107 results about "Ghosting" patented technology

The present invention relates to digital image processing and high dynamic range image processing. In order to detect and correct ghosting generated in the image fusing process, the present invention adopts the technical scheme that a multi-exposure image fusion ghosting removing method based on background modeling comprises the following steps of: firstly, converting images into gray-scale images and carrying out histogram equalization on a series of images with different exposures, so that the gray scales of the images are normalized into a range of 0 to 255 and are uniformly distributed; then a moving object is used as a singular point and obtaining a modeling result of removing a background image of the moving object by utilizing median filtering; respectively subtracting the background image from the images comprising the moving object to obtain the images which only comprise the moving object after the background is removed; and by morphological operations comprising erosion and dilation, carrying out precision processing on the moving object images, using an obtained result as a weight map and then according to the weight map, carrying out fusion on the images. The multi-exposure image fusion ghosting removing method based on background modeling is mainly applied to digital image processing.

The invention belongs to the digital image processing field and relates to a processing method for retaining a ghosting-free moving object in high-dynamic range image fusion. The objective of the invention is to remove ghosting and retain a high-exposure degree moving object in an image fusion process. The method includes the following steps that: images are converted into gray-scale images, and histogram equalization is carried out on the series of gray-scale images; with a moving object adopted as a singular point, a modeling result of a moving object-removed background image is obtained by utilizing median filtering; the background image is subtracted from the images comprising the moving object, so that images which only comprise the moving object after the background is removed can be obtained; and through morphological operations comprising erosion and dilation, precision processing is performed on the moving object images, large values are assigned for weight portions corresponding to the pixels of the moving object to be retained by using the coordinate information of the pixels of the moving object to be retained, and an obtained result is adopted as a weight map, and then, fusion is performed on the images according to the weight map. The processing method of the invention is mainly applied to digital image processing.

The invention discloses a 3D (three-dimensional) stereo video single-file double-video stream recording method. In file recording, the method comprises the steps of: defining the ID of a left video stream as N and the ID of a right video stream as N+1, and synthesizing the two video streams in one video file; in synthesis of 3D video file, the method comprises the steps of: taking the start timestamp of the video file as a uniform reference, and adding a timestamp mark to each video frame of the two video streams such that the video frames corresponding to the left and right video streams have the same timestamp. In the invention, the IDs of the left and right video streams are defined by an odd number and an even number so that the left and right video streams can be automatically identified when played; and moreover, the timestamp mark is added to each image frame by taking the start timestamp of the file as reference so as to achieve accurate time synchronization of the left and right images, obtain strong third dimension of the images and prevent ghost and the like. According to the invention, the file format adopts standard formats such as ASF, MKV, AVI and the like for writing; and when in playing, modification of a player is not required so as to facilitate use.

A compression algorithm, utilizing a new concept called “shared pixel parts”, allows for transmitting, receiving, and displaying full HD stereoscopic 3-D video with no loss of resolution or frames to either eye, using a single conventional TV channel and any type of conventional display without alteration. Depending on the type of display, viewers wear either passive polarized glasses, a new type of static colored-filter glasses, or a new type of active colored-filter glasses to view the same data signal showing virtually ghost-free full-color images with great depth, a wide angle of view, and a bright flickerless image which doesn't produce any discomfort even after extended viewing. The signal is also compatible with current 3-D-ready TVs. The compression algorithm can also be used to transmit and display a 2-D image with double the resolution of the display being used for viewing, with no increase of required transmission bandwidth.

The invention relates to the field of liquid crystal televisions, and discloses a ghost image elimination method for a shutter type three-dimensional (3D) liquid crystal television. The method specifically comprises the following steps of: setting N paths of pulse width modulation (PWM) circuits and N groups of light-emitting diode (LED) backlight units, wherein the N paths of PWM circuits correspond to the N groups of LED backlight units respectively; dividing each frame of image into N areas corresponding to the LED backlight units, and after a frame of image is prepared, controlling the Nth path of PWM circuit by using a master chip to turn on the Nth LED backlight unit and display the Nth area of the first frame of image, and preparing the data of the (N-1)th area of a second frame; and finishing the display of the N areas within the display time of each frame of image. The image of each area is displayed in turn to finish round robin within the display time of each frame of image, the corresponding contents of the next frame are prepared when the image of the previous frame is displayed, and the preparation time of each frame of image is the sum of (N-1)/N of the image display time and blanking time. Therefore, the slow response of liquid crystal molecules is greatly improved, and the ghost images of the shutter type 3D liquid crystal television are effectively eliminated without influence on watching effects.

Faced by the problems that in the foreground extraction of a visual background extractor (ViBe), ghosting arises easily and that when an object with dense foreground passes a ghosting area, it requires a long time to eliminate the ghost, the invention provides a ViBe-oriented foreground ghosting removal method in order to suppress the generation of ghosting and remove the ghosting rapidly in the foreground extraction. According to the invention, a second frame detects the number and the positions of the ghosting generated in the model initializing process, on the basis of which, an intersection set between the background area of each frame after the second frame and the detected background area is obtained. The obtained set represents the background points blocked by the foreground of the first frame but not blocked by the current frame. The re-initialization operations are made to the local model of the pixels in the set and after the operations, the set of the ghosting areas is updated until the ghosting is completely removed. The method of the invention can remarkably suppress and remove ghosting generated from the foreground extraction by a visual background extractor.

The invention provides a network model training method, an image processing method and related equipment, and relates to the technical field of images. The network model training method comprises the steps: obtaining a first training image pair; training an initial generator by using the first training image pair to obtain an intermediate generator; obtaining a second training image pair; and training an initial network model by using the first training image pair and the second training image pair to obtain a first target network model. A deep learning method is utilized, demosaicing, noise reduction and super-resolution are carried out in a combined mode, and in the process of converting the image of an RAW domain into the image of an RGB domain, the purposes of reducing noise, reducing ghosting and improving the definition of the image are achieved.

The invention provides an image processing method and device applied to a VR scene and electronic equipment, and the method comprises the steps: obtaining the first head posture information of a userat a first moment, and employing a virtual camera to shoot and obtain the corresponding first image information according to the first head posture information; Generating a first head posture matrixaccording to the first head posture information, and caching the first image information according to the first head posture matrix; Obtaining a second head posture matrix corresponding to second headposture information of the user at the moment when the target image is displayed on the display screen; And generating a target image displayed on the display screen according to the first head posture matrix, the second head posture matrix and the cached first image information. According to the embodiment of the invention, the target image conforming to the current head posture of the user canbe obtained, the consistency of the image frame rate and the display timeliness can be ensured, the defects of image jitter and ghosting are overcome, and the watching experience of the user is improved.

The invention discloses a real-time video stitching algorithm based on optimal stitching line updating, and belongs to the technical field of digital image processing. According to the real-time videostitching scheme provided by the invention, an image registration technology based on feature points is used, and internal and external parameters of each camera are estimated through matching feature points of different view angle images. When image fusion is carried out between different visual angles, an image fusion algorithm based on an optimal suture line is adopted, and a background elimination algorithm based on KNN and a suture line updating algorithm based on dynamic programming are used for avoiding discontinuity and ghosting caused by the fact that a moving object penetrates through the suture line in order to solve the problems of ghosting and discontinuity possibly generated by the moving object. And finally, splicing seams are eliminated by using an image fusion algorithm based on a convolution pyramid. By adopting the video stitching scheme designed by the invention, a good large-view-field image can be provided under the condition of ensuring the stitching real-time performance.

The invention discloses a live broadcast method at a dynamic viewing angle of 0-360 degrees. The method comprises the following steps: firstly, acquiring a 360-degree panoramic image through a 360-degree panoramic camera; then, cutting the 360-degree panoramic image; reserving a required visual angle scene of any angle between 0 degree and 360 degrees as a visual area; replacing the cut image datawith black, setting the black image data as a shielding area, transmitting the data of the visible area part and the data of the shielding area part to a cloud server after the data of the visible area part and the data of the shielding area part are encoded by an encoder for video encoding, and obtaining a viewable video at any angle between 0 degree and 360 degrees through video decoding and 360-degree video rendering. The 360-degree panoramic camera is used for collecting 360-degree videos, the videos at any angle of 0-360 degrees are achieved through a video cutting and filling algorithmand a video center point moving algorithm, then 360-degree impression experience is achieved through a video encoding and decoding technology, the phenomena of image splicing, ghosting and the like can be avoided, and the impression of a user is improved.

Surveillance techniques to monitor areas to identify persons, faces of persons and fake items are disclosed. Monitoring and recording using remote surveillance units (RSUs) of audio, video, infrared, ultraviolet, infrared and/or lidar is disclosed. The analysis and display of surveillance data with means for analysis and utilizing ghosting is disclosed. In another embodiment, detection of damage by (1) discoloration of a damaged responsive surface or (2) release of a notice gas from a damaged responsive surface is disclosed. The detection of the proximity of an object near a responsive surface due to changes in capacitance is disclosed. In another embodiment, the use of multiple interconnected RSUs with inverted audio alarm means is disclosed.

The embodiment of the invention discloses an image processing method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring a plurality of images with different exposure parameters, then, respectively carrying out adaptive segmentation on each image; dividing each image into a plurality of subareas; combining the subareas with the same image content into a subarea set to acquire a plurality of subarea sets, then aligning each subarea set, carrying out high dynamic range synthesis on each subarea set to obtain a corresponding high dynamic synthesis subarea, and finally generating a high dynamic synthesis image of a plurality of images according to the high dynamic synthesis subareas corresponding to the plurality of subarea sets. In the embodiment of the invention, the electronic equipment performs local region alignment on the image so as to eliminate ghosting, so that the quality of high dynamic range synthesis is improved when highdynamic range synthesis is performed.

The embodiment of the invention provides an image noise reduction method and device, a terminal and a storage medium, and the method comprises the steps: obtaining a first image group, wherein the first image group comprises a reference image and a plurality of frames of alignment images aligned with the reference image, each alignment image comprises a plurality of target points, the plurality of target points of each aligned image are in one-to-one correspondence with the plurality of feature points in the reference image; rejecting the first ghosting image according to local difference information between the alignment image and the reference image; removing the second ghosting image according to the overall difference information between the alignment image and the reference image; and fusing the aligned image without the first ghosting and the second ghosting with the reference image, and outputting a final image. By using the advantages of the multi-frame image, the noise reduction and dynamic range improvement are realized, the rapid operation can be realized, the real-time purpose is achieved, the motion scene can be properly processed, the ghosting is well inhibited, and the image quality is not influenced.

The invention discloses a multi-sensor real-time fusion method for a mobile robot remote takeover scene. According to the multi-sensor real-time fusion method, the laser radar and the multiple camerasare used for collecting data, the defects of a single sensor are overcome, and the laser radar and the multiple cameras are fused to generate a video stream with depth information, so that an operator knows distance information of surrounding objects. Aiming at the problems of splicing seams and ghosting in a linear fusion method, the invention provides a linear fusion method with a threshold value, so that the splicing seams and the ghosting are eliminated while the real-time performance is ensured, and the splicing effect is improved. In combination with the actual situation of an intelligent mobile robot, real-time fusion of multiple paths of high-definition video streams with depth information is completed through offline parameter updating and adoption of a specific parallelization method, the time delay of data transmission is reduced while redundant information of multiple sensors is removed, and the real-time fusion method has actual application value.

The invention discloses a video fusion algorithm based on a dynamic optimal suture line and an improved fade-in and fade-out method, and belongs to the technical field of video information processing. The method comprises the steps of: acquiring a real-time video stream with an overlapping area through a multi-view camera system; extracting feature points of the overlapped region, performing rough matching and purification on the feature points, and calculating an internal reference matrix and a rotation matrix of each camera by combining a beam adjustment method; performing spherical projection on images between different visual angles and twisting the images to the same reference plane, and then splicing video images by adopting an optimal suture line algorithm, aiming at the situation that a moving object passes through an image overlapping area to generate double images and blurring; adopting an optimal suture line dynamic updating algorithm to avoid ghosting and blurring phenomena generated when a moving object penetrates through a suture line; and obtaining a wide-view-field and high-resolution panoramic spliced video with a good visual effect by using an improved fade-in and fade-out fusion algorithm. The steps are simple, and the spliced image with a good visual effect can be quickly obtained.

The embodiment of the invention discloses a method for generating an HDR image and electronic equipment, and the method comprises the steps: the method is applied to the electronic equipment comprising a first camera and a second camera, the electronic equipment displays a first interface, and the first interface comprises a preview image and a first control; detecting a first operation on the first control; in response to the first operation, the first camera collects a first original image, and the second camera collects a second original image; performing first processing on the first original image to obtain a first image; performing second processing on the second original image to obtain a second image; according to the method, the first image and the second image are subjected to weighted fusion to obtain the HDR image, the images with different exposure coefficients are synchronously acquired based on the double cameras, and the HDR image is obtained based on fusion of the images with different exposure coefficients, and the images are acquired synchronously, so that exposure duration difference does not exist, and the problem of motion ghosting can be effectively avoided.

The invention discloses a station caption adjustment method, an OLED television and a storage medium. The method comprises the following steps: reading a video pixel brightness parameter in real time,the video pixel brightness parameter being a pixel color brightness parameter value of each pixel in a screen; comparing the change of the video pixel brightness parameter of the video area in the set time, and determining a fixed station caption area; and adjusting the pixel brightness of the fixed station caption area, and carrying out marginalization processing on the fixed station caption area. According to the invention, the fixed station caption area in the playing picture is detected in real time; the pixel brightness of the determined fixed station caption area is reduced, marginalization processing is carried out on the fixed station caption area, the boundary of the fixed station caption area is weakened, static ghosting generated when the OLED screen television is played in thefixed station caption area for a long time is eliminated, and the picture playing quality is improved.

The invention discloses a ghost weakening system and a ghost weakening method for electrophoretic electronic paper. The ghost weakening system for the electrophoretic electronic paper comprises a maincontrol module, a camera module, a display module, a storage module, a driving module and a PC terminal. The weakening method for the ghost weakening system comprises the steps that the camera moduleshoots an image of the electronic paper display screen and stores the image in the storage module; image preprocessing is carried out on an image through a PC (Personal Computer) end, wherein the image preprocessing comprises primary gray scale transformation, histogram equalization, secondary gray scale transformation, binaryzation and median filtering; acquiring an image, then performing edgeextraction on the image, cutting out an area with characters, performing character recognition, and finally reversely driving the recognized characters once through a driving module to weaken a ghostphenomenon. According to the invention, the characters of the image ghosting are recognized and are reversely driven by the driving module once, so that the effect of weakening the ghosting is achieved, and the reading experience of a user is greatly improved.

The invention provides a nighttime monitoring video quality improvement method assisted by daytime image reference. The method is based on a homomorphic filtering basic theory. Different from a traditional homomorphic filtering algorithm, the method has the advantages that daytime background information in the same scene is used as a reflection component to calculate an illuminance component of anight background, similar pixel blocks of the foreground are found around the night foreground, the illuminance component of the foreground is constructed according to the similar block relation of the foreground and the background, and the purpose of enhancing the night foreground is achieved. According to the method, the problems of enhancement failure and image discordance in an extremely darkarea are well solved. According to the method, the phenomena of'lumps' and'halo 'in an SSR algorithm and an MSRCR algorithm are eliminated, the problem of'ghosting' in a Denoising algorithm is eliminated, the problem of color distortion in a classic algorithm in the prior art is greatly improved, the contrast of images after quality improvement is enhanced obviously, and the overall quality of a night monitoring video is greatly improved.

The invention provides a video image processing method and a video image processing device; a 3D video image comprising a left viewpoint image and a right viewpoint image is input; a first compensation image and a second compensation image are obtained according to the left viewpoint image and the right viewpoint image, and inserted into the left viewpoint image and the right viewpoint image; four-frame images of the left viewpoint image, the first compensation image, the right viewpoint image and the second compensation image are displayed after frequency multiplication line-by-line refreshing; and the four-frame images are accumulated to form the left viewpoint image or the right viewpoint image. According to the system and the method, the technical problem that the ghosting phenomenon exists in an existing shutter 3D display technology while the watching is performed by naked eyes is solved; a 2D vide image without ghosting can be obtained simultaneously when a 3D video image is displayed, and therefore, 2D and 3D are compatibly displayed simultaneously; and the purpose that 2D and 3D images can be simultaneously watched by different crowds can be achieved.