62results about How to "Improve blur" patented technology

An imaging device for imaging an object to generate a color image data, comprising: a solid-state imaging element comprising: a photoelectric conversion layer laminated above a semiconductor substrate; photoelectric conversion elements comprising at least two kinds of photoelectric conversion elements aligned on the semiconductor substrate for detecting colors different from that to be detected by the photoelectric conversion layer; and a signal reading circuit formed on the semiconductor substrate for reading out color signals according to signal charge accumulated in the photoelectric conversion elements and signal charge generated in the photoelectric conversion layer; and a color signal generating unit for generating a color signal constituting one-pixel data in the color image data on the basis of signal charge accumulated in a first photoelectric conversion element corresponding the one-pixel data among the photoelectric conversion elements and signal charge accumulated in at least one second photoelectric conversion element, adjacent to the first photoelectric conversion element, for detecting the same color as that of the first photoelectric conversion element.

The invention discloses an anti-glare flame-retardant polycarbonate material which comprises the following components by weight percent: 85-95 percent of polycarbonate, 2-12 percent of flame retardant, 0.1-1 percent of flame-retardant synergist, 0.1-5 percent of anti-glare agent, 0.3-2 percent of anti-ultraviolet agent, 0.05-3 percent of blue light absorbent, 0.2-3 percent of antioxidant and 0.2-2 percent of lubricating agent. The anti-glare flame-retardant polycarbonate material not only has an excellent anti-glare function, but also has higher light transmittance, and is capable of absorbing the blue wave band of the LED lighting spectrum at the same time. The invention further discloses a preparation method of the anti-glare flame-retardant polycarbonate material, which comprises the following steps: firstly preparing flame-retardant masterbatches, organic silicon microbead anti-glare masterbatches and blue light absorption masterbatches, and then preparing the anti-glare flame-retardant polycarbonate material, so that the performance anti-glare flame-retardant polycarbonate material is improved. The anti-glare flame-retardant polycarbonate material is particularly suitable for preparing an optical housing for LED lighting.

The invention provides an electronic image stabilization method based on inertial navigation information. The method comprises the following steps that: firstly, feature points of each image of a video are extracted, subsequently the overall motion estimation is carried out on two adjacent frames by using a feature point matching algorithm; if the overall motion estimation has larger errors, the overall motion is estimated by using motion information obtained by an inertial navigation system so as to compensate the errors caused by the fact that the overall motion is estimated by merely using an image; the overall motion estimation adopts an affine transformation model; then motion filtering is carried out on the model in combination with the image and the inertial navigation information so as to distinguish the active motion and a trembling component of a camera; and moreover, a rotational component introduced because of the trembling is determined; and furthermore an original video image is compensated by using the obtained trembling and rotational components so as to obtain a stable video; and the image restoration treatment is carried out on each frame of image of the video in combination with the image information and the inertial navigation prediction motion information. The method needs no complex hardware and is good in electronic image stabilization; and the calculation precision is improved.

The invention discloses a restraining method on edge Halo effects during the process of resetting a projections onto convex sets (POCS) super-resolution image, and belongs to the technical field of image processing techniques. The restraining method on the edge Halo effects during the process of resetting the POCS super-resolution image comprises a step of reading the sequence of a low-resolution image, a step of acquiring a high-resolution initial estimated value of convex sets by applying of a wavelet bicubic interpolation, a step of carrying out motion estimation to achieve low-resolution image registration by applying of a pre-filtering sub pixel iteration method, a step of determining a point spread function with edge-preserving characteristics by comprehensively considering space position information and grey information restraint of high-resolution estimation image pixels, a step of determining a length-variable loose projection parameter value according to relevance between all low-resolution observation frames and reference frames, and a step of resetting super-resolution by applying of a POCS method and combining of the high-resolution initial estimated value, the point spread function with the edge-preserving characteristics and the length-variable loose projection parameter value. Therefore, the image space resolution is improved and the ideal high-resolution image is obtained. By means of the restraining method, the edge Halo effects in the reset image can be reduced to a very large extent.

The invention relates to a water surface and underwater target classification method based on interference fringes and deep learning. Firstly, a Gaussian beam acoustic model is used for simulation toobtain a large number of acoustic field interference fringe images, which comprise a clear image, a blurred image and an image with random deep errors, as a training set for a deep belief network (DBN) and a CNN; the well-trained DBN is then used as a front-end processing module of the CNN, and beams are added at the same time to form a pre-processing module of the system; and finally, a pre-processed image is inputted to the DBN, and the inputted fringe image is thus autonomously classified. The method solves the problem that the training set sample is too small in deep learning by simulatinga large number of acoustic field interference fringe images. The DBN model is designed to optimize the blur and fringe shift problems in the actual input image. The CNN is designed to autonomously classify water surface and underwater targets.

A pixel unit of an organic electroluminescent display having a light emitting area and a transparent area is provided. The pixel unit includes a scan line, a data line, a control element electrically connected to the scan line and the data line, a power line electrically connected to the control element and an organic light-emitting diode disposed inside the light emitting area and electrically connected to the power line and the control element. The power line includes a main portion and a plurality of branch portions connected to the main portion. The branch portions are disposed inside the transparent area and electrically independent from the data line. A width of each of the branch portions is identical to a minimum line width of the data line.

The invention relates to an edge detail protection-based infrared image super-resolution reconstruction method and belongs to the image processing technical field. On the basis of a traditional POCS (projection onto convex set) reconstruction algorithm, the edge detail protection-based infrared image super-resolution reconstruction method is provided in order to solve the problem that the traditional POCS reconstruction algorithm heavily relies on high-resolution initial estimation images and ringing phenomena in reconstructed images. According to the method, a low-resolution image sequence is read; a high-resolution image initial estimation value is obtained through using an iterative curvature interpolation method; and the high-frequency detail region and smooth region of an infrared image are iteratively corrected through adopting different point spread functions, so that the edge oscillation effect of a reconstructed image can be decreased. With the method adopted, an image edge detail-well protected reconstruction result can be obtained.

Operating a display of a user equipment. The present invention relates to a method for operating a display (2) of a user equipment (1). The display (2) is configured to display an object (50). A permission is assigned to the object (50), which defines an allowance for a person (11-13) to view the object (50). According to the method, a gaze position (63) of a person (12) looking at the display (2) is determined, which defines a position on the display (2) at which the person (12) is looking. A distance between the gaze position (63) and a position of the object (50) is determined and furthermore, it is determined, if the person (12) is allowed to view the object based on the permission assigned to the object (50). If the person (12) is not allowed to view the object (50), the object (50) is concealed depending on the determined distance between the gaze position (63) and the position of the object (50).

The invention discloses a three-frequency ambiguity solution method for Beidou precise point positioning. The three-frequency ambiguity solution method for Beidou precise point positioning comprises the steps of: (1), obtaining the floating point ambiguity of the first EWL by using a HMW combination, obtaining a fixed solution after rounding, and obtaining a high-precision observation value; (2),introducing two pseudo-distance combination observation values, setting a constraint condition by using the first high-precision extra-wide-lane observation value and the two pseudo-distance observation values at the same time, obtaining the floating point ambiguity of the second EWL, and obtaining a fixed solution after rounding; (3), performing linear combination by using the two high-precisionEWL fixed solutions so as to obtain the WL combined fixed ambiguity, and obtaining the WL high-precision observation value; and (4), setting a constraint condition by using a wide-lane observation value and a pseudo-distance observation value at the same time, so that an NL combined ambiguity is obtained. The three-frequency ambiguity solution method for Beidou precise point positioning in the invention analyzes signals in three frequencies of B1I, B2a and B3I; the high-precision floating point solution is obtained; therefore, the rounded fixed solution can have relatively high precision; andthus, a reliable locating result can be obtained.

An image processing apparatus and a radiation tomography apparatus capable of acquiring a tomographic image of high visibility are provided. The image processing apparatus initially generates a map m showing an extending direction of a linear structural object seen in an original image P0. Then, the apparatus generates an extrapolated image P1 by adding the area positioned at the side section of the original image P0 to the side section so as to extend the linear structural object seen in the side section by referring to the map m. By adding the side section to the original image P0 while shifting the pattern of the side section so as to extend the linear structural object seen in the side section of the image, the subject image can be naturally compensated. After this operation, by generating the tomographic image D, a tomographic image D having improved visibility can be generated.

The invention relates to a spaceflight optical remote sensing high-density quantization information processing method, and belongs to the technology of spaceflight remote sensing. The ground resolution of a conventional spatial optical remote sensor is difficult to improve at an on-orbit imaging stage because of the limit of the size of a detector, and a large amount of noise is generated on a focal plane because of the degeneration of the detector, thereby causing spot noise in an obtained image. An ideal effect is difficult to achieve through a conventional digital image processing method. Starting from the quantum state of information, the method gives consideration to the superposition state information, achieves the de-noising through the filtering in a quantum space, synthesizes a plurality of pieces of quantum information, and finally improves the information amount obtained by a system. The method is suitable for a satellite degeneration image or an image which is difficult to process on the ground, and hopefully, the method can be applied to the satellite-borne processing after the calculation complexity is reduced.

The present application is applicable to the field of image proces technology, and provices a method for generating a high dynamic range image, a mobile terminal and a computer readable storage medium. The method comprises: obtaining a preview picture collected by a camera of the mobile terminal, and identifying foreground objects in the preview screen, based on the foreground objective, selectinga preset number of cameras from a plurality of cameras of the mobile terminal as a target camera, and determining the exposure time corresponding to each target camera, controlling the target camerato acquire images based on the corresponding exposure time, obtaining the image corresponding to each exposure time, and synthesizing the image corresponding to each exposure time to obtain a high dynamic range image. The application can reduce the shooting time of the high dynamic range image and solve the problem of blurring the outline of a local object.

The invention discloses a method and an apparatus for optimizing a motion image in a virtual reality scene. The method comprises the steps that: a rendering thread renders contents to be displayed frame by frame, is in a rendering state in the rendering process, and enters a waiting state every after rendering one frame; a send-to-screen thread reads a newest rendering result frame by frame according to the same frequency with a screen refresh rate; the send-to-screen thread, according to a reading result, judges a state of the rendering thread, when judging that the rendering process is in the waiting state, sends the reading result to a screen for display after processing the reading result and awakens the rendering thread, and when judging that the rendering process is in the rendering state, executes a one-time screen clearing operation; and after being awakened, the rendering thread carries out rendering on a next frame of contents. According to the technical scheme disclosed by the invention, by asynchronously refreshing the rendering thread and the send-to-screen thread, the problems of reduction of system performance and low rendering frequency which are brought in when a screen refreshing rate is improved are effectively solved, so that an afterglow effect is effectively reduced, and the problem of image blurring under the motion condition is improved.

The invention provides an image processing method, an image processing device, a computer readable storage medium and electronic equipment, and relates to the technical field of image processing. The image processing method disclosed by the invention comprises the following steps: acquiring continuous multi-frame images; extracting foreground region images of the at least two frames of images; according to the maximum pixel value of each foreground sub-region in each foreground region image, fusing the foreground sub-regions to form a foreground fused image, wherein the foreground sub-regions are sub-regions of the foreground region image; and generating a target image based on the foreground fusion image and the background region image of the at least one frame of image. The technical scheme of simulating the long exposure effect is achieved through multi-frame image fusion, the method and device are suitable for electronic equipment such as a smart phone, a clear and continuous motion track can be reserved in the finally output target image, and meanwhile the influence of noise is reduced.

An image processing method includes acquiring first information that indicates a size of an image of a specific subject in a target image relative to a size of the target image, acquiring second information that indicates a size of the specific subject, acquiring third information with which an angle of view of the target image can be determined, and estimating a subject distance, which is a distance from an image pickup apparatus that has generated the target image to the specific subject on the basis of the first information, the second information and the third information.

The invention relates to a logic encryption defense method based on key gate position selection. The method mainly comprises two parts: data preprocessing and a key gate position encryption algorithm.Specifically, in the encryption process, the positions of the secret key doors are marked in a data preprocessing mode, paths with too long delay are removed, and then a two-stage encryption algorithm is adopted for implanting the positions of the secret key doors and a main guide door. Finally, the Hamming distance and an area overhead are calculated to evaluate the security standard of the encrypted circuit. According to the method, the gate-level netlist encryption effect can be improved; the Hamming distance is greater than 50% through a dominant gate implantation algorithm; the output ambiguity is maximized, the security protection circuit is simple in structure, can resist secret key sensitization attacks, simultaneously reduces area overhead, comprehensively improves security evaluation requirements of logic encryption, reduces problems of IC theft and IP piracy in a chip design stage, and can effectively prevent design information from being maliciously stolen, so that the security protection circuit can be universally applied to logic encryption, and has relatively strong practicability.

The invention provides an optical scanning device and an image forming device. The optical scanning device comprises a light source used for diverging a light beam, a first optical unit used for enabling the light beam emitted by the light source to be collimated in a main scanning direction and to be focused in an auxiliary scanning direction, an optical deflector used for deflecting the light beam emitted by the light source, and an imaging optical system used for guiding the light beam deflected by the optical deflector to a scanned surface to be imaged. When the optical deflector is used for deflecting the light beam emitted by the light source at a maximum deflection angle, the emergent light beam of the imaging optical system forms a maximum incident included angle phi max together with a normal line of the scanned surface in the main scanning direction, and the light spot inclination rate, defined in the description, of a light spot formed on the scanned surface by the imaging optical system meets a condition defined in the description, so that the possibility of edge blur of a transfer image is reduced, and the optical scanning imaging quality is improved.

The invention discloses an image style conversion and model training method and device. The method comprises the steps of: respectively inputting a first style image and a second style image into a content encoder and a style encoder in an encoder network, and respectively extracting a content encoding feature image and a style encoding feature image; respectively inputting the style and the content coding feature image into a decoder network to obtain a target image converted from a first style to a second style. According to the method, an image style conversion model formed by the coder anddecoder networks is obtained by pre-training an image training sample comprising a plurality of first and second style images and instance images cut from the image training sample. By applying the method, the style conversion adaptability to various different scenes can be improved, the problems of image blurring and poor instance effect in the image after style conversion are improved, and meanwhile, coarse-grained and fine-grained high-quality image style migration is realized.

The invention provides a liquid crystal display driving method, which comprises the steps of dividing a liquid crystal display into a plurality of display areas, and driving the plurality of display areas in the partition mode; setting a backlight coefficient corresponding to the backlight of each display area in the liquid crystal display as a first backlight coefficient within the front half time of the displaying process of one frame of a picture on the liquid crystal display, wherein the first backlight coefficient is smaller than or equal to a maximum backlight coefficient in the backlight corresponding to each display area of the liquid crystal display; setting a backlight coefficient corresponding to the backlight of each display area in the liquid crystal display as a second backlight coefficient within the latter half time of the displaying process of one frame of the picture on the liquid crystal display, wherein the second backlight coefficient is smaller than the first threshold value. In this way, a preset picture can be inserted into the frame of the picture during the latter half time of the displaying process of the picture on the liquid crystal display. The difference between the gray scale of a preset picture and the grey scale of a full-black picture is within a preset range. By adopting the technical scheme provided by the invention, the picture fuzziness when a moving picture is displayed on the liquid crystal display can be improved.

Operating a display of a user equipment. The present invention relates to a method for operating a display (2) of a user equipment (1). The display (2) is configured to display an object (50). A permission is assigned to the object (50), which defines an allowance for a person (11-13) to view the object (50). According to the method, a gaze position (63) of a person (12) looking at the display (2) is determined, which defines a position on the display (2) at which the person (12) is looking. A distance between the gaze position (63) and a position of the object (50) is determined and furthermore, it is determined, if the person (12) is allowed to view the object based on the permission assigned to the object (50). If the person (12) is not allowed to view the object (50), the object (50) is concealed depending on the determined distance between the gaze position (63) and the position of the object (50).

The invention belongs to the technical field of image processing, and relates to a license plate image processing method based on an attention mechanism, which comprises the following steps of: enlarging a receptive field by using convolution kernels of different sizes, processing features by using instance standardization, and recalibrating a mean value and a variance of the image features under the condition of not being influenced by the batch processing size; a cross-stage attention module is designed between different stages, and multi-scale residual features of the previous and later stages are fused, so that the loss of the features during transmission can be made up, more valuable image features can be amplified, the network gradually pays attention to feature information facilitating deblurring in the training process, and the deblurring efficiency is improved. Meanwhile, a local attention mechanism is used in the stage, network learning is assisted for deblurring, an energy loss function is improved, the overall training effect of a network model is improved, and a good effect is achieved in the aspect of reconstructing a clear license plate image.

The present invention reduces moving image blurring in a hold-response-type display device. 1 frame is divided into 3 fields. Assuming the gradation-brightness characteristic of a first field as 1 g, the gradation-brightness characteristic of a second field as 2 g, and the gradation-brightness characteristic of a third field as 3 g, the third field is set at an initial stage or at a final stage of the frame. Due to such setting, the moving image blurring can be effectively reduced up to the relatively high brightness.

The invention discloses a modified full-bio-based PBAT biodegradable plastic and a preparation method thereo. The preparation method comprises the following preparation processes of (1) taking phytic acid, itaconic acid, epichlorohydrin, sodium hydroxide and feather meal to prepare a bio-based flame retardant, (2) taking cellulose acetate, lactic acid, wheat gluten protein and epoxidized soybean oil to obtain the bio-based polymer, and (3) melting and blending poly(butylene adipate-co-terephthalate), polylactic acid, the bio-based polymer and the bio-based flame retardant to obtain the bio-based degradable plastic. By adding the PLA, the bio-based flame retardant and the bio-based polymer into a PBAT system, fuzziness of a phase interface between the PLA and PBAT components can be promoted, the interface bonding effect between the two phases is enhanced, the compatibility is effectively improved, a blend is dispersed more uniformly, the comprehensive mechanical property of the system can be improved, and further improvement of the flame retardant property of the system is promoted.

The invention discloses an image processing method for converting a 2-dimensional image into a 3-dimensional image, comprising the following steps: amplifying an original 2-dimensional image; sharpening the depth map of the 2-dimensional image; and according to the amplified 2-dimensional image and the sharpened depth map, carrying out steps of motion, broken hole compensation, reduction, synthesis and the like on the 2-dimensional image to obtain the 3-dimensional image. In the step of the broken hole compensation, the image processing method disclosed by the invention generates the gray scales of broken holes according to the average value or mid-value of the gray scales of a plurality of pixels at the same side of the broken holes, thus blurry or broken phenomenon on images or characters in the 3-dimensional image can be improved.

The invention belongs to the technical field of stereo projection and provides a stereo projection system with a single projector and dual liquid crystal light valves. Along the direction of light path, the stereo projection system sequentially comprises a projector, a linear polarizer, a first optical self-compensated bend (OCB) mode liquid crystal light valve, a second OCB mode liquid crystal light valve, a polarization state converter, a projector screen, circular polarized glasses and a synchronous circuit. Stereo projection at higher frame rates is achieved by means of the stereo projection system with the single projector and the dual OCB mode liquid crystal light valves, the dynamic compensation driving technology and the dual OCB mode liquid crystal light valves capable of adjusting the polarization state of light at a high speed. Besides, the structure is simple, the cost is low, and by using one projector to match with the dual liquid crystal light valves and a mating driving circuit, audiences can watch stereo projection through wearing pairs of circular polarized glasses.

The inventioni provides a surface-modified metal oxide powder obtained by subjecting a metal oxide powder to a surface treatment, characterized in that the ratio of the initial frictional charge (A) measured at 1 minute after the powder is dispersed in a ferrite carrier to the later frictional charge (B) measured at 30 minutes thereafter, A / B, is 1.0-2.0 and that the later frictional charge (B) is -300 to +300 [mu]C / g. The surface-modified metal oxide powder is less electrifiable and has better stability over time of frictional charge than conventional ones. Consequently, the powder, when used as an external additive, stabilizes various properties including flowability and electrification characteristics without considerably reducing the properties of a developer or the like. Thus, problems such as blurring and a decrease in image quality can be minimized.