30results about How to "Fast and accurate registration" patented technology

The invention provides a two-dimensional and three-dimensional medical image registration method, which can quickly carry out registration on an intra-operative two-dimensional X-ray image and a preoperative three-dimensional CT (Computed Tomography) image. The two-dimensional and three-dimensional medical image registration method comprises the following steps: carrying out filtering preprocessing on the three-dimensional CT image, projecting the three-dimensional CT image in two mutually-orthogonal plane systems to obtain two corresponding digitally reconstructed radiograph (DRR) sets according to a maximum intensity projection (MIP) algorithm; obtaining two two-dimensional X-ray images on another two orthogonal planes, and carrying out the filtering preprocessing on the two-dimensional X-ray image; carrying out traversal on the two DRR sets, and carrying out registration on the two DRR sets and another two orthogonal two-dimensional X-ray images so as to determine a corresponding most similar DRR as well as the in-plane position coordinate, the in-plane rotation angle and an out-of-plane rotation angle of the DRR; converting the coordinate and the angles to be under a three-dimensional CT image coordinate system to obtain six registration parameters. According to the method, the X-ray image and the CT image which is generated in advance can be subjected to accurate and quick real-time registration.

A multi-modality image registration method includes the following steps: sampling reference images and images to be registered according to a first sampling proportion to obtain the reference images and the images to be registered with first resolution ratio, adopting a B spline surface as a deformation model to perform matching on the reference images and the images to be registered, calculating matching measure and obtaining first matching parameters, and increasing sampling proportion to repeat the processes until the matching measure reaches a reset threshold value, wherein in matching, matching parameters of last time are used as initial matching parameters after each-time sampling proportion is increased. An operation navigation method based on multi-modality image registration is further provided.

An image registering method includes extracting a first cross-section image from first volume data about a target object, which is acquired via a first image acquisition modality; displaying the first cross-section image; acquiring second volume data about the target object, the second volume data including a second cross-section image corresponding to the first cross-section image and position information of a probe, via a second image acquisition modality using the probe; and registering the first volume data with the second volume data, based on the first cross-section image, the second cross-section image, and the position information of the probe.

The invention relates to an automatic registration algorithm for a carrier terminal of a low-voltage power distribution network. The registration process of an entire power grid electric meter is completed in a wave form at multiple stages of adjacent node registration and non-adjacent node registration, adjacent and non-adjacent node supplementary registration and the like by taking each desk area concentrator as a center on the basis of registered domain carrier nodes, the proximity of power grid registration nodes is measured by using a signal quality parameter carried by a power line narrow-band carrier signal during desk area-crossing transmission, and the parameters are collected, calculated and compared by using a home domain determining algorithm, so that a home desk area of carrier nodes is accurately determined; and moreover, conflicts are handled and the priorities of messages are sequenced by adopting high-efficiency conflict avoidance based on a physical layer CSMA (Carrier Sense Multiple Access) technology and a multi-priority carrier packet receiving-transmitting control algorithm, so that communication nodes of all kinds in the desk area compete to use the same communication medium simultaneously. The automatic registration algorithm can be applied to power grid meter reading network electric meter registration of a plurality of kinds, and is particularly suitable for power line carrier remote meter reading network electric meter registration in complex communication environments.

A method registers data using a set of primitives including points and planes. First, the method selects a first set of primitives from the data in a first coordinate system, wherein the first set of primitives includes at least three primitives and at least one plane. A transformation is predicted from the first coordinate system to a second coordinate system. The first set of primitives is transformed to the second coordinate system using the transformation. A second set of primitives is determined according to the first set of primitives transformed to the second coordinate system. Then, the second coordinate system is registered with the first coordinate system using the first set of primitives in the first coordinate system and the second set of primitives in the second coordinate system. The registration can he used to track a pose of a camera acquiring the data.

An image registration method includes acquiring first image data for a target object that includes first coordinate information; acquiring second image data for the target object that includes second coordinate information, by using a probe; and registering the first image data with the second image data, using the first coordinate information and the second coordinate information. According to the image registration method, image registration between a plurality of pieces of volume data adjusted so that their coordinate axes correspond to each other is performed, whereby a high-quality registered image may be quickly and simply obtained.

The invention provides an automatic door which comprises an automatic door body and a video monitoring device mounted on the automatic door body. The video monitoring device comprises an image collecting module, an image initial processing module, a filtering module, a background modeling module, a background reducing module and a feature matching module. The image collecting module is used for collecting video monitor images. The image initial processing module is used for carrying out movement target initial detection on initial video image sequence information and outputting effective video image sequence information containing a movement target. The filtering module is used for receiving the effective video image sequence information and filtering the background shape of the effective video image sequence information. The background modeling module is used for establishing a filtered background model, the background model is composed of shape context histograms of N expression peripheral points with weights. The background reducing module is used for carrying out attribute classification on the peripheral points on the current frame of image and reducing the peripheral points belonging to the background. According to the automatic door, false detection on the movement target caused by background noise and vidicon shaking can be reduced to the maximum degree, and good real-time performance is achieved.

The invention discloses a network fast access system comprising a server and a mobile terminal. The server acquires and stores communication operator frequency band frequency point configuration information and also monitors the change of the operator intermediate frequency band frequency point configuration information in real time and transmits the information to the mobile terminal. The mobileterminal comprises a receiving module which is used for receiving the changed frequency band frequency point configuration information transmitted by the transmitting module, a second storage module which is used for storing the frequency band frequency point configuration information, an address recognition module which is used for recognizing the position of the mobile terminal, a screening module which is used for screening the corresponding address operator frequency band frequency point configuration information, a trigger module which is used for issuing the network access instruction and a network registration unit which is used for accessing to the local network. The invention also provides an access method based on the network fast access system. Compared with the systems and themethods in the prior art, the network fast access system is simple in structure, the access method is high in reliability, and the user can fast and accurately access to the communication network after arriving at the new address.

The invention provides an unmanned reservoir monitoring system which comprises the components of (1), an image acquisition module which is used for acquiring a video monitoring image; (2), an image preprocessing module which is used for performing preliminary moving object detection on original video image sequence information and outputting effective video image sequence information that comprises the moving object; (3), a filtering module which is used for receiving the effective video image sequence information and filtering the background shape of the effective video image sequence information; (4), a background modeling module which is used for establishing a background model after filtering, wherein the background model is composed of N shape context histograms which comprise weight and represent edge points; (5) a background subtraction module which is used for performing attribute classification on the edge points on the current image frame and subtracting the edge points that belong to the background; (6), a characteristic matching module; and (7), an alarm module. The unmanned reservoir monitoring system can maximally reduce false detection of the moving object caused by background noise and camera jitter and furthermore realizes high real-time performance.

The invention discloses a three-dimensional image iterative registration method, which comprises the following steps: (1) acquiring a reference image and a target image, calculating the central coordinates of each vertebral section of the reference image and the target image, and obtaining an initial transformation relationship between the reference image and the target image; (2) generating a plurality of reference cross-section images in the reference image, and intercepting the target image according to the step (1) to obtain a corresponding target cross-section image; (3) performing optimal similarity matching on a certain target cross-section image and the corresponding reference cross-section image to obtain the transformation quantity, and calculating to obtain the pose of the corresponding target cross-section coordinate system; (4) judging whether the pose increments of the plurality of target section coordinate systems are all smaller than a set threshold, and if so, calculating a final transformation relation between the reference image coordinate system and the target image coordinate system; otherwise, updating the transformation quantity and the pose of the corresponding target section coordinate system, and repeating the step. According to the invention, accurate registration between the two vertebral section images can be rapidly completed, the efficiency is high, the adaptability is high, and a good registration effect can be achieved for poor image data.

A multi-modality image registration method includes the following steps: sampling reference images and images to be registered according to a first sampling proportion to obtain the reference images and the images to be registered with first resolution ratio, adopting a B spline surface as a deformation model to perform matching on the reference images and the images to be registered, calculating matching measure and obtaining first matching parameters, and increasing sampling proportion to repeat the processes until the matching measure reaches a reset threshold value, wherein in matching, matching parameters of last time are used as initial matching parameters after each-time sampling proportion is increased. An operation navigation method based on multi-modality image registration is further provided.

The invention provides an unmanned warehouse monitoring system, and the system comprises (1), an image collection module which is used for collecting a video monitoring image; (2), an image initial processing module which is used for the initial detection processing of a moving object of original video image sequence information, and the outputting of effective video image sequence information comprising the moving object; (3), a filtering module which is used for receiving the effective video image sequence information and carrying out the filtering of the background shape of the effective video image sequence information; (4), a background modeling module which is used for building a background model after filtering, wherein the background model is set to consist of N shape context histograms which represent edge points and have weights; (5), a background reduction module which is used for carrying out the attribute classification of the edge points of a current image frame, and reducing the edge points belonging to the background; (6), a feature matching module; (7), an alarm module. The device can reduce the wrong detection, caused by background noise and the jittering of a camera, of the moving object, and is good in instantaneity.

An image registration method includes acquiring first image data for a target object that includes first coordinate information; acquiring second image data for the target object that includes second coordinate information, by using a probe; and registering the first image data with the second image data, using the first coordinate information and the second coordinate information. According to the image registration method, image registration between a plurality of pieces of volume data adjusted so that their coordinate axes correspond to each other is performed, whereby a high-quality registered image may be quickly and simply obtained.

The invention provides an intelligent zebra crossing guardrail, which comprises an intelligent zebra crossing guardrail body and a video monitoring device installed on the intelligent zebra crossing guardrail body, wherein the video monitoring device comprises an image acquisition module (1), an images first processing module (2), a filtering module (3), a background modeling module (4), a background subtraction module (5) and a feature matching module (6); the image acquisition module (1) is used for acquiring a video monitored image; the images first processing module (2) is used for carrying out preliminary detection processing of a moving object on original video image sequence information, and outputting effective video image sequence information containing the moving object; the filtering module (3) is used for receiving the effective video image sequence information and filtering background shape of the effective video image sequence information; the background modeling module (4) is used for establishing a background model after filtering, wherein the background model is designed to comprises N weighted shape context histograms representing peripheral points; and the background subtraction module (5) is used for carrying out attribute classification on the peripheral points on a current frame image and subtracting the peripheral points belong to the background. The intelligent zebra crossing guardrail can reduce error detection of the moving object caused by background noise and video camera jitter to the greatest extent, and has good real-time performance.

The invention provides a monitoring device for a playground. The monitoring device comprises the components of (1), an image acquiring module which is used for acquiring a video monitoring image; (2), an image preprocessing module which is used for performing moving object preliminary detecting on original video image sequence information and outputting effective video image sequence information that comprises a moving object; (3), a filtering module which is used for receiving the effective video image sequence information and filtering the background shape of the effective video image sequence information; (4), a background modeling module which is used for establishing a background model after filtering, wherein the background module is composed of N shape context histograms which comprise weights and represent edge points; (5), a background subtraction module which is used for performing attribute classifying on the edge points on the current image frame and subtracting the edge points that belong to the background; and (6), a characteristic matching module. The monitoring device can maximally reduce moving object error detection caused by background noise and camera jitter, and furthermore has high real-time performance.

The invention provides a guide system, and the system comprises (1), an image collection module which is used for collecting a video monitoring image; (2), an image initial processing module which is used for the initial detection processing of a moving object of original video image sequence information, and the outputting of effective video image sequence information comprising the moving object; (3), a filtering module which is used for receiving the effective video image sequence information and carrying out the filtering of the background shape of the effective video image sequence information; (4), a background modeling module which is used for building a background model after filtering, wherein the background model is set to consist of N shape context histograms which represent edge points and have weights; (5), a background reduction module which is used for carrying out the attribute classification of the edge points of a current image frame, and reducing the edge points belonging to the background; (6), a feature matching module; (7), a guide module. The system can reduce the wrong detection, caused by background noise and the jittering of a camera, of the moving object, and is good in instantaneity.

The invention provides a moving object shooting method. The shooting method is implemented through the electric connection of a shutter and a video monitoring device. The video monitoring device comprises (1) an image acquiring module for acquiring a video monitoring image; (2) an image initial processing module for performing the initial detection processing of the moving object on the original video image sequence information and outputting the valid video image sequence information containing the moving object; (3) a filter module for receiving the valid video image sequence information and performing the filtering processing on a background shape of the valid video image sequence information; (4) a background modeling module for establishing a filtered background model, wherein the background model is composed of N shape context histograms with weights and used for representing peripheral points; (5) a background substraction module for classifying attributes of the peripheral points on the current frame image and substracting the peripheral points belonging to the background; and (6) a feature matching module. Through the adoption of the method provided by the invention, the false detection on the moving target caused by the background noise and shaking of a camera can be maximally reduced, and the instantaneity is good.

The invention provides a fouling monitoring device comprising (1) an image acquisition module which is used for acquiring video monitoring images; (2) an image preliminary processing module which is used for performing moving target preliminary detection processing on original video image sequence information and outputting effective video image sequence information including a moving target; (3) a filtering module which is used for receiving the effective video image sequence information and performing filtering processing on the background shape of the effective video image sequence information; (4) a background modeling module which is used for establishing a background model after filtering, wherein the background model is arranged to be composed of N shape context histograms which have weight and represent edge points; (5) a background subtraction module which is used for performing attributive classification on the edge points of a current frame of image and subtracting the edge points belonging to the background; and (6) a feature matching module. False detection of the moving target caused by background noise and camera jittering can be reduced to the largest extent and the real-time performance is great.

The invention discloses an aircraft C-shaped beam registration method based on local invariance. The method comprises the following steps: acquiring actual measurement point cloud data of an aircraftC-shaped beam; carrying out coarse registration on the C-shaped beam digital-analog and actual measurement point cloud data by utilizing a rapid point feature histogram feature description method; carrying out segmentation processing and plane extraction on the roughly registered actual measurement point cloud data, segmenting the C-shaped beam into three planes, and judging the bottom surface ofthe C-shaped beam from the three planes; taking the bottom surface of the C-shaped beam as a constraint of C-shaped beam digital-analog registration, carrying out fine registration by adopting part ofpoint cloud data of the bottom surface of the C-shaped beam and the C-shaped beam digital-analog registration, and calculating to obtain a transformation matrix of actually measured point cloud dataof the bottom surface of the C-shaped beam and the C-shaped beam digital-analog registration; and performing position transformation on the roughly registered aircraft C-shaped beam actual measurementpoint cloud data by adopting the transformation matrix to obtain a final registration result. According to the method, the success rate and the accuracy of registration of the airplane C-shaped beamactual measurement data and the standard digital-analog before detection under the deformation condition are greatly improved.

The invention provides a monitoring device for searching a moving object, and the device comprises (1), an image collection module which is used for collecting a video monitoring image; (2), an image initial processing module which is used for the initial detection processing of a moving object of original video image sequence information, and the outputting of effective video image sequence information comprising the moving object; (3), a filtering module which is used for receiving the effective video image sequence information and carrying out the filtering of the background shape of the effective video image sequence information; (4), a background modeling module which is used for building a background model after filtering, wherein the background model is set to consist of N shape context histograms which represent edge points and have weights; (5), a background reduction module which is used for carrying out the attribute classification of the edge points of a current image frame, and reducing the edge points belonging to the background; (6), a feature matching module. The device can reduce the wrong detection, caused by background noise and the jittering of a camera, of the moving object, and is good in instantaneity.

The invention provides a finishing line watch stopping device which comprises a finishing line watch stopping device and a video monitoring device which is electrically connected with the finishing line watch stopping device, wherein the video monitoring device comprises the components of (1), an image acquisition module which is used for acquiring a video monitoring image; (2) an image preprocessing module which is used for performing moving object preliminary detection on original video image sequence information and outputting effective video image sequence information that comprises a moving object; (3), a filtering module which is used for receiving the effective video image sequence information and filtering the background shape of the effective video image sequence information; (4), a background modeling module which is used for establishing a background model after filtering, wherein the background module is composed of N shape context histograms which comprise weights and represent edge points; (5), a background subtraction module which is used for performing attribute classifying on the edge points on the current image frame and subtracting the edge points that belong to the background; and (6), a characteristic matching module. The finishing line watch stopping device can maximally reduce moving object error detection caused by background noise and camera jitter, and furthermore has high real-time performance.

The present invention relates to an automatic registration algorithm for a low-voltage distribution network carrier terminal. It uses each district concentrator as a centre, relies on a domain carrier node which has finished registration, and falls into neighbourhood node registration and non-neighbourhood node registration and neighbourhood and non-neighbourhood node supplementary registration, etc. to complete the registration process of the electric meters of the overall electric grid in a multi-stage wave manner, measures the proximity of the electric-grid registration nodes by virtue of signal quality parameters carried by power-line narrow-band carrier signals during cross-district propagation, and collects, calculates and compares these parameters through a home-domain confirmation algorithm, thereby accurately determining the home district of a carrier node. In addition, a high-efficiency conflict avoidance and multi-priority carrier packet transceiving control algorithm based on a physical-layer CSMA technology is adopted to process the priorities of conflict and discharging messages so that all types of communication nodes in a district simultaneously compete to use one and the same communication medium. The present invention can be used for electric-meter registration of a plurality of types of electric-grid meter-reading networks, and is in particular suitable for electric-meter registration of an electric-power carrier remote meter-reading network in a relatively complicated communication environment.

The invention provides a registration method for a circuit design layout and an electron microscope scanning image. The registration method comprises the following steps: S1, providing the circuit design layout and the electron microscope scanning image to be registered; S2, processing the circuit design layout to be registered to obtain a binary image of the design layout; processing the electronmicroscope scanning image to be registered to obtain an electron microscope scanning binary image; S3, performingGaussian filtering on the design layout binary image and the electron microscope scanning binary image so that the gray value of the central axis of the area corresponding to the design pattern and the scanning pattern is enabled to be maximum; and S4, performing registration accordingto the central axes of the design pattern and the scanning pattern. The invention further provides a registration system of the circuit design layout and the electron microscope scanning image, the circuit design layout, an imaging error calculation method of the circuit design layout and electronic equipment. The circuit design layout, the electron microscope scanning image registration method and system, the circuit design layout, the imaging error calculation method of the circuit design layout and the electronic equipment have the advantages of accurate registration, accurate error calculation and the like.

The invention provides an image splicing method and an image splicing device, which can realize rapid and accurate image registration and splicing. The method comprises the steps of obtaining a to-be-spliced image and a reference image; carrying out feature point detection on the to-be-spliced image and the reference image and calculating acceleration robust features; aiming at the to-be-spliced image and the reference image, taking each detected feature point as a center, selecting feature image blocks with a preset size, extracting convolutional neural network features according to the selected feature image blocks, fusing the obtained acceleration robust features and the convolutional neural network features, and generating features of the to-be-spliced image and features of the reference image; calculating the matching degree between the features of the fused to-be-spliced image and the features of the reference image, and carrying out the transformation matrix estimation, therebyachieving the registration of the to-be-spliced image and the reference image; and fusing the registered to-be-spliced image and the reference image to obtain a complete spliced image. The invention relates to the field of image processing.

The present invention proposes a navigation system for orthopedic surgery based on multi-mode image fusion, including: a binocular vision positioning camera for real-time acquisition of the spatial coordinates of passive infrared reflective marking balls placed on surgical instruments and the skin surface of subjects; The segmentation module is used to accurately segment the spine part in the initial image; the spatial coordinate registration module is used to register the spatial coordinates and image coordinates; the surgical instrument calibration module is used to establish the coordinates of the tip of the surgical instrument and place it at its end The conversion relationship of passive infrared reflective marker spherical coordinates; the multimodal image registration module is used to register the intraoperative cone-beam CT image to the preoperative CT image; the visual interface module is used to establish a three-dimensional image for intraoperative Real-time display of the position of the surgical instrument in the image of the subject. The present invention can accurately locate the real-time position of the surgical instrument in the subject during the operation, provide accurate and reliable navigation for orthopedic surgery, effectively reduce the operation risk and operation time, and the system is easy to operate and has high applicability.

The invention provides a red light running monitoring device which comprises the components of (1), an image acquisition module which is used for acquiring a video monitoring image; (2), an image preliminary processing module which is used for performing preliminary detecting of a motion target on original video image sequence information and outputting effective video image sequence information that comprises the motion target; (3), a filtering module which is used for receiving the effective video image sequence information and filtering background shape of the effective video image sequence information; (4), a background modeling module which is used for establishing a background model after filtering, wherein the background model is composed of N context histograms which comprise weights and represent the shapes of edge points; (5), a background subtraction module which is used for performing attribute classification on the edge points on the current image frame and subtracting the edge points that belong to the background; and (6), a characteristic matching module. The red light running monitoring device can maximally reduce motion target misdetection caused by background noise and camera jitter and furthermore realizes high real-time performance.

The invention provides a video monitoring device capable of detecting a moving object. The video monitoring device comprises the components of (1), an image acquisition module which is used for acquiring a video monitoring image; (2), an image preprocessing module which is used for performing moving object preliminary detection processing on original video image sequence information and outputting effective video image sequence information that comprises the moving object; (3), a filtering module which is used for receiving the effective video image sequence information and performing filtering on the background image of the effective video image sequence information; (4), a background modeling module which is used for establishing a background model after filtering, wherein the background model is composed of N shape context histograms which comprise the weights and represent edge points; (5), a background subtracting module which is used for performing attribute classification on the edge points on the current image frame and subtracting the edge points that belong to the background; and (6), a characteristic matching module. The video monitoring device can maximally reduce moving object false detection caused by background noise and camera jitter and furthermore realize high real-time performance.

The invention provides a device for remotely acquiring useful monitoring information. The device comprises a remote transmission device and a video monitoring device, wherein the video monitoring device comprises the components of (1), an image acquisition module which is used for acquiring a video monitoring image; (2), an image preprocessing module which is used for performing moving object preliminary detection on original video image sequence information and outputting effective video image sequence information that comprises a moving object; (3), a filtering module which is used for receiving the effective video image sequence information and filtering the background shape of the effective video image sequence information; (4), a background modeling module which is used for establishing a background model after filtering, wherein the background module is composed of N shape context histograms which comprise weights and represent edge points; (5), a background subtraction module which is used for performing attribute classifying on the edge points on the current image frame and subtracting the edge points that belong to the background; and (6), a characteristic matching module. The device can maximally reduce moving object error detection caused by background noise and camera jitter, and furthermore has high real-time performance.

The invention provides an intelligent monitoring early-warning device, and the device comprises (1), an image collection module which is used for collecting a video monitoring image; (2), an image initial processing module which is used for the initial detection processing of a moving object of original video image sequence information, and the outputting of effective video image sequence information comprising the moving object; (3), a filtering module which is used for receiving the effective video image sequence information and carrying out the filtering of the background shape of the effective video image sequence information; (4), a background modeling module which is used for building a background model after filtering, wherein the background model is set to consist of N shape context histograms which represent edge points and have weights; (5), a background reduction module which is used for carrying out the attribute classification of the edge points of a current image frame, and reducing the edge points belonging to the background; (6), a feature matching module; (7), an alarm module. The device can reduce the wrong detection, caused by background noise and the jittering of a camera, of the moving object, and is good in instantaneity.