A three-dimensional CT image target size measurement method based on projection transformation

Abstract

The application discloses a kind of three-dimensional CT image target object size measurement method based on projection transformation, the starting point and end point position of target object in three-dimensional space are determined by calibration technique, the accuracy of scale method is greatly improved, more meet the demand of first-line graph determination.This application method does not limit the category of measured object, and two point selections are used to locate three-dimensional space point coordinates, and then the starting point and end point of the measurement target are determined.This method excludes the system error introduced by projecting three-dimensional coordinates to a two-dimensional screen, is more universal and accurate.When point selection, the parallel ability of graphics hardware is fully utilized, so that the calibration operation of measurement point is assisted by the imaging process of three-dimensional CT image, without introducing pre-processing and early identification, and without depth test, measurement calculation is completed with minimum algorithm complexity and time overhead.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of three-dimensional CT image processing, and particularly relates to a three-dimensional CT image target object size measurement method based on projection transformation. BACKGROUND

[0002] CT technology has been widely used in medical, industrial, military, geological and other industries due to its great advantages in material detection. With the increasing requirements of airport security accuracy at home and abroad, CT equipment is gradually replacing traditional perspective equipment, and three-dimensional CT images are gradually replacing two-dimensional perspective images, becoming the main basis for security officers to judge. In general three-dimensional CT images, objects in the row package are randomly stacked and mutually occluded. The row package image is mapped to a two-dimensional screen through an imaging algorithm, and it is particularly important to quickly and accurately judge the size of the target object in the row package to assist the security officer in reading.

[0003] Current three-dimensional image measurement is mostly performed by drawing a two-dimensional scale on the screen for rough size estimation, and the scale changes when scaling. This method is easy to implement, but due to the projection principle, there is an unavoidable system error between the two-dimensional size and the three-dimensional space size, and the error is large, which makes it difficult to achieve the expected measurement effect in actual application.

[0004] Chinese patent application CN114403854A discloses a coronoid process parameter measurement method and device based on CT three-dimensional reconstruction image, and provides a method for measuring coronoid process parameters in the medical field. CT data is acquired for reconstruction, and a proximal ulna three-dimensional model is extracted. Reference points are determined, including: standard lateral first ulna dorsal point A, standard lateral second ulna dorsal point B, coronoid process tip point, eagle beak tip point, trochlear fossa lowest point C, coronoid process distal end bevel change point D, coronoid process front and innermost point E, first coronoid process width point F, second coronoid process width point G, and coronoid process proximal ulna diaphysis extension point H. The measurement parameters are determined according to the reference points, and the coronoid process parameters are measured by using the measurement parameters. The main shortcomings of this scheme are: 1) only supporting the measurement of coronoid process parameters in medical CT, not involving the measurement of other materials; 2) the measurement depends on the extraction of the proximal ulna three-dimensional model after reconstruction, and then the reference points are determined, and the measurement parameters are determined according to the reference points. It is a model calculation depending on the reference object, and the universality is not enough for arbitrary object measurement.

[0005] Chinese patent application CN105787919A discloses a security CT three-dimensional image operation method and device, gives a method for measuring CT image article size, provides a security CT three-dimensional image; the article image in the three-dimensional image is selected and response is made based on the selection. The core is that the three-dimensional volume rendering is carried out on the security CT three-dimensional image, and the position information of the object surface in the three-dimensional image is obtained; the normal direction change value between the start point and the end point of the selected part is calculated, when the change value is less than the set angle threshold, the surface depth is inquired to convert the start point and the end point into three-dimensional coordinates, and the coordinate conversion direction size measurement value to be measured is output. The acquisition of the normal direction information depends on the position of the first time that the current light hits the opaque area in the security CT image, the position is recorded as the article image surface depth, the gradient value of the voxel at the position is calculated, and the normal direction information is recorded. The scheme has the following deficiencies: 1) it depends on the precalculation of the object surface position information, which increases the time overhead; 2) the algorithm takes the first hit as the basis for estimating the three-dimensional coordinate, and the size calculation depends on the gradient change, so the accuracy of this method is greatly affected by the specific luggage article. SUMMARY

[0006] In view of the deficiencies of the prior art, the present application aims to provide a three-dimensional CT image target object size measurement method based on projection transformation.

[0007] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme:

[0008] A three-dimensional CT image target object size measurement method based on projection transformation, comprising the following steps:

[0009] (1) using a three-dimensional CT image imaging algorithm, a three-dimensional image of a luggage bag freely rotatable by dragging a mouse is presented on a screen;

[0010] (2) a user labels a measurement start point at one viewing angle of the three-dimensional image of the luggage bag, and a corresponding projection cuboid region in the three-dimensional space is calculated and obtained according to the position selected by the user on the screen; after the user rotates the three-dimensional image of the luggage bag, the user selects a measurement start point again in the projection cuboid region determined by the user when selecting the measurement start point for the first time at another viewing angle, and another projection cuboid region in the three-dimensional space is calculated and obtained according to the position selected by the user on the screen; the range of intersection of the two calculated projection cuboid regions is taken as a candidate point region of the measurement start point, and the average value of the coordinates of the candidate point region of the measurement start point is taken as the measurement start point coordinate;

[0011] (3) the user marks the measurement end point under one perspective of the three-dimensional image of the row package, a corresponding projection cuboid region in the three-dimensional space is calculated according to the position selected by the user on the screen; after the user rotates the three-dimensional image of the row package, the user selects the measurement end point again on the projection cuboid region determined by the user when selecting the measurement end point for the first time under another perspective, another corresponding projection cuboid region in the three-dimensional space is calculated according to the position selected by the user on the screen; the intersection range of the two calculated projection cuboid regions is taken as the candidate point region of the measurement end point, and the average of the coordinates of the candidate point region of the measurement end point is taken as the measurement end point coordinate;

[0012] (4) the measurement size of the target object is calculated according to the measurement start point coordinate and the measurement end point coordinate.

[0013] Further, in step (1), according to the imaging principle of volume rendering, a ray is emitted from a pixel point on the screen, the ray intersects with the three-dimensional CT row package image, and the intersection part is sampled and synthesized to form the final screen image.

[0014] Further, the projection cuboid region, the candidate point region, the measurement start point and the measurement end point are displayed in a set form on the screen.

[0015] Further, the method for averaging the space coordinates of the candidate point region of the measurement start point and the measurement end point adopts the mean value method, the weighted mean value or the extreme value method.

[0016] The application also provides a computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, and the computer program is executed by a processor to realize the above method.

[0017] The application also provides a computer device, which comprises a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program to realize the above method.

[0018] The application has the following beneficial effects: the application determines the start point and the end point positions of the target object in the three-dimensional space through the calibration technology, the accuracy is greatly improved compared with the method of the scale, and the demand of the first-line judgment is better met. The application does not limit the category of the measurement object, and the three-dimensional space point coordinates are positioned by twice selection, and then the start point and the end point of the measurement target are determined. This method excludes the system error introduced by the projection of the three-dimensional coordinates to the two-dimensional screen, is more universal and accurate. The parallel capability of the graphic hardware is fully utilized when selecting, so that the calibration operation of the measurement point is assisted by the imaging process of the three-dimensional CT image, without introducing the pretreatment and the early identification, and without the need of depth test, so that the measurement calculation is completed with the minimum algorithm complexity and time consumption. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1The overall method flowchart of the embodiment of the present application;

[0020] Figure 2 The schematic diagram of the principle of the three-dimensional imaging algorithm in the embodiment of the present application;

[0021] Figure 3 The schematic diagram of the principle of perspective projection;

[0022] Figure 4 The schematic diagram of the embodiment of the present application for calculating the corresponding projection cube 1 according to the position of the screen point selection;

[0023] Figure 5 The schematic diagram of the embodiment of the present application for calculating the corresponding projection cube 2 according to the position of the screen point selection;

[0024] Figure 6 The schematic diagram of the embodiment of the present application for determining the measured size of the target object according to the three-dimensional coordinates. DETAILED DESCRIPTION

[0025] The present application will be further described below in conjunction with the accompanying drawings. It should be noted that the present embodiment is based on the technical solution, and gives detailed implementation and specific operation process, but the protection scope of the present application is not limited to the present embodiment.

[0026] The present embodiment provides a three-dimensional CT image target object size measurement method based on projection transformation, as shown in Figure 1 The method comprises the following steps:

[0027] (1) using a three-dimensional CT image imaging algorithm, a row package three-dimensional image that can be freely rotated by dragging a mouse is presented on a screen;

[0028] (2) a user labels a measurement starting point at one viewing angle of the row package three-dimensional image, and calculates and obtains a corresponding projection cube region in a three-dimensional space according to the position of the point selection on the screen; after the user rotates the row package three-dimensional image, the user selects a measurement starting point again in the projection cube region determined by the user when the measurement starting point is selected for the first time at another viewing angle, and calculates and obtains another corresponding projection cube region in a three-dimensional space according to the position of the point selection on the screen; the range of the intersection of the two calculated projection cube regions is taken as a candidate point region of the measurement starting point, and the coordinate mean value of the candidate point region of the measurement starting point is taken as the measurement starting point coordinate.

[0029] (3) The user calibrates the measurement endpoint from one perspective of the baggage 3D image, and calculates a corresponding projection cube region in 3D space for the position selected by the user on the screen; after the user rotates the baggage 3D image, the user selects the measurement endpoint again from another perspective on the projection cube region determined when the user first selected the measurement endpoint, and calculates another corresponding projection cube region in 3D space for the position selected by the user on the screen; the area where the projection cube regions obtained by the two calculations intersect is taken as the candidate point region of the measurement endpoint, and the average coordinate of the candidate point region of the measurement endpoint is taken as the coordinate of the measurement endpoint.

[0030] (4) Calculate the measurement dimensions of the target object based on the coordinates of the measurement start point and the measurement end point.

[0031] In this embodiment, step (1) is specifically based on the imaging principle of volume rendering, such as... Figure 2 As shown, rays are emitted from the pixels of the screen, intersecting with the 3D CT scan image. The intersecting parts are sampled and synthesized to form the final screen image.

[0032] It should be noted that the process of emitting rays from the screen is related to the projection method. This embodiment uses perspective projection, such as... Figure 3 As shown, perspective projection is the projection method that best conforms to the principle of human eye observation. The area that the human eye can observe is defined as a visual cone, and the range between the near plane and the far plane is the visible range. In the method of this embodiment, the near plane can be understood as the screen.

[0033] In this embodiment, in step (2), after setting the observation range and angle in perspective projection, the process of rotating and scaling the three-dimensional image of the baggage and then imaging it on the screen is as follows: Figure 4 As shown, the target objects in three-dimensional space, after perspective projection calculation, correspond to a two-dimensional screen area connected by projection lines. After the user selects the measurement starting point on the screen, a corresponding projected cube area in three-dimensional space can be obtained, such as... Figure 4 The area marked by the solid line is denoted as Projected Cube 1. After rotating the 3D image of the baggage by a certain angle, within the area of ​​the already determined Projected Cube 1, the measurement starting point is selected again. During the refresh and redraw calculation, Projected Cube 2 can be determined to be related to it, such as... Figure 5 As shown. The three-dimensional spatial region where projected cube 1 and projected cube 2 intersect serves as the candidate point region for the measurement starting point. The average spatial coordinates of the candidate point region for the measurement starting point are calculated to form the coordinates of the measurement starting point, as shown. Figure 6 As shown.

[0034] In the embodiment, the principle of implementing step (3) is the same as that of step (2), the user clicks the measurement end point on the screen, clicks the measurement end point again after rotating the row package three-dimensional image by a certain angle, the projection cubes formed by the two times of calculation intersect to generate a candidate point area of the measurement end point, the spatial coordinates of the candidate point area of the measurement end point are averaged to form the coordinate of the measurement end point.

[0035] It should be noted that the method of the embodiment is not only suitable for perspective projection, but also suitable for orthographic projection.

[0036] In the embodiment, the display forms of the projection cube area, the candidate point area, the measurement start point, the measurement end point and the like can be in specific forms according to requirements.

[0037] In the embodiment, the method of averaging the spatial coordinates of the candidate point area of the measurement start point and the measurement end point can adopt mean value method, weighted mean value, extreme value method and the like. After the measurement size of the target object is obtained, the numerical value can be displayed on the interface in the form of text prompt, graphic prompt, scale prompt and the like.

[0038] The method of the embodiment is not only suitable for three-dimensional CT images generated by a dual-energy CT system, but also suitable for three-dimensional CT images generated by a single-energy CT system; not only suitable for a CT system in a slip ring mode, but also suitable for a static CT system; not only suitable for a single-row detector CT system, but also suitable for a multi-row detector CT system.

[0039] For those skilled in the art, various corresponding changes and modifications can be given according to the above technical solutions and concepts, and all these changes and modifications should be included in the protection scope of the claims of the present application.

Claims

1. A method for measuring the size of a target object in a three-dimensional CT image based on projection transformation, characterized in that, Includes the following steps: (1) Using a three-dimensional CT image imaging algorithm, a three-dimensional image of a baggage that can be freely rotated by dragging the mouse is presented on the screen; (2) The user calibrates the measurement starting point from one perspective of the three-dimensional image of the baggage, and calculates and obtains a corresponding projected cube region in three-dimensional space for the position selected by the user on the screen. After the user rotates the 3D image of the baggage, from another perspective, the user selects the measurement starting point again in a projected cube region determined when the user first selected the measurement starting point. Another projected cube region in 3D space is calculated for the position selected by the user on the screen. The area where the two calculated projected cube regions intersect is taken as the candidate point region of the measurement starting point, and the average coordinate of the candidate point region of the measurement starting point is taken as the coordinate of the measurement starting point. (3) The user calibrates the measurement endpoint from one perspective of the baggage 3D image. A corresponding projected cube region in 3D space is calculated for the position selected by the user on the screen. After the user rotates the baggage 3D image, the user selects the measurement endpoint again from another perspective on the projected cube region determined when the user first selected the measurement endpoint. Another corresponding projected cube region in 3D space is calculated for the position selected by the user on the screen. The area where the two calculated projected cube regions intersect is taken as the candidate point region of the measurement endpoint. The average coordinate of the candidate point region of the measurement endpoint is taken as the coordinate of the measurement endpoint. (4) Calculate the measurement dimensions of the target object based on the coordinates of the measurement start point and the measurement end point.

2. The method according to claim 1, characterized in that, In step (1), specifically based on the imaging principle of volume rendering, rays are emitted from the pixels of the screen, and the rays intersect with the three-dimensional CT row image. The intersecting part is sampled and synthesized to form the final screen image.

3. The method according to claim 1, characterized in that, The projection cube region, candidate point region, measurement start point, and measurement end point are displayed on the screen in a predefined format.

4. The method according to claim 1, characterized in that, The method for calculating the average spatial coordinates of candidate point areas at the starting and ending points of the measurement can be the mean method, weighted mean method, or extreme value method.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-4.

6. A computer device, characterized in that, It includes a processor and a memory, the memory being used to store a computer program; the processor being used to execute the computer program to implement the method of any one of claims 1-4.

Images