84results about How to "Efficient splicing" patented technology

A system and method for providing connectivity to a foreign network for a device configured for communication over a home network without reconfiguring the device include intercepting packets transmitted by the device, selectively modifying intercepted packets which are incompatible with the foreign network to be compatible with network settings of the foreign network, and selectively providing network services for the device corresponding to network services available on the home network to reduce the delay associated with accessing the network services from the foreign network, or to provide network services otherwise inaccessible from the foreign network. Network services are provided by or through a configuration adapter connected to the device or to the foreign network. The configuration adapter accommodates incompatibilities resulting from proxy server requests, domain name server requests, and / or outgoing email service requests to provide transparent network access for mobile users without reconfiguration of the users computing device.

The invention discloses a three-dimensional image segmentation method and system based on a full convolutional neural network, and the method comprises the following steps: step1, collecting and obtaining a sequence image, and carrying out the marking of the sequence image, and obtaining training sample data; step 2, performing normalization preprocessing on the training sample data obtained in the step 1; step 3, applying the sample data processed in the step 2 to carry out supervised training on pre-constructed 3-D full convolution residual U-net network model, training to a preset convergence condition, and obtaining a trained three-dimensional image segmentation model; and step 4, after normalization processing is carried out on the sequence image data to be segmented, inputting the sequence image data into the three-dimensional image segmentation model trained in the step 3, and obtaining a sequence image segmentation result. The continuity information of the sequence can be fullyutilized, and a relatively good result can be obtained in three-dimensional image segmentation.

The present invention provides a system and method for mosaicing multiple input images, captured by one or more remote sensors, into a seamless mosaic of an area of interest. Each set of input images captured by the remote sensors within a capture interval are ortho-rectified and mosaiced together into a composite image. Successive composite images, along a given flight path, are then mosaiced together to form a strip. Adjacent strips are then mosaiced together to form a final image of the area of interest.

The invention provides an image connection method based on a SURF algorithm, relating to the technical field of image processing. The problem of an unideal image connection processing effect caused by a low feature point extraction speed and low robustness in by using a SIFT algorithm and a Harris algorithms is solved. Matlab is applied to drive an industrial camera, and the calibration of a camera is completed so as to use the camera to carry out video recording. A gray world method is used to carry out white balance adjustment on each frame of image in a recorded video. The SURF algorithm is applied to carry out feature point extraction on two photos which are recorded at the same time after the white balance adjustment. A RANSAC algorithm is applied to carry out mismatching feature point pair removing on the two images whose feature points are marked. The interpolation is used, the images which are processed by the RANSAC algorithm are connected together, the image connection is completed, and an image with a larger viewing angle is obtained. The method is especially suitable for the image connection under an industrial camera platform.

The invention discloses an OCT eye fundus image data registration method, which comprises the following steps: at first, extracting retina edges of an OCT eye fundus image by virtue of a Canny edge detection method, and collecting the edges in a format of point cloud data; then, extracting characteristic points of the point cloud data by adopting a space grid division method; next, calculating a transformation matrix of point clouds to be registered to eliminate obvious position errors by virtue of an SVD (singular value decomposition) algorithm; finally, performing accurate registration by virtue of an improved iterative closest point algorithm, and applying an obtained rotation matrix and translation matrix to the original OCT eye fundus image to obtain a final result. When a dense point cloud with a relatively large volume of data is processed, the method has obvious advantages in terms of time complexity and registration accuracy. Under most conditions, the efficiency of a conventional iterative closest point algorithm is improved by 70 percent by the method, not only can the OCT eye fundus image be effectively registered and spliced, but also the accuracy of a large-vision eye fundus retina accuracy is ensured.

The invention provides a video image stitching method. The method comprises the following steps: obtaining a plurality of video images from a video file; identifying video images, the subtitles of which are different, from the plurality of video images, wherein the subtitle of the video image obtained through next identification is different from that of the video image obtained through the previous identification in the video images, the subtitles of which are different; and carrying out image stitching on the video images, the subtitles of which are different. The invention also provides a video image stitching device. The method and device can device can realize image stitching of the video images, the subtitles of which are different, to facilitate a user to check and share the subtitles in the video.

The invention belongs to the technical field of stereo images, being able to realize effective stitching of stereo images, avoiding local deformation and reasonably reducing the vertical parallax. The invention relates to a stereo image stitching method based on content of images. The method comprises the steps: utilizing a consistent homography matrix after optimization to perform pre-deformation on the left and right images; utilizing an image tuning and warping method based on contents to optimize image stitching and avoid distortion of vision; and at last, utilizing a lookup algorithm based on kerf and a multi-tape mixed algorithm to acquire the final image stitching result. The stereo image stitching method based on content of images is mainly used for image processing.

The present invention provides a system and method for mosaicing multiple input images, captured by one or more remote sensors, into a seamless mosaic of an area of interest. Each set of input images captured by the remote sensors within a capture interval are ortho-rectified and mosaiced together into a composite image. Successive composite images, along a given flight path, are then mosaiced together to form a strip. Adjacent strips are then mosaiced together to form a final image of the area of interest.

The embodiment of the invention provides a splicing method and a splicing device of Lidar measurement data. The method mainly comprises the following steps of: carrying out multiple times of data acquisition on the same object in difference positions by utilizing LiDAR so as to acquire multiple point cloud data, carrying out two-to-two paring treatment on the multiple point cloud data, and uniformly segmenting two paired point cloud data to generate multiple segmenting units; selecting multiple feature units from a candidate feature unit set of the two paired point cloud data to form a matched feature unit set; forming a transformation matrix R according to a rotating shaft L and a rotating angle theta, and solving the values of the rotating shaft L and the rotating angle theta according to the principle of minimal error sum of squares of all the matched feature units; transforming all data points in the first point cloud data by virtue of the transformation matrix R and adding the obtained transformation data into the second clod point data to be combined into a complete data set. According to the embodiment of the invention, the feature regions of all the cloud point data and multiple matched feature unit pairs can be accurately and quickly determined based on the working parameters of LiDAE acquisition data, so as to quickly carry out effective splicing on the point cloud data subjected to multiple times of LiDAR sampling.

The invention relates to the field of video processing, and discloses a video stitching method and device, electronic equipment and a computer readable storage medium, and the method comprises the steps: detecting the similarity between a first image and a second image, wherein the first image is the last frame of image of a first to-be-stitched video, and the second image is the first frame of image of a second to-be-stitched video; then, if the similarity meets a preset condition, determining a motion vector of the first image relative to the second image; according to the motion vector, determining at least one compensation frame between the first image and the second image, and splicing the first image and the second image based on at least one compensation frame so as to splice the first to-be-spliced video and the second to-be-spliced video. According to the method provided by the embodiment of the invention, the smooth compensation can be efficiently carried out on jitter occurring at the joint in real time, so smooth transition can be carried out among multiple video segments, and the occurrence of image jitter or image hopping after the multiple video segments are splicedis effectively reduced.

The invention discloses a method for aerial photography map splicing based on super-pixels and SIFT. The method comprises the steps that (1) super-pixel self-adaptation segmentation is conducted on aerial photography maps; (2) three to ten super-pixel pre-matching pairs are obtained by using the color normalization average value of a plurality of super-pixels obtained in the step (1) as guidance information, SIFT features of the super-pixel pre-matching pairs are extracted, and the super-pixel pre-matching pairs which fail in matching are removed after matching of the SIFT features is conducted; (3) a supervisory mechanism is generated in an on-line mode, and matching point pairs meeting the requirements are obtained; (4) a transformation matrix is calculated through the matching point pairs which are accurately selected in the step (3), images are fused, and image splicing is achieved. The method for aerial photography map splicing based on the super-pixels and SIFT achieves rapid and efficient splicing of the aerial photography maps, and the effectiveness of the method is verified through experiments.

The invention provides a method and device for quickly splicing a three-dimensional model of a house at a mobile terminal. The method comprises the following steps of acquiring point cloud data shot at each preset shooting point in the house; sampling the acquired point cloud data to obtain sampled point cloud data; adding RGB information to the sampled point cloud data in order to generate pointcloud data with color information; and rendering and splicing the point cloud data with the color information at the mobile terminal in order to obtain a preliminary three-dimensional model of the house. According to the method and the device, the splicing of the three-dimensional model of the house can be quickly and efficiently completed at the mobile terminal.

The invention provides an automatic detecting and recognizing method of scroll captions in videos and belongs to the field of video retrieval. The automatic detecting and recognizing method includes steps of (1) detecting a scroll caption area; (2) filtering the detected vertical scroll caption area by a self-adaption detecting window so as to reduce repetition rate of caption recognition; (3) stitching multiframe images comprising horizontal captions into a frame of images comprising a complete scroll caption, inputting the frame of images as OCR (optical character recognition) software after binaryzation and recognizing the captions. Kinetic characteristics of the scroll captions in videos are sufficiently taken into consideration, fixed captions and scroll captions in the videos can be differentiated, the horizontal scroll captions are stitched, and the vertical scroll captions are filtered by the caption selection area algorithm. Accordingly, better recognition performances including higher recall ratio, precision ratio and lower repetition rate are obtained, and action of the scroll caption information of the videos in video retrieval is sufficiently given into play.

The invention discloses a sewage filtration device convenient to mount and clear. The device comprises a first mounting rack, filter nets, a first mounting rod, a splicing groove, first mounting rotating shafts and fixing plates, wherein the filter nets are mounted at the upper part and the lower part of the first mounting rack, the splicing groove is formed in one side of the first mounting rod,and each fixing plate is welded on one side of each of the first mounting rotating shaft. The sewage filtration device is scientific and reasonable in structure and safe and convenient to use; a grating and the filter nets are arranged, large-size solid wastes in sewage are cleared by the grating, part of small-size solid wastes in the sewage are cleared by the filter nets, and following sewage treatment is facilitated; second mounting rotating shafts and a power rotating shaft are arranged, the second mounting rotating shafts and the first mounting rack can be effectively driven to rotate through the power rotating shaft, one filter net with excessive solid wastes attached to the surfaces is rotated to the upper part, clearing of the solid wastes is facilitated, and besides, the other filter net can continuously have a function of filtering the small-size solid wastes.

The invention discloses a panorama video stitching method and system. The method includes the steps that firstly, a client side uploads multiple routes of synchronized panorama videos to cloud; the cloud completes stitching of the multiple routes of panorama videos for the received multiple routes of panorama videos and stores the stitched panorama videos; the client side downloads the stitched panorama videos from the cloud. According to the method, the process of stitching the panorama videos is completed through the cloud, the problem that the calculation capacity of the client side is not enough is solved, and rapid and efficient panorama video stitching is achieved.

A method for splicing reinforcement panels using a UV curable adhesive. The method includes the steps of: (a) aligning facing ends of two reinforcement panels; (b) applying adhesive on said facing ends of said two reinforcement panels; (c) butting together said facing ends of said two reinforcement panels; and (d) applying an ultraviolet light source to the adhesive. The method eliminates waste, reduces splicing time and improves the mechanical properties of the resultant mat.

The invention provides a method for splicing video panoramas with large view of field based on a hybrid projection transformation model. The method comprises the following five steps: step 1, performing downsampling on a video to reduce the amount of computation; step 2, detecting and extracting Sift feature points; step 3, clustering the feature points, performing region division on the image byusing an SUM method, and matching the feature points; step 4, calculating a hybrid projection matrix; and step 5, performing iterative operation to achieve registration and fusion of subsequent imagesequences and finally splicing the panoramas. By adoption of the method provided by the invention, the problem of large error of the traditional single projection matrix method and the problem of timewasting in the multi-projection matrix method are solved, and the method has a wide application background.

The invention discloses a broken grain splicing method which is used for processing a fingerprint image with broken grains and comprises the following steps of: 1. obtaining a directional diagram of the fingerprint image; 2. computing a broken grain ridgeline period and a broken grain curvature according to the fingerprint image and the directional diagram thereof; 3. drawing the effective area of a broken grain ridgeline according to the broken grain ridgeline period and the broken grain curvature; and 4. fitting the broken grain ridgeline in the effective area. By adopting the broken grain splicing method, the broken grains can be effectively spliced, false characteristic points caused by the broken grains are greatly decreased, the validity of fingerprint characteristic point matching is further improved, and meanwhile, because the broken grains are effectively spliced, the passing rate of ridgeline comparison is improved.

The invention discloses a fish eye monitoring image splicing method based on combination and matching of multiple feature points. The method comprises the following steps: carrying out distortion correction on an image by applying a barrel-shaped distortion model and Hough line detection; extracting N feature points from the selected areas of the two images and respectively solving feature descriptors of the feature points; pre-matching the N feature points by using a violent matching method or a fast nearest neighbor search method; performing confidence ranking on the pre-matching pairs of the N feature points; selecting two pairs of matching pairs with the highest confidence coefficient of each feature point to form an optimization set, searching for and reserving four groups of optimalmatching pairs, and eliminating other matching pairs; selecting a matching pair with a sub-high confidence coefficient in the obsoleted feature points, adding the matching pair into the optimization set, and performing optimization again until a termination condition is met; and calculating a homography matrix according to the four groups of matching pairs obtained through optimization, transforming the image and realizing splicing. According to the method, the homography matrix is calculated through the combination of the matching pairs among the multiple feature points, so that the selectionaccuracy of the matching pairs is remarkably improved, and then effective splicing of two monitoring images is realized.

The invention provides a honeycomb pore profiled steel sheet bidirectional overlapped floor slab, and belongs to the field of buildings. The honeycomb pore profiled steel sheet bidirectional overlapped floor slab is formed by pouring a concrete layer after a plurality of honeycomb pore profiled steel sheet-concrete prefabricated bottom plates are spliced. The honeycomb pore profiled steel sheet-concrete prefabricated bottom plates comprise honeycomb pore profiled steel sheets with flanges on the two sides and concrete bottom plates. The honeycomb pore profiled steel sheets are wave-shaped thin steel plates, and a plurality of evenly-distributed hollow-out holes are formed in the honeycomb pore profiled steel sheets. The lower flanges of the honeycomb pore profiled steel sheets and the concrete bottom plates are poured into a whole. The lower flanges of the honeycomb pore profiled steel sheets exceed the edges of the concrete bottom plates. Beveled edges tilting downwards are formed in the two side edges of the concrete bottom plates at the splicing positions. The honeycomb pore profiled steel sheet bidirectional overlapped floor slab has the characteristics of being low in dead weight, high in strength, easy to construct and free of formwork erecting, and effectively solves the splicing problem of the overlapped floor slab. Horizontal reinforced bars penetrate through holes of the profiled steel sheets so that the structural characteristics of the bidirectional floor slab can be achieved, the height of structural layers can be effectively reduced, dead weight is reduced, and good economic benefits are achieved.

The invention discloses a shape-optimized rectangular panoramic image construction method based on feature selection, and the method comprises the steps of carrying out the splicing of two overlappedimages which are shot in a non-fixed mode based on a common camera, and obtaining a rectangular panoramic image which is not distorted in shape and is not lost in information; taking the shot image asan input image, performing feature point matching on the input image, and screening mismatched feature points and isolated feature points by using a clustering algorithm; performing image registration operation on the input image overlapping area by utilizing the screened feature points; gradually changing distortion transformation is carried out in the non-overlapping area so as to reduce perspective distortion generated by the image; and establishing a grid optimization model based on boundary constraint and central region reservation, obtaining an optimal rectangular grid vertex, and finally obtaining a final rectangular panoramic image through an interpolation method. According to the invention, the rectangular panoramic image is generated under the condition that the information lossis reduced as much as possible through non-fixed shooting of the image with the overlapped area.

The invention relates to the technical field of geographic information, in particular to a gridding coding method of urban three-dimensional live-action data. The method comprises the following steps:S1, acquiring an overall area of a three-dimensional real scene of a city, and gridding the overall area by taking one point of the overall area as an original point and using a set segmentation gridunit; S2, taking a point of a set position in each gridding after gridding as a position point, and then setting a grid name according to the coordinate of the position point of each gridding; and S3, setting a storage path of the city three-dimensional live-action data, enabling the subdirectory to comprise folders named by the grid names obtained in the step S2, and then storing the city three-dimensional live-action data into the corresponding grid name folders according to the grid position information obtained by the city three-dimensional live-action data. According to the method, the stored urban three-dimensional live-action data is enabled to carry geographic space recognition information.

The invention discloses a full-automatic efficient packaging box production line. The full-automatic efficient packaging box production line comprises a paper feeding device, a gluing device and a bonding device, wherein the paper feeding device, the gluing device and the bonding device are sequentially connected through an annular conveyor belt. The gluing device comprises a vertically-liftable glue gun with a rectangular glue spray opening. The glue gun is connected with a glue pool. The top of the glue gun is connected with a vertical lifting device. By the adoption of the full-automatic efficient packaging box production line, the application range is wide, the requirements of paper with different sizes and thicknesses can be met, the paper jamming phenomenon is avoided effectively, and splicing of a packaging box can be completed efficiently.

A throat plate switching mechanism includes a first throat plate and a second throat plate. The first throat plate is formed with a first needle hole having an elongated hole shape at a stitch point. The second throat plate is formed with a second needle hole. A width of the second needle hole is narrower than a width of the first needle hole. The second throat plate is equipped in the first throat plate such that the second throat plate is switched into a use position at which the second needle hole overlaps the first needle hole, and a retreated position at which the second throat plate does not cover the first needle hole.

The invention discloses a positioning method of a scanning sampling starting point. The method comprises arranging a light source and a scanning mirror opposite to the light source, perpendicularly emitting lights emitted by the light source to the scanning mirror, rotating the scanning mirror so that an A-C light scanning zone from outside to inside is formed below the scanning mirror, arranginga single point detector at the inner side of the light scanning zone, detecting and acquiring light C of the scanning mirror through the single point detector, converting a light signal into an electric signal and outputting an image information collection triggering signal through an output end of the single point detector. The positioning method ensures effective joining of images acquired by each frame, improves the photoelectric recognition precision, realizes more accurate scanning imaging starting point positioning, prevents the occurrence of the starting point drift in the existing practical projects and further improves the accuracy of image acquisition.

The invention discloses a CUDA-based PET system matrix calculation method. The method comprises the following steps: firstly, dividing a detector system according to a fixed angle to obtain response line equations of subsets after division; then, allocating a grid to each angle in CUDA, allocating a block to each sub-response line at each angle, transmitting pixel coordinate data of the reconstructed image to a shared memory of each block, and calculating a probability weight of each pixel rapidly and parallelly by threads in the blocks, wherein the grids are sequentially and serially executed, and different blocks in the same grid are quickly and parallelly executed; and finally, splicing the probability weights within angles to obtain a system matrix subset, and splicing the system matrix subset to obtain a system matrix. According to the method, on the premise of ensuring the accuracy of system matrix elements, the creation efficiency is improved, and the problems caused by low system matrix creation speed, limited GPU video memory resources and large system matrix data volume are solved.

The invention discloses a hashtable-based scattered line segment quick connecting method. The method comprises the steps of (1) inputting a scattered line segment sequence; (2) establishing linked point pairs and saving the linked point pairs into a hashtable map; (3) newly establishing a current contour and finding out one unused linked point in the map as the current contour starting point p; (4) adding the point p into the current contour, querying a next point d of the current contour according to the point p to enable p to be equal to d; (5) repeating the step (4) until the current contour is closed, and adding the contour into the contour sequence; (6) repeating the steps (3)-(5) until all the linked points in the map are used; and (7) outputting the connected contour sequence. According to the hashtable-based scattered line segment quick connecting method, by establishing the two special data structures of line segment vortexes and a hashtable, the topological relation between the vortexes of the stored line segments is made complete. Quick connection of scattered line segments is realized through hashtable querying. The method is stable, reliable and high in robustness.

The invention discloses a multifunctional mobile kitchen, which comprises a cabinet body, wherein universal wheels are arranged at the bottom end of the cabinet body; the cabinet body is successively,from top down, provided with a meal platform, a cooking platform, an oven and a storage cabinet; the meal platform comprises a chopping board, and an insertion gap is arranged between the rear edge of the top end panel and the rear side plate; the top end panel of the cooking platform is an integral microlite panel. The left side of the microlite panel is a high temperature operation area and theright side is a low temperature operation area; the oven is arranged in the cabinet body and below the cooking platform; the storage cabinet comprises a storage drawer, wherein a deep ultraviolet LEDlamp is arranged in the cabinet body, the deep ultraviolet LED lamp illuminates towards the drawer that is in a push-in state; the mobile kitchen fully considers the factors of the cooking process and the moving, so that the household kitchen is no longer limited to a corner, and the mobile kitchen has the functions of moving, preparing food, cooking, storing, platform expanding and the like, which make cooking become funny. The small space of the mobile kitchen is fully used to carry out kitchen work.

The embodiment of the invention discloses a multifunctional mechanized assembled teaching blackboard device. The multifunctional mechanized assembled teaching blackboard device comprises a supportingrack embedded into a wall body and a combined blackboard arranged on the supporting rack, wherein the combined blackboard comprises a side pull combination block positioned at the upper half part anda functional combination block positioned at the lower half part; the side pull combination block at least comprises two side pull plates; the functional combination block at least comprises multiplegroups of rotated plates capable of turning over towards the outer side; the supporting rack comprises a first side frame for movably mounting the side pull combination block, a second side frame formovably mounting the functional combination block as well as a rail rack for storing the side pull plates pulled from the first side frame to the outer side for moving; self-ejection type side pull plate supplying cavities which are in one-to-one correspondence to the side pull plates are formed in the first side frame; and self-ejection type rotated plate supplying cavities which are in one-to-one correspondence to various groups of rotated plates are formed in the second side frame. The multifunctional mechanized assembled teaching blackboard device realizes the effects of enlarging the teaching and writing space, increasing the answering space and improving the teaching quality and the teaching efficiency.