the sternums is removed, segmentation processing is performed again to remove unnecessary soft tissue, and regions of interest, including trachea and blood vessels, inside the sternums are obtained; volume rendering and surface rendering are performed. Medical image segmentation has great significance in three-dimensional visualization, three-dimensional positioning, operation planning formulation and computer assisted diagnosis. With the adoption of the automatic segmentation method, the effect of subjective factors of an observer is completely avoided, the data processing speed is increased, and the repeatability is good.

The embodiment of the invention discloses an image processing method, device and equipment, and belongs to the technical field of computer vision correlation in artificial intelligence. The method comprises the following steps: carrying out candidate segmentation identification on an original image by adopting a first segmentation model to obtain a candidate foreground image region and a candidatebackground image region of the original image; recombining the candidate foreground image region, the candidate background image region and the original image to obtain a recombined image, wherein the pixel points in the recombined image and the pixel points in the original image have a one-to-one correspondence relationship; and carrying out region segmentation identification on the recombined image by adopting a second segmentation model to obtain a target foreground image region and a target background image region of the original image. According to the invention, the accuracy of image segmentation can be improved.

The invention discloses a monolithic macro-channel heat sink for a semiconductor laser. The monolithic macro-channel heat sink comprises a heat sink body, wherein the heat sink body is provided with a chip installation region and a water cooling region; the chip installation region is positioned at one end of the heat sink body; the water cooling region is positioned in the position near the chip installation region; the water cooling region is provided with a grid which penetrates through the heat sink body; the grid is composed of at least two parallelly-arranged grid holes; each grid hole comprises two mutually parallel side walls, and a connecting end for connecting the two side walls. The monolithic macro-channel heat sink has an efficient and uniform heat radiation effect. The invention further correspondingly provides the semiconductor laser with the monolithic macro-channel heat sink.

The invention relates to a real time wide view ultrasonic imaging device and method. The method comprises the steps of acquiring an image, registering the image, fusing the image and carrying out edge optimization. The device comprises a wide view image acquisition module, a wide view image registration module, a wide view image fusion module and a wide view image edge optimization module. The method and the device which are provided by the invention have the advantages that ultrasonic image information of a continuous sequence is directly acquired from the image acquisition module for carrying out image registration and image fusion, the ultrasonic image information does not need to be acquired from a static video, then real time map making is realized by combining an efficient algorithm in the invention, and great help is provided for operation and diagnosis of a user.

The invention discloses a method for preparing a dodecagonal zinc oxide micron rod. The method carries out secondary growth by integrating a hydrothermal method and a gas phase transmission method, and comprises the following steps that: first, a hexagonal zinc oxide micron rod is prepared through the hydrothermal method; then secondary epitaxial growth is carried out through the gas phase transmission method; and in the gas phase transmission method, because the side atoms of the hexagonal zinc oxide micron rod are desorbed, a cross sectional structure turns from a hexagon to a dodecagon. The method is simple and easy to operate, has low manufacturing cost, and can realized controlled preparation. The dodecagonal zinc oxide micron rod can be applied in designing an ultraviolet micro-laser with ultra-high quality factors and an ultra-low threshold.

The invention provides a high-resolution image building extraction precision processing method based on prior vector guidance, and the method comprises the steps: carrying out the data preprocessing of a prior vector and high-resolution remote sensing image data based on grating vector consistency analysis, and obtaining a data set after geometric registration and correction; fully considering spectral and spatial characteristics of ground features based on an FNEA image segmentation algorithm, carrying out image fine segmentation on a high-resolution remote sensing image within the constraint of a prior vector boundary by setting scale parameters to obtain a series of ground feature category image spots with different sizes, and obtaining ground feature category segmented image spots under grating vector suite; and obtaining a final result of building target extraction by adopting a majority voting method, enhancing the homogeneity and integrity of a building object roof extraction result within the constraint of a priori GIS land use vector boundary, and obtaining a deep network model building extraction precision processing result under the constraint of the priori GIS land use vector. According to the method, the building extraction quality of the high-resolution remote sensing image based on the deep learning model is remarkably improved.

An image vectorization method based on sub-pixels is a method capable of automatically converting any raster image into a vectorized image. The method comprises the following steps: firstly, segmenting an image by using an L0-based gradient minimization method to obtain a segmented image; secondly, providing an image boundary extraction method based on sub-pixels, so that the extracted boundary isaccurate to the sub-pixels; then, carrying out spline fitting on the extracted serialized boundary by using a spline fitting formula provided by the invention; and finally, drawing a vector diagram through the obtained splines through calculation. The method has the advantages that the vector diagram consistent with the content of the input image can be automatically generated, the smoothness ofthe spline can be guaranteed while the spline fitting accuracy is guaranteed, and more importantly, the boundary extraction method based on the sub-pixels is provided and is a great improvement on a traditional boundary extraction method.

The invention discloses an illumination compensation method and device for a dot matrix two-dimensional code. The method includes the steps of processing an original dot matrix two-dimensional code image by a binarization method and a morphological method, so that no hole is present in a dot matrix in the image to facilitate the comparison according to the processing result of binarization and theresult of morphological processing on the binarized image so as to quickly and accurately determine positions of holes of the dot matrix; eroding or swelling the original dot matrix two-dimensional code image to fill the holes of the dot matrix; determining an expected pixel value of the processed image according to the positions of the holes of the dot matrix, and compensating the expected pixelvalue to the holes of the original dot matrix two-dimensional code image to obtain a complete filled dot matrix two-dimensional code image; and smoothing the filled dot matrix two-dimensional code image to obtain a dot matrix two-dimensional code image with uniformly distributed grayscale. Thus, the method provided by the application can clearly distinguish the dot matrix from the background in the image and avoid affecting the subsequent signal extraction.