CN110033465A - A kind of real-time three-dimensional method for reconstructing applied to binocular endoscope medical image - Google Patents
A kind of real-time three-dimensional method for reconstructing applied to binocular endoscope medical image Download PDFInfo
- Publication number
- CN110033465A CN110033465A CN201910315817.0A CN201910315817A CN110033465A CN 110033465 A CN110033465 A CN 110033465A CN 201910315817 A CN201910315817 A CN 201910315817A CN 110033465 A CN110033465 A CN 110033465A
- Authority
- CN
- China
- Prior art keywords
- dimensional
- point
- image
- segmentation
- camera
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The present invention relates to a kind of real-time three-dimensional method for reconstructing applied to binocular endoscope medical image, this method passes through super-pixel segmentation first, and by color complex region, the boundary of different zones constitutes the three-dimensional framework of organ.Then according to epipolar line restriction principle, to the profile information that left side visual angle was photographed, corresponding polar curve on LOOK RIGHT is successively found.In order to obtain accurately matching double points, described in conjunction with operator with ORB feature, quick and precisely in positioning camera corresponding region intersection point position, pass through corresponding position relationship calculate boundary skeleton three-dimensional data.Coordinate relativeness is finally obtained using SFS method inside subregion after singulation, in conjunction with different colours gradient difference, the three-dimensional coordinate information between each region is extrapolated, obtains organ whole three-dimensional appearance coordinate in scene.The present invention solves the problems, such as the reconstruction of endoscope high-precision three-dimensional, and easy to operate compared with existing three-dimensional rebuilding method, high reliablity, operation risk is low, alleviates the pain of patient.
Description
Technical field
The present invention relates to a kind of real-time three-dimensional reconstructions applied to binocular endoscope medical image to obtain method, more specifically
It says, can be used in showing the accurate three-dimensional appearance of organ in endoscopic images the present invention relates to one kind and sit calibration method.
Background technique
Nowadays, there are about 17,500,000 people to die of heart disease every year in the whole world, has accounted for the 30% of whole death tolls.And China
Cardiovascular disease number has reached 2.9 hundred million people, and the death rate is much higher than other diseases, it is seen that its influence to people's health is big.
Traditional modus operandi needs to split thoracic cavity, and breast bone cutting has an immense impact on to the respiratory function of patient.Due to Intersternal incision
Tension is higher, so that the patient that constitution is poor, post-operative recovery are very difficult.
Minimally Invasive Surgery mode can not only reduce the risk of operation, more can be reduced the pain of Case treatment.Endoscope is micro-
The signal of interest acquisition mode of invasive procedures, doctor no longer need out chest, it is only necessary to make a call to 3 apertures on the wall of the chest, place thoracic cavity respectively
Mirror imaging device, ultrasonic scalpel and operation waste absorption plant.After operation, superficial wound can self-heal,
The wound and pain for greatly reducing patient, also shorten postoperative rehabilitation duration.
Accurate correspondence of the conventional two-dimensional endoscope since intuitive three dimensional local information can not be generated in surgical doctor brain
Relationship needs expertly carry out the operation of key position for a long time using it by the doctor trained.Existing two dimension endoscope
There are still following risks in use:
(1) two-dimentional endoscope lacks picture depth sense, cause doctor in the course of surgery to important anatomical structure and its
Opposite position generates visual erroneous judgement.And due to the missing of sense of depth, doctor can not accurately judge access site
The depth, easily because operation error causes accidental haemorrhage.
(2) two-dimentional endoscopic images distortion is larger, and human tissue structure is again sufficiently complex, thus will affect the smoothness of operation
Property and progress, extend operating time.
Three-dimensional endoscope bring human body three-dimensional spatial operation has started revolutionary variation in medical domain, uses three-dimensional
The Minimally Invasive Surgery of technology will become the mainstream of operation.The use of three-dimensional endoscope significantly reduces pain in the art of patient, also contracts
Short postoperative healing time.It, can be with if the three-dimensional data of key position can be obtained while obtaining 3-dimensional image
Operating time is greatly shortened, operation risk is reduced.The real-time three-dimensional method for reconstructing of endoscopic medical image proposed by the present invention, just
Put forward to solve the above-mentioned problems.
The present invention devises a kind of super-pixel segmentation and generates what three-dimensional framework was merged with SFS (Shape From Shading)
Quick three-dimensional reconstructing method, since the construction of internal organs is different, shade is distributed also not identical, single three-dimensional rebuilding method
It is difficult to obtain whole three-dimensional appearance information.The present invention passes through superpixel segmentation method first, and color complex region is divided
It cuts, the boundary of different zones constitutes the three-dimensional framework of organ.Then according to epipolar line restriction principle, the wheel that left side visual angle was photographed
Wide information successively finds corresponding polar curve on LOOK RIGHT.In order to obtain accurately left and right matching double points, need and ORB
(Oriented FAST and Rotated BRIEF) feature describes operator combination, quick and precisely positions corresponding in the camera of left and right
The position of region intersection point calculates the three-dimensional data of boundary skeleton by the corresponding position relationship of left and right camera.Finally, by three-dimensional bone
Frame is merged with SFS, and existing SFS algorithm reconstruction precision height relies on the source model of its imaging, only to the region of solid colour
With preferable 3-D effect, but for color different zones, there are relatively large deviations for three-dimensional data.The present invention is in super-pixel point
The coordinate relativeness in region is obtained using SFS method inside subregion after cutting, using the three-dimensional framework coordinate of generation as base
Standard successively extrapolates the three-dimensional coordinate information between each region in conjunction with the gradient difference of different colours, and then obtains in scene
Whole three-dimensional appearance coordinates of organ.Realize that real-time three-dimensional of the endoscope in surgical scene is rebuild, for doctor provide it is accurate with
Effective navigation information.
Summary of the invention
The present invention devises a kind of real-time three-dimensional reconstruction applied to endoscopic medical image and obtains method, and this method can answer
It, can for the three-dimensional coordinate of key position can be obtained while obtaining 3-dimensional image using in the operation of three-dimensional endoscope
To greatly shorten operating time, operation risk is reduced.
The hardware device that the endoscopic medical image real-time three-dimensional is rebuild includes:
LED cold light source one；
Optics hard stem endoscope two；
For establishing the calibrating platform of high-precision coordinate benchmark；
For acquiring two, the 1200*1600 industry color camera of image；
For precision controlling, Image Acquisition and the computer of data processing；
For placing the scanning platform of the light source and the camera.
The real-time three-dimensional reconstruction of endoscopic medical image designed by the present invention obtains method, specific steps are as follows:
Step 1: binocular camera being demarcated, if the coordinate system where left camera A is OaXaYaZaIf where right camera B
Coordinate system be ObXbYbZb, the spin matrix between two cameras is R, translation matrix T, the formula of calibration such as formula (1) institute
Show,r11-r33Spin matrix for the right camera B relative to the left camera A
Component, tx, ty, tzTranslation matrix component for the right camera B relative to the left camera A；
Step 2: the detecting lenses of binocular endoscope being protruded into patient body, to obtain organ surface image, and are peeped interior
The organ surface image of mirror acquisition carries out denoising and picture smooth treatment using median filtering method, protects the detailed information of image；
Step 3: being divided using SLIC (Simple Linear Iterative Clustering) superpixel segmentation method
The organ surface image that step 2 obtains is thin by organ surface image first by the color of adjacent pixel, brightness, textural characteristics
It is divided into multiple subregions, then all subregion image is transformed into CIE-Lab color space from RGB color, according to super-pixel
Number evenly distributes seed point in image, and three-dimensional colouring information and two-dimensional is utilized in the contiguous range of seed point
Spatial positional information calculates the pixel each searched to the distance of the seed point, to cluster to pixel, and passes through
The destination number in super-pixel segmentation region, controls the size of cut zone, is finally iterated optimization and enhancing connectivity, obtains
Organ surface image after segmentation, distance are calculated as shown in formula (2), dcRepresent color distance, dsRepresent space length, NsGeneration
Maximum space distance, N in table classcRepresent maximum color distance；
Step 4: the organ surface image after step 3 segmentation according to outer limit matching principle, collects point in left camera
Organ surface image segmentation boundary selected point after cutting collects in right camera and determines polar curve on the organ surface image after segmentation
And polar curve and partitioning boundary intersection point, obtain accurately left and right matching double points, then with ORB (Oriented FAST and
Rotated BRIEF) feature describe operator combination, position left and right camera in corresponding region intersection point position, choose intersection point in
It is match point with highest point is spent, if the slope of left camera selected point is ka, then the slope k of opposite match pointbIt can be by formula
(3) it obtains, kaFor the slope of certain point P in left camera acquired image skeleton, kbFor in right camera acquired image skeleton
With the slope of the P point corresponding points, this step is repeated, the three-dimensional coordinate of whole skeleton positions, structure can be obtained
At boundary skeleton three-dimensional coordinate, the three-dimensional coordinate information on each subregion boundary is recorded；
Step 5: the inside of subregion after each obtained super-pixel segmentation of step 4, first with SFS (Shape
From Shading) three-dimensional reconstruction is carried out, linear approach three-dimensional modeling is chosen, surface graded p and q is carried out using finite-difference method
Then discrete approximation carries out linearization process according to formula (4) on the direction height Z (x, y), finally obtain the coordinate of partial points
Variation relation；
In formula (4):
Step 6: step 5 changes in coordinates relationship obtained and the obtained boundary skeleton coordinate information of step 4 are carried out
Fusion calculates organ surface global three-dimensional coordinate；Operation finishes.
The beneficial effects of the present invention are: the super-pixel segmentation proposed through the invention is merged with the ORB under epipolar line restriction
The quick three-dimensional reconstructing method that the generation method and three-dimensional framework of three-dimensional framework are merged with SFS, can both reduce three-dimensional reconstruction
Feature Points Matching time, and the quantity and accuracy of Feature Points Matching can be improved.It is with the three-dimensional framework coordinate of generation
Benchmark both can successively extrapolate the three-dimensional coordinate information between each region, in turn in conjunction with the gradient difference of different colours
Whole three-dimensional appearance coordinates of organ in scene can be obtained.
Detailed description of the invention
Fig. 1: three-dimensional rebuilding method flow chart；
Comparison diagram before and after Fig. 2: SLIC superpixel segmentation method；
(a) original image before dividing；
(b) picture after dividing；
(c) three-dimensional framework picture in boundary is generated after dividing；
Each subregion image after Fig. 3: SFS processing；
(a) overall region image；
(b) borderline region image；
(c) non-boundary area image.
Specific embodiment
The real-time three-dimensional reconstruction of endoscopic medical image designed by the present invention obtains method, and the three-dimensional rebuilding method is such as
Shown in Fig. 1, concrete operations are as follows:
Building for binocular endoscope is completed, and binocular camera is demarcated, if left camera A coordinate system is OaXaYaZaIf
Coordinate system where right camera B is ObXbYbZb, the spin matrix between two cameras is R, translation matrix T, the formula of calibration
As shown in formula (5),r11-r33It is the right camera B relative to the left phase
The spin matrix component of machine A, tx, ty, tzTranslation matrix component for the right camera B relative to the left camera A.
The detecting lenses of binocular endoscope are protruded into patient body, to obtain organ surface image, and endoscope are acquired
Organ surface image denoised using median filtering, picture smooth treatment, protect the detailed information of image.
Divide organ surface image using SLIC superpixel segmentation method, before and after super-pixel segmentation comparison diagram such as Fig. 2 (a) and
Shown in Fig. 2 (b)；
(1) initialization seed point.According to the super-pixel number of setting, seed point is uniformly distributed in image.If picture
A total of N number of pixel, pre-segmentation are the super-pixel of K identical size, then the size of each super-pixel block is N/K, then phase
The distance of adjacent seed point is approximately A；
(2) seed point is reselected in n × n neighborhood of seed point.Specific method is to calculate all pixels in the neighborhood
The gradient value of point, moves on to the smallest place of neighborhood inside gradient for seed point；
It (3) is each pixel distribution class label in the neighborhood around each seed point.The search range of SLIC limits
For 2S × 2S, can be restrained with accelerating algorithm；
(4) distance metric, including color distance and space length.For each pixel searched, it is calculated separately
With the distance of the seed point.Shown in distance calculating method such as formula (6) and formula (7), wherein dcRepresent color distance, dsIt represents
Space length, NSIt is maximum space distance in class, is defined as NS=S=sqrt (N/K) is suitable for each cluster.Maximum color
Distance Nc, we take a fixed constant m to replace, and the value of m takes 10.Shown in final distance metric such as formula (8), due to each
Pixel can all be searched by multiple seed points, so each pixel has one at a distance from surrounding seed point, take minimum
It is worth cluster centre of the corresponding seed point as the pixel；
(5) iteration optimization.Picture can obtain more satisfactory effect after practice 10 iteration of discovery, so the number of iterations takes
10；
(6) enhance connectivity.The flaw occurred by above-mentioned iteration optimization: there are more connection situations, super-pixel size mistake
Small and single super-pixel is cut into multiple discontinuous super-pixel, can be solved by enhancing connectivity.
In binocular stereo vision measurement, Stereo matching is key technology, and epipolar-line constraint plays an important role.Observe scene
Two image center C of point0And C1Connection space three-dimensional point X is tracked to the straight line of image center, spatial point X can be found one
Point p in width image.On the contrary, the corresponding points q in another piece image can be found by point p.Along this straight line another
One image surface scans for, and straight line forms an imaginary straight line L in another image surface, this straight line is referred to as point p's
Polar curve.One endpoint of the polar curve is projected as boundary with the infinite point on raw observation line, another endpoint is in former camera
The heart is projected as boundary the 2nd image surface, is pole e.Two-dimensional image point p in one visual angle is mapped to by basis matrix F
On polar curve in another visual angle.If the slope of left camera selected point is ka, then the slope k of match point on the other sidebIt can be by public affairs
Formula (9) obtains.kaFor the slope of certain point P in left camera acquired image skeleton, kbFor in right camera acquired image with
The slope of P point corresponding points, r11-r33Spin matrix component for the right camera B relative to the left camera A, tx, ty,
tzTranslation matrix component for the right camera B relative to the left camera A.
In the unconspicuous organ surface image of processing feature, the algorithm combined is matched with ORB using epipolar line restriction, it can
Effectively to make up the matching error using ORB matching algorithm, matching precision is improved, epipolar line restriction and ORB can be effectively played
With the advantage for combining algorithm.It by ORB Feature Extraction Feature point and is matched and is screened, extract a certain region in left camera
Intersection position, by its available position EP point in another camera of camera calibration parameter, calculating is intersected with the polar curve
All areas boundary point position ORB feature, the ORB feature of itself and point to be matched in left camera is subjected to similarity pair
Than finding out the highest point of similarity in right camera and being used as feature to be matched.This step is repeated, whole skeleton institutes can be obtained
Three-dimensional coordinate in position constitutes boundary skeleton three-dimensional coordinate, records the three-dimensional coordinate information on each subregion boundary, boundary three
It ties up shown in skeleton picture such as Fig. 2 (c)；
The detection process of ORB feature are as follows:
(1) pixel p is chosen in the picture, if its brightness is Ip；
(2) a threshold value T is set, is worth for Ip20%；
(3) centered on p, 16 pixels on the circle that radius is 3 pixels are chosen；
(4) if there is the brightness of continuous 12 points to be greater than I on the circle chosenP+ T is less than IP- T, p are considered as feature
Point.
The three-D imaging method merged using three-dimensional framework with SFS.SFS algorithm is a kind of quickly and effectively three-dimensional reconstruction side
Method, but the current reconstruction precision height of this method relies on the source model of its imaging, and due to the construction of internal organs difference,
The distribution of its shade is not also identical, and the recovery deviation of three-dimensional data is be easy to cause using a kind of source model.Super-pixel segmentation
Region afterwards reduces the influence of complex colors background, then carries out SFS three-dimensional reconstruction to different zones after segmentation, and weight can be improved
Build precision.In order to realize the coordinate unification of different zones, three-dimensional coordinate is carried out to different regions on the basis of three-dimensional framework
Fusion, to obtain the accurate three-dimensional point cloud of color change complex object.
Linear approach three-dimensional modeling is chosen, SFS used herein is to carry out using finite-difference method to surface graded p and q
Then discrete approximation carries out linearization process in height Z-direction, this method arithmetic speed is fast, for any reflection letter
Number is all suitable for.Discrete approximation is used for p and q, as shown in formula (10) and formula (11):
For the tonal gradation E (x, y) of some pixel (x, y) and given image, function f is about height in formula (12)
Degree figure Zn-1Linear approximation can use Taylor series expansion, then go to solve using Jacobi iteration, can be obtained by simplification:
Then, for Z (x, y)=Zn(x, y), the height map of nth iteration can directly press formula (13) solution:
In formula (13):
Now, if the initial estimate of all pixels point is Z0(x, y)=0, height Z can be passed through by formula (13)
Iteration obtains.Each subregion image is as shown in Figure 3 after SFS processing.
Obtained boundary skeleton coordinate information after changes in coordinates relationship obtained after SFS and super-pixel segmentation is carried out
Fusion calculates global three-dimensional coordinate.Operation finishes.
The present invention and the difference of existing three-dimensional rebuilding method maximum have at following 3 points:
(1) it proposes super-pixel segmentation and merges three-dimensional framework generation method with the ORB under epipolar line restriction, this method is by face
Color complex region is split, and the boundary of different zones constitutes the three-dimensional framework of organ.This method can both reduce three-dimensional reconstruction
The Feature Points Matching time, and the accuracy of Feature Points Matching can be improved.
(2) rapid three dimensional imaging process that three-dimensional framework is merged with SFS is proposed, this method is sat with the three-dimensional framework generated
It is designated as benchmark, in conjunction with the gradient difference of different colours, successively extrapolates the three-dimensional coordinate information between each region, and then obtain
Whole three-dimensional appearance coordinates of organ in scene.
(3) solve the problems, such as that three-dimensional endoscope high-precision three-dimensional is rebuild, boosting three-dimensional reconstruction is in medical treatment and industrial circle
Development and application.This research is low to the skill requirement of doctor, and operation is more simple, and reliability also correspondinglys increase；With existing biography
System surgical operation is compared, this research can mitigate the pain of patient, shortens operating time, reduces operation risk.In addition, this research
New algorithm and theoretical research foundation are provided for the high-precision endoscope three-dimensional reconstruction of other field.
In conclusion the advantages of three-dimensional rebuilding method of the present invention, is:
(1) it can get accurately three-dimensional coordinate data；
(2) influence for reducing complex colors background, it is fast to different zones three-dimensional reconstruction speed after segmentation, and rebuild essence
Degree greatly promotes.
Three-dimensional medical endoscope indagation can not only shorten the training of doctors time, reduce operating time, can also solve me
The key technology difficulty that state's Minimally Invasive Surgery is promoted.Meanwhile the three-dimensional endoscope hand constructed by this 3 D stereo restoration methods
The popularization of art technology can push the development of precisely medical treatment and virtual reality medical procedure, promote China's Medical Instruments industry
Progress.
Schematically the present invention and embodiments thereof are described above, this describes no limitation, institute in attached drawing
What is shown is also one of embodiments of the present invention.So not departed from if those of ordinary skill in the art are inspired by it
In the case where the invention objective, each component layouts mode of the same item or other forms that take other form, without
Creative designs technical solution similar with the technical solution and embodiment, is within the scope of protection of the invention.
Claims (1)
1. the present invention devises a kind of real-time three-dimensional method for reconstructing applied to binocular endoscope medical image, characterized in that packet
Containing steps are as follows:
Step 1: binocular camera being demarcated, if the coordinate system where left camera A is OaXaYaZaIf the seat where right camera B
Mark system is ObXbYbZb, the spin matrix between two cameras is R, translation matrix T, shown in the formula of calibration such as formula (1),r11-r33Spin matrix point for the right camera B relative to the left camera A
Amount, tx, ty, tzTranslation matrix component for the right camera B relative to the left camera A；
Step 2: the detecting lenses of binocular endoscope being protruded into patient body, to obtain organ surface image, and endoscope are adopted
The organ surface image of collection carries out denoising and picture smooth treatment using median filtering method, protects the detailed information of image；
Step 3: utilizing SLIC (Simple Linear Iterative Clustering) superpixel segmentation method segmentation step 2
Organ surface image is subdivided by obtained organ surface image first by the color of adjacent pixel, brightness, textural characteristics
Multiple subregions, then all subregion image is transformed into CIE-Lab color space from RGB color, according to super-pixel number,
Seed point is evenly distributed in image, and three-dimensional colouring information and two-dimensional space bit are utilized in the contiguous range of seed point
Confidence breath calculates the pixel each searched to the distance of the seed point, to cluster to pixel, and passes through super-pixel
The destination number of cut zone, controls the size of cut zone, optimization and enhancing connectivity is finally iterated, after obtaining segmentation
Organ surface image, distance calculates as shown in formula (2), dcRepresent color distance, dsRepresent space length, NsIt represents in class
Maximum space distance, NcRepresent maximum color distance；
Step 4: by the organ surface image after step 3 segmentation, according to outer limit matching principle, after left camera collects segmentation
Organ surface image segmentation boundary selected point, right camera collect segmentation after organ surface image on determine polar curve and
Polar curve and partitioning boundary intersection point, obtain accurately left and right matching double points, then with ORB (Oriented FAST and Rotated
BRIEF) feature describes operator combination, positions the position of corresponding region intersection point in the camera of left and right, chooses matching degree highest in intersection point
Point be match point, if the slope of left camera selected point be ka, then the slope k of opposite match pointbIt can be obtained by formula (3), ka
For the slope of certain point P in left camera acquired image skeleton, kbFor in right camera acquired image skeleton with the P
The slope of point corresponding points, repeats this step, can obtain the three-dimensional coordinate of whole skeleton positions, constitutes boundary skeleton
Three-dimensional coordinate records the three-dimensional coordinate information on each subregion boundary；
Step 5: the inside of subregion after each obtained super-pixel segmentation of step 4, first with SFS (Shape
From Shading) three-dimensional reconstruction is carried out, linear approach three-dimensional modeling is chosen, surface graded p and q is carried out using finite-difference method
Then discrete approximation carries out linearization process according to formula (4) on the direction height Z (x, y), finally obtain the coordinate of partial points
Variation relation；
In formula (4):
Step 6: step 5 changes in coordinates relationship obtained is merged with the obtained boundary skeleton coordinate information of step 4,
Calculate organ surface global three-dimensional coordinate；Operation finishes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910315817.0A CN110033465B (en) | 2019-04-18 | 2019-04-18 | Real-time three-dimensional reconstruction method applied to binocular endoscopic medical image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910315817.0A CN110033465B (en) | 2019-04-18 | 2019-04-18 | Real-time three-dimensional reconstruction method applied to binocular endoscopic medical image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110033465A true CN110033465A (en) | 2019-07-19 |
| CN110033465B CN110033465B (en) | 2023-04-25 |
Family
ID=67239087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910315817.0A Active CN110033465B (en) | 2019-04-18 | 2019-04-18 | Real-time three-dimensional reconstruction method applied to binocular endoscopic medical image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110033465B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807803A (en) * | 2019-10-11 | 2020-02-18 | 北京文香信息技术有限公司 | Camera positioning method, device, equipment and storage medium |
| CN110992431A (en) * | 2019-12-16 | 2020-04-10 | 电子科技大学 | Combined three-dimensional reconstruction method for binocular endoscope soft tissue image |
| CN111080778A (en) * | 2019-12-23 | 2020-04-28 | 电子科技大学 | Online three-dimensional reconstruction method of binocular endoscope soft tissue image |
| CN111598939A (en) * | 2020-05-22 | 2020-08-28 | 中原工学院 | Human body circumference measuring method based on multi-vision system |
| CN113034387A (en) * | 2021-03-05 | 2021-06-25 | 成都国科微电子有限公司 | Image denoising method, device, equipment and medium |
| CN113052956A (en) * | 2021-03-19 | 2021-06-29 | 安翰科技(武汉)股份有限公司 | Method, device and medium for constructing film reading model based on capsule endoscope |
| CN114078139A (en) * | 2021-11-25 | 2022-02-22 | 四川长虹电器股份有限公司 | Image post-processing method based on portrait segmentation model generation result |
| CN114531767A (en) * | 2022-04-20 | 2022-05-24 | 深圳市宝润科技有限公司 | Visual X-ray positioning method and system for handheld X-ray machine |
| WO2022127533A1 (en) * | 2020-12-18 | 2022-06-23 | 安翰科技(武汉)股份有限公司 | Capsule endoscope image three-dimensional reconstruction method, electronic device, and readable storage medium |
| CN116797744A (en) * | 2023-08-29 | 2023-09-22 | 武汉大势智慧科技有限公司 | Multi-time-phase live-action three-dimensional model construction method, system and terminal equipment |
| CN114078139B (en) * | 2021-11-25 | 2024-04-16 | 四川长虹电器股份有限公司 | Image post-processing method based on human image segmentation model generation result |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101587082A (en) * | 2009-06-24 | 2009-11-25 | 天津工业大学 | Quick three-dimensional reconstructing method applied for detecting fabric defect |
| WO2015024407A1 (en) * | 2013-08-19 | 2015-02-26 | 国家电网公司 | Power robot based binocular vision navigation system and method based on |
| CN107767442A (en) * | 2017-10-16 | 2018-03-06 | 浙江工业大学 | A kind of foot type three-dimensional reconstruction and measuring method based on Kinect and binocular vision |
| CN107907048A (en) * | 2017-06-30 | 2018-04-13 | 长沙湘计海盾科技有限公司 | A kind of binocular stereo vision method for three-dimensional measurement based on line-structured light scanning |
| WO2018107427A1 (en) * | 2016-12-15 | 2018-06-21 | 深圳大学 | Rapid corresponding point matching method and device for phase-mapping assisted three-dimensional imaging system |
| CN108335350A (en) * | 2018-02-06 | 2018-07-27 | 聊城大学 | The three-dimensional rebuilding method of binocular stereo vision |
-
2019
- 2019-04-18 CN CN201910315817.0A patent/CN110033465B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101587082A (en) * | 2009-06-24 | 2009-11-25 | 天津工业大学 | Quick three-dimensional reconstructing method applied for detecting fabric defect |
| WO2015024407A1 (en) * | 2013-08-19 | 2015-02-26 | 国家电网公司 | Power robot based binocular vision navigation system and method based on |
| WO2018107427A1 (en) * | 2016-12-15 | 2018-06-21 | 深圳大学 | Rapid corresponding point matching method and device for phase-mapping assisted three-dimensional imaging system |
| CN107907048A (en) * | 2017-06-30 | 2018-04-13 | 长沙湘计海盾科技有限公司 | A kind of binocular stereo vision method for three-dimensional measurement based on line-structured light scanning |
| CN107767442A (en) * | 2017-10-16 | 2018-03-06 | 浙江工业大学 | A kind of foot type three-dimensional reconstruction and measuring method based on Kinect and binocular vision |
| CN108335350A (en) * | 2018-02-06 | 2018-07-27 | 聊城大学 | The three-dimensional rebuilding method of binocular stereo vision |
Non-Patent Citations (2)
| Title |
|---|
| 宋丽梅;陈昌曼;张亮;董霄;: "高精度全局解相在多频率三维测量中的应用" * |
| 董霄;宋丽梅;: "一种基于相位法的彩色三维形貌测量方法" * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807803A (en) * | 2019-10-11 | 2020-02-18 | 北京文香信息技术有限公司 | Camera positioning method, device, equipment and storage medium |
| CN110992431A (en) * | 2019-12-16 | 2020-04-10 | 电子科技大学 | Combined three-dimensional reconstruction method for binocular endoscope soft tissue image |
| CN111080778A (en) * | 2019-12-23 | 2020-04-28 | 电子科技大学 | Online three-dimensional reconstruction method of binocular endoscope soft tissue image |
| CN111080778B (en) * | 2019-12-23 | 2023-03-31 | 电子科技大学 | Online three-dimensional reconstruction method of binocular endoscope soft tissue image |
| CN111598939A (en) * | 2020-05-22 | 2020-08-28 | 中原工学院 | Human body circumference measuring method based on multi-vision system |
| CN111598939B (en) * | 2020-05-22 | 2021-01-26 | 中原工学院 | Human body circumference measuring method based on multi-vision system |
| WO2022127533A1 (en) * | 2020-12-18 | 2022-06-23 | 安翰科技(武汉)股份有限公司 | Capsule endoscope image three-dimensional reconstruction method, electronic device, and readable storage medium |
| CN113034387A (en) * | 2021-03-05 | 2021-06-25 | 成都国科微电子有限公司 | Image denoising method, device, equipment and medium |
| CN113034387B (en) * | 2021-03-05 | 2023-07-14 | 成都国科微电子有限公司 | Image denoising method, device, equipment and medium |
| CN113052956A (en) * | 2021-03-19 | 2021-06-29 | 安翰科技(武汉)股份有限公司 | Method, device and medium for constructing film reading model based on capsule endoscope |
| CN114078139A (en) * | 2021-11-25 | 2022-02-22 | 四川长虹电器股份有限公司 | Image post-processing method based on portrait segmentation model generation result |
| CN114078139B (en) * | 2021-11-25 | 2024-04-16 | 四川长虹电器股份有限公司 | Image post-processing method based on human image segmentation model generation result |
| CN114531767A (en) * | 2022-04-20 | 2022-05-24 | 深圳市宝润科技有限公司 | Visual X-ray positioning method and system for handheld X-ray machine |
| CN116797744A (en) * | 2023-08-29 | 2023-09-22 | 武汉大势智慧科技有限公司 | Multi-time-phase live-action three-dimensional model construction method, system and terminal equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110033465B (en) | 2023-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110033465A (en) | A kind of real-time three-dimensional method for reconstructing applied to binocular endoscope medical image | |
| Meinzer et al. | Computerized planning of liver surgery—an overview | |
| CN109785374A (en) | A kind of automatic unmarked method for registering images in real time of dentistry augmented reality surgical navigational | |
| US20140369584A1 (en) | Method And Apparatus For Determining Tumor Shift During Surgery Using A Stereo-Optical Three-Dimensional Surface-Mapping System | |
| CN107016666B (en) | Image post-processing method for pelvis CT three-dimensional reconstruction based on coordinate system | |
| CN104462650A (en) | Materialized heart 3D model manufacturing method capable of achieving internal and external structures | |
| CN106327479B (en) | The device and method that congenital heart disease art medium vessels recognize under angiography intermediary | |
| CN105096270B (en) | A kind of calcified plaque minimizing technology in coronary artery three-dimensional reconstruction | |
| CN108369736A (en) | Method and system for the tissue volume for calculating excision according to image data in 2D/2.5D arts | |
| CN107049489B (en) | A kind of operation piloting method and system | |
| CN106960439B (en) | A kind of vertebrae identification device and method | |
| CN111080778A (en) | Online three-dimensional reconstruction method of binocular endoscope soft tissue image | |
| CN106236264A (en) | The gastrointestinal procedures air navigation aid of optically-based tracking and images match and system | |
| CN116421313A (en) | Augmented reality fusion method in navigation of lung tumor resection operation under thoracoscope | |
| Ali et al. | ArthroNet: a monocular depth estimation technique with 3D segmented maps for knee arthroscopy | |
| CN114711962A (en) | Augmented reality operation planning navigation system and method | |
| CN107204045A (en) | Virtual endoscope system based on CT images | |
| CN111658142A (en) | MR-based focus holographic navigation method and system | |
| CN109934798A (en) | Internal object information labeling method and device, electronic equipment, storage medium | |
| CN108510506A (en) | A kind of tubular structure image partition method | |
| Li et al. | A fully automatic surgical registration method for percutaneous abdominal puncture surgical navigation | |
| CN115619790B (en) | Hybrid perspective method, system and equipment based on binocular positioning | |
| Andrea et al. | Validation of stereo vision based liver surface reconstruction for image guided surgery | |
| US20220175457A1 (en) | Endoscopic image registration system for robotic surgery | |
| CN113648060B (en) | Ultrasonic guided soft tissue deformation tracking method, device, storage medium and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |