122results about How to "Improve registration speed" patented technology

The invention discloses an indoor scene 3D reconstruction method based on Kinect and solves the technical problem of the real-time reconstruction of an indoor scene 3D model and avoidance of excessive redundant points. The method comprises steps of: obtaining the depth data of an object by using Kinect and de-nosing and down-sampling the depth data; obtaining the point cloud data of a current frame and calculating the vector normal of each point in the frame; using a TSDF algorithm to establish a global data cube, and using a ray casting algorithm to calculate predicted point cloud data; calculating a point cloud registration matrix by using an ICP algorithm and the predicted point cloud data, fusing the obtained point cloud data of each frame into the global data cube, and fusing the point cloud data frame by frame until a good fusion effect is obtained; rendering the point cloud data with an isosurface extraction algorithm and constructing the 3D model of the object. The method improves the registration speed and the registration precision, is fast in fusion speed and few in redundancy points, and can be used for real-time reconstruction of the indoor scene.

The invention discloses an automatic depth-image registration method, which comprises the following steps: proceeding triangularization method for depth image; transmitting point cloud data into triangle grid data; finding the boundary point in the depth image according to the triangle grid data; calculating the character value of non-boundary vertex in the depth image; extracting character point set with projected surface character; finding the mating relationship as corresponding point among character points of two depth images; finding at least three couples of corresponding vertex; calculating the estimating value of movement of two depth images according to the corresponding point; adopting modified ICP algorism to optimize the result; finishing the precise registration of two depth images. The invention improves the speed and precision of registration, which reduces the calculating quantity with simple course.

The invention discloses a liver multi-phase CT image fusion method which comprises the steps of firstly performing coarse registering on a source image sequence according to a multi-resolution CT image registering method based on a combined histogram, then realizing automatic liver image dividing according to a region growing algorithm in combination with confidence connection and liver image dividing based on a gradient vector flow snake model, and effectively extracting the edge information of the liver; performing blood vessel extraction based on a directional region growing algorithm on the liver image, performing free deformation transformation based on a B spline and liver non-rigid registering based on space weighting mutual information on a liver essential image, thereby accurately finding an image pair at a same position of a space; and finally performing image fusion based on wavelet transformation. The liver multi-phase CT image fusion method aims at the characteristics of a liver CT image. An image separation process and an image registering process are combined with the image fusion process, thereby greatly improving fusion precision.

The invention relates to a multi-resolution medical image registration method based on the quantum-behaved particle swarm optimization, which is characterized in that the method comprises the following step of: firstly removing the backgrounds of the two images to be registered to keep the images off the interference of noise; next obtaining the images with low resolution from the two background-removed images via the wavelet transform, using the low resolution to be the object, taking the normalized mutual information as the objective function, using the quantum-behaved particle swarm optimization, then the images with the high resolution to be the object, and obtaining the rotation amount and the translation amount between the two images to be registered to finish the image registration by using the Powell method. With solving a plurality of local extrema based on the objective function of the mutual information, the invention greatly improves the registration precision and speed, reaching up to the sub-pixel level; and is widely applicable to the fields of the image discrimination of the clinical diagnosis, the framing of the radiation treatment, the image guided surgical, etc.

The invention discloses a product packaging surface defect detection and classification method based on machine vision. The method comprises the steps of 1, acquiring a high-definition color image of defect-free product packaging, making the high-definition color image into a standard image, conducting real-time shooting with a camera, conducting online acquisition of a high-definition color image of product packaging to be detected, and making the high-definition color image as an image to be detected; 2, conducting image matching on the image to be detected and the standard image based on SURF algorithm; 3, conducting difference image operation on the two images matched in step 2 to obtain a defective image; 4, conducting feature extraction on the defective image to obtain the geometrical features and color features of the defective image; 5, classifying product packaging surface defects by means of RBF neural network algorithm. Automatic defect detection and classification are conducted by means of a machine vision system, human factor interference can be avoided, labor cost is reduced greatly, and then huge hidden cost caused by training and management when artificial detection is adopted is avoided.

A depth image automatic registration process combining with texture information is mainly used to reconstruct the three-dimensional model of various actual objects, which comprises the steps: firstly extracting a texture image from scanning data or generating a texture image according a depth image, secondly extracting interest pixels in the texture image based on the characteristics of SIFT, and finding out candidate sets therein matched with the interest pixels through the crosscheck method, thirdly finding out correct matched pixel pairs in the candidate sets according to geometry information constraint, fourthly finding out matched apex pairs corresponding to the matched pixel pairs in a three-dimensional space, and calculating the rigid replacement matrix between two depth images, fifthly optimizing the result through using the improved ICP algorism, sixthly dividing the input sequences of several depth images into a plurality of bar type subsequences based on the registrations of two depth images, finally combining the subsequences through the strategy of forward searching, and constructing an integral three-dimensional model. The invention can fast register large-scale three-dimensional scanning data to generate a three-dimensional model, and has strong anti-noise capability and excellent universality.

The invention relates to a hand-held three-dimensional surface information extraction method and an extractor thereof. The method comprises the following steps of: 1) randomly arranging a plurality of circular markers on a scanned object, symmetrically arranging two cameras on the left and right sides of a linear laser projector and arranging an image acquisition and processing card having two parallel paths of working threads; 2) acquiring left-eye and right-eye two-dimensional surface information captured by the cameras on the left and right sides by using one path of working thread in the image acquisition and processing card; 3) extracting the markers in the left-eye and right-eye two-dimensional surface information and performing stereoscopic vision matching to reconstruct three-dimensional coordinates and a topological structure of the markers in the left-eye and right-eye two-dimensional surface information by an elliptic object image processing method by using the other path of working thread in the image acquisition and processing card, computing a registration matrix of scanning data in the whole scanning data in a current posture by using the reconstructed topological structure of the three-dimensional markers, extracting laser streak points from the left-eye and right-eye two-dimensional surface information, reconstructing the three-dimensional coordinates of the laser streak points in the left-eye and right-eye two-dimensional surface information and performing left and right eye fusion on the three-dimensional coordinates of the two laser streak points and then registering the three-dimensional coordinates of the laser streak points subjected to the left and right eye fusion into the whole scanning data of the object to be scanned by using the registration matrix in the current posture; and 4) repeating the steps 2) to 3) till all the scanning data of the object to be scanned is acquired.

The invention belongs to the technical field of computer vision, and particularly relates to a point cloud automatic registration method based on a local curved surface feature histogram. The method comprises the following steps: carrying out cyclic voxel filtering on a source point cloud, downsampling the source point cloud to a designated point number, and carrying out voxel filtering on a target point cloud according to the size of an obtained voxel; carrying out key point searching and feature description on the basis of points with the maximum neighborhood curvature mean value of pre-keypoints, detecting the points with the curvature larger than 0.02 and serve as the pre-key points, calculating the curvature mean value of the neighborhood points of the points, and classifying the points with the maximum local curved surface curvature mean value as the key points; calculating a feature descriptor of the local curved surface histogram according to the relationship between the center of gravity of the point cloud in the neighborhood of the key point and the normal and distance of each point in the neighborhood; and calculating a mutual corresponding relation between the source point cloud feature descriptor and the target point cloud feature descriptor, removing an error corresponding relation according to random sampling consistency, and estimating an optimal coordinate transformation matrix according to SVD. Application prospect is wide.

The invention discloses an image registration method and apparatus. The image registration method includes the following steps: 1) performing extraction of image characteristics; 2) introducing a gray moment into an LDB (Local Difference Binary) descriptor, constructing the main direction of the characteristic points, rotating the neighborhood image blocks of the characteristic points to the main direction when describing the characteristic points, and then constructing a descriptor; 3) performing characteristic matching on images; and 4) looking for the optimum matching point. The image registration method and apparatus can further improve the speed and the accuracy during the image registration process, and can utilize the Hamming distance to measure the similarity between the image characteristics, thus reducing the computational complexity and improving the registering speed.

The method comprises: according to user information, generating and saving the fast registration information; when user needs to make registration, a fast registration is provided for user according to the fast registration information. The invention also provides an apparatus for fast registration. According to the registration information registered in first time, in the subsequent registration process, the only thing to do is to read the first registration information.

The invention discloses a dual-flat panel-based two-dimensional to three-dimensional medical image registering method and a dual-flat panel-based two-dimensional to three-dimensional medical image registering system. The method comprises the following steps of: generating the three-dimensional image of an object to be imaged, and generating a first and a second data reception register (DRR) imagelibraries in the planar outer corner directions of a first imaging plane and a second imaging plane off line; acquiring the first and second X-ray images of the object to be imaged in the first imaging plane and the second imaging plane, respectively; estimating the position parameters of the first and second X-ray images with the DRR images in the first and second DRR image libraries generated off line, respectively; adjusting the three-dimensional image according to the parameter estimation results of the first and second X-ray images, respectively, and then generating the first and second DRR image libraries on line; and estimating the position parameters of the first and second X-ray images with the DRR images in the first and second DRR image libraries generated on line, respectively. By adopting the method and the system of the invention, the complexity of the image registration can be reduced, and the speed, the precision and the success ratio of registration can be improved.

The invention relates to a coordinate system registration method, a robot control method and a device, equipment and a medium. The method comprises the following steps: acquiring a first position coordinate of each correction element of a coordinate system registration tool in an image coordinate system; wherein the coordinate system registration tool comprises at least four correction elements which are fixed in position and are not in a same plane; acquiring a second position coordinate of each correction element in the robot coordinate system by controlling an execution tail end connected with an operation arm of the robot to be in contact with the correction element; determining a conversion relationship between the image coordinate system and the robot coordinate system according to each first position coordinate and each second position coordinate; wherein the conversion relation is used for converting the target position of the target object from the image coordinate system to the robot coordinate system so as to obtain the position coordinate of the target position under the robot coordinate system. According to the method, the coordinate system registration tool is used, the operation steps and the calculation method are simple, and the registration speed between the robot coordinate system and the image coordinate system is greatly increased.

The invention belongs to the medical image processing and application field and relates to a global optimization-based neural navigation automatic space registration method. The method includes the following steps that: the point cloud of the whole head surface of a patient is obtained through scanning, and the point cloud is converted into a patient space; normalization processing is performed on the point cloud in the patient space and point cloud in an image space reconstructed by an image device, and the coordinates of the point cloud in the patient space and the point cloud in the image space reconstructed are unified to be within [- 1, 1] <3>; a Go-ICP algorithm-based global optimization spatial registration method is adopted to perform automatic registration on the two point clouds; and the two point clouds are fused, and are unified into the same space. As indicated by clinical practice results, the method of the invention has the advantages of simple implementation, convenient use and reliable precision, and is convenient to apply to clinical application, can be integrated in an existing navigation system and can improve the registration accuracy of the navigation system and reduce registration time.

The invention provides a high-precision point cloud registration method for three-dimensional laser scanning. The method comprises steps of providing strategy of two steps from preposed coarse registration to iterative fine registration; carrying out curvature-based feature extraction on scattered point clouds which are obtained through three-dimensional laser scanning and have different visual angles to obtain feature points; and calculating coordinate transformation parameters of the matched feature point pairs to obtain a preposed coarse registration result, eliminating wrong homonymous points by using a distance constraint condition for the condition that the two-point cloud data is not in a subset relationship, and then performing iterative fine registration to obtain a final registration result. Feature point extraction is rapid and accurate, feature point matching accuracy is high, speed is high, coarse registration provides a good initial position for fine registration, after iterative fine registration is iterated to a certain number of times, an accurate result can reach very high precision, registration precision can reach the millimeter level, the method can be used forcomplex point cloud data, and engineering application requirements are met. The point cloud registration speed and precision are greatly improved, and the registration method is wide in application range.

The invention brings forward a novel image registering method. The method comprises the following steps: establishing gray-scale-based mutual information registering adaptation value function; establishing and initializing a group, wherein four dimensions of each individual in the group respectively represents horizontal translation, vertical translation, a rotation angle and a zoom coefficient of a floating image; according to the mutual information registering adaptation value function, calculating a fitness value of each individual, and calculating an optimal position of the whole group; by use of an iteration mechanism of a differential evolution algorithm, updating a position vector of each individual, and updating the optimal position of the whole group; determining whether conditions for executing an alternative strategy is satisfied, and if so, executing the corresponding alternative strategy; and repeatedly executing the aforementioned steps until maximum iteration frequency Tmax of the differential evolution algorithm is satisfied. The novel image registering method has the advantages of good registering stability and high precision, greatly improves the performance of an image registering algorithm and lays a reliable foundation for subsequent image processing work.

The invention discloses an image registration method and an image registration device. The image registration method comprises the following steps: acquiring a first image and a second image of a target object, wherein, and the spatial resolution of the first image and the spatial resolution of the second image are the same, the first image and the second image are images collected by a first device and a second device respectively in a state that the imaging planes of the first and second devices are parallel to each other; determining the image registration information according to the target object in the first image and the second image; carrying out image registration on the images collected by the first device and the second device according to the image registration information. Theimage registration method and image registration device solve the technical problem that the image registration can only be carried out on each single image in the prior art.

The invention provides a method and apparatus for invited registration of nodes. The method comprises the following steps: a trusted node sends registration invitation information to a first node, andreceives a registration parameter sent by the first node, wherein the trusted node is a registered node in a blockchain network, and the first node is a node to be registered; the trusted node generates a network address corresponding to the first node according to the registration parameter; the trusted node sends a registration request of the first node to a master node, wherein the registration request is used for indicating the master node to verify the trusted node, and writes the network address corresponding to the first node into the blockchain network. According to the method and apparatus provided by the invention, the node to be registered is pre-registered via the registered node so as to assist the master node to complete the registration process of the node, thereby reducingthe service volume of the master node, and improving the registration speed of the node.

The invention relates to a two-dimensional and three-dimensional medical image registration method and a system, wherein the method comprises the steps of: generating a three-dimensional image of a three-dimensional imaged body, and generating a DRR (data receiving register) image library in an offline manner along the direction of a rotating angle outside the plane; acquiring an X-ray image of the three-dimensional imaged body; taking the X-ray image as a registered image, taking the DRR image in the DRR image library generated in an offline manner as a reference image, and respectively carrying out estimation on values such as a transition parameter in the plane, and / or a rotating angle parameter in the plane, and / or a rotating angle parameter outside the plane; taking a newest parameter estimation result as a reference position to carry out adjustment on a three-dimensional image, and generating the DRR image library along the rotating angle directions outside the two planes; taking the X-ray image as the registered image, taking the DRR image in the newest DRR image library generated in an online manner along the rotating angle directions outside the two planes as the reference image, and respectively carrying out estimation on values such as the transition parameter in the plane, and / or the rotating angle parameter in the plane, and / or the rotating angle parameter outsidethe plane.

The invention belongs to the field of medical image analysis and application, relates to a method for registering a stereoscopic target image to a template image, and specifically relates to the method for registering a stereoscopic medical image rapidly, accurately and non-linearly based on sparse representation. The sparse combination coefficient can be obtained by utilizing detected mark corresponding points of the template image and the target image and by searching an established deformation field dictionary and corresponding point dictionary. The coefficient can fuse corresponding examples of the deformation field dictionary, so that a final deformation field of the target image is obtained, and the target image is registered to the template image. The method has a relatively-good application in a clinic environment and can be used for registering brain nuclear magnetic resonance images rapidly and accurately in neurosciences, or used for accurately positioning the position of the prostate in prostate cancer radioactive therapy, and can realize more accurate registering more rapidly.

The invention relates to a multi-resolution non-rigid head medical image registration method on the basis of image edges, which pertains to the field of information fusion. The method utilizes searches gradually becoming accurate to transform parameters during the registration process. At first, wavelet transformation is used for detecting the edge of an image; a edge pyramid image is formed on the basis of an edge image; then relatively wide search is carried out in the lowest scale level (the roughest level) of the edge pyramid image; the best transformation for registering the two images is found; each subsequent level takes the search result of the former level as a center; and search scope is reduced to continually search until the search in the highest scale level is carried out. Compared with the registration directly using the original image, the method can reduce calculated amount, improve accuracy and can register images of different spatial resolution.

The invention discloses a quick ICP method for three-dimensional laser radar point cloud matching. In terms of characteristics of original point cloud data of three-dimensional laser radar, the method is used for quickly calculating the closest point between two point clouds under a polar coordinate system. The three-dimensional laser radar point cloud registration speed is increased according to the method. The registration speed of the original point cloud data of the three-dimensional laser radar is greatly increased, namely time complexity is reduced from O (DNcNm) of the classic ICP method to O (DNc), wherein Nc is the number of current point cloud data, and Nm is the number of model point set data.

The embodiment of the invention discloses a network registration method and device. The network registration method comprises the following steps: receiving an information acquisition request sent by a terminal, wherein the information acquisition request carries current geographic position information of the terminal; then, acquiring preset network feature information which corresponds to the geographic position information, wherein the preset network feature information includes a preset network type; and transmitting the preset network feature information to the terminal in order that the terminal performs network registration according to the preset network feature information. According to the scheme, the network feature information can be provided for the terminal in order that the terminal can directly perform network registration according to the network feature information without performing complicated network searching operations. Compared with the prior art, the network registration method and device have the advantages that the network registration speed and the network connection speed are increased.

The invention discloses an image registration method and device. The method comprises a step of acquiring a reference image and a to-be-registered image as a current reference image and a current to-be-registered image, a step of determining an image segmentation result, a step of taking two mutually corresponding image blocks in the current reference image and the current to-be-registered image as a reference image block and a to-be-registered image block, a step of determining a similarity degree of the corresponding image blocks, a step of determining a target displacement amount of the to-be-registered image block according to the similarity, a step of allowing the to-be-registered image block to move according to the target displacement amount to complete registration of this time, judging whether a registration end condition is satisfied or not, ending the registration if so, taking a to-be-registered image after the registration of this time if not, and determining an image segmentation result again to start a next registration operation. According to the image registration method and device, the image registration speed can be improved, and the accuracy of an image registration result can be improved.

The invention discloses a radio frequency device, an authentication server and an authentication method. The radio frequency device comprises a first authentication unit. The first authentication unit comprises a PUF sub unit, an encoder and a second processing sub unit, wherein the PUF sub unit is suitable for processing a first challenge when the first challenge is acquired and outputting corresponding real response data; the encoder is suitable for carrying out encoding and processing on the real response data, acquiring corresponding auxiliary data and transmitting the auxiliary data to the authentication server; and the second processing sub unit is suitable for processing the real response data, acquiring corresponding first processing data and transmitting the data to the authentication server. When the above radio frequency device is applied, the safety during the authentication process between the radio frequency device and the authentication server can be improved.

The present invention relates to na encrypt server activation method of the Dotnet LAN software. The present invention belongs to the LAN server technical field. The encrypt server activation method of the Dotnet LAN software includes an encrypt server and an activating terrace and is characterized in that the server collects computer hardware physical information, counts the hardware physical information by encryption, and educes the requiring registration documents, then the activates the terrace and generates a registration decode. The activation of server by encryption can be realized by educing the registration documents and leading in the registration documents. The present invention improves the encryption safety performance and registration speed of the LAN software with high reliability, low cost and low error rate, making the validity of the encryption, the capability and stability of the product guaranteed.

The invention discloses a multi-supervision image super-resolution reconstruction method based on a generative adversarial network. The method comprises the following steps: registering slice images with different resolutions of the same slice; making a training data set by using the registered slice images; on the training data set, training a generative adversarial model by using a multi-supervision multi-stage generation model; and reconstructing the low-resolution image into a high-resolution image by using the trained generative adversarial model. According to the method, the image shot under the low-power lens is reconstructed into the high-resolution image, the imaging time can be saved, the hardware space for storing the image can also be saved, and blurring and artifacts of commonmethods on pathological data are overcome.

The invention relates to the field of digital image processing, and provides an image registration method to provide an accurate and quick solution for the image registration under similar transformation. The adopted technical scheme is that the image registration method based on similar feature triangles comprises the following steps: 1) extracting and matching feature points; 2) randomly selecting samples to form feature triangles; 3) pre-checking the similar feature triangles; 4) selecting sample points according to a maximum Euclidean distance method; 5) obtaining a transformational model through transformational model solving and all-data checking. The method is mainly applied to the digital image processing.

The invention provides an image registration method and device, wherein the method comprises the following steps of: acquiring a registration reference image and an image to be registered, performinginitial registration on the image to be registered, and obtaining an initial registration image; dividing a first local image block and a second local image block corresponding to each other in the initial registration image and the registration reference image according to a predetermined division strategy; performing local registration on the first local image block and the second local image block to obtain a deformation field of the first local image block; applying the deformation field of the first local image block to the initial registration image to obtain an accurate registration image. By implementing the embodiment of the present application, the registration accuracy can be effectively improved, and the registration time can be reduced.

The embodiment of the invention discloses a network registration method comprising the steps that when startup of a mobile terminal is detected, RPLMN recorded by the mobile terminal is read, and PLMN of current candidate cells is obtained through network scanning; the first time of startup of the mobile terminal in a roaming location is determined according to the recorded RPLMN and the PLMN of the current candidate cells; and when the first time of startup of the mobile terminal in the roaming location is determined, the PLMN of the current candidate cells is applied to perform network registration. The embodiment of the invention also discloses the mobile terminal. With application of the network registration method and the mobile terminal, the steps of registration of a roaming location network can be simplified in the first time of startup in the roaming location so that speed of network registration can be enhanced.

The invention provides a method and a device for generating a domain name based on a real-name registration system to solve the problems that the domain name registering speed is slow, the identity reality of a registered enterprise corresponding to the domain name cannot be effectively validated, and the integrity degree of the enterprise is easily affected in the existing domain name generating method. The method for generating the domain name comprises the steps of receiving an organization code number input by a register applicant; validating the identity reality of the registered enterprise according to the one-to-one corresponding relation of the organization code number stored in advance and the enterprise name; when the identity of the registered enterprise is true, taking the organization code number of the registered enterprise as a domain name subject, and generating the domain name of the registered enterprise according to a domain name generating rule. Due to the technical scheme, website subject identity and an entity identity are matched; the integrity degree of the enterprise is enhanced, and the problem that the domain name system and the network security are threatened by unreality and inaccuracy of the domain name information is solved.