424results about How to "Realize splicing" patented technology

The invention provides a method of reconstructing a dentition through a dental CT (Computed Tomography) image and three-dimensional scanning data. The method includes the steps of: preprocessing the CT image, and removing soft-tissue pixels; using an image segmentation method to obtain all individual tooth contours; using a surface drawing method to obtain a dentition model of the CT image; and establishing a segmentation surface to extract all dental crown models of the CT image; obtaining a three-dimensional scanning model of a dental jaw, and obtaining independent dental crown models afterprocessing; registering the dental crown models of the CT image and the dental crown models of the scanning data; and carrying out fusion and reconstruction on dental root models of the CT image and the dental crown models of the scanning data to effectively construct a three-dimensional model of the dentition with dental crown information with high precision. The method solves the existing technical problem that voxel resolution of patient complete-dental-jaw three-dimensional-models extracted through CT images is low, and can be used for formulation assisting of invisible-orthodontics schemes of physicians, assisted design of orthoses and further biomechanical analysis.

The invention discloses a semantics-based article information tracking and tracing method for the Internet of things. Tracking of article information from upstream to downstream and tracing of the article information from downstream to upstream of a certain specific article unit in any link of the article distribution are realized by respectively recording a base attribute and a status attribute of an article and a uniquely coded mapping relation of the article among an upstream and downstream relation of the article distribution, access addresses of the article information of upstream and downstream nodes and the upstream and downstream nodes in a corresponding semantic label on each node which the article passes through. The method solves the tracking chain breaking problem caused by hidden label data of the article due to the operation of assembly, package, partition and the like in the article distribution process. By using the method, the upstream node can know the destination of the product at any time and implement accurate product recall in case of a quality accident; and the downstream node can rapidly judge the origin of the article and rapidly trace the responsibility cause in case of the quality accident.

The present invention relates to a three-dimensional modeling method and device for a geological prospecting ore body, wherein the method includes: Step 1, delimiting a single project ore body according to the exploration project, and obtaining the upper and lower boundaries of the ore body; Step 2, projecting the prospecting project onto the prospecting project On the profile, connect the upper and lower boundaries of the ore body on the exploration profile to obtain the geological body boundary of the ore body in the exploration profile; step 3, connect the geological body boundaries between the exploration profiles with a three-dimensional curved surface to obtain the three-dimensional reconstructed geological body. The invention can realize geological three-dimensional modeling from drilling to profile to curved surface to finally generate solid.

The invention discloses an OLED splicing display screen which is spliced by plural soft OLED display screens or hard OLED display screens. The splicing to the soft OLED display screen is that the edge non-image display zone of each OLED display screen is bended to the back face of the OLED display screen according to a certain angle. The splicing to the hard OLED display screen is that a wire is lead out from the hard screen lateral face. The LED display screens are tightly adhered together according to the matrix arraying mode of M rows multiplying by N lines. The image on each OLED display screen and the image displayed on the other OLED display screen are entirely linked by splicing plural OLED display screens and fixing on a supporter. The invention has low cost and can be produced in batch.

The invention provides a map building method and system based on data fusion of two laser radars as well as a map. The method comprises a step of building a point cloud map and also comprises the following steps: performing calibration and data fusion on the two laser radars; performing interframe point cloud coarse registration by adopting a rapid space model point cloud registration algorithm; and performing pose correction by adopting the nearest neighbour iteration, so that optimal interframe transformation is obtained and accurate interframe point cloud splicing is realized. The method provided by the invention firstly performs space and time synchronization on the double laser radars and performs the data fusion on the laser radars; then the interframe point cloud coarse registrationis performed by adopting the rapid space model point cloud registration algorithm, and point cloud registration correction is performed by utilizing the nearest neighbour iteration point cloud registration, so that high-precision interframe point cloud splicing is realized; and finally building of the point cloud map is completed.

The invention discloses a curved-surface bionic compound eye imaging device for a zoom lens array, and belongs to the field of lens bionic compound eye imaging devices according to a curved-surface array. The curved surface of a glass spherical shell of the device is provided with a lens array which consists of single eyes at a plurality of layers, wherein the curvatures and focal lengths of the single-eye lenses at all layers change layer by layer, but all the single-eye lenses are focused on the surface of a unique plane image sensor, thereby enabling the single-eye lens located at the edge of a view field to be able to be effectively focused, and avoiding a defocused state. According to the invention, a sieve-mesh-shaped metal hole array covers the surface of an image sensor, and each sieve mesh is enabled to be corresponding to one independent monocular lens optical path, thereby achieving an effect of segmenting the view field, achieving the integrated jointing of all monocular lens images, and preventing the interference and impact among the adjacent monocular lenses. The device is also high in resolution, is high in instantaneity, is easy to miniaturize, is easy to become light, and is convenient for manufacturing and maintenance.

The invention provides a method and a device for removing glasses from a human face image, and a method and a device for wearing the glasses in the human face image. The removal method comprises the following steps of: acquiring a glasses-worn eye area image and a non-eye area image from a glasses-worn human face image; matching the glasses-worn eye area image by using a classifier to acquire a corresponding eye area type, wherein the classifier is used for multiple types of eye areas which are constructed according to glasses-free human face image samples, and the multiple types of eye areas are classified according to eye shapes; acquiring a base vector corresponding to the eye area type, wherein the base vector is acquired by performing principal component analysis (PCA) study aiming at the sample of the eye area type; mapping the glasses-worn eye area image to the base vector, and reconstructing to acquire a glasses-free eye area image; and splicing the non-eye area image with the glasses-free eye area image to acquire a glasses-removed human face image. By the method, the glasses can be relatively effectively removed from the human face image.

The invention provides a free-moving type three-dimensional scanning method, belongs to the field of optical detection, and relates to a full three-dimensional skeleton surveying method, a handheld measuring device and a multi-angle splicing method based on texture, the curvature of surface and normal directions. Mobile handheld high-precision three-dimensional scanning is a problem of the scanning field, according to the free-moving type three-dimensional scanning method, a three-frequency color fringe projection method is utilized to project three-frequency color fringes of 1 : 4 :12 to the surface of a tested object through a projection device, and a camera is used for shooting a deformed fringe pattern in a single-frame exposure mode. On one hand, a three-frequency phase unwrapping method is adopted by the deformed fringe pattern to calculate out high-frequency carrier frequency to unwrap a phase, and system calibration parameters are used for obtaining a three-dimensional point cloud of the tested object; on the other hand, a texture pattern of the surface of the tested object can be recovered through the deformed fringe pattern. Handheld scanning has a higher requirement for multi-angle splicing, and by means of the free-moving type three-dimensional scanning method, non-mark splicing of multi-angle point clouds is achieved according to the color texture characteristics, measurement of the curvature of surface and restraint of the normal directions.

The invention discloses a method and device for scanning multiple identity cards. The method comprises the following steps of: scanning the both surfaces of the multiple identity cards in a scanning area; performing edge extraction on scan images; performing the procedures of positioning, cutting and correcting by rotation on the scan images subjected to the edge extraction, thus obtaining a scan image of each identity card; performing characteristic extraction on each scan image of the identity card to obtain a characteristic pattern of the scan image of each identity card; comparing the characteristic pattern of the scan image of each identity card with a characteristic database storing standard characteristic patterns of the identity cards to perform orientation judgment and orientation correction on the scan image of each identity card, thus obtaining a corrected scan image of each identity card; and splicing a front image and a back image of each corrected scan image of each identity card, and then outputting a spliced image. The method and the device ensure that customers can conveniently perform double-face scanning operation on the multiple identity cards, are high in scanning speed and efficiency, provide good user experience, and can be widely applied to the field of scanning.

The invention relates to a crack splicing method in bridge quality detection. The machine vision technology, modern sensing technology, data analysis processing technology, calculation technology canbe effectively fused into bridge quality detection. Aiming at the acquired local image of the crack; the splicing of crack images is accurately completed. The automatic detection of the whole crack ofthe bridge is met; the safe use of the bridge is ensured. Method of installing an installation device, Collecting crack images, Crack image stitching, carrying out graying on the spliced image; grayscale transformation and image filtering preprocessing are carried out. The method is scientific. The bridge crack automatic splicing device is simple in structure, easy to operate, good in detectioneffect and capable of rapidly and accurately achieving automatic splicing of the same crack at different visual angles, the problem that long cracks cannot be contained in crack images is solved, themaximum crack width and the automatic detection level of the real position of the maximum crack width in bridge quality detection are further improved, and smoothness and safety of traffic transportation are effectively guaranteed.

The invention discloses a method and a device for splicing strips of an airborne laser scanning system. The method comprises: using a three-dimensional normal distribution transformation algorithm as a splicing algorithm in strip splicing, dividing voxel units in point cloud data of the airborne laser scanning system, establishing normal distribution in the voxel units, mean values and variance of the normal distribution representing characteristics of the voxel units; and then establishing a similarity function, and obtaining offset of strips when value of the similarity function is maximum; and adjusting adjacent strips according to the offset, to finally splice a plurality of strips together. In addition, through contrast experiment analysis, methods and parameters on three aspects of sample rate before strip splicing, data organization method in a strip splicing process, and discretization are selected to be more proper. The method has higher robustness, shorter splicing time, and better splicing precision, and memory utilization rate of a computer is lower.

The invention relates to an accurate positioning method for a circular marker in large-sized curved plate measurement in a shipbuilding manufacture process. Related equipment of the accurate positioning method for the circular marker in large-sized curved plate measurement comprises a global measuring device and a local measuring device; the global measuring device is formed by two industrial cameras and a high-performance server which can store and analyze; the local measuring device is formed by two industrial cameras and a projector. A specific marker is installed on the local measuring device to accurately measure the spatial position of the local measuring device. A vision measurement based circular marker positioning technology comprises the accurate positioning method for the circular marker installed on a measuring device. The method can automatically, timely, conveniently and accurately position the circular marker on the measuring device without being in contact with a steel plate surface.

The invention discloses a method and a system for correcting and jointing tilted scanning and imaging data of a linear array tower. The method comprises the following steps of: arranging a linear array scanner on the tower for performing multi-angle tilted scanning and imaging, thus obtaining a plurality of single-scene images; performing geometric distortion correction without controlling points for the plurality of single-scene images according to a geometrical relationship between an imaging area and the scanner during tilted imaging; extracting homonymy controlling points in the corrected single-scene images; performing image registration according to the extracted homonymy controlling points; and jointing the registered images under a custom coordinate system. By virtue of the method and the system for correcting and jointing tilted scanning and imaging data of the linear array tower, the problems of geometric distortion correction carrying multispectral imaging data and jointing of multi-field images can be solved; the implementation method is quick and practical.

The invention discloses a narrow-view-field double-camera image fusion method based on a large virtual camera. The narrow-view-field double-camera image fusion method comprises the steps of constructing the large virtual camera according to two single cameras, and constructing geometric imaging parameters of the large virtual camera according to geometric imaging parameters of the two single cameras; constructing corresponding geometric imaging models according to the geometric imaging parameters of the two cameras and the large virtual cameras; calculating and outputting a rational polynomial model coefficient corresponding to the large virtual camera; respectively performing indirect-method geometric correction on images of the two single cameras according to a coordinate forward calculation process and a coordinate back calculation process based on the geometric imaging models to obtain two images under an image coordinate system of the large virtual camera, and obtaining a fused image of the large virtual camera. According to the narrow-view-field double-camera image fusion method based on the large virtual camera, the concept of the large virtual camera is ingeniously used; high-precision fusion of the images of the two cameras in a narrow view field is realized, and the rational polynomial model coefficient corresponding to the large virtual camera is supplied; furthermore, the processing procedure is fully automatic, and manual intervention is not needed; the narrow-view-field double-camera image fusion method is applicable to a ground preprocessing procedure.

The invention discloses an image body identification, correction and registration method comprising the following steps: target detection: performing threshold segmentation by observing the features of an image, and roughly distinguishing between the foreground and the background; binarization: getting a black-and-white binary image, removing noise points through expansion, and retaining a compartment part in the original image; image feature extraction: positioning the four corners of the compartment based on the mathematical properties of trapezoid; perspective transformation: stretching an irregular convex quadrilateral to a regular rectangle through a transformation matrix; and registration and splicing: combining multiple box body images, and for the images with common image features, registering and splicing multiple box body images on the side by use of an Opencv open-source library Stitcher class or augmenting the images based on the properties of the matrix. The image body identification, correction and registration method is beneficial to eliminating background interference, reducing the sample set of box body identification and improving the efficiency of identification.

The application relates to an image scanning outputting method and system, a computer device and a storage medium. The method includes: a plurality of to-be-spliced images are obtained by a pluralityof image sensors, wherein the two adjacent to-be-spliced images are overlapped partially at least; the positions of other to-be-spliced images are adjusted by using one to-be-spliced image as a reference according to a preset coordinate distance; and the overlapped parts of the other to-be-spliced images are cut based on the preset coordinate distance to obtain a spliced image. According to the invention, the multiple to-be-spliced images are obtained by the plurality of image sensors; the positions of the to-be-spliced images are adjusted, the overlapped parts of the to-be-spliced images arecut off, and then the plurality of to-be-spliced images are spliced to realize splicing. Therefore, the collection gap between the plurality of to-be-spliced images is eliminated; image missing as well as repeated image splicing is avoided, so that the spliced image is complete.

The invention relates to a method for splicing maskless digital projection lithography pattern. The method comprises the steps of S1 segmenting a pattern to be etched to obtain multiple segmented frames of sub-pattern, wherein each frame of sub-pattern is of same size; S2 setting a template size which is the same with the size of each frame of sub-pattern; S3 multiplying each segmented frame of sub-pattern by the template to obtain a boundary splicing region of the sub-pattern, and employing a gray modulation template to modulate the gray value of the boundary splicing region to obtain the modulated sub-pattern; S4 carrying out exposure on the modulated sub-pattern frame-by-frame with digital micromirror display to realize the splicing of the reproduction transfer of the modulated sub-pattern which is displayed by the digital micromirror to a substrate. The method provided by the invention can fairly solve the splicing problem which exists when a pattern with a large area is etched, thus the etching quality of the digital projection lithography pattern is improved.

An aspheric surface adjustment error compensation method for sub-aperture stitching detection relates to the field of optical detection, and includes the steps: setting an interferometer to lead the curvature radius of a reference sphere wave front of the interferometer to be matched with the closest spherical radius of a central area of a to-be-detected aspheric surface; adjusting the relative position relation of the interferometer with the to-be-detected aspheric surface, so that the standard spherical wave front of the interferometer is aligned with a to-be-detected aspheric surface area; by means of a pattern search error compensation method, eliminating adjustment errors generated when the standard spherical surface detects the to-be-detected aspheric surface with large deviation amount; and realizing full-aperture stitching detection to obtain a precise surface-shaped result by means of the sub-aperture stitching detection technology. By means of the pattern search error compensation method, the adjustment errors resulting from nonalignment of stitching measurement positions can be effectively eliminated from measured sub-aperture phase data, so that stitching of multiple sub-apertures can be effectively realized, and aspheric surface full-aperture surface-shaped stitching detection can be precisely completed.

The invention discloses a method and system for split screen display of display information in an equipment room. The method includes the following steps that a main display device transfers original display information to multiple vice display devices which join together into a complete screen; the central processing unit extracts corresponding display sprites from the original display information for each vice display device based on the number of the vice display devices and the positions on the complete screen; the central processing unit resizes multiple display sprites to the corresponding display size of each vice display device based on display sizes of display sprites; and the multiple processed display sprites are played on the corresponding vice display devices. The original display information can be segmented into multiple display sprites through a segmentation module, and eventually displayed on multiple vice display devices which are joined together and form a complete picture. The image mosaic is achieved, display of a larger screen is achieved, and a variety of display effects can be achieved.

The invention relates to a two-dimensional panoramic image acquisition method and device, equipment and a storage medium, and the method comprises the steps: collecting to-be-spliced images accordingto a preset row and column collection mode, and enabling the proportion of the overlapped area of every two adjacent to-be-spliced images to be greater than a first preset proportion threshold value;for every two adjacent to-be-spliced images, selecting an overlapped area with a second preset proportion threshold value as an interest area, extracting feature points in the interest area, and determining the main direction of the feature points; describing the feature points to obtain a description operator, and performing feature point matching according to the Hamming distance by applying thedescription operator to obtain a feature point pair so as to obtain a target to-be-spliced image after feature point matching; aiming at the target to-be-spliced image of each row, carrying out registration and fusion to obtain a row target image splicing result image; and aiming at each line of target image splicing result image, carrying out registration and fusion to obtain a two-dimensional panoramic image. The precision of the panoramic image in the industrial application is improved, and the flexibility of obtaining the panoramic image is improved.

The invention discloses a multi-robot SLAM (Simultaneous Localization And Mapping) algorithm based on sub-map feature matching. The multi-robot SLAM algorithm comprises the steps of: the step 1, acquiring environment information by adopting a plurality of robots, and creating a sub-map sequence according to the environment information; the step 2, matching sub-maps created by a single robot, and constructing the front end of a single robot SLAM algorithm; the step 3, matching the sub-map sequences created by different robots based on a grid map fusion algorithm of the maximum public subgraph,and storing all the matching results; the step 4, calculating the relative poses among the robots according to the results in the step 3; the step 5, constructing multi-robot closed-loop constraints according to the calculation results in the step 3 and the step 4; the step 6, constructing the front end of the multi-robot SLAM algorithm according to the results in the step 2 and the step 5; the step 7, performing optimization of the rear end of the multi-robot SLAM; and the step 8, creating a global grid map. The multi-robot SLAM algorithm based on sub-map feature matching can solve the problem of modeling a large-scale environment or an environment with high requirement on working efficiency.

The present invention discloses a train wheel tread extraction and splicing method based on image segmentation and image fusion. The method comprises the steps of: dividing a wheel tread primary areaaccording to the feature of wheel image collection device fixed point imaging, employing a gray-scale projection technology, further locating a tread area according to the size of the gray projectionvalue and the variation rate, on this basis, employing an edge detection method to detect a wheel flange wire, completing wheel tread extraction, and finally, calculating the minimum smooth width according to the gray difference of a wheel tread overlapping area, splicing the wheel tread through a pixel matching method based on a space, and generating a 360-degree wheel tread panoramic image. Thewheel tread extraction and splicing method based on image processing belongs to the pre-processing technology of a locomotive damaging detection system, can effectively and rapidly extract the locomotive tread area, can improve the preprocessing speed of the locomotive damaging detection system, and has high practicability and wide application prospect.

The invention relates to the field of image processing, in particular to a real-time splicing method for unmanned aerial vehicle aerial stereo images. According to the splicing method for the unmanned aerial vehicle aerial stereo images, an inter-frame interval of key aerial stereo images is adjusted and extracted in real time according to the overlapping ratio of adjacent to-be-spliced aerial stereo images; a transformation matrix of every two adjacent frames is obtained according to a matching relation between homonymy points of every two adjacent frames, and the serial aerial stereo images are transformed to a coordinate system of reference images; and reference frames are adjusted in real time according to set reference frame adjustment judgment conditions, so that accumulative errors generated in the process that the aerial stereo images are transformed to a coordinate system of the reference frames are dispersed to all frames of images, and the situation that serious distortion of subsequent images is caused by error accumulation is avoided.

The invention relates to a virtual sound field real-time interaction system based on a binaural effect, and the system comprises an audio signal collection module which is used for collecting multiple paths of audio signals; a sound field rendering module which is used for performing sound field rendering on the audio signals and acquiring multiple paths of audio signals with reconstructed sound field information based on a binaural effect; a virtual sound field interaction module which is used for splicing multiple paths of audio signals with reconstructed sound field information in real time according to the head parameters and the user orientation information, and constructing a real-time sound effect in the whole sound field; and the output module is used for outputting a real-time sound effect according to the individual adjustment parameters. Compared with the prior art, audio is processed from the sound and the space sense of the sound field, the original sound field is highly restored and reproduced, and the method has the advantages of flexibility, interactivity, user friendliness and the like.

The invention provides a hand-held three-dimensional real-time scanning method. The method comprises steps that (1), a measurement angle of a measured object is selected, and an adjacent angle superposition rate is made to be greater than 60%; (2), a measurement mode is selected, and inching measurement or continuous measurement is carried out; (3), the exposure time of a camera is adjusted according to the information of different-angle illumination and reflectivity; (4), scanning measurement is carried out by the camera according to the selected measurement mode and the selected measurement angle, namely shooting; (5), the three frequency colorful phase shift profilometry and a colorful stripe decoupling method are employed to acquire three-dimensional point cloud and a texture graph of each angle; and (6), the acquired different-angle three-dimensional point cloud acquired in the step (5) are spliced to be an integral model. The method is advantaged in that the method is applicable to most measurement objects, especially dynamic objects, the method needs no artificial mark points, and a multi-angle measurement splicing problem can be excellently solved.

The invention relates to a method for realizing the regulating and the splicing of an extra-long specification X-ray image in the medical radiology image field. The regulating and splicing method has the steps as follows: in the first step, two tailored version steel rules are put on two sides of the radiographed position of a patient in a parallel way, holes are drilled on the tailored version steel rules, scales are marked on drilling points in sequence, and the drilling points are marking points for the regulating and the splicing of the image; in the second step, subsection images are acquired and transmitted to a computer, and the examined position and the marking points on the steel rules are displayed in the images; in the third step, the subsection images are regulated to obtain regulated sequence images to be spliced; in the fourth step, an image splicing coordinate system is set up, and two adjacent subsection images are spliced respectively according to the order of the image sequence; in the fifth step, two steel rules are selected and put to the three marking points in an overlapping area of each image, and are matched with the triangle between the two images formed by the three marking points; taking one image as a reference image, the splicing between two adjacent subsection images is completed through coordinate transformation. The method is simple, convenient and practical, and has high precision and low cost.

The invention discloses a catadioptric long-focal-length large-view-field small-size imaging optical system. The catadioptric long-focal-length large-view-field small-size imaging optical system comprises a front group positive-focal-power reflection type optical group and a rear group negative-focal-power transmission type optical group. The front group positive-focal-power optical system comprises a concave primary mirror reflector and a convex secondary mirror reflector; and the rear group negative-focal-power transmission type optical group is composed of a positive lens group, a negativelens group and a small focal power meniscus lens. An object imaging light beam is reflected by the concave primary mirror reflector and the convex secondary mirror reflector in sequence, then enters atransmission-type system, is refracted by the positive lens group, the negative lens group and the small focal power meniscus lens in sequence, and then is imaged on an image plane. In the invention,requirements of a high resolution, a large breadth range and a small size and weight of a space camera are met.

The invention provides an image splicing method and device, electronic equipment and a computer readable storage medium, and relates to the field of aerial photography image splicing. The image splicing method comprises the following steps: acquiring a first image and a second image; wherein the first image is an image shot by a camera of the unmanned aerial vehicle at a first position, and the second image is an image shot by the camera at a second position; acquiring attitude information of the unmanned aerial vehicle according to the first position and the second position; wherein the attitude information represents angle information of the unmanned aerial vehicle moving from the first position to the second position; and splicing the first image and the second image according to the attitude information to obtain a spliced image. According to the method, the to-be-spliced images with the position information are acquired, the attitude information of the unmanned aerial vehicle is acquired, splicing of the multiple images is further realized, and splicing of the images acquired by the unmanned aerial vehicle is realized without depending on information of an inertial measurementodometer.

The invention belongs to the technical field of marine coral biological research, and particularly relates to a coral distribution and health condition evaluation method based on deep clustering analysis. Acquired video images are spliced into a large image through an improved image splicing algorithm, target recognition detection is performed on warm water coral through a deep learning algorithm,then image segmentation is performed on a recognition target through a k-means segmentation algorithm, and finally coral distribution and health conditions are researched. According to the method provided by the invention, the image splicing efficiency is improved; smooth seamless splicing of the coral images is realized; the coral in the image is rapidly detected by using the YOLOV3, the coral coverage rate is improved by calculating the coral area through image segmentation, so that the overall calculation rate of coral distribution and health conditions is greatly improved, and a good effect is achieved for researching the coral reef distribution condition and health conditions.

The invention relates to a television wall splicing system and method based on a Display Port. The system comprises a signal source input module and N splicing control units. The signal source input module is an HDMI signal conversion module and used for converting input DVI / HDMI signals into HDMI signals. Each splicing control unit comprises a DP signal conversion module, an image selecting drive module and a display module, wherein the DP signal conversion module, the image selecting drive module and the display module are sequentially connected. The image selecting drive module of the mth splicing control unit is further connected to the DP signal conversion module of the (m+1)th splicing control unit, and the HDMI signal conversion module is connected to the DP signal conversion module and the image selecting drive module of the first splicing control unit. According to the television wall splicing system and method, the circuit structure is simple, cost is low, popularization is easy, splicing of large liquid crystal television walls is achieved, and good market application prospects are achieved.