CN113920012A - High-reliability and fault-tolerance skin CT large-view-field image splicing method and system - Google Patents

High-reliability and fault-tolerance skin CT large-view-field image splicing method and system Download PDF

Info

Publication number
CN113920012A
CN113920012A CN202111152470.6A CN202111152470A CN113920012A CN 113920012 A CN113920012 A CN 113920012A CN 202111152470 A CN202111152470 A CN 202111152470A CN 113920012 A CN113920012 A CN 113920012A
Authority
CN
China
Prior art keywords
image
skin
spliced
splicing
images
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
Application number
CN202111152470.6A
Other languages
Chinese (zh)
Inventor
陈振鑫
党世杰
王佳俊
张洪远
杨雨澄
赵凌霄
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Suzhou Institute of Biomedical Engineering and Technology of CAS
Original Assignee
Suzhou Institute of Biomedical Engineering and Technology of CAS
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Suzhou Institute of Biomedical Engineering and Technology of CAS filed Critical Suzhou Institute of Biomedical Engineering and Technology of CAS
Priority to CN202111152470.6A priority Critical patent/CN113920012A/en
Publication of CN113920012A publication Critical patent/CN113920012A/en
Pending legal-status Critical Current

Links

Images

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
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • 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
    • G06T2207/10072—Tomographic images
    • G06T2207/10081—Computed x-ray tomography [CT]
    • 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
    • G06T2207/30088—Skin; Dermal

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Image Processing (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

The invention discloses a high-reliability and fault-tolerant skin CT large-view-field image splicing method and a system, wherein the method comprises the following steps: 1) manually determining the starting point and the size of a splicing region, setting M lines of multiplied by N columns of skin CT images to be spliced according to the shape of a grid, and ensuring that two adjacent images, namely a left image, a right image, an upper image and a lower image, have overlapping regions; 2) and sequentially scanning each grid of the first row from the left to the right from the starting point of the splicing area by the objective lens of the skin mirror, then switching to the next row and carrying out reverse movement, circulating in this way, finishing the scanning of all the grids along the S-shaped track, and simultaneously acquiring images and splicing. By introducing a fault-tolerant mechanism, the invention can ensure that the whole splicing failure caused by the local image splicing error can not be caused in the large-view field splicing process, but the error area is limited in a local small range, thereby improving the robustness and efficiency of large-view field image splicing.

Description

High-reliability and fault-tolerance skin CT large-view-field image splicing method and system
Technical Field
The invention relates to the field of skin CT detection, in particular to a high-reliability and fault-tolerant skin CT large-view-field image splicing method and system.
Background
The skin CT imaging is based on the confocal principle, and an in-vivo scanning device is added on the basis of a confocal laser scanning microscope, so that the cell characteristics of different structural layers of the skin can be observed noninvasively.
Skin CT obtains a gray scale image by scanning the subcutaneous tissue, which is generally a gray scale image with varying shades based on the differences in the refractive index of light in microstructures in the skin tissue, such as melanin, oxygenated hemoglobin, and organelles. In the skin CT scanning process, if the laser is too strong (too weak), the whole image is too bright (dark), so that different tissue structures cannot be correctly distinguished, and therefore, the laser intensity needs to be adjusted according to different subcutaneous depths, so as to obtain a high-quality skin CT image with appropriate brightness.
Skin CT is a technique that improves resolution by sacrificing the field of view, usually a localized area, is captured. When skin CT is used for diagnosis, not only local observation but also comprehensive detection and analysis are required, and therefore, in order to acquire a large-field skin CT image, a plurality of skin CT images need to be spliced.
The existing image stitching algorithms mainly comprise the following steps:
firstly, the displacement vector of the image to be spliced is searched based on the matching of the image feature points. And calculating the integral displacement of the images by extracting the characteristic points of the overlapping area of the two images and matching. Common image feature operators include HARRISS, SURF, SIFT and the like, and have good effect on natural images with strong features. The skin CT image has the characteristics of large noise, unstable feature details and the like, so that the feature point matching mode has the difficulties of difficult extraction of feature points, large matching error and the like for the skin CT image.
And secondly, searching a displacement vector of the image to be spliced based on a phase correlation method. And the displacement vectors of the two images to be matched are calculated in the frequency domain, so that the image splicing is realized in the airspace. The method is easy to generate pixel-level errors, and when the magnification factor of a microscope image is high, the errors are easy to amplify, so that the splicing visual effect is influenced.
Thirdly, calculating the correlation degree of the overlapped area under different displacement vectors by traversing the displacement vectors, and splicing by taking the displacement vector with the highest correlation degree as the displacement vector of the image to be spliced. Although the method is suitable for splicing the skin CT images, the calculation amount is too large, and the real-time property of splicing is seriously influenced.
Skin CT is a device for real-time detection of human skin, and imaging is susceptible to external shaking and internal skin tissue movement, so that CT images in the same area have certain changes in different periods. In addition, the microscope objective needs to be moved in the splicing process, usually, the motor drives the objective to move and shoot, and the motor also generates certain deviation in the process of moving fixed steps, so that the images cannot be accurately spliced.
A more reliable solution is now needed.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a method and a system for stitching skin CT large view field images with high reliability and fault tolerance, aiming at the defects in the prior art. By introducing a fault-tolerant mechanism, the invention can ensure that the whole splicing failure caused by the local image splicing error can not be caused in the large-view field splicing process, but the error area is limited in a local small range, thereby improving the robustness of large-view field image splicing.
In order to achieve the purpose, the invention adopts the technical scheme that: a high-reliability and fault-tolerant skin CT large-visual-field image stitching method comprises the following steps:
1) manually determining the starting point and the size of a splicing region, setting M lines of multiplied by N columns of skin CT images to be spliced according to the shape of a grid, and ensuring that two adjacent images, namely a left image, a right image, an upper image and a lower image, have overlapping regions;
2) sequentially scanning each grid of the first row from the left to the right from the starting point of the splicing area by the objective lens of the skin mirror, then switching to the next row and carrying out reverse movement, and repeating the steps so as to complete the scanning of all the grids along the S-shaped track;
in the process of scanning by the objective lens of the skin mirror, the skin CT image of the current grid is firstly acquired, and then the acquired skin CT image is spliced to the spliced skin CT image with the large visual field before the current grid is scanned, so that the images are acquired and spliced simultaneously until the acquisition and splicing of M rows of skin CT images with N columns are completed, and the complete skin CT image with the large visual field is obtained.
Preferably, in the step 2), after the skin CT image of the current grid is acquired, it is first detected whether the brightness of the current image is within a normal range, so as to determine whether the intensity of the incident laser of the skin mirror needs to be adjusted, so as to ensure that an image with brightness meeting the requirement is acquired;
judging whether the brightness of the image is in a normal range or not through the image gray mean value mean and the average deviation mdev, wherein the formula is as follows:
mean＝∑(xi-ref)/N；
mdev＝∑|(i-ref)-mean|×His[i]/∑His[i]；
wherein, the image gray mean value mean is the mean value of the image minus the reference value, xiSetting a gray value of a current image pixel as ref, and setting the number of the image pixels as N; his [ i ]]Representing an image gray level histogram, wherein i ranges from 0 to 255;
setting deviation threshold value T according to skin CT image characteristicsmIf mdev is less than T, indicating that the brightness is normal; if mdev is less than T, brightness is abnormal, and whether the image is too bright or too dark is judged according to mean value: if mean is greater than 0, the image is over bright, the laser intensity needs to be reduced, and if mean is less than 0, the image is over dark, the laser intensity needs to be increased; and after the laser is adjusted, the skin CT image of the current grid is shot again, and whether the image brightness is normal or not is calculated and judged until the image with the brightness meeting the requirement is obtained, and then splicing is started.
Preferably, each image is given a splicing mark bit, if the image is successfully spliced, the image is marked as correct, and if the image is unsuccessfully spliced, the image is marked as error;
the method for judging the success or failure of image splicing comprises the following steps:
when the image which is wrong with the splicing mark position is spliced, directly marking the image to be spliced which is spliced with the spliced image as the error;
when the image which is correct to the splicing mark position is spliced, the maximum value Max of the matching degree Ncc of the overlapped area of the image to be spliced and the spliced image after splicing is calculatedNccWith a set threshold value T of degree of matchNSize of (Max)Ncc＞TNAnd if not, marking the image to be spliced as an error.
Preferably, the calculation formula of the matching degree Ncc of the two images in the overlapping area is:
Figure BDA0003287606540000031
where f represents the pre-match image pixel value, t represents the template image pixel value, μ represents the pixel mean, σ represents the standard deviation, and n represents the total number of pixels.
Preferably, the method for stitching the first row image, the first column image and the last column image comprises the following steps: when the first row of images are spliced, splicing the current image with the previous image in the moving direction, and when the first column or the last column of images are spliced, splicing the current image with the image which is in the same column as the current image and is positioned on the row of the current image;
if the spliced image splicing mark bit is wrong, directly marking the current image as wrong;
if the spliced image splicing mark bit is correct, calculating the maximum value Max of the matching degree Ncc of the overlapped area of the two spliced imagesNccIf Max isNcc＞TNThe current picture is marked as correct, otherwise the current picture is marked as wrong.
Preferably, the method for stitching the images at the positions except the first row image, the first column image and the last column image comprises the following steps:
recording the current image as P0, recording the previous image in the moving direction of the current image as P1, and recording the image which is in the same column as the previous image and is positioned at the upper line of the current image as P2;
if one image in the P1 and the P2 is marked as correct and one image in the P2 is marked as wrong, the P0 selects the image marked as correct to be spliced, and then the maximum value Max of the matching degree Ncc of the overlapped area of the two images is calculated after the splicingNccIf Max isNcc＞TNIf the current image is correct, marking the current image as wrong;
if both P1 and P2 are marked as correct, calculating the maximum value Ncc01 of the matching degree of the overlapped region after P0 and P1 are spliced and the maximum value Ncc02 of the matching degree of the overlapped region after P0 and P2 are spliced, when Ncc01 is larger than Ncc02, P0 selects to be spliced with P1, and when Ncc01 > TNIf so, marking the current image as correct, otherwise, marking the current image as wrong; when Ncc01 is not greater than Ncc02, P0 chooses to splice with P2, and when Ncc02 > TNIf so, marking the current image as correct, otherwise, marking the current image as wrong;
if both P1 and P2 are marked as error, then P0 splices according to the preset offset and marks the current image as error.
Preferably, when splicing, calculating the optimal displacement vector of the current image, and then splicing the current image according to the optimal displacement vector movement;
for the spliced images, carrying out weighted fusion on the overlapped areas of the two spliced images through a gray-scale weighted average fusion algorithm, taking the weighted-fusion pixel value result as the pixel value of the overlapped area, wherein the weighted average fusion algorithm formula is as follows:
V＝ai×V1+(1-ai)×V2,ai＝w-i/w；
wherein V is the fused pixel value, V1、V2Respectively representing the pixel values of the image to be stitched and the image to be stitched, aiAnd representing the weight, w represents the width of fusion, and i represents the abscissa of the current pixel point.
The invention also provides a high-reliability and fault-tolerant skin CT large-view-field image splicing system which adopts the method to splice the skin CT large-view-field images.
The invention also provides a storage medium having a computer program stored thereon, characterized in that the program is adapted to carry out the method as described above when executed.
The invention also provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method as described above when executing the computer program.
The invention has the beneficial effects that:
according to the high-reliability and fault-tolerant skin CT large-view-field image splicing method, by introducing a fault-tolerant mechanism, the effect that partial image splicing errors are caused by shaking and the like is limited in a small-range area in the large-view-field skin CT splicing process, and the integrity and robustness of large-view-field splicing can be guaranteed;
by introducing a GPU acceleration mechanism, the calculation speed of traversal search is greatly increased, the calculation efficiency of the algorithm is obviously increased, the skin CT image can be spliced while scanning, the shaking of a patient during the skin CT image acquisition is reduced, and the splicing accuracy is improved.
Drawings
FIG. 1 is a flow chart of the high reliability and fault tolerance method for stitching large field of view images of skin CT according to the present invention;
fig. 2 is a schematic diagram of a splicing movement strategy in embodiment 1 of the present invention;
FIG. 3 is a schematic diagram of splicing in embodiment 1 of the present invention;
fig. 4(a) - (f) are a process diagram for stitching 4 x 4 skin CT images in an embodiment.
Detailed Description
The present invention is further described in detail below with reference to examples so that those skilled in the art can practice the invention with reference to the description.
It will be understood that terms such as "having," "including," and "comprising," as used herein, do not preclude the presence or addition of one or more other elements or groups thereof.
Example 1
Referring to fig. 1, the method for stitching a skin CT large visual field image with high reliability and fault tolerance according to this embodiment includes:
firstly, manually determining the starting point and the size of a splicing region, setting M lines of multiplied by N columns of skin CT images to be spliced according to the shape of a grid, and ensuring that two images adjacent to each other left and right and up and down have an overlapping region;
secondly, sequentially scanning each grid of the first row from the left to the right from the starting point of the splicing area by the objective lens of the skin mirror, then switching to the next row and carrying out reverse movement, and repeating the steps so as to complete the scanning of all the grids along the S-shaped track, and referring to fig. 2;
in the process of scanning by the objective lens of the skin mirror, the skin CT image of the current grid is firstly acquired, and then the acquired skin CT image is spliced to the spliced skin CT image with the large visual field before the current grid is scanned, so that the images are acquired and spliced simultaneously until the acquisition and splicing of M rows of skin CT images with N columns are completed, and the complete skin CT image with the large visual field is obtained.
In the step 2), after the skin CT image of the current grid is obtained, firstly, whether the brightness of the current image is in a normal range is detected, and whether the incident laser intensity of the skin mirror needs to be adjusted is judged, so that the image with the brightness meeting the requirement is obtained;
whether the brightness of the image is in a normal range is judged through the image gray level mean and the average deviation mdev, when the brightness is abnormal, the mean deviates from a mean point (which can be assumed as 128), and the average deviation is also small, and the specific formula is as follows:
mean＝∑(xi-ref)/N；
mdev＝∑|(i-ref)-mean|×His[i]/∑His[i]；
wherein, the image gray mean value mean is the mean value of the image minus the reference value, xiFor the gray value of the current image pixel, ref is a set reference value (usually 128), and N is the number of image pixels; his [ i ]]Representing image gray scaleIn the block diagram, the range of i is 0-255;
setting deviation threshold value T according to skin CT image characteristicsmIf mdev is less than T, indicating that the brightness is normal; if mdev is less than T, brightness is abnormal, and whether the image is too bright or too dark is judged according to mean value: if mean is greater than 0, the image is over bright, the laser intensity needs to be reduced, and if mean is less than 0, the image is over dark, the laser intensity needs to be increased; and after the laser is adjusted, the skin CT image of the current grid is shot again, and whether the image brightness is normal or not is calculated and judged until the image with the brightness meeting the requirement is obtained, and then splicing is started.
The step of performing formal splicing comprises:
1. and preprocessing the original skin CT image acquired by shooting. Because the skin CT image has high imaging noise, Gaussian filtering smoothing is required to be carried out on the skin CT image; in addition, when the skin CT image is shot, laser adjustment is carried out on the basis of the integral gray value of the image, and local over-brightness or over-darkness of the image may occur, so that the image is enhanced in a self-adaptive contrast-limiting histogram equalization mode, and the characteristics of the image are highlighted;
2. and traversing and matching the preprocessed images to be spliced to realize splicing.
2-1, introducing a fault-tolerant mechanism, firstly, giving a splicing mark bit to each image, if the image is successfully spliced, marking the image as correct, and if the image is unsuccessfully spliced, marking the image as wrong;
the method for judging the success or failure of image splicing comprises the following steps:
when the image which is wrong with the splicing mark position is spliced, directly marking the image to be spliced which is spliced with the spliced image as the error;
when the image which is correct to the splicing mark position is spliced, the maximum value Max of the matching degree Ncc of the overlapped area of the image to be spliced and the spliced image after splicing is calculatedNccWith a set threshold value T of degree of matchNSize of (Max)Ncc＞TNAnd if not, marking the image to be spliced as an error.
The method comprises the following steps of judging the matching degree Ncc of two images in an overlapped region by adopting a GPU accelerated Normalized Cross-Correlation (Ncc), wherein the Ncc is calculated according to the following formula:
Figure BDA0003287606540000071
where f represents the pre-match image pixel value, t represents the template image pixel value, μ represents the pixel mean, σ represents the standard deviation, and n represents the total number of pixels.
2-2, the method for splicing the first row image, the first column image and the last column image comprises the following steps: when the first row of images are spliced, splicing the current image with the previous image in the moving direction, and when the first column or the last column of images are spliced, splicing the current image with the image which is in the same column as the current image and is positioned on the row of the current image;
if the spliced image splicing mark bit is wrong, directly marking the current image as wrong; the first row image, the first column image and the last column image are generally stable, and splicing errors are not easy to occur;
if the spliced image splicing mark bit is correct, calculating the maximum value Max of the matching degree Ncc of the overlapped area of the two spliced imagesNccIf Max isNcc＞TNThe current picture is marked as correct, otherwise the current picture is marked as wrong.
2-2, the method for splicing the images at the rest positions except the first row of images, the first column of images and the last column of images comprises the following steps:
referring to fig. 3, let the current image be P0, the image immediately before the current image in the moving direction be P1, and the image in the same column as the previous image and on the same row as the current image be P2;
if one image in the P1 and the P2 is marked as correct and one image in the P2 is marked as wrong, the P0 selects the image marked as correct to be spliced, and then the maximum value Max of the matching degree Ncc of the overlapped area of the two images is calculated after the splicingNccIf Max isNcc＞TNThen mark the current imageMarking the image as correct, otherwise marking the current image as wrong;
if both P1 and P2 are marked as correct, calculating the maximum value Ncc01 of the matching degree of the overlapped region after P0 and P1 are spliced and the maximum value Ncc02 of the matching degree of the overlapped region after P0 and P2 are spliced, when Ncc01 is larger than Ncc02, P0 selects to be spliced with P1, and when Ncc01 > TNIf so, marking the current image as correct, otherwise, marking the current image as wrong; when Ncc01 is not greater than Ncc02, P0 chooses to splice with P2, and when Ncc02 > TNIf so, marking the current image as correct, otherwise, marking the current image as wrong;
if both P1 and P2 are marked as error, then P0 splices according to the preset offset and marks the current image as error.
The P0 is not only singly spliced with the P1 or the P2, but an image with better splicing effect is selected from the two images for splicing, so that on one hand, splicing can be more accurate, and on the other hand, when a splicing error occurs in the splicing process of the P1 or the P2, the P0 can select the splicing mark bit as a correct image for splicing, so that the error is prevented from being continuously transmitted, and the error is limited to a current splicing error area.
2-3, after the images to be stitched are determined, calculating an optimal displacement vector for stitching the current image and the images to be stitched, and then moving the current image to be overlaid on the images to be stitched according to the optimal displacement vector, for example, referring to the figure, if the images to be stitched are P1, if the coordinates of the upper left point of the image to be stitched P1 on the stitched whole image are (x1, y1), and the optimal displacement vector is (x, y), the coordinates of the upper left point of the image to be stitched P0 on the stitched image are (x1+ x, y1+ y).
2-4, performing weighted fusion on the overlapped area of the spliced two images through a gray-scale weighted average fusion algorithm for the spliced images, taking the result of the weighted fusion pixel value as the pixel value of the overlapped area, wherein the formula of the weighted average fusion algorithm is as follows:
V＝ai×V1+(1-ai)×V2,ai＝w-i/w；
wherein the content of the first and second substances,v is the fused pixel value, V1、V2Respectively representing the pixel values of the image to be stitched and the image to be stitched, aiAnd representing the weight, w represents the width of fusion, and i represents the abscissa of the current pixel point.
Referring to fig. 4(a) - (f), in an embodiment, the procedure of stitching skin CT images of 4 × 4 size, each image has a width and a height of 1000 × 1000 pixels. The graph (a) is that the splicing starts, the background layer is generated, and a first image is placed at a fixed position; the image (b) is a first line of spliced images, and the first line of images is only matched and spliced with the previous image in the current movement direction; the graph (c) and the graph (e) are image splicing during line feed, and the current image is only spliced with the corresponding position of the previous line during line feed; the image (d) is a general splicing process, the current image is respectively matched with the previous image at the corresponding position of the previous line and the previous image in the current movement direction, and the image with higher matching degree is preferentially selected for splicing; and (f) is the final result after the whole splicing process is completed.
Example 2
The embodiment provides a high-reliability and fault-tolerant skin CT large-view-field image stitching system, which performs stitching on a skin CT large-view-field image by using the method of embodiment 1.
The present embodiment also provides a storage medium having stored thereon a computer program for implementing the method of embodiment 1 when executed.
The present embodiment also provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor implements the method of embodiment 1 when executing the computer program.
While embodiments of the invention have been disclosed above, it is not limited to the applications listed in the description and the embodiments, which are fully applicable in all kinds of fields of application of the invention, and further modifications may readily be effected by those skilled in the art, so that the invention is not limited to the specific details without departing from the general concept defined by the claims and the scope of equivalents.

Claims (10)

1. A high-reliability and fault-tolerant skin CT large-visual-field image stitching method is characterized by comprising the following steps of:
1) manually determining the starting point and the size of a splicing region, setting M lines of multiplied by N columns of skin CT images to be spliced according to the shape of a grid, and ensuring that two adjacent images, namely a left image, a right image, an upper image and a lower image, have overlapping regions;
2) sequentially scanning each grid of the first row from the left to the right from the starting point of the splicing area by the objective lens of the skin mirror, then switching to the next row and carrying out reverse movement, and repeating the steps so as to complete the scanning of all the grids along the S-shaped track;
in the process of scanning by the objective lens of the skin mirror, the skin CT image of the current grid is firstly acquired, and then the acquired skin CT image is spliced to the spliced skin CT image with the large visual field before the current grid is scanned, so that the images are acquired and spliced simultaneously until the acquisition and splicing of M rows of skin CT images with N columns are completed, and the complete skin CT image with the large visual field is obtained.
2. The method for stitching the skin CT large-view-field images with high reliability and fault tolerance according to claim 1, wherein in the step 2), after the skin CT image of the current grid is obtained, whether the brightness of the current image is in a normal range is detected, so as to determine whether the incident laser intensity of a skin mirror needs to be adjusted to ensure that the obtained image with the brightness meeting the requirement;
judging whether the brightness of the image is in a normal range or not through the image gray mean value mean and the average deviation mdev, wherein the formula is as follows:
mean＝∑(xi-ref)/N；
mdev＝∑|(i-ref)-mean|×His[i]/∑His[i]；
wherein, the image gray mean value mean is the mean value of the image minus the reference value, xiSetting a gray value of a current image pixel as ref, and setting the number of the image pixels as N; his [ i ]]Representing an image gray level histogram, wherein i ranges from 0 to 255;
setting deviation threshold value T according to skin CT image characteristicsmAnd, if mdev is less than T,indicating that the brightness is normal; if mdev is less than T, brightness is abnormal, and whether the image is too bright or too dark is judged according to mean value: if mean is greater than 0, the image is over bright, the laser intensity needs to be reduced, and if mean is less than 0, the image is over dark, the laser intensity needs to be increased; and after the laser is adjusted, the skin CT image of the current grid is shot again, and whether the image brightness is normal or not is calculated and judged until the image with the brightness meeting the requirement is obtained, and then splicing is started.
3. The method for stitching high-reliability and fault-tolerant skin CT large-view images according to claim 2, wherein each image is assigned with a stitching flag bit, if the image stitching is successful, the image is marked as correct, and if the image stitching is failed, the image is marked as error;
the method for judging the success or failure of image splicing comprises the following steps:
when the image which is wrong with the splicing mark position is spliced, directly marking the image to be spliced which is spliced with the spliced image as the error;
when the image which is correct to the splicing mark position is spliced, the maximum value Max of the matching degree Ncc of the overlapped area of the image to be spliced and the spliced image after splicing is calculatedNccWith a set threshold value T of degree of matchNSize of (Max)Ncc＞TNAnd if not, marking the image to be spliced as an error.
4. The method for stitching skin CT large-field-of-view images with high reliability and fault tolerance according to claim 3, wherein the degree of matching Ncc of the two images in the overlapping region is calculated by the following formula:
Figure FDA0003287606530000021
where f represents the pre-match image pixel value, t represents the template image pixel value, μ represents the pixel mean, σ represents the standard deviation, and n represents the total number of pixels.
5. The method for stitching the skin CT large visual field image with high reliability and fault tolerance according to claim 4, wherein the stitching method of the first row image, the first column image and the last column image is as follows: when the first row of images are spliced, splicing the current image with the previous image in the moving direction, and when the first column or the last column of images are spliced, splicing the current image with the image which is in the same column as the current image and is positioned on the row of the current image;
if the spliced image splicing mark bit is wrong, directly marking the current image as wrong;
if the spliced image splicing mark bit is correct, calculating the maximum value Max of the matching degree Ncc of the overlapped area of the two spliced imagesNccIf Max isNcc＞TNThe current picture is marked as correct, otherwise the current picture is marked as wrong.
6. The method for stitching the skin CT large visual field images with high reliability and fault tolerance according to claim 5, wherein the method for stitching the images at the rest positions except the first row image, the first column image and the last column image comprises the following steps:
recording the current image as P0, recording the previous image in the moving direction of the current image as P1, and recording the image which is in the same column as the previous image and is positioned at the upper line of the current image as P2;
if one image in the P1 and the P2 is marked as correct and one image in the P2 is marked as wrong, the P0 selects the image marked as correct to be spliced, and then the maximum value Max of the matching degree Ncc of the overlapped area of the two images is calculated after the splicingNccIf Max isNcc＞TNIf the current image is correct, marking the current image as wrong;
if both P1 and P2 are marked as correct, calculating the maximum value Ncc01 of the matching degree of the overlapped region after P0 and P1 are spliced and the maximum value Ncc02 of the matching degree of the overlapped region after P0 and P2 are spliced, when Ncc01 is larger than Ncc02, P0 selects to be spliced with P1, and when Ncc01 > TNWhen it is correct, the current image is marked as correct, otherwiseMarking the current image as an error; when Ncc01 is not greater than Ncc02, P0 chooses to splice with P2, and when Ncc02 > TNIf so, marking the current image as correct, otherwise, marking the current image as wrong;
if both P1 and P2 are marked as error, then P0 splices according to the preset offset and marks the current image as error.
7. The method for stitching the skin CT large visual field images with high reliability and fault tolerance as claimed in claim 6, wherein when stitching, the optimal displacement vector of the current image is calculated, and then the current image is stitched according to the movement of the optimal displacement vector;
for the spliced images, carrying out weighted fusion on the overlapped areas of the two spliced images through a gray-scale weighted average fusion algorithm, taking the weighted-fusion pixel value result as the pixel value of the overlapped area, wherein the weighted average fusion algorithm formula is as follows:
V＝ai×V1+(1-ai)×V2,ai＝w-i/w；
wherein V is the fused pixel value, V1、V2Respectively representing the pixel values of the image to be stitched and the image to be stitched, aiAnd representing the weight, w represents the width of fusion, and i represents the abscissa of the current pixel point.
8. A high-reliability and fault-tolerant skin CT large-field image stitching system, which is characterized in that the method according to any one of claims 1 to 7 is adopted to perform the stitching of the skin CT large-field image.
9. A storage medium on which a computer program is stored, characterized in that the program is adapted to carry out the method of any one of claims 1-7 when executed.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1-7 when executing the computer program.
CN202111152470.6A 2021-09-29 2021-09-29 High-reliability and fault-tolerance skin CT large-view-field image splicing method and system Pending CN113920012A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111152470.6A CN113920012A (en) 2021-09-29 2021-09-29 High-reliability and fault-tolerance skin CT large-view-field image splicing method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111152470.6A CN113920012A (en) 2021-09-29 2021-09-29 High-reliability and fault-tolerance skin CT large-view-field image splicing method and system

Publications (1)

Publication Number Publication Date
CN113920012A true CN113920012A (en) 2022-01-11

Family

ID=79237093

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111152470.6A Pending CN113920012A (en) 2021-09-29 2021-09-29 High-reliability and fault-tolerance skin CT large-view-field image splicing method and system

Country Status (1)

Country Link
CN (1) CN113920012A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114594038A (en) * 2022-05-09 2022-06-07 天津立中车轮有限公司 Scanning splicing application method for accurately measuring and calculating porosity

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114594038A (en) * 2022-05-09 2022-06-07 天津立中车轮有限公司 Scanning splicing application method for accurately measuring and calculating porosity
CN114594038B (en) * 2022-05-09 2022-08-09 天津立中车轮有限公司 Scanning splicing application method for accurately measuring and calculating porosity

Similar Documents

Publication Publication Date Title
US20220157047A1 (en) Feature Point Detection
EP2188779B1 (en) Extraction method of tongue region using graph-based approach and geometric properties
CN113538522B (en) Instrument vision tracking method for laparoscopic minimally invasive surgery
Ramli et al. Feature‐based retinal image registration using D‐saddle feature
CN112164043A (en) Method and system for splicing multiple fundus images
JP2000182050A (en) Individual identifying device for animal
CN113190115B (en) Eyeball tracking method and virtual reality equipment
CN113920012A (en) High-reliability and fault-tolerance skin CT large-view-field image splicing method and system
JP5278984B2 (en) Image analysis apparatus and image analysis program
JP2008198078A (en) Image detection device
KR100647750B1 (en) Image processing apparatus
CN113781523B (en) Football detection tracking method and device, electronic equipment and storage medium
Yedidya et al. Tracking of blood vessels in retinal images using Kalman filter
JP2008287354A (en) Registering device, collating device, program, and data structure
CN114022514A (en) Real-time sight line inference method integrating head posture and eyeball tracking
CN113643184A (en) Optical coherence tomography-based fundus blood vessel display method, system and medium
CN116703837B (en) MRI image-based rotator cuff injury intelligent identification method and device
CN116883372A (en) Method and system for adaptively identifying tumor based on blood vessel region image
KR20090110348A (en) Object shape generating method, object shape generating device and program
CN114693912B (en) Endoscopy system having eyeball tracking function, storage medium, and apparatus
JP4664805B2 (en) Face edge detection device, face edge detection method, and program
Xi et al. Multi-retinal images stitching based on the maximum fusion and correction ratio of gray average
CN117257204B (en) Endoscope control assembly control method and system
Otero-Mateo et al. A fast and robust iris segmentation method
CN117726656B (en) Target tracking method, device, system and medium based on super-resolution image

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