CN109345458A - One kind splicing ultrasound image method based on improved harris Corner Detection - Google Patents
One kind splicing ultrasound image method based on improved harris Corner Detection Download PDFInfo
- Publication number
- CN109345458A CN109345458A CN201811246341.1A CN201811246341A CN109345458A CN 109345458 A CN109345458 A CN 109345458A CN 201811246341 A CN201811246341 A CN 201811246341A CN 109345458 A CN109345458 A CN 109345458A
- Authority
- CN
- China
- Prior art keywords
- image
- class
- ultrasound image
- ultrasound
- value
- 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.)
- Pending
Links
- 238000002604 ultrasonography Methods 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000001514 detection method Methods 0.000 title claims abstract description 28
- 230000003044 adaptive effect Effects 0.000 claims abstract description 10
- 230000001225 therapeutic effect Effects 0.000 claims abstract description 4
- 241000566145 Otus Species 0.000 claims description 4
- 206010028980 Neoplasm Diseases 0.000 abstract description 4
- 230000007704 transition Effects 0.000 abstract description 2
- 241001269238 Data Species 0.000 abstract 1
- 230000003187 abdominal effect Effects 0.000 abstract 1
- 238000007408 cone-beam computed tomography Methods 0.000 abstract 1
- 238000003384 imaging method Methods 0.000 description 7
- 238000001959 radiotherapy Methods 0.000 description 5
- 208000006678 Abdominal Neoplasms Diseases 0.000 description 2
- 210000001015 abdomen Anatomy 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001803 electron scattering Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 210000004872 soft tissue Anatomy 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/32—Indexing scheme for image data processing or generation, in general involving image mosaicing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10132—Ultrasound image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention discloses a kind of Harris operator ultrasound image joining methods chosen based on adaptive threshold T, the following steps are included: S1. image obtains: in CT simulator locating room and therapeutic room, successively acquisition CT and ultrasound image, this studies all image datas volume data acquired from three abdominal post-operation tumor patient scanning, every patient chooses 10 ultrasounds and the layer images of CBCT, S2. the image obtained in S1 is pre-processed: 400 × 400 two-dimensional ultrasonic image is divided into 4 × 4 non-overlapping subgraphs first before doing ultrasound image splicing, calculate angle point receptance function R Distribution value in subgraph, it prepares to obtain adaptive threshold, realize the smooth transition between two width ultrasound stitching images, eliminate splicing gap, the accuracy of image mosaic is better than traditional The method of Harris Corner Detection.
Description
Method field
The present invention relates to a kind of ultrasound image joining methods, and in particular to one kind is based on improved harris Corner Detection
Splice ultrasound image method.
Background method
In clinical radiation therapy, conical beam CT (CBCT) is used as common image guidance device, has been widely used in image and has drawn
Lead radiotherapy and adaptive radiation therapy.However CBCT soft tissue resolution is low, and marksmanship gunnery hardening effect, electron scattering cause
Artifact phenomenon make rebuild after picture quality it is poor.Radiotherapy technology, ultrasonic in combination CBCT are guided compared to traditional CBCT
The image guidance techniques for carrying out the verifying of pendulum position can reduce the Set-up errors of abdominal tumor.But utilize the imaging of ultrasonic guidance radiotherapy
Image acquired in equipment is narrow beam imaging, is only its localized imaging region compared to CBCT image, for pelvic cavity and abdominal tumor,
By tumor target and organ cannot be jeopardized be entirely included in areas imaging, cause CBCT on the basis of tumor target GTV with
The image of ultrasonic both modalities which carries out generating registration error with punctual, and then influences the order of accuarcy that patient finally puts position verifying.
Therefore, the areas imaging for increasing ultrasound image is the solution for improving ultrasonic in combination CBCT guidance pendulum position verifying precision.
Image mosaic technology, which refers to, is combined into one group of image with lap by image registration, image co-registration
The new images at the big visual angle of one width, interior new images include all pixels information spliced in preceding image.Corner Detection is spelled as image
The common method of connection technology, can be divided into following two categories: one, the Corner Detection based on image border, such method is mainly according to inspection
The edge crossing point measured determines angle point, depends on image segmentation and edge extracting, and computationally intensive and serious forgiveness is smaller,
Use scope is relatively limited to, and edge detection operator mainly includes LOG operator, Canny operator and Sobel operator etc.；Two, based on figure
As the Corner Detection of gray scale, by searching the pixel in image at the maximum of local gray level variation as angle point, based on ash
Spending detective operators mainly includes Moravec operator, Harris operator and Susan operator etc..Wherein Harris operator is used as and is based on
Image grayscale detects the common method of angle point, insensitive to the variation of image rotation, grey scale change and noise, and calculates relatively simple
A large amount of characteristic point that is single, being extracted in complicated human tissue structure image.But traditional Harris Corner Detection Algorithm
Threshold value need to be manually set when extracting angle point, does not have scale invariability, if threshold value setting is excessive, will lead to lacking for part angle point
It loses, if threshold value setting is too small, will lead to a large amount of pseudo- angle points of generation.
Therefore, we design a kind of based on improved harris Corner Detection splicing ultrasound image method.
Summary of the invention
The purpose of the present invention is to solve disadvantages present in existing joining method, and the one kind proposed is based on improved
Harris Corner Detection splices ultrasound image method.
To achieve the goals above, present invention employs following method schemes:
One kind splicing ultrasound image method based on improved harris Corner Detection, comprising the following steps:
S1. image obtains: in CT simulator locating room and therapeutic room, successively acquiring CT and ultrasound image.
S2. the image obtained in S1 is pre-processed: first by the two of 400 × 400 before doing ultrasound image splicing
Ultrasound Image Segmentation is tieed up into 4 × 4 non-overlapping subgraphs, calculates angle point receptance function R Distribution value in subgraph, to obtain adaptive thresholding
Value is prepared.
S3. using the improved Harris angle point merging algorithm for images chosen based on adaptive threshold to pretreated in S2
Image is screened: dividing principle using OTUS, the angle point number of detection is determined by optimal T value and splices two width ultrasound figures
Picture.
Preferably, it is improved to traditional based on Harris corner feature registration Algorithm in the S3, if R value is i
Pixel number be ni, sum of all pixels N, piIndicate the probability that i occurs in 4 × 4 subgraphs, specific algorithm are as follows:
Given initial threshold T, is divided into two class of A, B for the pixel in image:
A=0,1 ..., T } B=T+1, T+2 ..., K } (3)
The then probability that this A class and B class pixel R value occur are as follows:
R value mean value in A class and B class and entire 4 × 4 subgraph is respectively as follows:
The variance of A class and B class is respectively as follows:
The between-group variance of A class and B classAnd intra-class varianceFor
OTUS is divided principle and is pointed out,Smaller, the R value in group is more similar,Bigger, two groups of difference is bigger, because
Corresponding T is as optimal threshold, objective function when maximum are as follows:
The invention has the advantages that: the abdomens for using improved Harris Corner Detection Algorithm being ultrasonic in combination CBCT guidance
Tumour puts position verifying and provides the ultrasound image information in the wide visual field, and realizes the smooth transition between two width ultrasound stitching images,
Image slot is eliminated, the accuracy of image mosaic splices the method for ultrasound image better than traditional Harris Corner Detection.
Detailed description of the invention
The following further describes the present invention with reference to the drawings.
Fig. 1 is ultrasound image stitching image.Scheme a, b is ultrasound image to be spliced, and figure c is to utilize the improved angle Harris
The splicing ultrasound image that point detection algorithm obtains, figure d are the splicing ultrasound image that traditional Harris Corner Detection Algorithm obtains；
Fig. 2 is the lab diagram for splicing ultrasound image and goldstandard ultrasound image.Fig. 2 a is improved Harris Corner Detection
The pixel difference figure of splicing ultrasound image and goldstandard ultrasound image that algorithm obtains, Fig. 2 b are the calculation of tradition Harris Corner Detection
The pixel difference figure of splicing ultrasound image and goldstandard ultrasound image that method obtains
Fig. 3 is the lab diagram that the spliced ultrasound of patient is registrated fusion with CBCT.Fig. 3 a, 3b are respectively that bladder maximum is cut
Two ultrasound images to be spliced in face, Fig. 3 c, 3d are respectively to utilize improved Harris Corner Detection Algorithm and tradition
The ultrasound image that Harris Corner Detection Algorithm is spliced, Fig. 3 e are the CBCT image of level identical as ultrasound image, Fig. 3 f
Fused image is registrated with CBCT for ultrasound.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, carries out to the method scheme in the embodiment of the present invention clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.
Embodiment:
(1) image obtains: before acquiring image, need to be carried out with calibration body mould to indoor laser lamp coordinate system and ultrasonic probe
Calibration, these calibrations may insure the coordinate position of each 3D ultrasound voxel of system acquisition relative to indoor laser lamp coordinate system
It is known.CT and ultrasound image in therapeutic room.After to be calibrated, ultrasound and CBCT image are successively acquired.
(2) image preprocessing: 400 × 400 two-dimensional ultrasonic image is divided into first before doing ultrasound image splicing
4 × 4 non-overlapping subgraphs calculate angle point receptance function R Distribution value in subgraph, prepare to obtain adaptive threshold.
(3) the improved Harris angle point merging algorithm for images chosen based on adaptive threshold: thought using the segmentation of OTSU
Think, R value replacement pixels value is responded with the angle point of pixel each in image, iterates and calculates T value, is constantly rejected on edge
Pseudo- angle point determines the angle point number of detection eventually by optimal T value and splices two width ultrasound images.
Steps are as follows: setting R value as the number of the pixel of i is ni, sum of all pixels N, piIndicate that i occurs in 4 × 4 subgraphs
Probability, formula is as follows:
Given initial threshold T, is divided into two classes for the pixel in image:
A=0,1 ..., T } B=T+1, T+2 ..., K } (3)
The then probability that these two types of pixel R values occur are as follows:
R value mean value in A class and B class and entire 4 × 4 subgraph is respectively as follows:
The variance of A class and B class is respectively as follows:
The between-group variance of A class and B classAnd intra-class varianceFor
OTSU points out,Smaller, the R value in group is more similar,Bigger, two groups of difference is bigger, becauseInstitute when maximum
Corresponding T is as optimal threshold, objective function are as follows:
(4) improved Harris Corner Detection stitching algorithm is obtained into ultrasound image and original Harris detection method respectively
Compared with the ultrasound image of acquisition is made the difference with goldstandard ultrasound image, pixel difference figure more tends to blue, then two width comparison charts
Difference it is smaller.It is registrated using the spliced ultrasound image of improved method with CBCT, verifying patient abdomen histoorgan exists
The integrality being imaged in ultrasound image.
The splicing ultrasound image that the application is obtained by the Harris angular-point detection method chosen based on adaptive threshold at
Picture has widened the areas imaging of ultrasound, solves the defect of ultrasonic device narrow beam imaging.It is obtained using improved Harris method
Splicing ultrasound image and goldstandard ultrasound image degree of similarity it is higher, and realize putting down between two width ultrasound stitching images
Degree of slipping over, eliminates gap, and the accuracy of image mosaic splices the method for ultrasound image better than traditional Harris Corner Detection.
More than, it is merely preferred embodiments of the present invention, but scope of protection of the present invention is not limited thereto, it is any
Be familiar with the method personnel in this method field the invention discloses method within the scope of, scheme and its invention according to the method for the present invention
Design is subject to equivalent substitution or change, should be covered by the protection scope of the present invention.
Claims (2)
1. one kind splices ultrasound image method based on improved harris Corner Detection, which comprises the following steps:
S1. image obtains: in CT simulator locating room and therapeutic room, successively acquiring CT and ultrasound image；
S2. the image obtained in S1 is pre-processed: first that 400 × 400 two dimension is super before doing ultrasound image splicing
Acoustic image is divided into 4 × 4 non-overlapping subgraphs, calculates angle point receptance function R Distribution value in subgraph, does to obtain adaptive threshold
Prepare；
S3. using the improved Harris angle point merging algorithm for images chosen based on adaptive threshold to image pretreated in S2
It is screened: principle is divided using OTUS, the angle point number of detection is determined by optimal T value and splices two width ultrasound images.
2. it is according to claim 1 a kind of based on improved harris Corner Detection splicing ultrasound image method, it is described
It is improved to traditional based on Harris corner feature registration Algorithm in S3, if the number for the pixel that R value is i is ni, as
Plain sum is N, piIndicate the probability that i occurs in 4 × 4 subgraphs, specific algorithm are as follows:
Given initial threshold T, is divided into two class of A, B for the pixel in image:
A=0,1 ..., T } B=T+1, T+2 ..., K } (3)
The then probability that this A class and B class pixel R value occur are as follows:
R value mean value in A class and B class and entire 4 × 4 subgraph is respectively as follows:
The variance of A class and B class is respectively as follows:
The between-group variance of A class and B classAnd intra-class varianceFor
OTUS is divided principle and is pointed out,Smaller, the R value in group is more similar,Bigger, two groups of difference is bigger, becauseIt is maximum
When corresponding T as optimal threshold, objective function are as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811246341.1A CN109345458A (en) | 2018-10-25 | 2018-10-25 | One kind splicing ultrasound image method based on improved harris Corner Detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811246341.1A CN109345458A (en) | 2018-10-25 | 2018-10-25 | One kind splicing ultrasound image method based on improved harris Corner Detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109345458A true CN109345458A (en) | 2019-02-15 |
Family
ID=65311661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811246341.1A Pending CN109345458A (en) | 2018-10-25 | 2018-10-25 | One kind splicing ultrasound image method based on improved harris Corner Detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109345458A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113609943A (en) * | 2021-07-27 | 2021-11-05 | 东风汽车有限公司东风日产乘用车公司 | Finger vein recognition method, electronic device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080095421A1 (en) * | 2006-10-20 | 2008-04-24 | Siemens Corporation Research, Inc. | Registering 2d and 3d data using 3d ultrasound data |
| CN102903085A (en) * | 2012-09-25 | 2013-01-30 | 福州大学 | Rapid image mosaic method based on corner matching |
| CN108492340A (en) * | 2018-01-31 | 2018-09-04 | 倪昕晔 | Method based on ultrasound image acquisition puppet CT |
-
2018
- 2018-10-25 CN CN201811246341.1A patent/CN109345458A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080095421A1 (en) * | 2006-10-20 | 2008-04-24 | Siemens Corporation Research, Inc. | Registering 2d and 3d data using 3d ultrasound data |
| CN102903085A (en) * | 2012-09-25 | 2013-01-30 | 福州大学 | Rapid image mosaic method based on corner matching |
| CN108492340A (en) * | 2018-01-31 | 2018-09-04 | 倪昕晔 | Method based on ultrasound image acquisition puppet CT |
Non-Patent Citations (3)
| Title |
|---|
| 刘文进等: "基于AP聚类图像分块的角点检测改进方法", 《计算机工程》 * |
| 邓小炼等: "像素自相关矩阵的阈值自适应角点检测算法", 《农业工程学报》 * |
| 霍东旭等: "基于改进Harris算法的全自动无缝图像拼接", 《黑龙江大学自然科学学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113609943A (en) * | 2021-07-27 | 2021-11-05 | 东风汽车有限公司东风日产乘用车公司 | Finger vein recognition method, electronic device and storage medium |
| CN113609943B (en) * | 2021-07-27 | 2024-05-17 | 东风汽车有限公司东风日产乘用车公司 | Finger vein recognition method, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8406495B2 (en) | MR segmentation using transmission data in hybrid nuclear/MR imaging | |
| ES2706542T3 (en) | Diagnostic system for multimodal mammography | |
| Brahme et al. | 4D laser camera for accurate patient positioning, collision avoidance, image fusion and adaptive approaches during diagnostic and therapeutic procedures | |
| US20150371420A1 (en) | Systems and methods for extending a field of view of medical images | |
| EP2245592B1 (en) | Image registration alignment metric | |
| Li et al. | Comprehensive evaluation of ten deformable image registration algorithms for contour propagation between CT and cone-beam CT images in adaptive head & neck radiotherapy | |
| Hacihaliloglu et al. | Automatic bone localization and fracture detection from volumetric ultrasound images using 3-D local phase features | |
| CA2743937A1 (en) | C-arm pose estimation using intensity-based registration of imaging modalities | |
| US20130266230A1 (en) | Image registration apparatus | |
| WO2013030707A1 (en) | Integration of user inputs and correction of deformation vector field in deformable image registration workflow | |
| Fontanarosa et al. | A CT based correction method for speed of sound aberration for ultrasound based image guided radiotherapy | |
| US11288848B2 (en) | Three-dimensional ultrasound image display method | |
| Paganelli et al. | Quantification of organ motion based on an adaptive image‐based scale invariant feature method | |
| King et al. | Registering preprocedure volumetric images with intraprocedure 3-D ultrasound using an ultrasound imaging model | |
| Koo et al. | Hierarchical CT to ultrasound registration of the lumbar spine: a comparison with other registration methods | |
| CN115311258B (en) | Method and system for automatically segmenting organs in SPECT planar image | |
| Sun et al. | Stepwise local stitching ultrasound image algorithms based on adaptive iterative threshold Harris corner features | |
| US20160110869A1 (en) | Cortical bone segmentation from mr dixon data | |
| CN109345458A (en) | One kind splicing ultrasound image method based on improved harris Corner Detection | |
| CN111275617B (en) | Automatic splicing method and system for ABUS breast ultrasound panorama and storage medium | |
| Hopp et al. | Segmentation of 3D ultrasound computer tomography reflection images using edge detection and surface fitting | |
| Grimwood et al. | Assisted probe positioning for ultrasound guided radiotherapy using image sequence classification | |
| JP2019536538A (en) | Bone and hard plaque segmentation in spectral CT | |
| Galdames et al. | Registration of renal SPECT and 2.5 D US images | |
| Smit et al. | Validation of 3D ultrasound for image registration during oncological liver surgery |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190215 |