CN107220933A

CN107220933A - A kind of reference line determines method and system

Info

Publication number: CN107220933A
Application number: CN201710329970.XA
Authority: CN
Inventors: 马庆贞; 王燕燕; 欧阳斌
Original assignee: Shanghai United Imaging Healthcare Co Ltd
Current assignee: Shanghai United Imaging Healthcare Co Ltd
Priority date: 2017-05-11
Filing date: 2017-05-11
Publication date: 2017-09-29
Anticipated expiration: 2037-05-11
Also published as: CN107220933B

Abstract

The present invention provides a kind of reference line and determines method and system, and methods described includes：Obtain at least two images to be spliced, at least two image correspondence to be spliced, first spatial positional information；Based on described at least two images to be spliced, a stitching image, the stitching image correspondence second space positional information are determined；Determine the splicing relation between described at least two images to be spliced；And based on the splicing relation between described at least two images to be spliced, first spatial positional information, and second space positional information, determine at least one reference line.

Description

A kind of reference line determines method and system

【Technical field】

Reference line is accurately positioned automatically for splicing result the present invention relates to a kind of Medical Imaging, more particularly to one kind Method and system.

【Background technology】

Medical Imaging, reference line has very important significance for the discovery, diagnosis and treatment of disease.Example Such as, in computer tomography (Computed Tomography, CT) image of the full vertebra of human body, user (such as doctor) energy Size, position and the position relationship between surrounding vital tissue of focus are enough accurately obtained using reference line.However, mesh Before, the display existence position of reference line on the medical image is inaccurate, and can not accurate adjustment the problem of.For example, due to Change in location in splicing, in some medical images (such as vertebra magnetic resonance imaging (Magnetic Resonance Imaging, MRI) image) on can not accurately show the position of reference line or reference line can not be adjusted.This problem So that user (such as doctor) can not accurately and quickly obtain the position of reference line, reduce user (such as doctor) and disease is examined Disconnected and therapeutic efficiency.Therefore a kind of method or system of reference line positioning are needed so that can accurately show in medical image Show and adjust the position of reference line.The method or system of this reference line positioning will effectively lift user (such as doctor) to disease The diagnosis of disease and therapeutic efficiency.

【The content of the invention】

A kind of one side of the application determines method there is provided a kind of MRI, CT and/or PET image reference line.Should Method includes the one or multi-step in following operation：Obtain at least two images to be spliced, at least two image to be spliced The first spatial positional information of correspondence；Based on described at least two images to be spliced, a stitching image, the stitching image are determined Correspondence second space positional information；Determine the splicing relation between described at least two images to be spliced；And based on it is described extremely Splicing relation between few two images to be spliced, first spatial positional information, and second space positional information, it is determined that extremely A few reference line.

Optionally, described at least two images to be spliced include at least one of CT images, MRI image or PET image.

Optionally, the first spatial positional information includes the positional information and directional information of described at least two images to be spliced At least one of；In positional information and directional information of the second space positional information including the stitching image at least It is a kind of.

Optionally, determine that the splicing relation between described at least two images to be spliced includes entering the image to be spliced At least one following operation of row：Translation；Rotation；Scaling；And shearing.Can be in described at least two images to be spliced It is any one or more translated, rotated, scaled and sheared in one or more.

Optionally, the splicing relation between at least two images to be spliced includes a registering matrix.

Optionally, based on the splicing relation between described at least two images to be spliced, at least one reference line bag is determined Include：Based on first spatial positional information and the second space positional information, described at least two images to be spliced are determined Intersection point between the stitching image；Based on the splicing relation between described at least two images to be spliced, described in adjustment extremely Few intersection point between two images to be spliced and the stitching image；And based on described at least two figures to be spliced after adjustment Intersection point between picture and the stitching image, determines at least one reference line.

Optionally, based on first spatial positional information and the second space positional information, determine described to be spliced Intersection point between image and the stitching image includes：Plane and the spliced map based on described at least two images to be spliced The plane of picture, determines the intersection point.

Optionally, determine that at least one reference line includes：Based on first spatial positional information and described second Spatial relation, determines the first reference line；Translation and the rotation behaviour that first reference line is carried out based on an object function At least one in work, obtains a regulation matrix；And first reference line is corrected based on the regulation matrix Obtain the second reference line.

According to some embodiments of the present application, there is provided a kind of determination of the reference line of CT images, MRI image or PET image System.The system can include a kind of computer-readable storage medium for being configured as storing executable module and a processing Device, the processor is able to carry out the executable module of the computer-readable storage media storage.Executable module includes： Image mosaic module and reference line determining module.Image mosaic module is configured as：At least two images to be spliced are obtained, it is described At least two first spatial positional informations of image correspondence to be spliced；Based on described at least two images to be spliced, a spelling is determined Map interlinking picture, the stitching image correspondence second space positional information；And between determination at least two image to be spliced Splicing relation；Reference line determining module, is configured as based on the splicing relation between described at least two images to be spliced, described First spatial positional information, and second space positional information, determine at least one reference line.

Optionally, the splicing relation between at least two images to be spliced includes carrying out at least one to the image to be spliced Individual following operation：Translation；Rotation；Scaling；And shearing.

Optionally, the splicing relation between described at least two images to be spliced, determines that at least one reference line includes：Base In first spatial positional information and the second space positional information, described at least two images to be spliced are determined and described Intersection point between stitching image；Based on the splicing relation between described at least two images to be spliced, adjustment described at least two Intersection point between image to be spliced and the stitching image；And based on described at least two images to be spliced after adjustment and institute The intersection point between stitching image is stated, at least one reference line is determined.

Optionally, it is described to be based on first spatial positional information and the second space positional information, it is determined that described treat Intersection point between stitching image and the stitching image includes：Plane and the spelling based on described at least two images to be spliced The plane of map interlinking picture, determines the intersection point.

Optionally, at least one reference line includes：Based on first spatial positional information and the second space Position relationship, determines the first reference line；In translation and rotation process that first reference line is carried out based on an object function At least one, obtain one regulation matrix；And acquisition is corrected to first reference line based on the regulation matrix Second reference line.

【Brief description of the drawings】

Fig. 1 is the schematic diagram of reference line alignment system；

According to some embodiments of the present application, shown in Fig. 2 is an illustrative diagram of processing equipment；

According to some embodiments of the present application, shown in Fig. 3 is a kind of exemplary process diagram of reference line positioning；

According to some embodiments of the present application, shown in Fig. 4 is to determine reference based on spatial positional information and splicing relation A kind of exemplary process diagram of line；

According to some embodiments of the present application, shown in Fig. 5 is a kind of exemplary process diagram of reference line correction；

According to some embodiments of the present application, shown in Fig. 6 is a kind of illustrative diagram of reference line regulation；

According to some embodiments of the present application, shown in Fig. 7 is a kind of framework of computer equipment of processing equipment, the shifting Dynamic equipment is implemented for the particular system for implementing to disclose in the application；

According to some embodiments of the present application, shown in Fig. 8 is a kind of structure of mobile device, and the mobile device can be used The particular system for implementing to disclose in the application in realization；And

According to some embodiments of the present application, shown in Fig. 9 A to 9C is that reference line is accurately positioned a kind of exemplary of result Schematic diagram.

【Embodiment】

Many details are elaborated in the following description to fully understand the present invention.But the present invention can be with Much it is different from other manner described here to implement, those skilled in the art can be in the situation without prejudice to intension of the present invention Under do similar popularization, therefore the present invention is not limited to the specific embodiments disclosed below.

As shown in the application and claims, unless context clearly points out exceptional situation, " one ", " one ", " one The word such as kind " and/or "the" not refers in particular to odd number, may also comprise plural number.It is, in general, that term " comprising " only points out bag with "comprising" Include clearly identify the step of and element, and these steps and element do not constitute one it is exclusive enumerate, method or equipment May also include other the step of or element.

According to some embodiments of the present application, shown in Fig. 1 is the schematic diagram of reference line alignment system.Reference line can refer to The intersection between second medical image and first medical image is shown on first medical image.Reference line is for description Lesion locations, determine it is all significant in terms of the selection PCI of therapeutic scheme and tumour.

Reference line alignment system 100 can include one or more imaging devices 110, one or more networks 120, one Or multiple processing equipment 130 and one or more databases 140.

Imaging device 110 can be scanned to detection object, obtain scan data, the scan data can pass through net Network 120 is sent to processing equipment 130 and is further processed, and can also store in database 140.The detection object can be with Including human body, animal etc..Imaging device 110 can include but is not limited to computer tomography (Computed Tomography, CT) equipment, magnetic resonance imaging (Magnetic Resonance Imaging, MRI) equipment or positron emission Fault imaging (Positron Emission Computed Tomography, PET) equipment.

Processing equipment 130 can be to the data of input (for example, being deposited in imaging device 110 is obtained and/or database 140 The scan data of storage, scan image) handled, analyze to generate result.For example, processing equipment 130 can be according to sweeping Retouch data generation scan image.In another example, processing equipment 130 can be split to scan image, obtain image segmentation result. The scan image can be two dimensional image or 3-D view.Processing equipment 130 can include processor and input/ Output device (not shown).In certain embodiments, the processor can be a server or a clothes Wu Qi groups.One server farm can be centralized, such as data center.One server farm can also be distribution Formula, such as one distributed system.The processor can be Cloud Server, file server, database server, FTP One or more of combinations in server, apps server, proxy server, mail server etc..The server Can be local or long-range.In certain embodiments, the server can be accessed in database 140 and stored Information (for example, being stored in the medical image in database 140), the information in imaging device 110 is (for example, imaging device 110 The medical image of shooting).In certain embodiments, the input/output device can also may be used to the processor input data To receive the data of the processor output, and the data of output are showed in forms such as numeral, character, image, sound. In certain embodiments, the input/output device can include but is not limited in input unit, output device etc. one kind or Several combinations.The input unit can include but is not limited to character entry apparatus (for example, keyboard), optical reader (for example, optical mark reader, optical character reader), image input unit (for example, Genius mouse, action bars, light pen), figure As input unit (for example, video camera, scanner, facsimile machine), analogue input unit (for example, language analog-to-digital conversion identifying system) One or more of combinations in.The output device can include but is not limited to display device, printing equipment, plotter, One or more of combinations in image output device, instantaneous speech power, magnetic recording system etc..In certain embodiments, locate Reason equipment 130 may further include storage device (not shown), and the storage device can store various information, example Such as, program and data etc..In certain embodiments, the intermediate data and/or result that processing equipment 130 is produced are (for example, sweep Tracing picture, image segmentation result etc.) it can be stored in the storage device of database 140 and/or processing equipment 130, can also Exported by input/output device.

Data storage device 140 can refer to the equipment with store function.Data storage device 140 can store from into The various data produced in being worked as the scan data and processing equipment 130 of the collection of equipment 110.Data storage device 140 can be with It is local or long-range.Data storage device 140 can include but is not limited to hierarchical database, network type number According to one or more of combination therein such as storehouse and relational database.Data storage device 140 can be by after information digitalization Again with using electricity, magnetically or optically etc. the storage device of mode is stored.Database 140 can be for the various information of storage, example Such as program and data.Data storage device 140 can be the equipment using electric energy mode storage information, such as various storages Device, random access memory (Random Access Memory (RAM)), read-only storage (Read Only Memory (ROM)) etc..Wherein random access memory include but is not limited to dekatron, selectron, delay line storage, WILLIAMS-DARLING Ton pipe, Dynamic RAM (DRAM), SRAM (SRAM), IGCT random access memory (T-RAM), zero capacitance are random One or more of combinations in memory (Z-RAM) etc..Read-only storage includes but is not limited to magnetic bubble memory, A.O. line and deposited Reservoir, thin-film memory, magnetic plated wire memeory, magnetic core internal memory, magnetic drum memory, CD drive, hard disk, tape, early stage are non- It is volatile memory (NVRAM), phase-change memory element, reluctance type random storage formula internal memory, ferroelectric random stored memory, non-volatile SRAM, flash memory, the electronics formula of erasing can make carbon copies read-only storage, Erasable Programmable Read Only Memory EPROM, programmable read only memory, Mask ROM, floating connection door random access memory, nanometer random access memory, racing track internal memory, in variable resistance type Deposit, one or more of combinations in programmable metallization unit etc..Data storage device 140 can be deposited using magnetic energy mode The equipment for storing up information, such as hard disk, floppy disk, tape, core memory, magnetic bubble memory, flash memory.Data storage device 140 Can be the equipment using optical mode storage information, such as CD or DVD.Data storage device 140 can utilize magneto-optic The equipment of mode storage information, such as magneto-optic disk.The access mode of data storage device 140 can be random storage, serial Access one or more of combinations in storage, read-only storage etc..Data storage device 140 can be impermanent memory storage Device or permanent memory memory.Above-mentioned storage device is to list some examples, and system is positioned in reference line The database that can be used in system 100 is not limited thereto.

Network 120 can be single network or multiple network combination.Network 120 can include but is not limited to LAN, wide area network, common network, dedicated network, WLAN, virtual network, city Metropolitan Area Network (MAN), public switch telephone network One or more of combinations in network etc..Network 120 can include multiple network access point, such as wired or WAP, base Stand or network exchange point, data source is connected network 120 by above access point and information is sent by network.

It should be noted that the description for service system above, only for convenience of description, can not be limited in the application Within the scope of illustrated embodiment.It is understood that for those skilled in the art, understanding the principle of the system Afterwards, modules can be combined in the case of without departing substantially from this principle, or constitutes subsystem and other moulds Block is connected, and the various modifications and variations in form and details are made to the implementation above method and systematic difference field.For example, In some embodiments, database 140 can be the cloud computing platform with data storage function, including but not limited to public cloud, Private clound, community cloud and mixed cloud etc..Such deformation, within the protection domain of the application.

According to some embodiments of the present application, shown in Fig. 2 is an illustrative diagram of processing equipment 130.Processing Equipment 130 can include image mosaic module 210, reference line determining module 220, and reference line correction module 230.Institute in Fig. 2 Stating module can be realized by the computer in such as Fig. 7 by CPU element 720.Between the module can directly with one another (and/ Or indirectly) connection.It should be evident that processing equipment 130 is only to represent some embodiments of the present application described in Fig. 2, for this For the those of ordinary skill in field, on the premise of not paying creative work, can be made according to Fig. 2 description modification, Increase and delete.For example, two of which module can be combined into a module, or one of module can be divided into two Or multiple modules.

In certain embodiments, image mosaic module 210 can produce one or more spliced maps based on image to be spliced Picture.Image mosaic module 210 can obtain the image to be spliced from imaging device 110.The image to be spliced obtained can be wrapped Include but be not limited to one or more medical images (such as CT images, MRI image, PET image).The image to be spliced can be with It is two dimensional image or 3-D view.The image to be spliced can include but is not limited to original image and/or processing figure Picture.The original image can refer to the image (such as medical image) directly obtained according to scan data.The processing image can To refer to the image that processing acquisition is carried out to original image.Processing to original image can include but is not limited to image enhaucament, figure As restructuring, three-dimensional reconstruction, image filtering, Image Coding (for example, compressed encoding), image format conversion, image rendering, image contracting One or more of combination therein such as put.

The number of the image to be spliced of the acquisition can be two or more.In certain embodiments, it is some Image to be spliced can be based on the scanning to the different angles of same detection object (such as human body, part of human body) and generate. In certain embodiments, based on multiple different angles, imaging device 110 can produce multiple images section.Each image sections can be with Including at least one image to be spliced, and the image to be spliced has related or similar spatial positional information.For example, Imaging device 110 can be based on 3 full vertebras of different angle shot human bodies, to obtain 3 image sections：Image sections A (such as cervical vertebra images Section), image sections B (such as thoracic vertebrae image sections) and image sections C (such as lumbar vertebrae image sections).Image sections A, image sections B and image sections C bags The amount of images to be spliced contained can be the same or different.

The stitching image can be two dimensional image or 3-D view.For example, stitching image can be displaying people The two dimensional image of body whole vertebra.For another example stitching image can be the 3-D view for showing human liver.In some embodiments In, image to be spliced or the image to be spliced of different modalities that the stitching image can be based on identical mode and generate.Example Such as, the stitching image can be based on different modalities medical image (such as MRI image, CT images, PET image) and give birth to Into.The medical image (or medical image of different modalities) of the identical mode can correspond to identical or different detection object And/or the different piece of same detection object.For example, the stitching image of the full vertebra of human body can be by MRI image (such as cervical vertebra image Section) and CT images (such as lumbar vertebrae image sections) be spliced.

In certain embodiments, reference line determining module 220 can determine reference line.The reference line can be used for one The position relationship between another image and current reference image is shown on individual reference picture.The reference line can include one The element such as bar or a plurality of straight line, line segment, curve, point.The reference line can include any number of pixels.In some embodiments In, reference line is generally represented with the straight line on reference picture, dotted line, line segment etc..The reference picture can be splicing Image or image to be spliced.For example, in skull CT scanning, imaging device 110 can produce cross-section bit image, the arrow of head Shape bit image or coronal bit image.Reference line determining module 220 can determine and/or show one or many on cross-section bit image Bar reference line, the reference line can be used for showing position relationship (such as Fig. 9 A between coronal bit image and current cross-section bit image It is shown).The cross-section bit image can be represented when detecting that object (for example, human body) is stood on ground, with sagittal image It is vertical with coronal bit image, it will detect that object is crosscutting for two-part image up and down.The sagittal image can be represented when inspection It is left and right two by detection object rip cutting by the fore-and-aft direction of detection object when survey object (for example, human body) is stood on ground The image divided.The coronal bit image can be represented when detecting that object (for example, human body) is stood on ground, by detection object Left and right directions, will detection object be cut to before and after two-part image.

In certain embodiments, reference line determining module 220 can be determined and shown on stitching image a plurality of to be spliced The reference line of image and current stitching image.For example, on the stitching image of the full vertebra of human body, reference line determining module 220 can To show 5 reference lines.The reference line can show the stitching image and 5 cross-section bit image (such as 5 waists of the full vertebra of human body The cross-section bit image of vertebra) between position relationship.User (such as doctor) can be based on 5 reference lines, analyze some lumbar vertebrae Cross-section bit image in the complete supraspinal accurate location (the cross-section bit image such as fracture lumbar vertebrae belongs to the 2nd lumbar vertebrae) of human body, so as to Diagnosis and successive treatment in disease.

In certain embodiments, reference line correction module 230 can be corrected to the position of reference line.In some implementations In example, reference line determining module 220 can show the corresponding reference line of image to be spliced on stitching image.Due in splicing During the generation of image, the position relationship between image to be spliced can change so that corresponding reference line position occurs It is inaccurate.For example, the reference line position shown on the stitching image of the full vertebra of human body may beyond vertebra top/bottom latitude (as being higher than the 1st cervical vertebra with reference to line position), or the angle of reference line and vertebra and the scanning angle of reality are not consistent (such as vertebra The reference line direction of cross-section bit image is parallel with vertebra direction or close to parallel) etc..Reference line correction module 230 can pass through Manual mode or automated manner with reference to line position to being corrected.

In certain embodiments, user (such as doctor) can pass through input output assembly 760 or input/output (input output (I/O)) unit 850 inputs correction instruction.The correction instruction can include translation, the rotation to reference line Turn, increase, delete.Reference line correction module 230 can obtain correction instruction and reference line is corrected accordingly. In some embodiments, user (such as doctor) can preset reference based on network 120, computer 700 or mobile device 800 The rule of line correction is (such as the parallel spacing of reference line more than the specific thickness or reference line of 5mm or other threshold values or reference line Specific length etc.).The rule corrected based on the reference line, reference line correction module 230 can carry out phase to reference line automatically The correction answered.

According to some embodiments of the present application, shown in Fig. 3 is a kind of exemplary process diagram of reference line positioning.Flow 300 can be realized by one or more hardware, software, firmware etc. and/or its combination.In certain embodiments, flow 300 can be with Processing equipment (such as the processing equipment 130 shown in Fig. 1) and/or calculating as one or more operation image concatenation modules 210 Equipment (such as the computer shown in Fig. 7-8) is realized.

In the step 310, image mosaic module 210 can obtain at least two images to be spliced.The image to be spliced It can be scan image.Scan image can include but is not limited to CT images, MRI image or PET image.The scan image can To be two dimensional image or 3-D view.The scan image can include but is not limited to original image and/or processing figure Picture.The original image can refer to the image directly obtained according to scan data.The processing image can refer to original image Carry out the image of processing acquisition.Processing to original image can include but is not limited to image enhaucament, image reorganization, Three-dimensional Gravity Build, image filtering, Image Coding (for example, compressed encoding), image format conversion, image rendering, image scaling etc. therein one Plant or several combinations.

Image enhaucament can represent to increase the contrast in the entirely or partially region of image.For example, in the vertebra of human body In MRI image, the contrast of vertebrae and peripheral nerve or soft tissue can be increased, an imaging technician or doctor can be quick Easily identify the border on vertebra side.For another example for craniocerebral MRI image, can increase some focuses (epileptic focus) or The brain tissue in person's critical function area, facilitates doctor to judge the cut coverage of operation, is reduced while maximum magnitude removal of lesions Damage to normal cerebral tissue's especially brain tissue in critical function area.In certain embodiments, MRI image enhancing can include One or more of combination therein such as contrast enhancing, noise removal, background removal, edge sharpening, filtering and wavelet transformation.

Image reorganization can represent to generate the image of any aspect according to existing MRI scan image.Three-dimensional reconstruction can be with table Show and three-dimensional reconstruction is carried out according to the two-dimensional scan image of liver, obtain the 3-D view of liver.For example, image mosaic module 210 Can be by the MRI image of vertebra from medical science digital image transmission form (DICOM (Digital Imaging and Communications in Medicine) form) it is converted into visual post process grid format (Visualization Toolkit Image, VTI form).

Image Coding may also be referred to as compression of images, can represent meeting certain picture quality (for example, signal to noise ratio) Under the conditions of, the information included in image or image is represented with less bit.Image rendering can be represented high-dimensional information Become low dimensional information, for example, three-dimensional information is become two-dimensional signal.

In certain embodiments, image mosaic module 210 can obtain scan data from imaging device 110, according to sweeping Retouch data reconstruction and obtain original image.The method that image mosaic wooden unit 210 is rebuild to MRI can include what is filled based on K spaces MRI method for reconstructing or the MRI method for reconstructing based on image area.It can be included based on the MRI method for reconstructing that K spaces are filled in half Fu Leaf imaging method, SMASH imaging methods, or AUTO-SMASH imaging methods etc..MRI method for reconstructing based on image area can utilize MRI The priori of certain in image carries out the reconstruction of MRI image, and this method can reduce the data scanning time, it is calm accelerate MRI into The process of picture.MRI method for reconstructing based on image area can according to sensitivity information different between coil come reconstruction image or Can be according to the openness carry out image reconstruction of MRI image (for example, utilizing compression sensing method to carry out the reconstruction of MRI image).

In step 320, image mosaic module 210 can obtain the spatial positional information of the image to be spliced.It is described Spatial positional information can include the sky under the three-dimensional coordinate information of image to be spliced, two-dimensional coordinate information, particular image format Between positional information etc..The spatial positional information of the image to be spliced can be obtained by the view data of arbitrary image form.Institute DICOM format, VTI forms etc. can be included but is not limited to by stating picture format.Described image data can include one or more Logical level.Described image data text physically may include file header and data set group.File header may include preamble (example Such as 128 bytes) and prefix (such as 4 bytes) composition.Preamble can be by fixed structure or without fixed structure.For example, in some realities Apply in example, byte is set to 00H during no content, and prefix is DICM character strings, for identifying DICOM file.Data set is saved All necessary informations of DICOM file are operated, are made up of multiple data elements.Each data element includes Tag, Vale Tetra- fields of Representation, Value Length, Value Field, save the element information form definition with Content.

The spatial positional information for the image to be spliced that image mosaic module 210 is obtained can include the position of image to be spliced Information (such as the image position (patient) in DICOM), the directional information of image to be spliced are (such as in DICOM Image orientation (patient)) etc. other information related to locus.The position letter of the image to be spliced Breath can include the related coordinate information of image to be spliced (such as picture of the first row first row (such as upper left corner) on image to be spliced The coordinate of element).The coordinate information can be included in one or more of arbitrary system coordinate.In certain embodiments, The positional information of the image to be spliced can include one or more of preferred coordinates system coordinate.For example, image to be spliced Spatial positional information can be expressed as O1 (x1, y1, z1).

The directional information of the image to be spliced can include the information in any image to be spliced with directional correlation.Example Such as, the directional information of the image to be spliced may include the one or more parts of image to be spliced (for example the first row pixel and/or First row pixel) corresponding directional information (one or more directions, represents that vectorial, the one or more direction in direction is corresponding Trigonometric function value etc.).For example, the directional information of image to be spliced can include image to be spliced on the first row direction vector and The direction vector of first row.For another example the directional information of image to be spliced can be expressed as the normal vector of image to be spliced.One In a little embodiments, the normal vector is the direction vector of the first row and the direction vector multiplication cross of first row.

In certain embodiments, image to be spliced can be expressed as such as equation (1)：

(x-x1) nx+ (y-y1) ny+ (z-z1) nz=0 (1)

Wherein, (x1, y1, z1) is the positional information of image to be spliced；And the side that (nx, ny, nz) is image to be spliced To information (such as normal vector).

In a step 330, image mosaic module 210 can determine a spliced map based at least two images to be spliced Picture.The image to be spliced can include the first image to be spliced, second image to be spliced etc..Image mosaic module 210 can be with By treat stitching image carry out translation, rotation, scaling or shearing etc. operation determine stitching image.For example, image is spelled Connection module 210 can set any one image to be spliced as fixed reference picture picture (the such as first image to be spliced), and other are treated Stitching image is translated or assisted and/or rotation process according to the fixed reference picture picture.

In 330, image mosaic module 210 can obtain the spatial positional information of stitching image.The space bit confidence Breath can include the space bit confidence under three-dimensional coordinate information, two-dimensional coordinate information or the particular image format of stitching image Breath etc..

The spatial positional information for the stitching image that image mosaic module 210 is obtained can include the positional information of stitching image (such as the image position patient in DICOM), directional information (such as image in DICOM of stitching image Orientation patient) etc..The positional information of the stitching image can include the related coordinate information of stitching image (such as the coordinate of the pixel of the first row first row (such as upper left corner) on stitching image).The coordinate information can be included in arbitrarily One or more of coordinate system coordinate.In certain embodiments, the positional information of the stitching image can include specific sit One or more of mark system coordinate.For example, the spatial positional information of image to be spliced can be expressed as O2 (x2, y2, z2).

The directional information of stitching image can include the information in any stitching image with directional correlation.For example, described spell The directional information of map interlinking picture may include that the one or more parts of stitching image (such as the first row pixel and/or first row pixel) are right The directional information (one or more directions, the corresponding trigonometric function value in vectorial, one or more direction for representing direction etc.) answered. For example, the directional information of stitching image can include the direction vector and the direction vector of first row of the first row on stitching image. For another example the directional information of stitching image can be expressed as the normal vector of image to be spliced.In certain embodiments, the normal direction Amount be the first row direction vector and first row direction vector multiplication cross result.

In certain embodiments, the spatial positional information of stitching image can be represented as such as equation (2)：

(x-x2) Nx+ (y-y2) Ny+ (z-z2) nz=0 (2)

Wherein, (x2, y2, z2) is the positional information of stitching image；And (Nx, Ny, Nz) believes for the direction of stitching image Breath.

In certain embodiments, image mosaic module 210 can determine stitching image based on stitching algorithm.The splicing is calculated Method can include the stitching algorithm based on the related stitching algorithm in region and feature based.The splicing related based on region is calculated Method can the gray value based on image to be spliced, the difference of its gray value is calculated using least square method；Based on the gray value Obtain the similarity degree of diversity judgement image overlapping region to be spliced；And determined based on the similarity degree between image to be spliced The scope of overlapping region and position, finally produce stitching image.The stitching algorithm of the feature based can include feature extraction And feature registration.The algorithm of characteristic matching can include crosscorrelation, range conversion, dynamic programming, structure matching, chain code correlation Scheduling algorithm etc..

In step 340, reference line determining module 220 can determine image to be spliced (the such as first image to be spliced and Second image to be spliced) between splicing relation.The splicing relation can be used to indicate that it is determined that being treated during stitching image Position relationship between stitching image.Position relationship between the image to be spliced can include translation, rotation, scaling or One kind in the operation such as shearing, it is several, or its combination.In certain embodiments, in the CT scan to the full vertebra of human body, image The first image to be spliced may be selected as fixed reference picture picture in concatenation module 210, and based on stitching algorithm stitching image.For example, Image mosaic module 210 can adjust position of second image to be spliced relative to fixed reference picture picture (the first image to be spliced) Put, the size of the second image to be spliced, direction etc. can also be adjusted.It is in course of adjustment, reference line determining module 220 can be true It is fixed and/or store position relationship between image to be spliced (between the such as first image to be spliced and the second image to be spliced).Institute A registering matrix can be expressed as by stating the position relationship between image to be spliced.In certain embodiments, registering matrix can be with It is 4*4 matrix：The first three rows of wherein registering matrix can represent affine transformation, wrap Include rotation, translation and shear.The fourth line of matrix can represent a projective transformation.

In 350, reference line determining module 220 can be based on the relative space position information between image to be spliced and splicing Relation determines reference line.In certain embodiments, reference line determining module 220 can be directly based upon the space of the first stitching image The splicing relation of positional information, the spatial positional information of the second stitching image and the second stitching image and the first stitching image is determined Reference line.For example, reference line determining module 220 can the positional information based on the first stitching image, the position of the second stitching image Registering matrix (such as 4*4 matrix) between information and the first image to be spliced and the second image to be spliced is directly spelled first Reference line is determined on map interlinking picture, the reference line may correspond to the second stitching image.

In certain embodiments, reference line determining module 220 can be based on image to be spliced spatial positional information and spelling The spatial positional information of map interlinking picture, determines initial reference line on stitching image；And based on the initial reference line and splicing Relation determines correction reference line.For example, in the CT images of the full vertebra of human body, reference line determining module 220 can be based on waiting to spell The three-dimensional coordinate (three-dimensional coordinate of such as top left corner pixel) and the spatial positional information of stitching image of map interlinking picture determine initial reference Line；And based on the registering matrix between the initial reference line and image to be spliced further determine that correction with reference to line position or Person's displaying correction reference line.

According to some embodiments of the present application, shown in Fig. 4 is to determine reference based on spatial relation and splicing relation A kind of exemplary process diagram of line.Flow 400 can be realized by one or more hardware, software, firmware etc. and/or its combination. In certain embodiments, flow 400 can by one or more operation image concatenation modules 210 processing equipment (such as Fig. 1 institutes The processing equipment 130 shown) and/or computing device (such as the computer shown in Fig. 7-8) realize.

In 410, reference line determining module 220 can determine the intersection point between image to be spliced and stitching image.One In a little embodiments, spatial information and the spatial information of stitching image that reference line determining module 220 can be based on image to be spliced, Determine the intersection point between image to be spliced and stitching image.Intersection point between the image to be spliced and stitching image can pass through The residing plane or plane equation of stitching image and image to be spliced is determined.In certain embodiments, reference line determines mould Block 220 can be based on stitching image spatial positional information (such as positional information or directional information) determine the residing flat of stitching image Face or plane equation.Reference line determining module 220 can also be based on image to be spliced spatial positional information (such as positional information Or directional information) determine the residing plane or plane equation of image to be spliced.The plane equation of the stitching image and/or treat The plane equation of stitching image can include intercept form plane equation, point French plane equation, general expression plane equation, normal form One or more of combinations of plane equation etc..For example, reference line determining module 220 can based on image to be spliced position believe Breath obtains the pixel coordinate O1 (x1, y1, z1) in the image upper left corner to be spliced；Directional information based on image to be spliced, which is obtained, to be waited to spell The normal vector (dx, dy, dz) of map interlinking picture.Reference line determining module 220 can also and based on pixel coordinate O1 (x1, y1, z1) and Normal vector (dx, dy, dz) determines the point French plane equation of image to be spliced.

In certain embodiments, the intersection point between image and stitching image to be spliced can be intersected or two based on face face The solution of plane equation is determined.In certain embodiments, reference line correction module 230 can calculate the plane side of image to be spliced Solution between journey and the plane equation of stitching image, and then determine the intersection point between image to be spliced and stitching image.At some In implementation, if not having intersection point between image waiting and stitching image, reference line correction module 230 can determine image to be spliced Parallel position, or otherwise processed are between stitching image.

In 420, reference line correction module 230 can be based on the splicing relation adjustment intersection point position between image to be spliced Put.In some implementations, reference line correction module 230 can be adjusted based on the registering relation antinode between image to be spliced It is whole, the position of intersecting point after being adjusted.For example, the spelling of the first image (fixed reference picture picture) to be spliced and the second image to be spliced Connect relation and represent or be stored in database 140 with registering matrix (such as 4*4 matrixes).Reference line correction module 230 can be calculated Solution between the corresponding plane equation of second image to be spliced and the corresponding plane equation of the first stitching image, and then determine first Intersection point between stitching image and the second stitching image.Adjusted based on registering matrix (such as 4*4 matrixes) on the first stitching image The intersection point of second image to be spliced.If for another example in splicing, the second image to be spliced is relative to the first image to be spliced (fixed reference picture picture) is shifted 5 units to X-axis positive direction, and reference line correction module 230 is determining stitching image and the , can be the intersection point to negative direction 5 units of translation after the intersection point of two figures to be spliced, the position of intersecting point after being adjusted.

In 430, reference line correction module 230 can determine to refer to line position based on the position of intersecting point after adjustment.One In a little embodiments, the intersection point of image and stitching image to be spliced can be multiple (such as 2).For example, reference line correction module Four sides of the plane of 230 planes and stitching image based on image to be spliced, can be relative at two of the plane of stitching image 2 intersection points are determined on parallel side.The position of 2 intersection points can be adjusted in 420, and then be determined after 2 adjustment Position of intersecting point.Intersection point after 2 adjustment based on stitching image, reference line correction module 230 is handed over after can connecting 2 adjustment Point, produces a line segment.The line segment can be used as corresponding reference line of the image to be spliced on stitching image.

In certain embodiments, the intersection point of image and stitching image to be spliced can be two or more points.Example Such as, the plane equation of plane equation and stitching image of the reference line correction module 230 based on image to be spliced, can be in spliced map As 100 intersection points of upper determination；And 100 intersection points are adjusted behind position, reference line correction module 230 can be with member-retaining portion Intersection point after adjustment can also delete the intersection point after partial adjustment on stitching image.Intersection point after the argument section adjustment Can be based on length or region that reference line is shown on stitching image etc..If for example, the length of reference line is by system 100 Be set to 5cm, reference line correction module 230 can delete more than 5cm outside intersection point, and the intersection point based on reservation determines one Line segment is used as reference line.

According to some embodiments of the present application, shown in Fig. 5 is a kind of exemplary process diagram of reference line correction.Flow 500 can be realized by one or more hardware, software, firmware etc. and/or its combination.In certain embodiments, flow 500 can be with Processing equipment (such as the processing equipment 130 shown in Fig. 1) and/or calculating as one or more operation image concatenation modules 210 Equipment (such as the computer shown in Fig. 7-8) is realized.

In 510, reference line correction module 230 can obtain at least two images to be spliced, and based on described at least two Individual image to be spliced, determines a stitching image.The image to be spliced can be scan image.Scan image can include but It is not limited to CT images, MRI image or PET image.The scan image can be two dimensional image or 3-D view.

In certain embodiments, image to be spliced can include multiple images section, and each image sections can include at least one Individual image to be spliced, and the image to be spliced has same or similar spatial positional information.For example, imaging device 110, based on 3 full vertebras of different angle shot human bodies, can obtain 3 image sections：Image sections A (such as cervical vertebra image sections), image Section B (such as thoracic vertebrae image sections) and image sections C (such as lumbar vertebrae image sections)；And image sections A, image sections B and image sections C include respectively 50,40 and 55 images to be spliced.

During stitching image is produced, image mosaic module 210 can determine that splicing is closed in units of described image section System.For example, the full vertebra CT images of human body can include 3 image sections：Image sections A, image sections B and image sections C；Each image sections Including at least an image to be spliced, and image to be spliced in same image sections has similar spatial positional information, For example with identical in-plane or scan mode, (such as image sections A can be that cross-section bit scan, image sections B can be arrows Shape bit scan or image sections C can be coronal scans)；And in splicing is carried out, image mosaic module 210 can be with Image sections A carries out the adjustment for such as translating, rotating, scale or shearing to image sections B or image sections C to fix reference image sections, And then obtain stitching image.

In 520, spatial relation that reference line determining module 220 can be based on image to be spliced and stitching image Spatial relation, it is determined that or/and the first reference line of display.First reference line can be initial reference line or Based on the reference line after initial reference line and splicing relation adjustment.In certain embodiments, reference line determining module 220 can be straight Connect the spatial relation based on image to be spliced and stitching image spatial relation determine image to be spliced plane and The plane of stitching image, and determine the intersection point between the plane；Reference line determining module 220 is directly determined based on the intersection point One line segment or straight line, are used as the first reference line.For example, reference line determining module 220 can be based on the 4th cross-section bit image of thoracic vertebrae Plane and the full vertebra stitching image of human body plane, determine the 4th cross-section bit image of thoracic vertebrae on the full vertebra stitching image of human body The first reference line, first reference line is vertical with the full vertebra direction of human body or connect by the 4th thoracic vertebrae on stitching image It is near vertical.

In certain embodiments, reference line determining module 220 can be based on image to be spliced spatial relation and spelling The spatial relation of map interlinking picture determines the plane of image to be spliced and the plane of stitching image, and determines between the plane Intersection point, and initial reference line.Reference line determining module 220 can adjust the position of the initial reference line based on splicing relation, And the position of the first reference line is determined based on the initial reference line after adjustment.For example, reference line determining module 220 can be based on matching somebody with somebody The initial reference line that quasi- matrix (such as 4*4 matrixes) is treated between the plane of stitching image and the plane of stitching image is adjusted, Produce the initial reference line after adjustment；And the first reference line is determined based on the initial reference line after the adjustment.

In 530, reference line correction module 230 can be corrected behaviour based on an object function to the first reference line Make, and obtain a regulation matrix.The correct operation can include to the translation of the first reference line, rotation, scaling or shear One or more combination in.The regulation matrix can be used to indicate that the correct operation to the first reference line.In some realities Apply in example, the regulation matrix can be stored in database 140 or network 120.In certain embodiments, in three dimensions In, if the plane of image to be spliced and the plane of stitching image are represented with homogeneous coordinates, the corresponding regulation square of the first reference line Battle array can include 4*4 matrix：WhereinIt can be used for Represent rotation, translation or shear, and (a₄₁ a₄₂ a₄₃ a₄₄) it can be used to indicate that projective transformation.

In 540, reference line correction module 230 can carry out school based on the regulation matrix to first reference line Just, and obtain the second reference line.In certain embodiments, reference line correction module 230 can be obtained based on regulation matrix the Two reference lines, second reference line can be expressed as equation (3)：

Y '=mY (3)

Wherein, Y ' represents the second reference line；Y represents the first reference line without corrected operation；And m is represented three Regulation matrix in dimension space, is a 4*4 matrix.

In certain embodiments, the second reference line can show on computer 700 or be set by network 120 in movement Standby 800 display.For example, in the MRI scan image of the full vertebra of human body, processing equipment 130 can be by network 120 spliced map Picture and the second reference line are shown on the mobile device 800 of user (such as surgeon) (such as smart mobile phone) so that Yong Huneng Enough carry out remote diagnosis and treatment；For another example in the skull CT scanning image of human body, processing equipment 130 can pass through network 120 stitching images and the second reference line are sent to high in the clouds；And user (such as patient) can obtain the splicing by high in the clouds Image and the second reference line, and be shown on smart mobile phone so that user can remotely receive image information.

According to some embodiments of the present application, shown in Fig. 6 is a kind of illustrative diagram of reference line regulation.Flow 600 can be realized by one or more hardware, software, firmware etc. and/or its combination.In certain embodiments, flow 600 can be with Processing equipment (such as the processing equipment 130 shown in Fig. 1) and/or calculating as one or more operation image concatenation modules 210 Equipment (such as the computer shown in Fig. 7-8) is realized.

In 610, processing equipment 130 can obtain at least two MRI images, and (such as the first MRI image, the 2nd MRI are schemed As etc.).At least two MRI image can be related to arbitrary scan object.For example, at least two MRI image can be wrapped The MRI scan image of the full vertebra containing human body, MRI scan image of head etc..In certain embodiments, the MRI of the full vertebra of human body Scan image can include at least three image sections：Cervical vertebra image sections, thoracic vertebrae image sections and lumbar vertebrae image sections.Described image section bag Containing one or more MRI images to be spliced, and each, the MRI image in image sections has phase in follow-up splicing Same or similar splicing relation.In certain embodiments, at least two MRI image can be different time scanning generation 's.For example, at least two MRI image can be included in preoperative vertebra MRI image and postoperative vertebra MRI figures Picture.

In 620, processing equipment 130 can obtain the spatial positional information of each MRI image.The spatial positional information It can include treating the space bit confidence under three-dimensional coordinate information, two-dimensional coordinate information or the particular image format of MRI image Breath.In certain embodiments, the spatial positional information for the stitching image that image mosaic module 210 is obtained can be included in position letter Cease (such as the image position patient under DICOM format), directional information (such as image under DICOM format Orientation patient) etc..In certain embodiments, processing equipment 130 can obtain the locus of different images section Information.For example, in the MRI scan image of the full vertebra of human body, processing equipment 130 can obtain cervical vertebra image sections, thoracic vertebrae image sections With the spatial positional information of lumbar vertebrae image sections, to determine each section after splicing relative to some fixation with reference to section position become Change.

In 630, processing equipment 130 can be based on stitching algorithm, determine a stitching image.The stitching algorithm can be with Including the stitching algorithm based on the related stitching algorithm in region and feature based.The stitching algorithm related based on region can be with Based on the gray value of image to be spliced, the difference of its gray value is calculated using least square method；Difference is obtained based on the gray value Judge the similarity degree of image overlapping region to be spliced；And overlay region between image to be spliced is determined based on the similarity degree The scope in domain and position, finally produce stitching image.The stitching algorithm of the feature based can include feature extraction and feature Registration.The algorithm of characteristic matching can include crosscorrelation, range conversion, dynamic programming, structure matching, the related scheduling algorithm of chain code Deng.For example, using based on the related stitching algorithm in region, processing equipment 130 can utilize at least two MRI images or multiple figures As section produces stitching image.The stitching image can be the coronal bit image of the full vertebra of human body.

In 640, processing equipment 130 can be based on the stitching image, determine the registering matrix between MRI image. In some embodiments, processing equipment 130 can be fixed reference with some MRI image or image sections (such as cervical vertebra image sections), Determine the splicing relation of other MRI images or image sections (such as thoracic vertebrae image sections or lumbar vertebrae image sections).For example, processing equipment 130 can be using the first MRI image as fixed reference.The splicing relation can be expressed as registering matrix.In certain embodiments, exist In the splicing of three dimensions, processing equipment 130 can determine the plane before the splicing of each MRI image and spliced flat Face, and the position relationship based on each MRI image relative to fixed reference picture picture, it is determined that partially registering matrix.In some embodiments In, it is corresponding after concatenation is carried out if the plane of MRI image is indicated using homogeneous coordinates in three dimensions Registering matrix can be expressed as 4*4 matrix, such as equation (4)：

P '=nP (3)

Wherein, P ' represents the MRI image after concatenation；P represents MRI image not Jing Guo concatenation；With It is a 4*4 matrix and n represents registering matrix in three dimensions.

In 650, processing equipment 130 can be based on the registering matrix and spatial relation, determine the first reference line. Processing equipment 130 can determine the first reference line of MRI image on full vertebra stitching image.In certain embodiments, handle Equipment 130 can be based on MRI image and full vertebra stitching image, determine the plane equation and full vertebra stitching image of MRI image Plane equation；And plane equation and the plane equation of full vertebra stitching image based on MRI image, processing equipment 130 can To determine the intersection point between MRI image and full vertebra stitching image.For example, based on the second MRI image in cervical vertebra image sections Spatial positional information (such as positional information and/or directional information), determines the first plane equation；Sky based on full vertebra stitching image Between positional information, determine the second plane equation；And based on the first plane equation and the second plane equation, processing equipment 130 can To determine the intersection point between the second MRI image in cervical vertebra image sections and full vertebra stitching image.In certain embodiments, it is based on The intersection point, processing equipment 130 directly can determine the first reference line on full vertebra stitching image.In certain embodiments, Based on the intersection point and registering matrix, processing equipment 130 can further adjust position of intersecting point, and determine the first intersection.

In 660, processing equipment 130 can obtain the regulation matrix of the first reference line.In certain embodiments, user (such as doctor) can manually be adjusted to first with reference to line position.The manual regulation can include putting down the first reference line Move, rotate, scale or shear etc..User (such as doctor) can input hand by the input output assembly 760 in computer 700 The instruction of dynamic regulation.Processing equipment 130 can receive the instruction adjusted manually, realize to first with reference to the manual of line position Regulation.In certain embodiments, the manual regulation can also include to the deletion of the first reference line, increase, form is set etc. One or more of combinations.For example, doctor can input the instruction adjusted manually by input output assembly 760, to delete One or more first reference lines.In certain embodiments, processing equipment 130 can the setting based on user be automatic joins to first Line position is examined to be adjusted.For example, the object function that processing equipment 130 can be set based on a user is on stitching image First is adjusted with reference to line position.In certain embodiments, processing equipment 130 can be based on to the first tune with reference to line position Section, determines a regulation matrix.The regulation matrix can be 4*4 matrix.The matrix of the 4*4 can be stored in database 140 or stored beyond the clouds by network 120.

In 670, processing equipment 130 can be based on the regulation matrix, update the position of the first reference line, produce second Reference line.In certain embodiments, user can choose whether to carry out reference line renewal.For example user can by input/it is defeated Go out the selection of component 760 and update the first reference line, processing equipment 130 can send regulation matrix to computer 700 or movement and set On standby 800；And based on the regulation matrix, processing equipment 130 can update first and refer to line position, produce second and refer to Line, second reference line can more accurately show the position relationship on MRI image and full vertebra stitching image, be easy to use Family (such as doctor) carries out the diagnosis and treatment of disease.

According to some embodiments of the present application, Fig. 7 is a kind of framework of computer equipment, and this computer equipment can be real Reality applies the particular system disclosed in the application.Particular system in the present embodiment explains one using functional block diagram and includes use The hardware platform at family interface.This computer can be the computer of a general purpose, or a meter for having a specific purpose Calculation machine.Two kinds of computers can realize the particular system in the present embodiment.Computer 700 is provided in which can implement current description Any component of information required for on-demand service.For example：Processing equipment 130 can be passed through by the computer of such as computer 700 Its hardware device, software program, firmware and their combination are realized.For convenience's sake, a meter is only depicted in Fig. 7 Calculation machine, but the correlation computer function of the information required for offer on-demand service described by the present embodiment be can be to be distributed Mode, implemented by one group of similar platform, the processing load of decentralized system.

Computer 700 can include COM1 750, and what is be attached thereto is the network for realizing data communication.Computer 700 Central processing system (CPU) unit, which can also be included, is used for execute program instructions, is made up of one or more processors.Show The computer platform of example includes internal communication bus 710, various forms of program storage units and data storage element, Such as hard disk 770, read-only storage (ROM) 730, random access memory (RAM) 740 can be configured as computer disposal And/or the various data files that communication is used, and the possible programmed instruction performed by CPU.Computer 700 can also include One input output assembly 760, supports the input/output data between computer and other assemblies (such as user interface 780) Stream.Computer 700 can also receive program and data by communication network.

According to some embodiments of the present application, Fig. 8 is a kind of structure of mobile device, and the mobile device can realize implementation The particular system disclosed in the application.In this example, for showing that the user equipment with interaction locations relevant information is a shifting Dynamic equipment 800, can include but is not limited to, smart mobile phone, tablet personal computer, music player, portable game, global positioning system System (Global Positioning System, GPS) receiver, wearable computing devices (such as glasses, wrist-watch), Huo Zheqi His form.Mobile device 800 in this example can include one or more central processing units (CPUs) 840, graphics processor (Graphical Processing Units, GPUs) 830, display unit 820, internal memory 860, antenna 810 are (for example, channel radio Believe unit), memory cell 890, and one or more input/output (input output (I/O)) unit 850.It is any its His suitable component, can include but is not limited to system bus or controller (not shown on figure), it is also possible to be included in movement In equipment 800.As shown in figure 8, a Mobile operating system 870, such as iOS, Android, Windows Phone, Yi Jiyi Individual or multiple applications 880 can be loaded into internal memory 860 from memory cell 890, and performed by central processing unit 840.Using 880 can include a browser or other Mobile solutions for being adapted to reference line correlation is received and handled on mobile device 800. User can be obtained and provided to processing equipment on the interaction of reference line relevant information by input/output equipment 850 130, and/or system 100 other assemblies, for example：Pass through network 120.

In order to realize different modules, unit and their function described in disclosure before, computer is hard Part platform is used as the hardware platform of one or more elements described above (for example：Processing equipment 130, and/or system 100 other assemblies).Hardware elements, operating system and the program language of this kind of computer be in nature it is common, can be with It is assumed that those skilled in the art are familiar with these technologies enough, can be using needed for technology described herein offer on-demand service The information wanted.One computer comprising user interface element can be used as personal computer (personal computer (PC)) or other kinds of work station or terminal device, it can also be used after appropriately programmed as server.It is considered that General operation of the those skilled in the art to such structure, program and this kind of computer equipment is all known, therefore institute There is accompanying drawing also all not need extra explanation.

According to some embodiments of the present application, shown in Fig. 9 A to 9C is that reference line is accurately positioned a kind of exemplary of result Schematic diagram.Fig. 9 A are the cross-section bit images of head；Fig. 9 B are the sagittal images of head；And Fig. 9 C are the coronal bitmaps of head Picture.In figure 9 a, 910 be cross-section bit image and head of the processing equipment 130 based on head coronal bit image determine reference Line, the reference line illustrates intersection of the coronal bit image (as shown in Figure 9 C) of head on the cross-section bit image of head. In Fig. 9 B, 920 be the reference line of the coronal bit image determination of sagittal image and head of the processing equipment 130 based on head, institute State reference line and illustrate intersection of the coronal bit image (as shown in Figure 9 C) of head on the sagittal image of head.

It is realized foregoing has outlined the different aspect of the method for the information required for offer on-demand service and/or by program The method of his step.Program part in technology be considered in the form of executable code and/or related data and " product " or " product " existed, is participated in or realizes by computer-readable medium.Tangible, permanent storage Jie Matter can include the internal memory or memory used in the module of any computer, processor or similar devices or correlation.For example it is each Store function can be provided for software by planting semiconductor memory, tape drive, disc driver or similar any time Equipment.

All softwares or a part therein there may come a time when to be communicated by network, such as internet or other communication networks Network.Software from a computer equipment or processor can be loaded into another by such communication.For example：From on-demand service system A management server or host computer be loaded onto the hardware platform of a computer environment, or other realize the meter of system Calculation machine environment, or the system to providing the related similar functions of information required for on-demand service.Therefore, it is another to transmit The medium of software element can also be used as the physical connection between local devices, such as light wave, electric wave, electromagnetic wave, pass through Cable, optical cable or air, which are realized, to be propagated.For the physical medium such as similar devices such as cable, wireless connection or optical cable of carrier wave, It can also be considered as the medium for carrying software.Usage herein is unless limited tangible " storage " medium, and other are represented The term of computer or machine " computer-readable recording medium " all represents the medium participated in during any instruction of computing device.

Therefore, a computer-readable medium may have diversified forms, include but is not limited to, tangible storage medium, Carrier media or physical transmission medium.Stable storage medium can include：CD or disk, and other computers or similar Used in equipment, the storage system of the system component described in figure can be realized.Unstable storage medium includes dynamic The main memory of internal memory, such as computer platform.Tangible transmission medium includes coaxial cable, copper cable and optical fiber, including meter The circuit of calculation machine internal system formation bus.Carrier wave transmission media can transmit electric signal, electromagnetic signal, acoustic signals or light wave Signal, these signals can be as produced by radio frequency or the method for infrared data communication.Common computer-readable Jie Matter includes hard disk, floppy disk, tape, any other magnetic medium；CD-ROM, DVD, DVD-ROM, any other optical medium；Wear Kong Ka, any other include the physical storage medium of small hole pattern；RAM, PROM, EPROM, FLASH-EPROM, any other is deposited Reservoir piece or tape；Transmit the carrier wave of data or instruction, cable or transmit carrier wave attachment means, any other can be using meter Calculate the machine-readable program code taken and/or data.In the form of these computer-readable mediums, there can be many kinds to appear in processing Device is among execute instruction, the process of the one or more results of transmission.

It will be appreciated by those skilled in the art that a variety of variations and modifications can occur in content disclosed herein.For example, Different system component described above is all realized by hardware device, but may also only pass through the solution party of software Case is achieved.For example：The installation system on existing server.In addition, the offer of positional information described herein may It is to be achieved by a firmware, the combination of firmware/software, the combination of firmware/hardware or the combination of hardware/firmware/software.

Although the present invention is disclosed as above with preferred embodiment, it is not for limiting the present invention, any this area Technical staff without departing from the spirit and scope of the present invention, may be by the methods and techniques content of the disclosure above to this hair Bright technical scheme makes possible variation and modification, therefore, every content without departing from technical solution of the present invention, according to the present invention Any simple modifications, equivalents, and modifications made to above example of technical spirit, belong to technical solution of the present invention Protection domain.

Claims

1. a kind of reference line determines method, including：

Obtain at least two images to be spliced, at least two image correspondence to be spliced, first spatial positional information；

Based on described at least two images to be spliced, a stitching image, the stitching image correspondence second space position are determined Information；

Determine the splicing relation between described at least two images to be spliced；And

Based on the splicing relation between described at least two images to be spliced, first spatial positional information and second space Positional information, determine at least one reference line.

2. reference line as claimed in claim 1 determines method, it is characterised in that：First spatial positional information includes described At least one of positional information and directional information of at least two images to be spliced；The second space positional information includes institute State at least one of positional information and directional information of stitching image.

3. reference line as claimed in claim 1 determines method, it is characterised in that：Determine described at least two images to be spliced it Between splicing relation include carrying out the image to be spliced at least one and following operate：Translation；Rotation；Scaling；And shearing.

4. reference line as claimed in claim 1 determines method, it is characterised in that：Between at least two image to be spliced Splicing relation includes a registering matrix.

5. reference line as claimed in claim 1 determines method, it is characterised in that：It is described to be based on described at least two figures to be spliced Splicing relation, first spatial positional information and second space positional information as between determine at least one reference line bag Include：

Based on first spatial positional information and the second space positional information, described at least two images to be spliced are determined Intersection point between the stitching image；

Based on the splicing relation between described at least two images to be spliced, at least two image to be spliced is adjusted and described Intersection point between stitching image；And

Based on the intersection point between described at least two images to be spliced and the stitching image after adjustment, it is determined that described at least one Bar reference line.

6. reference line as claimed in claim 5 determines method, it is characterised in that：It is described to be based on first spatial positional information With the second space positional information, determine that the intersection point between the image to be spliced and the stitching image includes：

The plane of plane and the stitching image based on described at least two images to be spliced, determines the intersection point.

7. reference line as claimed in claim 1 determines method, it is characterised in that：At least one reference line of the determination includes：

Based on first spatial positional information and the second space position relationship, the first reference line is determined；

At least one in the translation and rotation process of first reference line is carried out based on an object function, a tune is obtained Save matrix；And

The second reference line of acquisition is corrected to first reference line based on the regulation matrix.

8. a kind of reference line determines system, including：The memory of processor and the computer instruction that is stored with, the processor is held The step in claim 1-7 any one methods describeds is realized during the row computer instruction.

9. a kind of reference line determines system, including：

A kind of computer-readable storage medium, is configured as storing executable module, including：

Image mosaic module, is configured as：

Reference line determining module, is configured as based on the splicing relation between described at least two images to be spliced, described first Spatial positional information, and second space positional information, determine at least one reference line.

One processor, the processor is able to carry out the executable module of the computer-readable storage media storage.

10. reference line as claimed in claim 9 determines system, it is characterised in that：It is described to be spliced based on described at least two Splicing relation between image, determines that at least one reference line includes：

11. reference line as claimed in claim 10 determines system, it is characterised in that：It is described to be based on the first space bit confidence Breath and the second space positional information, determine the intersection point bag between described at least two images to be spliced and the stitching image Include：

12. reference line as claimed in claim 9 determines system, it is characterised in that：Wherein determine at least one reference line Including：