49results about How to "Solve the ghosting problem" patented technology

The invention relates to the field of image application, and discloses an image processing method and device and a terminal which are used for solving the technical problems that 'ghost images' are produced when multiple low-resolution images are combined into high-resolution images. The method is applied to the terminal comprising a first camera and a second camera, wherein the first camera and the second camera are located on the same side face of the terminal. The method comprises the steps that a first image, obtained by the first camera, of a first region is obtained, and a second image, obtained by the second camera at the same moment, of a second region is obtained; the first image is used as a reference image so as to compensate for the second image in a horizontal-motion mode; the first image and the second image which is compensated for in the horizontal-motion mode are combined into a third image, and the resolution of the third image is higher than that of the first image and that of the second image.

The embodiment of the invention provides a three-dimensional panoramic video stream generating method and equipment and a video conference method and equipment. The three-dimensional panoramic video stream generating method comprises the following steps of: acquiring the depth information of at least two video images; acquiring a plurality of image data of a depth level from a corresponding video image according to the depth information of each video image; and jointing the video image data according to the image data of the depth level to generate a three-dimensional panoramic video stream. The invention provides a three-dimensional panoramic seamless high-resolution telepresence conference video image for users by utilizing a multi-display mode based on different display equipment.

The invention discloses side information generating system and method based on distribution type video encoding, which belongs to the technical field of video encoding and mainly solves the problem that the original image information is lacked when the distribution type video encoding system is used to generate side information. The method has an encoder relevance calculating part, a decoder movement evaluating part, a movement smoothing part and a movement compensating part. The relevance among the original image frames is calculated by the encoder, and sent to the decoder. On the basis of relevance, forward and backward movements are evaluated by front and next decoded key frames, and the obtained movement vectors are detected and corrected. Then, the accurate movement vectors are obtained by the bidirectional movement evaluation and movement smoothing of an overlapped block. Finally, the side information is generated by the bidirectional movement compensation of the overlapped block. The invention greatly improves the object and subject quality of the side information and further enhances the quality of the decoded and restored image, and can be applied to the simple video communication system of the terminal equipment.

The invention relates to a method and system for binocular fish-eye image splicing and unfolding correction. The method for binocular fish-eye image splicing and unfolding correction includes the steps: utilizing a binocular fish-eye camera to obtain an original binocular fish-eye image, and extracting the left and right two effective circular imaging areas from the original binocular fish-eye image; extracting the characteristic points of the overlapped part in the two effective areas, and matching the characteristic points in pairs; respectively mapping the two effective area images to obtain two space coordinate unit spherical surface images, performing matrix singular value decomposition on the three dimensional space spherical surface, obtaining the rotation relation, that is, the rotation matrix, between the two images through calculation, and correcting one of the space coordinate unit spherical surface images through the rotation matrix; and finally respectively mapping the twospace coordinate unit spherical surface images to obtain two panoramic coordinate images, and splicing the two panoramic coordinate images into a complete 360 DEG panoramic image. The method and system for binocular fish-eye image splicing and unfolding correction can solve the problem that the fish-eye imaging sensors are not in parallel during the assembling process, thus causing errors, and have the advantages of being good in the splicing effect after correction, being high in processing timeliness, and being easy to implement.

The invention discloses a head mounted type display device based on holographic waveguide. The head mounted type display device comprises inlet coupling grating (1), left visual field deflection grating (2), right visual field deflection grating (3), outlet coupling grating (4), and rectangular waveguide (5). The inlet coupling grating (1), the left visual field deflection grating (2), the right visual field deflection grating (3), and the outlet coupling grating (4) are attached to the upper surface or the lower surface of the rectangular waveguide (5). Whether the inlet coupling grating (1), the left visual field deflection grating (2), the right visual field deflection grating (3), the outlet coupling grating (4), and the rectangular waveguide (5) are attached to the upper surface or the lower surface is determined by an exit entrance pupil light design direction. By adopting a pupil remodeling way, a visual field separation problem of a conventional two-dimensional pupil dilation way under a large visual field angle condition is solved.

The invention discloses an immersion type ultra-high-definition video processing system and an immersion type ultra-high-definition video processing method. The system comprises four subsystems: a video fusion multi-image multi-pixel level geometric correction system, an image edge fused three-dimensional smooth curve color correction system, a distributed multi-video fusion equipment control device and a distributed video synchronous broadcasting and browsing and editing system. The method provides solutions to the problem of deformation of a projected image in video fusion, the problem of communication in the distributed system, the problem of adaptive adjustment of fusion band parameters and the problem of synchronous broadcasting of the distributed system. According to the system and the method, the broadcasting effect of the video fusion system is increased, and users feel immersive.

The invention discloses a display system with an optical lens and display screen and an image display method of the display system. The image display method comprises the following steps that first, the optical lens is reversed in a virtual mode according to the optical parameters of the optical lens to obtain a virtual reversed lens, the objective image of the reversed lens is reversed to the objective image of the optical lens; second, the distortion coefficients of the three color components of RGB are calculated according to the optical parameters of the reverted lens after the color components pass through the reverted lens, and serve as a1, a2 and a3 respectively; third, pixel points of a predictive display image are amplified a1 times, a2 times and a3 times respectively, so that the three RGB color components of each pixel point is obtained to form a compensatory image after the pixel point is distorted, and a compensatory image is formed; fourth, after the compensatory image is output to the display screen and displayed, the compensatory image is displayed as a normal image through the optical lens. According to the display system with the optical lens and the display screen and the image display method of the display system, the problem that blur or ghosting or distortion of an image may happen is solved, and an algorithm can be carried out fast and conveniently.

The invention discloses a depth-based light field splicing method. The depth-based light field splicing method comprises: inputting a light field to be spliced and a sub-aperture image of the light field, and estimating a light field depth of the sub-aperture image of the light field to obtain a depth map of the light field; extracting feature points in the sub-aperture image of the light field, matching the feature points to obtain feature point pairs, and screening the feature point pairs to obtain matching feature point pairs; performing 4D meshing on the light field to be spliced, and predicting a global homography transformation matrix according to the matching feature point pairs; establishing a weight matrix according to the depth and position relationship between the feature pointsand center points of the meshes; and predicting the optimal homography transformation matrix of each mesh according to the global homography transformation matrix and the weight matrix, mapping the light field according to the optimal homography transformation matrix of each mesh in the light field, fusing the light field, and obtaining a light filed splicing result. The invention solves the problem of dislocation and ghosting caused by large parallax changes, and realizes a precise parallax-tolerant light field splicing method.

The invention discloses a naked eye three-dimensional (3D) image processing method and a naked eye 3D image processing device. The method is applied to electronic display equipment, and comprises the steps of acquiring human eye position information of N users, wherein N is an integer of more than or equal to 1; judging whether N visual area pairs which correspond to the N users are in a front view area according to the human eye position information of N users; adjusting at least one visual area image in a visual area formed by adjacent visual areas and the ith visual area pair on two sides of the ith visual area pair if M visual areas in the N visual area pairs are positioned in the front view area, wherein M is more than or equal to 1 and less than or equal to N, i is more than or equal to 1 and less than or equal to M, and i and M are integers; and adjusting at least one visual area image in a visual area formed by adjacent visual areas and the kth visual area pair on two sides of the kth visual area pair if p visual areas in the N visual area pairs are positioned in an inversion area, wherein P is more than or equal to 1 and less than or equal to N, k is more than or equal to 1 and less than or equal to M, and k and P are integers.

The invention discloses a solvent type acrylic pressure sensitive adhesive and a preparation method thereof. The solvent type acrylic pressure sensitive adhesive of the formula is prepared from, by weight, 1-3% of acrylic acid, 0.5-8% of hydroxyethyl acrylate, 0.5-8% of hydroxy-propyl acrylate, 0.5-8% of hydroxy-butyl acrylate, 24-71% of iso-cctyl acrylate, 20-71% of butyl acrylate, 20-70% of solvents and 1-4%o of initiators. An ethyl acetate, acetone and toluene mixed system is adopted for the solvent, and ethyl acetate, acetone and toluene are mixed according to the proportion of 9:0.5:0.5. The initiators are prepared from azodiisobutyronitrile and bis(4-tert-butylcyclohexyl)peroxydicarbonate. The acrylic adhesive for PE protective films is provided, and the protective films are mainly used for protecting screens in the mobile phone producing process. The adhesive has good adhesive force on base materials and meanwhile well solves the 'ghost' problem.

The invention relates to a high-quality 3D texture reconstruction method. According to the high-quality 3D texture reconstruction method, a fuzzy detection module and a dynamic color weight distribution function are additionally adopted in texture reconstruction; a fuzzy detection algorithm is adopted to set a certain fuzzy threshold value, so that fuzzy pictures inputted into a system are removed; and the dynamic color weight distribution function is designed so as to dynamically distribute color weights for newly acquired color pictures, so that color information stored in voxels can be updated. With the high-quality 3D texture reconstruction method adopted, the problem of color edge ghosting can be solved, and the texture quality of a reconstructed 3D model can be greatly improved to a certain extent.

The invention discloses a lens for 3D glasses. In a direction from a display screen to an external light source, the lens comprises the following multiple structures: a lower polaroid, a lower glass substrate, a lower transparent conductive film, a lower insulating layer, a lower orientation layer, a liquid crystal layer, sealing adhesive, an upper orientation layer, an upper insulating layer, anupper transparent conductive film, an upper glass substrate and an upper polaroid. The absorption axes of the upper polaroid and lower polaroid are vertical to each other, and each liquid crystal lamination of the liquid crystal layer is in 90-degree spiral arrangement. The lens for 3D glasses has the advantages of simple structure, good contrast ratio and capability of reducing light leak rate and improving viewing quality.

The invention relates to a glasses television, in particular to a transparent glasses television which can simultaneously watch the external real scenery and virtual images. The glasses television comprises an eyeglasses type front frame and ear handles, a display device, an optical lens group and a reflection mirror are arranged on the front frame, a prism combined part image synthesizer consists of two prisms, and the hypotenuses of the two prisms are adhered by 45 degrees. The glasses television has the advantages that: the problems of difficult assembly and adjustment of a single reflection surface and the dislocation problem of the virtual images and the external real scenery on an optical axis are solved; the ghost image problem is solved; the visual field of a system is increased without the great increase of the size, and a condenser component allows the NA of the system to be very large, thereby improving the observation subjective brightness and having great observation effect and good image quality; the light intensity and the contrast of the virtual images and the external real scenery can be adjusted, the virtual images and the real environment have the similar spatial depth to human eyes.

The embodiment of the invention discloses a frame insertion image processing method and device. The method comprises the following steps: inputting two reference frames before and after a frame insertion position into an optical flow estimation network to obtain two groups of optical flow fields between a first reference frame and a second reference frame; processing the two groups of optical flowfields by using an optical flow interpolation neural network, and deducing to obtain optical flow data between the two reference frames and a middle frame at the pin position; deducing by using a weight estimation neural network according to the two groups of optical flow fields, the optical flow data and two reference frames to obtain weight distribution for combining the two reference frames; and winding the two reference frames according to the optical flow data, and performing weighted summation on the wound reference frames by using the weight distribution to obtain a target intermediateframe. According to the method, the optical flow data between the reference frame and the intermediate frame can be effectively reconstructed, the stable and reliable intermediate frame can be obtained through combination, and the problems of ghosting, blurring and the like in the frame insertion image are well solved.

The embodiment of the invention provides an image splicing method and device. The method comprises the following steps: obtaining a basic splicing image and a to-be-spliced image; extracting feature points of the basic spliced image and the to-be-spliced image; determining a plurality of matched feature point pairs, according to the determined plurality of matching feature point pairs, calculatinga homography matrix of each image to be spliced; according to the homography matrix of each to-be-spliced image, splicing the to-be-spliced images; determining an overlapping area of each to-be-spliced image, according to the re-projection error of the matched feature point pair; carrying out first grid establishment on the overlapped area of each to-be-spliced image; and for each first grid of each to-be-spliced image, adjusting the homography matrix of the to-be-spliced image according to the weight of the first grid, obtaining the homography matrix of the first grid, and splicing the basicspliced image and at least one to-be-spliced image according to the homography matrix of each to-be-spliced image and the homography matrix of each first grid. By applying the embodiment of the invention, the problems of dislocation and ghosting of image splicing are solved.

A method for adjusting the definition of a 3D video, comprising the steps of: A. separating the left and right eye images according to the format of the 3D image in the input 3D video; B. respectively scaling the separated left and right eye images; D. Output the sharpness-adjusted left and right eye images according to the specified field frequency. Through the above method, the problem of image ghosting caused by adjusting the clarity of the 3D video signal output by interleaving in the prior art can be well solved, so that the 3D image output by horizontal interleaving can effectively improve the clarity without improving Ghosting due to sharpness.

The invention discloses a 4K real-time video panoramic stitching method based on GPU acceleration. The method comprises the following steps that hardware equipment takes streams and directly transmitsthe streams to a GPU video memory; calculating camera parameters and storing the parameters; calculating image brightness and chromaticity correction parameters; calculating a mask pattern, and performing image fusion on subsequent frames by using the mask pattern; mapping the images to a projection plane by using the calculated camera parameters, and superposing the overlapping areas of the twoimages together; performing compensation processing on the image by using the brightness and chroma correction parameters calculated in the step 3; and fusing the images by adopting an image pyramid and an optimal splicing technology to obtain a panoramic spliced image for display, output and storage. According to the method, the high-frequency detail information of the image is extracted by adopting a Gaussian multilayer pyramid method for fusion, and all operations are accelerated by using the GPU, so that the problems of ghosting and ghosting in the splicing process are solved, and the real-time requirement is met.

A system for adjusting three dimensional (3D) video sharpness comprises: a 3D video L / R separating unit, a left and right eye image interlacing synthesis unit and a sharpness adjustment unit. The 3D video L / R separating unit is used to acquire a 3D video and separate a left eye image from a right eye image according to a format of the 3D image in the 3D video. The left and right eye image interlacing synthesis unit synthesizes and interlaces the left eye image and the right eye image into image signals. The sharpness adjustment unit improves the sharpness of the interlaced image signals along an interlacing direction. In the prior art, a sharpness promotion direction and the image interlacing direction are not the same so that a problem of double image can be generated. By using the method of the invention, the above problem can be effectively solved. The sharpness of the interlaced output 3D image can be effectively improved and the double image can not be generated.

The invention discloses a medium-wave infrared target simulation system using an image space telecentric beam path and belongs to the field of infrared target simulation. The problem that a light spot(ghost image) shown by cold light rays irradiated by a cold stop in an existing detection system on a detection face covers a target objective image, and the detection precision of the detection system is affected is solved. The target simulation system is used for generating a simulation target with the infinite the target position and is further used for making cold light rays irradiated by thecold stop in an infrared target detector return by the same way the cold light rays come through a target source, an amplified light spot is formed on the detection face of the infrared target detector, the target objective image is imaged on the detection face with the amplified light spot as the background, and the influences of the ghost image on the target objective image are eliminated. Thetarget source includes an interference target source and a to-be-detected target source. The forming process of the infrared simulation target includes that after the inference target source and the to-be-detected target source are coupled through the image space telecentric beam path, beam expanding is conducted, and therefore the infrared simulation target with the infinite target position is generated. The medium-wave infrared target simulation system is mainly used for generating the infrared simulation target.

The invention provides a mechanical key detection chip, the chip can convert the impedance value of a mechanical key into a corresponding digital value, the chip has a cascade expansion function in a scan chain mode, serial expansion use can be facilitated, the chip integrates a light emitting diode drive, and by means of the characteristics, the chip can be applied to the field of mechanical keyboards, and the mechanical keyboards which are efficient, reliable and low in cost are developed.

The invention discloses a light guiding lens and a light emitting diode (LED) lamp. The light guiding lens and the LED lamp comprise a strip-shaped body which is composed of an upper surface, a first side surface, a second side surface and a lower surface, and the strip-shaped body can extend along the length direction. The first side surface is vertically arranged. The second side surface is outwardly and obliquely arranged relative to the first side surface. The lower bottom edge of the second side surface is at least lower than the first side surface. Micro convex lenses are evenly distributed on the second side surface. The light guiding lens and the LED lamp can avoid a dazzling effect on eyes by strong directional irradiating light rays of a lamp bead, achieves the technical effect of restraining the dazzling phenomenon, has the advantages of being good in heat dissipation effect and lighting effect and soft in light, and achieves the purpose of a good lighting effect.

The invention discloses an image splicing method, electronic equipment and a computer readable storage medium, and the method comprises the steps: obtaining at least four original images, and enabling the partial regions of the adjacent original images to coincide; querying a mapping table according to a splicing template to obtain pixel coordinates of the original image, and performing calculation according to the pixel coordinates of the original image to obtain a conversion image; the mapping table comprises a coordinate mapping relation between an original image and corresponding points on the splicing template; and according to a vehicle model, the conversion image and a pre-created weight template, carrying out splicing to generate a panoramic view. According to the method, the image splicing efficiency is improved, the problems of ghosting and gaps in a splicing result can be effectively solved, and the two adjacent images are gradually transited during fusion splicing.

The invention relates to a shipborne rock core Ct transmission scanning imaging device and belongs to the technical field of natural gas hydrate basic physical property measurement. The device comprises a CT scanning support, a CT scanning system and a computer system; the CT scanning system is arranged at the center of the CT scanning support, and a passage for scanning a to-be-tested article isarranged at the center of the CT scanning system. due to the structure setting of the CT scanning support and the CT scanning system, the structure of the original CT transmission scanning imaging device is simplified; the CT scanning system solves the problems of fuzziness and double image during scanning imaging; a radiation isolation device effectively prevents leakage of scanning radiation; and drive systems at two sides of the CT scanning system ensure that the rock core sample feeding process is smooth, and the definition and accuracy of scanning imaging are further promoted.

The invention relates to the technical field of schlieren optical imaging, in particular to a folding schlieren apparatus and a gas temperature field visualization method based on the folding schlieren apparatus. The schlieren apparatus comprises a light source as well as a beam splitter, a convex lens and a plane mirror which are sequentially arranged at the rear of the light source; a space between the convex lens and the plane mirror is used for setting a testing region; reflected light via the plane mirror is converted in light path by virtue of the beam splitter after passing through theconvex lens, and a knife edge used for cutting a light beam is arranged on a path where the beam splitter converts light; and the schlieren apparatus also comprises an imaging device used for performing imaging on light emergent from the knife edge. The light given off by the light source generates parallel light after passing through the beam splitter and the convex lens, the parallel light passes through the testing region, is reflected by the plane mirror and then respectively passes through the testing region and the convex lens, the light path is converted by virtue of the beam splitter,and the light is convergent at the knife edge and is finally received by the imaging device for imaging. The folding schlieren apparatus can improve definition of a schlieren image and solves a doubleimage problem.

The invention discloses a method for synthesizing multi-viewpoint stereoscopic images, comprising: the arrangement of multi-viewpoint stereoscopic images is composed of a front view area and an inversion area, and the front view area is larger than the inversion area. Let the number of viewpoints of the naked-eye 3D display device be N d , the number of viewpoints in the emmetropic area is N s , the number of viewpoints in the inversion zone is N i , then N s >N i and N d =N s +N i . Viewpoint number corresponding to the display pixel or sub-pixel of the display device in the naked-eye 3D display device Among them, W p is the width of the display pixel or sub-pixel, W a is the tilt offset of each row of the raster, W d is the sum of the widths of several pixels or sub-pixels corresponding to the raster unit, m is the row number of the pixel or sub-pixel, and n is the column number. The method for synthesizing multi-viewpoint stereoscopic images proposed by the present invention can solve the problem of ghosting and improve the problem of viewing zone shear, especially on naked-eye 3D display devices using similar ultra-multi-viewpoint display technologies.

The invention is suitable for the technical field of image processing, and provides a dynamic image processing method and device, and the method comprises the steps of obtaining the pixel values of all pixels in a to-be-processed image; counting the number of pixel points of which the pixel values are in a preset interval; according to the number, adjusting pixel values of all pixel points in theto-be-processed image to obtain a first image and a second image; and fusing the to-be-processed image, the first image and the second image to obtain a target image. Through the scheme, the target image is formed by fusing different exposure images of the same image. Compared with the prior art, the image is formed by fusing different exposure images of the same image, so that the problems of ghosting, blurring or boundary traces and the like on the HDR image can be solved.

The invention relates to the technical field of vehicle lighting, and discloses a vehicle LED light source based on ceramic fluorescent piece packaging and a packaging process of the vehicle LED lightsource. The LED light source comprises an aluminum nitride ceramic makeup support which is internally provided with a fixed plate. A positive pin and a negative pin are respectively arranged at the positions, located on the two sides of the fixed plate, of the aluminum nitride ceramic makeup support, and three inverted LED wafers distributed at equal intervals are welded to the top of the aluminum nitride ceramic makeup support through coated scaling powder. According to the invention, inorganic ceramic fluorescent pieces are uniformly and accurately attached to the surfaces of three invertedLED wafers which are connected in series, and then the curing and molding are performed. Moreover, the distance between the wafers is controlled to be less than 120 microns through a reflow solderingeutectic process, so the problems of double images and projection shadows of a multi-wafer light source module are well solved, and the high-reliability vehicle LED light source packaged through theceramic fluorescent pieces has mass production feasibility and can meet the high-reliability requirement of the market.