35results about How to "Eliminate color differences" patented technology

The invention relates to a geocoding-free rapid image splicing method of a low-altitude unmanned plane. A key frame image is extracted from a video to carry out geocoding-free rapid image splicing to generate a panoramic image. The method comprises the following steps: extracting a key frame image from a video stream to obtain all to-be-spliced images; carrying out feature point extraction including establishing a multi-dimensional space unit, constructing a pyramid image, determining a key point position, and removing an unstable point on all to-be-spliced images and carrying out screening on the key point by using a SUSAN algorithm to obtain a final feature point; carrying out feature matching according to respective feature points of a corresponding reference image and a to-be-matched image; and carrying out smoothening on the to-be-spliced images by using a gain compensation method, carrying out weighted average fusion on spliced overlapped areas to eliminate seams and sawteeth, and carrying out splicing to generate a panoramic image. According to the technical scheme, the method can be applied to fields of emergency surveying and mapping, disaster prevention and mitigation, land monitoring, emergency monitoring, pipeline layout and inspection of the oil sector, and circuit layout and power line inspection of the power sector and the like.

The invention discloses a real-time stitching method for a high-definition video. The method mainly comprises the following steps of synchronous video decoding, initialization, frame-by-frame stitching and the like. Firstly, a high-definition original video of a network camera is collected, and the video is synchronously decoded. For a first frame of video image, the initialization is carried outand includes the steps of video registration based on sub-regions, finding of an optimal stitching line set, generation of a look-up table and the like. After the initialization is completed, the frame-by-frame stitching is carried out and includes the steps of coordinate mapping of video images, detection of a moving object, stitching line update, video fusion and the like, and finally a stitchedvideo is output. In addition, a parallelization strategy is adopted for each step in video decoding and frame-by-frame stitching stages. According to the method, the real-time requirement of high-definition video stitching can be met, and a stitching frame rate is not less than 30 fps for a double-path 720P video; and meanwhile, the phenomena of ghosting, blurring, exposure inconsistence and thelike in the stitched video are avoided.

The invention relates to an image fusion method for a dome screen system. The image fusion method for a dome screen system is characterized by determining the projection area of each projector by means of a mesh model of the dome screen body so as to enable each projector to have adjustability for the projection position; performing edge fusion on each projection area by means of polar coordinatesgeometrical correction processing, so as to enable the projection area image at different position and in different size to achieve the effect of accurate splicing; and respectively performing colorcalibration of the projectors and fusion band blanking operation of projection images by means of unified color processing and light leakage compensation processing, thus effectively avoiding color variation and luminance variation between each edge fusion image. By means of the steps provided by the image fusion method for a dome screen system, the stereo projection image on the dome screen bodycan achieve the display effect of complete frames, unified color and suitable luminance, and comfortable and perfect film watching experience can be brought to spectators. The image fusion device employing the image fusion method for a dome screen system can generate a standard fusion image with better display effect, and the dome screen system employing the image fusion method can have excellentprojection effect and can improve the film watching experience of spectators.

The invention relates to a 3D stereo image color correction method based on matching and optimization. The method comprises the following steps: 1, registering a reference image and a target image byusing a scene mapping algorithm to obtain a matching image of the reference image and the target image; 2, generating a confidence coefficient diagram of the matching image and the target image by using a structural similarity algorithm; 3, correcting the target image by using a global color correction algorithm; 4, performing image initialization on pixels with relatively low confidence coefficients in the matching image by using a correction result of the global color correction algorithm according to the confidence coefficient diagram to generate an initial color correction result image; and 5, improving the color consistency of the initial color correction result image by using an optimization algorithm to obtain a finally optimized color correction result image. By adoption of the method, a good correction effect of stereo image color difference can be achieved, and the corrected target image and the reference image maintain the color consistency.

The invention discloses an intelligent water-based ink color blending working process. The intelligent water-based ink color blending working process includes the steps of (1) system startup; (2) initial color blending; (3) initial formula preparation; (4) initial formula production; (5) formula correction; (6) corrected formula preparation; (7) corrected formula production; and (8) circulated correction and storage. According to the intelligent water-based ink color blending working process, through mutual combination of color blending software and an ink mixing machine, intelligent, standard and datamation color blending balance is achieved, the color difference obtained after color blending is eliminated to the maximum degree, the colors obtained after color blending are consistent, and the colors on specific products can keep consistent to the maximum degree.

The invention relates to a color-difference-eliminating light-emitting chip grading method and application method of graded light-emitting chips. The light-emitting chip grading method comprises the steps of: testing CIE1931 color coordinates of an LED light-emitting chip; determining the CIE1931 color coordinate range of the three primary colors of red, green and blue which human eyes can recognize according to MacAdam color ellipse tolerance range in the chromaticity theory and screen display effect and determining grades according to the range R (Delta x, Delta y), G (Delta x, Delta y), B (Delta x, Delta y). The application method of the light-emitting chip in each grade is as follows: the LED display screen is made by using the light-emitting chip belonging to the same grade or the light-emitting chip with increasing or decreasing grades. The invention can eliminate the color difference of the COB packaged display screen or make the color of the display screen gradually change, thereby increasing the comfort of viewing by human eyes.

The invention discloses a multi-high-resolution remote sensing image mosaic method considering surface feature category difference. The method comprises the steps of performing geometric registrationand overlapping region coordinate determination on high-resolution images; performing color uniformization on the whole input image by using a histogram matching method by adopting an integral and local color matching strategy, and further performing local color uniformization on different ground objects by using a moment matching method; generating an initial seam line network by utilizing the Voronoi diagram; endowing different ground objects in the overlapping area with different weights, searching for a path with the minimum total weight in the buffer area based on a dynamic planning method, and taking the path as an optimized seam line network; and eliminating splicing seams by using an inverse distance weighted image fusion algorithm to obtain a final mosaic image. According to the method, the tone consistency of the mosaic image on the whole can be ensured, the local chromatic aberration can be remarkably weakened, the joint line network bypassing the building can be automatically generated, the error accumulation of searching the joint lines one by one is effectively avoided, and the mosaic image with visual nature is quickly obtained.

The invention relates to a stereo video color correction algorithm based on homography matrixes. The algorithm comprises the following steps: firstly, detecting Harris corner feature points in all video frames of a target video, and matching and tracking the feature points between all adjacent video frames by using a Kanade-Lucas-Tomasi (KLT) algorithm; secondly, obtaining motion trajectories of each feature point according to the matching feature points between adjacent video frames, and then selecting a feature points; then, calculating the homography matrixes between two adjacent frames through the selected a feature points by using a random sample consensus algorithm (RANSAC); finally, correcting the color of the first frame in the target video by using an image color correction algorithm; and correcting the colors of the second frame and subsequent frames after the second frame in the target video according to the solved homography matrixes. The algorithm can achieve good color correction results for color-distorted stereoscopic videos, so that the corrected target video and reference video maintain time and spatial color consistency.

The invention discloses a panoramic stitching colour correction method. The panoramic stitching colour correction method comprises the following steps of: (S1), determining the position of an overlapped region of two original images; (S2), screening pixel points in the overlapped region; (S3), obtaining a colour deviation conversion parameter M for imaging the same object in a practical scene by the two original images in the overlapped region by calculation through the pixel points obtained by screening; and (S4), performing colour conversion of the two original images by utilizing the colour deviation conversion parameter M. By means of the panoramic stitching colour correction method disclosed by the invention, the technical problem that colour deviation still exists after an image stitching gap is processed in the prior art can be solved; and thus, the final image cannot generate obvious colour deviation when panoramic stitching is carried out.

The invention discloses automatic spraying camouflage coating for manipulators and a method for preparing the automatic spraying camouflage coating. The method includes (1), adding 35.00-42.00 parts of acrylic resin, 0.50-1.00 part of dispersing agents, 1.00-3.00 parts of anti-sagging agents, 1.50-2.50 parts of extinction powder, 0.10-5.00 parts of electrically conductive optical materials, 7.00-18.00 parts of pigment, 17.00-23.00 parts of fillers and 6.00-8.50 parts of mixed solvents into an accessory tank and uniformly stirring the acrylic resin, the dispersing agents, the anti-sagging agents, the extinction powder, the electrically conductive optical materials, the pigment, the fillers and the mixed solvents to obtain first mixtures; (2), grinding the first mixtures by the aid of a sand mill until the fineness of the first mixtures reaches 18-22 micrometers; (3), adding 4.00-6.00 parts of special modified polyester compounds, 3.00-5.00 parts of silicon dioxide nanometer composite materials and 6.00-8.50 parts of mixed solvents into paste and uniformly stirring the special modified polyester compounds, the silicon dioxide nanometer composite materials, the mixed solvents and the paste to obtain second mixtures; (4), regulating the viscosity of the second mixtures until the viscosity of the second mixtures reaches coating-4 cup 65-80 S; (5), filtering and packaging the automatic spraying camouflage coating. The automatic spraying camouflage coating and the method have the advantages that the automatic spraying camouflage coating is reasonable in design and has tone which is consistent with natural backgrounds, anti-infrared camouflage effects of spectral reflection curves can be realized by the automatic spraying camouflage coating, and problems in the aspects of sagging, drying speeds, surface smoothness and the like in automatic paint spraying procedures can be solved.

The invention relates to an LED light emitting chip hybrid gear dividing method. The method comprises the steps that according to the MacAdam color ellipse wide capacity range combined with the screendisplay effect, the CIE1931 color coordinate range which can be recognized by human eyes of three primary colors is determined, and gears are divided according to the range; light emitting chips areplanned into the corresponding gear according to different color coordinates thereof; the gears are divided into main gears and disperse gears according to the proportion of the number of light emitting chips in each gear, and the light emitting chips in a number of disperse gears are fully mixed and merged into a merged gear; the light emitting chips in the merged gear are placed on a merged gear blue film and provided for a downstream display manufacturer for use in mid-low end displays; and the light emitting chips belonging to the same main gear are placed on the same blue film and provided for a downstream display manufacturer for use in high-end small-pitch displays. According to the invention, the consistency of the color purity of high-end small-pitch displays is high, and the problems of large number, difficult management and poor availability of disperse gears of small share are solved.

The invention discloses an original image color adjustment method and device for a panorama camera. The method comprises the steps: obtaining a superposed area of a first original image and a second original image, which are collected by the panorama camera; detecting and obtaining feature points of the superposed area of the first and second original images, and describing the feature points through descriptors; carrying out the matching of the feature points of the first and second original images according to the descriptors of the feature points, and obtaining a plurality of groups of matched feature points; selecting one of the first and second original images as a reference image, selecting the other original image as a to-be-adjusted image, and obtaining a color adjustment parameteraccording to a YUV value of each group of matched feature points; and adjusting the color of the to-be-adjusted image according to the color adjustment parameter. Therefore, before the splicing of the original images of the panorama camera into a panorama image, the method achieves the adjustment of the colors of the original images, and improves the fusion effect of the panorama image.

The invention relates to a space-time consistency-based stereoscopic video color correction method. The method comprises the steps of firstly, selecting out a pair of key frames from a reference videoframe and target video frames every a frames, and selecting k pairs in each time sliding window; secondly, for each pair of the key frames, obtaining an initial color correction result frame of the target video key frames by adopting an image color correction algorithm, and then obtaining a color correction result video frame which is more consistent with the input target video key frames in structure through guided filtering calculation; thirdly, obtaining a color mapping function between the target video key frames and the target video key frames subjected to the color correction through improved histogram matching in the time sliding window; and finally, correcting the a target video frames located in the center of the time sliding window by using the color mapping function to obtain afinal target video frames subjected to the color correction. A good correction effect can be achieved for color difference of stereoscopic videos.

The invention provides a seam line automatic generation method based on a Dijkstra algorithm with weight. The seam line automatic generation method comprises the following steps that S1 an overlapped area between two adjacent images is calculated according to a digital ortho-image through ortho-rectification; S2 a Dijkstra algorithm reference point is calculated according to the overlapped area, and a digraph is generated; S3 weight parameters are set, the digraph shortest path is calculated through the Dijkstra algorithm based on the weight parameters, and a seam line is generated and displayed in real time; S4 the seam line is optimized, and a mosaic surface is generated and correlated with the images; and S5 the images correlated with the mosaic surface are outputted to the designated position according to control information. The seam line is extracted through the Dijkstra algorithm with the weight so that manmade objects can be effectively avoided by extraction of the seam line, color difference of water areas can be eliminated, large amount of manual editing work can be saved, the degree of automation of extraction of the seam line can be further enhanced and the extracted seam line is ensured to have great accuracy.

The invention discloses a panorama mosaic color correction method, comprising the following steps: step (S1), determining the position of the overlapping area of ​​two original images; step (S2), screening the pixels in the overlapping area; step (S3), Calculate and obtain the color deviation conversion parameter M of the same object in the actual scene in the overlapping area of ​​the two original images by using the filtered pixels; step (S4), use the color deviation conversion parameter M to perform color conversion on the two original images. The invention solves the technical problem in the prior art that there is still color deviation after processing the seams of image splicing, and ensures that the final image will not produce obvious traces of color difference during panorama splicing.

The invention discloses a panoramic image color correction method for an automobile data recorder, and the method comprises the following steps: obtaining a preset number of spliced image frames, andobtaining a template image according to the spliced image frames; superposing the template image with a to-be-corrected spliced image to obtain a superposed region; carrying out the discrimination ofthe Cr and Cd correlation degree of the obtained superposed region and the template image to determine whether to correct the Cr and Cd values of the superposed region till the color correction of allregions of the to-be-corrected spliced image is completed. The method provided by the invention can achieve the adjustment of the color gamut difference of the spliced image, can enable the color ofthe spliced image to be consistent, and facilitates the subsequent application of the spliced image in the automobile data recorder.

The invention provides touch display equipment. The touch display equipment comprises a transmitting cover plate, a shading layer, a touch induction layer and a color modulation layer, wherein the transmitting cover plate is provided with an inner surface; the inner surface is divided into a first region and a second region; the shading layer is arranged in the first region and not located in the second region of the inner surface; and the color modulation layer is located between the shading layer and the transmitting cover plate, so that the chromatic aberration between the first region and the second region is smaller than or equal to 3 when the touch display equipment is in a non-use state. Through the touch display equipment, the color difference between a visual region and a non-visual region can be lowered and even eliminated, and therefore the visibility of a boundary between the visual region and the non-visual region is lowered.

The invention provides a roughened epitaxial wafer-based electrode color difference improvement method of an LED chip. The surfaces of a GaN epitaxial wafer in abdicating regions for subsequently preparing electrodes are etched to flat, so that the surface roughness is the same; the electrodes are free of a color difference after being prepared; color difference exception is completely eliminated; the cost caused by the color difference exception is reduced; a transparent conductive layer and the electrodes are subsequently prepared; and the regions are in smooth connection, the current resistance is small, diffusion is fast, the light reflection probability is high and the photoelectricity of the LED chip is also improved to a certain extent.

The invention provides an image data processing method, an image data processing device and unmanned equipment. The image data processing method comprises the following steps: establishing a global digital orthoimage; the method comprises the following steps: acquiring a first digital orthoimage and a first weight map corresponding to N images in real time from the scene shooting of unmanned equipment, N being a positive integer; and updating the global digital orthoimage according to the first digital orthoimage and the first weight map to determine a target digital orthoimage in the updated global digital orthoimage. According to the invention, the color difference of the digital orthoimage can be effectively reduced, and the visual effect is improved.

The invention provides an intelligent prediction system for pathological indexes. Firstly, the shape categories of patterns in image blocks are classified, and then, the image blocks are input to pathological index predictors corresponding to different shape categories to obtain a classification result of each pathological index predictor. and finally, according to the number of the image blocks contained in different combinations formed by thee pathological index categories and the shape categories and the weight of each pathological index predictor, the score of a certain pathological section image belonging to each pathological index category is calculated, the pathological index category corresponding to the maximum score is taken as a final pathological index prediction result of a pathological section image. Therefore, the image blocks are classified more meticulously, and validity analysis is performed on the image blocks of various shapes and categories, so that the validity ofimage block-level prediction data and the predictors and the accuracy of a patient-level prediction result can be obviously improved.

The invention discloses a method for combined weaving of eucommia ulmoides twigs and wicker. The method includes the following operation steps that firstly, when the eucommia ulmoides twigs and the wicker after peel removing treatment are dried till the water content is 20% or below, the eucommia ulmoides twigs and the wicker are placed in soaking liquid, and are subjected to natural airing treatment for 20-24 hours after soaking treatment is performed for 5-6 hours; secondly, the eucommia ulmoides twigs are placed on a stainless steel net, the wicker is placed above the eucommia ulmoides twigs, the upper portion of a cooking boiler is covered with the stainless steel net, a cooking material is placed in the cooking boiler, a large fire is opened, and cooking treatment is performed for 50-60 min; and thirdly, the eucommia ulmoides twigs and the wicker after being subjected to treatment in the second step are placed into a drying oven again, after drying treatment, weaving forming is performed, and finished products are made. The method for combined weaving of the eucommia ulmoides twigs and the wicker is easy to operate, low in cost and green and environment-friendly, woven wickerproducts are remarkable in mould proof and insect proof effect, and it is avoided that due to the fact that a toxic mould proof agent is added, the quality and the safety performance of the wicker products are affected

The invention provides a face comparison method and device, and the method comprises the steps: obtaining a to-be-recognized face image; inputting the face image to be recognized into the face comparison model to obtain a face comparison result output by the face comparison model; wherein the face comparison model is obtained by training based on a sample image and a texture label corresponding to the sample image; the face comparison model is used for performing color adjustment and texture filling on face image features extracted based on the to-be-recognized face image and then performing face comparison with pre-stored face comparison features. According to the invention, color adjustment is carried out on the face image features by using the color features extracted by the face comparison model so as to eliminate color differences caused by image source differences, and texture filling is carried out on the face image features by using the texture features extracted by the face comparison model. Therefore, the loss of texture features of face image features caused by a color-removed region is avoided, and the accuracy of subsequent face comparison is improved.

The invention discloses a remote sensing image color consistency processing method based on an optimal path, belongs to the technical field of optical satellite remote sensing image processing, and solves the problem of low efficiency of color consistency processing of multiple remote sensing images in the prior art. The method comprises the following steps: determining a to-be-processed remote sensing image set, calculating effective boundaries of all remote sensing images in the set, and determining a spatial position relationship among the images in the set; selecting a color reference image from all the remote sensing images in the set; calculating an optimal uniform color path based on the spatial position relationship among the images in the set; based on the color reference image, performing brightness normalization preprocessing on all the images in the set; and performing color consistency processing on all the images subjected to brightness normalization preprocessing according to the sequence in the optimal uniform color path: performing color consistency processing on the node images according to the sequence in the optimal uniform color path, so that each node image is consistent with the reference image in color until all the node images in the path are completely processed.

The invention relates to a space-time consistency-based stereoscopic video color correction method. The method comprises the steps of firstly, selecting out a pair of key frames from a reference videoframe and target video frames every a frames, and selecting k pairs in each time sliding window; secondly, for each pair of the key frames, obtaining an initial color correction result frame of the target video key frames by adopting an image color correction algorithm, and then obtaining a color correction result video frame which is more consistent with the input target video key frames in structure through guided filtering calculation; thirdly, obtaining a color mapping function between the target video key frames and the target video key frames subjected to the color correction through improved histogram matching in the time sliding window; and finally, correcting the a target video frames located in the center of the time sliding window by using the color mapping function to obtain afinal target video frames subjected to the color correction. A good correction effect can be achieved for color difference of stereoscopic videos.

The invention relates to a method for preparing a nano-microlite plate by utilizing CRT (Cathode Ray Tube) screen glass. The method disclosed by the invention comprises the following steps: firstly, crushing screen glass of a waste cathode ray tube, an electronic gun of which is removed, and melting by adding auxiliary raw materials to prepare concentrated pigment; and then, mixing and forming CRT screen glass power and concentrated pigment powder according to the mass fraction, heating, crystallizing, and cooling to obtain nano-microlite. The nano-microlite plate prepared by using the preparation method disclosed by the invention is low in porosity and water absorption, high in strength, small in colour difference, simple in sintering process and low in cost; the utilization rate of the CRT screen glass is high; furthermore, pollution to the environment cannot be generated in the production and use processes; the nano-microlite plate is high in acid-base resistance property and abrasion resistance and free from radioactivity harm, and can be used as a novel high-grade decorative plate; the preparation method disclosed by the invention is simple in operation process and low in cost; the production cost of nano-microlite can be effectively reduced; the environment is protected; wastes are turned into wealth; and thus, the method disclosed by the invention is applied to large-scale industrialization production and has obvious economic benefits and environment-friendly benefits.

The invention provides a seam line automatic generation method based on a Dijkstra algorithm with weight. The seam line automatic generation method comprises the following steps that S1 an overlapped area between two adjacent images is calculated according to a digital ortho-image through ortho-rectification; S2 a Dijkstra algorithm reference point is calculated according to the overlapped area, and a digraph is generated; S3 weight parameters are set, the digraph shortest path is calculated through the Dijkstra algorithm based on the weight parameters, and a seam line is generated and displayed in real time; S4 the seam line is optimized, and a mosaic surface is generated and correlated with the images; and S5 the images correlated with the mosaic surface are outputted to the designated position according to control information. The seam line is extracted through the Dijkstra algorithm with the weight so that manmade objects can be effectively avoided by extraction of the seam line, color difference of water areas can be eliminated, large amount of manual editing work can be saved, the degree of automation of extraction of the seam line can be further enhanced and the extracted seam line is ensured to have great accuracy.

The invention relates to a stereo video color correction algorithm based on homography matrixes. The algorithm comprises the following steps: firstly, detecting Harris corner feature points in all video frames of a target video, and matching and tracking the feature points between all adjacent video frames by using a Kanade-Lucas-Tomasi (KLT) algorithm; secondly, obtaining motion trajectories of each feature point according to the matching feature points between adjacent video frames, and then selecting a feature points; then, calculating the homography matrixes between two adjacent frames through the selected a feature points by using a random sample consensus algorithm (RANSAC); finally, correcting the color of the first frame in the target video by using an image color correction algorithm; and correcting the colors of the second frame and subsequent frames after the second frame in the target video according to the solved homography matrixes. The algorithm can achieve good color correction results for color-distorted stereoscopic videos, so that the corrected target video and reference video maintain time and spatial color consistency.

In an image processing section, a foam toner signal generator generates a foam toner signal corresponding to a foam toner amount to be transferred to a three-dimensional print designation region discriminated at a three-dimensional print region discriminator, and a foam toner interpolation signal generator generates a foam toner interpolation signal corresponding to a foam toner amount to be transferred to an image region equal to or greater than a constant area in an ordinary print region discriminated at the three-dimensional print region discriminator. Foam toner of a predetermined amount corresponding to the foam toner signal is transferred to the three-dimensional print region, and foam toner of an amount that is less (of a lower height) than the predetermined amount and corresponds to the foam toner interpolation signal is transferred to the ordinary print region.