Two-dimensional image and three-dimensional model image registration error evaluation method based on contour curve shape features

Abstract

The invention discloses a two-dimensional image and three-dimensional model image registration error evaluation method based on contour curve shape features, and belongs to the technical field of image registration error evaluation. According to the invention, error evaluation is carried out by using the mismatching degree of shape features of two profile curves at the same spatial position after image registration; according to the technical key points, according to the theory that contour points with large curvatures serve as important feature points, the curvatures of all the contour points are calculated and serve as compensation parameters, the one-way mismatching degree and the maximum mismatching degree are obtained, and the overall shape matching degree of the curve is judged according to the one-way mismatching degree and the maximum mismatching degree; a center offset degree concept is defined, the two contour curves are divided into 10 areas, and compensation parameters are introduced to eliminate the influence that the number of contour points in the same area is inconsistent; the overall shape position offset distance is pre-judged through the offset distance of the top end and the bottom end of the contour curve, and finally whether the center offset degrees of all the areas meet the requirement or not is judged. According to the method, the mismatching degree of the contour curve shape features is taken as a judgment condition, and the image registration error evaluation efficiency can be improved.

Description

Technical Field

[0001] The patent of this invention relates to a method for evaluating the registration error between a two-dimensional image and a three-dimensional model image based on the shape features of a contour curve, and belongs to the technical field of image registration error evaluation. Background Art

[0002] Image registration technology is widely used in fields such as medical imaging, computer vision, and augmented reality. The accuracy and efficiency of image registration have become crucial. Current image registration error evaluation methods mainly focus on the image registration between two-dimensional images and the image registration between three-dimensional models. In addition, traditional image registration error evaluation methods tend to ignore the characteristics and structural information of the image itself. Since the registration of two-dimensional images and three-dimensional models involves the registration of images in different dimensions, there is no suitable image registration error evaluation method. This makes it impossible to guarantee the registration effect of two-dimensional images and three-dimensional models, hindering the application of two-dimensional image and three-dimensional model registration in medical imaging, computer vision, and augmented reality.

[0003] In order to evaluate the registration error between a two-dimensional image and a three-dimensional model image, the problem of dimensional unification needs to be addressed. A common method is to "upgrade" the two-dimensional image, but this method can only be achieved when the camera can capture multiple two-dimensional images at different angles and directions, which is often not possible in reality. Therefore, in order to evaluate the registration error between a two-dimensional image and a three-dimensional model image, it is necessary to project the three-dimensional model perpendicular to the two-dimensional image plane, obtain the edge contour of the three-dimensional model projection, and implement "dimensionality reduction" processing. The error is then evaluated using the edge contour points of the two-dimensional image and the edge contour points of the three-dimensional model projection. However, if the number of collected edge contour points of the two-dimensional image and the edge contour points of the three-dimensional model projection is different, it may be impossible to find matching feature contour points for evaluation, which in turn affects the evaluation results. In addition, existing quasi-error evaluation methods often ignore the influence of curvature on the reliability of neighboring point matching, resulting in mismatching of matching point pairs in areas with sudden changes in curvature, further reducing the accuracy of error evaluation. In summary, the field of image registration error evaluation technology urgently needs a method that can accurately and quantitatively evaluate the registration effect of two-dimensional images and three-dimensional model images. Summary of the Invention

[0004] In response to the above problems, the present invention proposes a method for evaluating the error in the registration of two-dimensional images and three-dimensional model images based on the shape features of contour curves. This method solves the problem in the current field of image registration error evaluation technology that there is a lack of error evaluation methods for the registration of two-dimensional images and three-dimensional model images based on the shape features of contour curves, and realizes a quantitative description of the registration effect of two-dimensional images and three-dimensional model images.

[0005] A method for evaluating the registration error between a two-dimensional image and a three-dimensional model image based on contour curve shape features is characterized in that the specific implementation process of the method is as follows:

[0006] Step 1: Import the registered 2D image contour point set and 3D model projection contour point set:

[0007] The lower left corner of the two-dimensional image is used as the coordinate origin, the right is the positive direction of the x-axis, the upward is the positive direction of the y-axis, and according to the right-hand rule, the vertical screen outward is the positive direction of the z-axis to establish a three-dimensional image registration error evaluation coordinate system w, obtain the contour curve information of the registered two-dimensional image, and input the contour point set of the registered two-dimensional image contour curve is the position information of the i-th contour point of the registered 2D image contour curve in the o-xy plane of the 3D image registration error evaluation coordinate system w. The value range of i is 1≤i≤n, where, is the x-axis coordinate of the i-th contour point of the registered 2D image contour curve located in the o-xy plane of the 3D image registration error evaluation coordinate system w, The y-axis coordinate of the i-th contour point of the registered two-dimensional image contour curve is located in the o-xy plane of the three-dimensional image registration error evaluation coordinate system w; the registered three-dimensional model is parallel projected onto the o-xy plane along the negative direction of the z-axis of the three-dimensional image registration error evaluation coordinate system w to obtain the projection contour curve information, and the contour point set of the registered three-dimensional model projection contour curve is input. is the position information of the jth contour point of the projection contour curve of the registered 3D model in the o-xy plane of the 3D image registration error evaluation coordinate system w. The value range of j is 1≤j≤m, where The jth contour point of the projection contour curve of the registered 3D model is located in the x-axis coordinate of the o-xy plane of the 3D image registration error evaluation coordinate system w. The jth contour point of the projection contour curve of the registered 3D model is located on the y-axis coordinate of the o-xy plane of the 3D image registration error evaluation coordinate system w;

[0008] Step 2: Calculate the maximum mismatch between the contour of the registered 2D image and the contour of the 3D model projection in the registered pose:

[0009] Use the K nearest neighbor method to find the contour points of the registered 2D image contour curve in the contour points of the registered 3D model projection contour curve. matching points Specifically: the contour points of the contour curve of the two-dimensional image after registration For query points, set the parameter K value to 1. The value of parameter K represents the number of neighboring points to be found near the query point. Setting the parameter K value to 1 means that the query point is found in the contour points of the projection contour curve of the 3D model after registration. The closest point to the Euclidean distance point That is, the contour point of the contour curve of the two-dimensional image after registration Matching point; the i-th contour point of the contour curve of the two-dimensional image after registration Matching points with the contour points of the projected contour curve of the registered 3D model The distance, using the symbol express,

[0010] Use the K nearest neighbor method to find the contour points of the registered 3D model projection contour curve in the contour points of the registered 2D image contour curve. matching points Specifically: Project the contour curve contour points of the registered 3D model For the query point, set the parameter K value to 1. The value of parameter K represents the number of neighboring points to be found near the query point. Setting the parameter K value to 1 means that the contour points of the contour curve of the two-dimensional image after registration are searched for the query point. The closest point to the Euclidean distance point That is the contour point of the projection contour curve of the three-dimensional model after registration The matching point of the 3D model projection contour curve after registration is the kth contour point Matching points with the contour points of the registered 2D image contour curve The distance, using the symbol express,

[0011] Calculate the curvature of the contour curve of the two-dimensional image after registration at the i-th contour point, and select Five contour points, when i=1 or i=2, select Five contour points, when i=n-1 or i=n, select Five contour points, using the least squares method to fit an expression f Tw (x) = a i0 +a i1 x+a i2 x 2 +a i3 x 3 +a i4 x 4 The polynomial curve is used to approximate the five selected contour points;

[0012]

[0013] The curvature of the i-th contour point of the registered two-dimensional image contour curve is

[0014] Calculate the curvature of the kth contour point of the projection contour curve of the 3D model after registration, and select Five contour points, when k = 1 or k = 2, select Five contour points, when k=m-1 or k=m, select Five contour points, using the least squares method to fit an expression f Th (x) = a k0 +a k1 x+a k2 x 2 +a k3 x 3 +a k4 x 4 The polynomial curve is used to approximate the five selected contour points;

[0015]

[0016] The curvature of the kth contour point of the projection contour curve of the 3D model after registration is

[0017] Define the contour points of the 2D image contour curve after registration To the matching points in the projection contour curve of the registered 3D model The one-way mismatch degree is represented by the symbol express, stipulate The maximum one-way mismatch between the contour curve of the registered 2D image and the contour curve of the registered 3D model projection is defined as Define the contour points of the projection contour curve of the 3D model after registration To the matching point in the contour curve of the registered 2D image The one-way mismatch degree is represented by the symbol express, stipulate The maximum one-way mismatch between the projection contour curve of the registered 3D model and the contour curve of the registered 2D image is defined as Define the maximum mismatch between the contour curve of the registered 2D image and the contour curve of the registered 3D model projection The upper limit of the maximum mismatch σ is specified as σ max ;

[0018] Specifically:

[0019] If σ≤σ max If established, jump to step 3;

[0020] If σ≤σ maxIf it does not hold, the image registration error evaluation ends, the maximum mismatch between the output registered 2D image contour curve and the registered 3D model projection contour curve exceeds the tolerance range, the evaluation fails, and the output value of σ;

[0021] Step 3: Calculate the center offset between the contour of the registered 2D image and the contour of the 3D model projection in the registered pose:

[0022] Define the minimum value of the y-axis of the contour point of the 2D image contour curve after registration The maximum value of the y-axis of the contour point of the contour curve of the two-dimensional image after registration by As a starting point, For the end point, Eleven line segments perpendicular to the y-axis are established for the step size to divide the contour curve of the registered two-dimensional image into 10 regions. The region close to the starting point is region 1, and the region close to the end point is region 10. The lower limit of the contour point screening threshold of the contour curve of the registered two-dimensional image is defined as The upper limit of the screening threshold of the contour point in the contour curve area of the two-dimensional image after registration is The contour point set of the contour curve area of the two-dimensional image after registration is defined as Where 1≤a≤10; the coordinates of the center point of the contour curve area a of the two-dimensional image after registration are defined as

[0023] Define the minimum y-axis value of the contour point of the projection contour curve of the 3D model after registration The maximum value of the y-axis of the contour point of the projection contour curve of the 3D model after registration by As a starting point, For the end point, For the step size, 11 line segments perpendicular to the y-axis are established to divide the projection contour curve of the registered 3D model into 10 regions. The region close to the starting point is region 1, and the region close to the end point is region 10. The lower limit of the contour point screening threshold of the projection contour curve of the registered 3D model is defined as The upper limit of the screening threshold for contour points in the projection contour curve area of the three-dimensional model after registration is The contour point set of the projection contour curve area of the 3D model after registration is defined as Among them, 1≤a≤10; the coordinates of the center point of the projection contour curve area a of the 3D model after registration are defined as

[0024] Define the top offset between the registered 2D image contour curve and the registered 3D model projection contour curve as The bottom offset between the registered 2D image contour curve and the registered 3D model projection contour curve is Top offset etop and bottom offset e bot The upper limit value is set to e max ; Define the center offset between the registered 2D image contour curve and the registered 3D model projection contour curve area a as Center offset The upper limit value is set as

[0025] a) Judge e top ≤e max Whether it is established, specifically:

[0026] If e top ≤e max If established, jump to step 3b);

[0027] If e top ≤e max If it is not established, the image registration error evaluation ends, the output of the registered two-dimensional image contour curve and the registered three-dimensional model projection contour curve top offset exceeds the tolerance range, the evaluation is unqualified, and the output is e top The value of

[0028] b) Judge e bot ≤e max Whether it is established, specifically:

[0029] If e bot ≤e max If established, jump to step 3c);

[0030] If e bot ≤e max If it is not established, the image registration error evaluation ends, the output of the registered two-dimensional image contour curve and the registered three-dimensional model projection contour curve bottom offset exceeds the tolerance range, the evaluation is unqualified, and the output is e bot The value of

[0031] c) Let a=1, i=1, α=1, j=1, δ=1, Jump to step 3d);

[0032] d) Judgment Whether it is established, specifically:

[0033] like If established, jump to step 3e);

[0034] like Not true, set i=i+1 and jump to step 3f);

[0035] e) Order Will Deposit Point Set In the example, let α = α + 1, i = i + 1, and jump to step 3 f);

[0036] f) Determine whether i≤n is true, specifically:

[0037] If i≤n holds, jump to step 3d);

[0038] If i≤n is not true, jump to step 3g);

[0039] g) Calculate the center point coordinates of the contour curve area a of the two-dimensional image after registration Jump to step 3h);

[0040] h) Judgment Whether it is established, specifically:

[0041] like If established, jump to step 3i);

[0042] like Not true, set j = j + 1, jump to step 3 j);

[0043] i) Order Will Deposit Point Set In the example, let δ = δ + 1, j = j + 1, and jump to step 3 j);

[0044] j) Determine whether j≤m is true, specifically:

[0045] If j≤m holds, jump to step 3h);

[0046] If j≤m does not hold, jump to step 3k);

[0047] k) Calculate the center point coordinates of the projection contour curve area a of the 3D model after registration Jump to step 31);

[0048] l) Calculate the center deviation between the contour curve of the registered two-dimensional image and the contour curve area a of the registered three-dimensional model projection Jump to step 3m);

[0049] m) Judgment Whether it is established, specifically:

[0050] like If established, jump to step 3n);

[0051] like If it is not established, the image registration error evaluation ends, and the center deviation of the output registered 2D image contour curve and the registered 3D model projection contour curve area a exceeds the tolerance range, the evaluation fails, and the output The value of

[0052] n) Order Jump to step 3o);

[0053] o) Judgment Whether it is established, specifically:

[0054] like If yes, set i=1, α=1, j=1, δ=1, a=a+1, and jump to step 3d);

[0055] like If it does not hold, the image registration error evaluation ends, the output image registration is qualified, the registration accuracy meets the requirements, and the output σ,e top , e bot , value.

[0056] The beneficial effects of the present invention are:

[0057] 1. Based on the theory that contour points with large curvature are important feature points in image registration, the present invention calculates the curvature of each contour point of the contour curve and uses the curvature as a compensation parameter to calculate the unidirectional mismatch respectively. The overall shape similarity of the contour curves of the two images after image registration is evaluated by calculating the maximum mismatch.

[0058] 2. When evaluating image registration errors, the present invention proposes the concept of center offset between the contour of the registered 2D image and the contour of the 3D model projection at the registered position, based on the principle that the shape features of the two contour curves at the same spatial position after registration should be the same. In order to eliminate the influence caused by the inconsistency of the number of contour points of the two contour curves in the same area, the concept of center offset between the registered 2D image contour and the 3D model projection at the registered position is introduced. Compensation parameter calculation: the center deviation between the contour curve of the two-dimensional image after registration and the contour curve area of the three-dimensional model after registration. Center offset The smaller the value, the more similar the shapes of the two contour curves are and the smaller the distance between them is.

[0059] 3. Compared with the invention patent "A method for evaluating the registration error of closed contour images with low curvature based on the error area ratio of circular domains" applied by the inventor on the same day, although both methods are used for the evaluation of the registration error of two-dimensional images and three-dimensional model images, the method mentioned in "A method for evaluating the registration error of closed contour images with low curvature based on the error area ratio of circular domains" is based on the premise that the registered image contour is closed and has low curvature, and then the error area ratio of the closed area is calculated by dynamically dividing the circular domain; this method considers the influence of curvature on the reliability of neighboring point matching by calculating the maximum mismatch and center offset, and takes into account both local and global evaluation effects; therefore, the proposal of this method compensates for the other method, thereby improving the series of methods for evaluating the registration error of two-dimensional images and three-dimensional model images.

[0060] 4. Compared with the invention patent "A method for evaluating the registration error of high-curvature contour images based on enclosing domain threshold" applied by the inventor on the same day, although both methods are used for the evaluation of the registration error of two-dimensional images and three-dimensional model images, the method mentioned in "A method for evaluating the registration error of high-curvature contour images based on enclosing domain threshold" is based on the premise that the registered image contour has high curvature. By calculating the proximity distance of high-curvature contour points, the proximity of projected contours and the area of error evaluation enclosing domain, it focuses on the error evaluation of local curvature mutation areas; this method considers the overall shape characteristics of the contour curve by calculating the maximum mismatch and center offset, taking into account both local and global evaluation effects; therefore, the proposal of this method compensates for the other method, thereby improving the series of methods for evaluating the registration error of two-dimensional images and three-dimensional model images. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] For ease of explanation, the present invention is described in detail with reference to the following specific implementations and accompanying drawings.

[0062] Figure 1 This is a flow chart of a method for evaluating the registration error between a two-dimensional image and a three-dimensional model image based on contour curve shape features;

[0063] Figure 2 Schematic diagram of the registration between the actual 2D image and the 3D model image in Example 2;

[0064] Figure 3 Schematic diagram of implementation example 2 of a method for evaluating registration errors between a two-dimensional image and a three-dimensional model image based on contour curve shape features;

[0065] Figure 4 Schematic diagram of the registration between the actual 2D image and the 3D model image in Example 3;

[0066] Figure 5 Schematic diagram of implementation example 3 of a method for evaluating registration errors between a two-dimensional image and a three-dimensional model image based on contour curve shape features; DETAILED DESCRIPTION

[0067] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is described below through the specific embodiments shown in the accompanying drawings. However, it should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present invention.

[0068] Example 1: Figure 1 、 Figure 2 、 Figure 3 、 Figure 4 、 Figure 5 As shown, this specific embodiment adopts the following technical solution: a method for evaluating the registration error between a two-dimensional image and a three-dimensional model image based on the shape characteristics of a contour curve. The specific implementation process of the method is as follows:

[0069] Step 1: Import the registered 2D image contour point set and 3D model projection contour point set:

[0070] The lower left corner of the two-dimensional image is used as the coordinate origin, the right is the positive direction of the x-axis, the upward is the positive direction of the y-axis, and according to the right-hand rule, the vertical screen outward is the positive direction of the z-axis to establish a three-dimensional image registration error evaluation coordinate system w, obtain the contour curve information of the registered two-dimensional image, and input the contour point set of the registered two-dimensional image contour curve is the position information of the i-th contour point of the registered 2D image contour curve in the o-xy plane of the 3D image registration error evaluation coordinate system w. The value range of i is 1≤i≤n, where, is the x-axis coordinate of the i-th contour point of the registered 2D image contour curve located in the o-xy plane of the 3D image registration error evaluation coordinate system w, The y-axis coordinate of the i-th contour point of the registered two-dimensional image contour curve is located in the o-xy plane of the three-dimensional image registration error evaluation coordinate system w; the registered three-dimensional model is parallel projected onto the o-xy plane along the negative direction of the z-axis of the three-dimensional image registration error evaluation coordinate system w to obtain the projection contour curve information, and the contour point set of the registered three-dimensional model projection contour curve is input. is the position information of the jth contour point of the projection contour curve of the registered 3D model in the o-xy plane of the 3D image registration error evaluation coordinate system w. The value range of j is 1≤j≤m, where The jth contour point of the projection contour curve of the registered 3D model is located in the x-axis coordinate of the o-xy plane of the 3D image registration error evaluation coordinate system w. The jth contour point of the projection contour curve of the registered 3D model is located on the y-axis coordinate of the o-xy plane of the 3D image registration error evaluation coordinate system w;

[0071] Step 2: Calculate the maximum mismatch between the contour of the registered 2D image and the contour of the 3D model projection in the registered pose:

[0072] Use the K nearest neighbor method to find the contour points of the registered 2D image contour curve in the contour points of the registered 3D model projection contour curve. matching points Specifically: the contour points of the contour curve of the two-dimensional image after registration For query points, set the parameter K value to 1. The value of parameter K represents the number of neighboring points to be found near the query point. Setting the parameter K value to 1 means that the query point is found in the contour points of the projection contour curve of the 3D model after registration. The closest point to the Euclidean distance point That is, the contour point of the contour curve of the two-dimensional image after registration Matching point; the i-th contour point of the contour curve of the two-dimensional image after registration Matching points with the contour points of the projected contour curve of the registered 3D model The distance, using the symbol express,

[0073] Use the K nearest neighbor method to find the contour points of the registered 3D model projection contour curve in the contour points of the registered 2D image contour curve. matching points Specifically: Project the contour curve contour points of the registered 3D model For the query point, set the parameter K value to 1. The value of parameter K represents the number of neighboring points to be found near the query point. Setting the parameter K value to 1 means that the contour points of the contour curve of the two-dimensional image after registration are searched for the query point. The closest point to the Euclidean distance point That is the contour point of the projection contour curve of the three-dimensional model after registration The matching point of the 3D model projection contour curve after registration is the kth contour point Matching points with the contour points of the registered 2D image contour curve The distance, using the symbol express,

[0074] Calculate the curvature of the contour curve of the two-dimensional image after registration at the i-th contour point, and select Five contour points, when i=1 or i=2, select Five contour points, when i=n-1 or i=n, select Five contour points, using the least squares method to fit an expression f Tw (x) = a i0 +a i1 x+a i2 x 2 +a i3 x 3 +a i4 x 4 The polynomial curve is used to approximate the five selected contour points;

[0075]

[0076] The curvature of the i-th contour point of the registered two-dimensional image contour curve is

[0077] Calculate the curvature of the kth contour point of the projection contour curve of the 3D model after registration, and select Five contour points, when k = 1 or k = 2, select Five contour points, when k=m-1 or k=m, select Five contour points, using the least squares method to fit an expression f Th (x) = a k0 +a k1 x+a k2 x 2 +a k3 x 3 +a k4 x 4 The polynomial curve is used to approximate the five selected contour points;

[0078]

[0079] The curvature of the kth contour point of the projection contour curve of the 3D model after registration is

[0080] Define the contour points of the 2D image contour curve after registration To the matching points in the projection contour curve of the registered 3D model The one-way mismatch degree is represented by the symbol express, stipulate The maximum one-way mismatch between the contour curve of the registered 2D image and the contour curve of the registered 3D model projection is defined as Define the contour points of the projection contour curve of the 3D model after registration To the matching point in the contour curve of the registered 2D image The one-way mismatch degree is represented by the symbol express, stipulate The maximum one-way mismatch between the projection contour curve of the registered 3D model and the contour curve of the registered 2D image is defined as Define the maximum mismatch between the contour curve of the registered 2D image and the contour curve of the registered 3D model projection The upper limit of the maximum mismatch σ is specified as σ max ;

[0081] Specifically:

[0082] If σ≤σ max If established, jump to step 3;

[0083] If σ≤σ max If it does not hold, the image registration error evaluation ends, the maximum mismatch between the output registered 2D image contour curve and the registered 3D model projection contour curve exceeds the tolerance range, the evaluation fails, and the output value of σ;

[0084] Step 3: Calculate the center offset between the contour of the registered 2D image and the contour of the 3D model projection in the registered pose:

[0085] Define the minimum value of the y-axis of the contour point of the 2D image contour curve after registration The maximum value of the y-axis of the contour point of the contour curve of the two-dimensional image after registration by As a starting point, For the end point, Eleven line segments perpendicular to the y-axis are established for the step size to divide the contour curve of the registered two-dimensional image into 10 regions. The region close to the starting point is region 1, and the region close to the end point is region 10. The lower limit of the contour point screening threshold of the contour curve of the registered two-dimensional image is defined as The upper limit of the screening threshold of the contour point in the contour curve area of the two-dimensional image after registration is The contour point set of the contour curve area of the two-dimensional image after registration is defined as Where 1≤a≤10; the coordinates of the center point of the contour curve area a of the two-dimensional image after registration are defined as

[0086] Define the minimum value of the y-axis of the contour point of the projection contour curve of the 3D model after registration The maximum value of the y-axis of the contour point of the projection contour curve of the 3D model after registration by As a starting point, For the end point, For the step size, 11 line segments perpendicular to the y-axis are established to divide the projection contour curve of the registered 3D model into 10 regions. The region close to the starting point is region 1, and the region close to the end point is region 10. The lower limit of the contour point screening threshold of the projection contour curve of the registered 3D model is defined as The upper limit of the screening threshold for contour points in the projection contour curve area of the three-dimensional model after registration is The contour point set of the projection contour curve area of the 3D model after registration is defined as Where 1≤a≤10; the coordinates of the center point of the projection contour curve area a of the 3D model after registration are defined as

[0087] Define the top offset between the registered 2D image contour curve and the registered 3D model projection contour curve as The bottom offset between the registered 2D image contour curve and the registered 3D model projection contour curve is Top offset e top and bottom offset e bot The upper limit value is set to e max ; Define the center offset between the registered 2D image contour curve and the registered 3D model projection contour curve area a as Center offset The upper limit value is set as

[0088] a) Judge e top ≤e max Whether it is established, specifically:

[0089] If e top ≤e max If established, jump to step 3b);

[0090] If e top ≤e max If it is not established, the image registration error evaluation ends, the output of the registered two-dimensional image contour curve and the registered three-dimensional model projection contour curve top offset exceeds the tolerance range, the evaluation is unqualified, and the output is e top The value of

[0091] b) Judge e bot ≤e max Whether it is established, specifically:

[0092] If e bot ≤e max If established, jump to step 3c);

[0093] If e bot ≤e maxIf it is not established, the image registration error evaluation ends, the output of the registered two-dimensional image contour curve and the registered three-dimensional model projection contour curve bottom offset exceeds the tolerance range, the evaluation is unqualified, and the output is e bot The value of

[0094] c) Let a=1, i=1, α=1, j=1, δ=1, Jump to step 3d);

[0095] d) Judgment Whether it is established, specifically:

[0096] like If established, jump to step 3e);

[0097] like Not true, set i=i+1 and jump to step 3f);

[0098] e) Order Will Deposit Point Set In the example, let α = α + 1, i = i + 1, and jump to step 3 f);

[0099] f) Determine whether i≤n is true, specifically:

[0100] If i≤n holds, jump to step 3d);

[0101] If i≤n is not true, jump to step 3g);

[0102] g) Calculate the center point coordinates of the contour curve area a of the two-dimensional image after registration Jump to step 3h);

[0103] h) Judgment Whether it is established, specifically:

[0104] like If established, jump to step 3i);

[0105] like Not true, set j = j + 1, jump to step 3 j);

[0106] i) Order Will Deposit Point Set In the example, let δ = δ + 1, j = j + 1, and jump to step 3 j);

[0107] j) Determine whether j≤m is true, specifically:

[0108] If j≤m holds, jump to step 3h);

[0109] If j≤m does not hold, jump to step 3k);

[0110] k) Calculate the center point coordinates of the projection contour curve area a of the 3D model after registration Jump to step 31);

[0111] l) Calculate the center deviation between the contour curve of the registered two-dimensional image and the contour curve area a of the registered three-dimensional model projection Jump to step 3m);

[0112] m) Judgment Whether it is established, specifically:

[0113] like If established, jump to step 3n);

[0114] like If it is not established, the image registration error evaluation ends, and the center deviation of the output registered 2D image contour curve and the registered 3D model projection contour curve area a exceeds the tolerance range, the evaluation fails, and the output The value of

[0115] n) Order Jump to step 3o);

[0116] o) Judgment Whether it is established, specifically:

[0117] like If yes, set i=1, α=1, j=1, δ=1, a=a+1, and jump to step 3d);

[0118] like If it does not hold, the image registration error evaluation ends, the output image registration is qualified, the registration accuracy meets the requirements, and the output σ,e top , e bot , value.

[0119] Implementation Example 2: Figure 3 As shown, the contour point set P of the two-dimensional image contour curve after input registration Tw , point set P Tw Contains n = 586 contour points; Input the contour point set P of the 3D model projection contour curve after registration Th , point set P Th Contains m = 724 contour points; specify σ max = 0.5; Use the K nearest neighbor method to find the contour points of the registered 2D image contour curve in the contour points of the registered 3D model projection contour curve matching points Calculate the distance between the i-th contour point of the registered 2D image contour curve and the matching point in the contour point set of the registered 3D model projection contour curve Calculate the curvature of the i-th contour point of the registered two-dimensional image contour curve Calculate the contour points of the 2D image contour curve after registration To the matching points in the projection contour curve of the registered 3D model One-way mismatch The maximum one-way mismatch between the contour curve of the registered two-dimensional image and the projection contour curve of the registered three-dimensional model is calculated as follows: Use the K nearest neighbor method to find the contour points of the registered 3D model projection contour curve in the contour points of the registered 2D image contour curve. matching points Calculate the distance between the kth contour point of the registered 3D model projection contour curve and the matching point in the contour point set of the registered 2D image contour curve Calculate the curvature of the kth contour point of the projection contour curve of the 3D model after registration Calculate the contour points of the projection contour curve of the 3D model after registration To the matching point in the contour curve of the registered 2D image One-way mismatch The maximum one-way mismatch between the projection contour curve of the registered 3D model and the contour curve of the registered 2D image is calculated as The maximum mismatch between the contour curve of the registered two-dimensional image and the contour curve of the registered three-dimensional model projection is σ=0.312; judge σ≤σ max Established, jump to step 3;

[0120] Regulation e max =0.2, Minimum value of the y-axis of the contour point of the 2D image contour curve after registration The maximum value of the y-axis of the contour point of the contour curve of the two-dimensional image after registration by As a starting point, As the end point, l Tw =0.893 as the step size, establish 11 line segments perpendicular to the y-axis to divide the contour curve of the registered 2D image into 10 areas, the area close to the starting point is area 1, and the area close to the end point is area 10; the minimum value of the contour point on the y-axis of the projection contour curve of the 3D model after registration The maximum value of the y-axis of the contour point of the projection contour curve of the 3D model after registration by As a starting point, As the end point, l Th=0.868 as the step size, establish 11 line segments perpendicular to the y-axis to divide the projection contour curve of the registered 3D model into 10 regions, the region close to the starting point is region 1, and the region close to the end point is region 10; calculate the offset between the top of the registered 2D image contour curve and the registered 3D model projection contour curve as e top =0.149; the bottom offset between the registered 2D image contour curve and the registered 3D model projection contour curve is e bot =0.097; judge e top ≤e max Established, e bot ≤e max Established; let a=1, i=1, α=1, j=1, δ=1,

[0121] judge Not true, let i = 2; judge i ≤ 586 is true; judge Not true, let i = 3; judge i ≤ 586 is true; when i = 213, judge Establish, order Will Deposit Point Set In the example, let α = 2, i = 214, and judge that i ≤ 586 is true; repeat the above steps until i = 587, and judge that i ≤ 586 is not true; calculate the coordinates of the center point of the contour curve area 1 of the two-dimensional image after registration

[0122] judge Not true, let j = 2; judge j ≤ 724 is true; judge Not true, let j = 3; judge j ≤ 724 is true; when j = 326, judge Establish, order Will Deposit Point Set In the example, let δ = 2, j = 327, and judge that j ≤ 724 is true; repeat the above steps until j = 725, and judge that j ≤ 724 is not true; calculate the coordinates of the center point of the projection contour curve area 1 of the 3D model after registration The center offset between the registered 2D image contour curve and the registered 3D model projection contour curve area 1 is calculated as follows: judge Establish, order judge Establish, let i = 1, α = 1, j = 1, δ = 1, a = 2; repeat the above steps until If it is not established, the image registration error evaluation ends; the image registration is qualified and the registration accuracy meets the requirements, σ=0.312, e top =0.149,e bot=0.097,

[0123] Implementation Example 3: Figure 5 As shown, the contour point set P of the two-dimensional image contour curve after input registration Tw , point set P Tw Contains n = 682 contour points; Input the contour point set P of the 3D model projection contour curve after registration Th , point set P Th Contains m = 835 contour points; specify σ max = 0.5; Use the K nearest neighbor method to find the contour points of the registered 2D image contour curve in the contour points of the registered 3D model projection contour curve matching points Calculate the distance between the i-th contour point of the registered 2D image contour curve and the matching point in the contour point set of the registered 3D model projection contour curve Calculate the curvature of the i-th contour point of the registered two-dimensional image contour curve Calculate the contour points of the 2D image contour curve after registration To the matching points in the projection contour curve of the registered 3D model One-way mismatch The maximum one-way mismatch between the contour curve of the registered two-dimensional image and the projection contour curve of the registered three-dimensional model is calculated as follows: Use the K nearest neighbor method to find the contour points of the registered 3D model projection contour curve in the contour points of the registered 2D image contour curve. matching points Calculate the distance between the kth contour point of the registered 3D model projection contour curve and the matching point in the contour point set of the registered 2D image contour curve Calculate the curvature of the kth contour point of the projection contour curve of the 3D model after registration Calculate the contour points of the projection contour curve of the 3D model after registration To the matching point in the contour curve of the registered 2D image One-way mismatch The maximum one-way mismatch between the projection contour curve of the registered 3D model and the contour curve of the registered 2D image is calculated as The maximum mismatch between the contour curve of the registered two-dimensional image and the contour curve of the registered three-dimensional model projection is σ=0.305; judge σ≤σ max Established, jump to step 3;

[0124] Regulation e max =0.2, Minimum value of the y-axis of the contour point of the 2D image contour curve after registration The maximum value of the y-axis of the contour point of the contour curve of the two-dimensional image after registration by As a starting point, As the end point, l Tw = 0.583 as the step size, establish 11 line segments perpendicular to the y-axis to divide the contour curve of the registered 2D image into 10 areas, the area close to the starting point is area 1, and the area close to the end point is area 10; the minimum value of the contour point on the y-axis of the projection contour curve of the 3D model after registration The maximum value of the y-axis of the contour point of the projection contour curve of the 3D model after registration by As a starting point, As the end point, l Th =0.589 as the step size, establish 11 line segments perpendicular to the y-axis to divide the projection contour curve of the registered 3D model into 10 regions, the region close to the starting point is region 1, and the region close to the end point is region 10; calculate the offset between the top of the registered 2D image contour curve and the registered 3D model projection contour curve as e top =0.053; the bottom offset between the registered 2D image contour curve and the registered 3D model projection contour curve is e bot =0.009; judge e top ≤e max Established, e bot ≤e max Established; let a=1, i=1, α=1, j=1, δ=1,

[0125] judge Not true, let i = 2; judge i ≤ 682 is true; judge Not true, let i = 3; judge i ≤ 586 is true; when i = 326, judge Establish, order Will Deposit Point Set In the example, let α = 2, i = 327, and judge that i ≤ 682 is true; repeat the above steps until i = 683, and judge that i ≤ 682 is not true; calculate the coordinates of the center point of the contour curve area 1 of the two-dimensional image after registration

[0126] judge Not true, let j = 2; judge j ≤ 835 is true; judge Not true, let j = 3; judge j ≤ 835 is true; when j = 411, judge Establish, order Will Deposit Point Set In the example, let δ = 2, j = 412, and judge that j ≤ 835 is true; repeat the above steps until j = 836, and judge that j ≤ 835 is not true; calculate the coordinates of the center point of the projection contour curve area 1 of the 3D model after registration The center offset between the registered 2D image contour curve and the registered 3D model projection contour curve area 1 is calculated as follows: judge Establish, order judge Establish, let i = 1, α = 1, j = 1, δ = 1, a = 2; repeat the above steps until If it is not established, the image registration error evaluation ends; the image registration is qualified and the registration accuracy meets the requirements, σ=0.305, e top =0.053,e bot =0.009,

Claims

1. A method for evaluating the registration error between a two-dimensional image and a three-dimensional model image based on contour curve shape features, characterized by: The specific implementation process of the method is as follows: Step 1: Import the registered 2D image contour point set and 3D model projection contour point set: The lower left corner of the two-dimensional image is used as the coordinate origin, the right is the positive direction of the x-axis, the upward is the positive direction of the y-axis, and according to the right-hand rule, the vertical screen outward is the positive direction of the z-axis to establish a three-dimensional image registration error evaluation coordinate system w, obtain the contour curve information of the registered two-dimensional image, and input the contour point set of the registered two-dimensional image contour curve is the position information of the i-th contour point of the registered 2D image contour curve in the o-xy plane of the 3D image registration error evaluation coordinate system w. The value range of i is 1≤i≤n, where, is the x-axis coordinate of the i-th contour point of the registered 2D image contour curve located in the o-xy plane of the 3D image registration error evaluation coordinate system w, The y-axis coordinate of the i-th contour point of the registered two-dimensional image contour curve is located in the o-xy plane of the three-dimensional image registration error evaluation coordinate system w; the registered three-dimensional model is parallel projected onto the o-xy plane along the negative direction of the z-axis of the three-dimensional image registration error evaluation coordinate system w to obtain the projection contour curve information, and the contour point set of the registered three-dimensional model projection contour curve is input. is the position information of the jth contour point of the projection contour curve of the registered 3D model in the o-xy plane of the 3D image registration error evaluation coordinate system w. The value range of j is 1≤j≤m, where The jth contour point of the projection contour curve of the registered 3D model is located in the x-axis coordinate of the o-xy plane of the 3D image registration error evaluation coordinate system w. The jth contour point of the projection contour curve of the registered 3D model is located on the y-axis coordinate of the o-xy plane of the 3D image registration error evaluation coordinate system w; Step 2: Calculate the maximum mismatch between the contour of the registered 2D image and the contour of the 3D model projection in the registered pose: Use the K nearest neighbor method to find the contour points of the registered 2D image contour curve in the contour points of the registered 3D model projection contour curve. matching points Specifically: the contour points of the contour curve of the two-dimensional image after registration For query points, set the parameter K value to 1. The value of parameter K represents the number of neighboring points to be found near the query point. Setting the parameter K value to 1 means that the query point is found in the contour points of the projection contour curve of the 3D model after registration. The closest point to the Euclidean distance point That is, the contour point of the contour curve of the two-dimensional image after registration Matching point; the i-th contour point of the contour curve of the two-dimensional image after registration Matching points with the contour points of the projected contour curve of the registered 3D model The distance, using the symbol express, Use the K nearest neighbor method to find the contour points of the registered 3D model projection contour curve in the contour points of the registered 2D image contour curve. matching points Specifically: Project the contour curve contour points of the registered 3D model For the query point, set the parameter K value to 1. The value of parameter K represents the number of neighboring points to be found near the query point. Setting the parameter K value to 1 means that the contour points of the contour curve of the two-dimensional image after registration are searched for the query point. The closest point to the Euclidean distance point That is the contour point of the projection contour curve of the three-dimensional model after registration The matching point of the 3D model projection contour curve after registration is the kth contour point Matching points with the contour points of the registered 2D image contour curve The distance, using the symbol express, Calculate the curvature of the contour curve of the two-dimensional image after registration at the i-th contour point, and select Five contour points, when i=1 or i=2, select Five contour points, when i=n-1 or i=n, select Five contour points, using the least squares method to fit an expression f Tw (x) = a i0 +a i1 x+a i2 x 2 +a i3 x 3 +a i4 x 4 The polynomial curve is used to approximate the five selected contour points; The curvature of the i-th contour point of the registered two-dimensional image contour curve is Calculate the curvature of the kth contour point of the projection contour curve of the 3D model after registration, and select Five contour points, when k = 1 or k = 2, select Five contour points, when k=m-1 or k=m, select Five contour points, using the least squares method to fit an expression f Th (x) = a k0 +a k1 x+a k2 x 2 +a k3 x 3 +a k4 x 4 The polynomial curve is used to approximate the five selected contour points; The curvature of the kth contour point of the projection contour curve of the 3D model after registration is Define the contour points of the 2D image contour curve after registration To the matching points in the projection contour curve of the registered 3D model The one-way mismatch degree is represented by the symbol express, stipulate The maximum one-way mismatch between the contour curve of the registered 2D image and the contour curve of the registered 3D model projection is defined as Define the contour points of the projection contour curve of the 3D model after registration To the matching point in the contour curve of the registered 2D image The one-way mismatch degree is represented by the symbol express, stipulate The maximum one-way mismatch between the projection contour curve of the registered 3D model and the contour curve of the registered 2D image is defined as Define the maximum mismatch between the contour curve of the registered 2D image and the contour curve of the registered 3D model projection The upper limit of the maximum mismatch σ is specified as σ max ; Specifically: If σ≤σ max If established, jump to step 3; If σ≤σ max If it does not hold, the image registration error evaluation ends, the maximum mismatch between the output registered 2D image contour curve and the registered 3D model projection contour curve exceeds the tolerance range, the evaluation fails, and the output value of σ; Step 3: Calculate the center offset between the contour of the registered 2D image and the contour of the 3D model projection in the registered pose: Define the minimum value of the y-axis of the contour point of the 2D image contour curve after registration The maximum value of the y-axis of the contour point of the contour curve of the two-dimensional image after registration by As a starting point, For the end point, Eleven line segments perpendicular to the y-axis are established for the step size to divide the contour curve of the registered two-dimensional image into 10 regions. The region close to the starting point is region 1, and the region close to the end point is region 10. The lower limit of the contour point screening threshold of the contour curve of the registered two-dimensional image is defined as The upper limit of the screening threshold of the contour point in the contour curve area of the two-dimensional image after registration is The contour point set of the contour curve area of the two-dimensional image after registration is defined as Where 1≤a≤10; the coordinates of the center point of the contour curve area a of the two-dimensional image after registration are defined as Define the minimum value of the y-axis of the contour point of the projection contour curve of the 3D model after registration The maximum value of the y-axis of the contour point of the projection contour curve of the 3D model after registration by As a starting point, For the end point, For the step size, 11 line segments perpendicular to the y-axis are established to divide the projection contour curve of the registered 3D model into 10 regions. The region close to the starting point is region 1, and the region close to the end point is region 10. The lower limit of the contour point screening threshold of the projection contour curve of the registered 3D model is defined as The upper limit of the screening threshold for contour points in the projection contour curve area of the three-dimensional model after registration is The contour point set of the projection contour curve area of the 3D model after registration is defined as Where 1≤a≤10; the coordinates of the center point of the projection contour curve area a of the 3D model after registration are defined as Define the top offset between the registered 2D image contour curve and the registered 3D model projection contour curve as The bottom offset between the registered 2D image contour curve and the registered 3D model projection contour curve is Top offset e top and bottom offset e bot The upper limit value is set to e max ; Define the center offset between the registered 2D image contour curve and the registered 3D model projection contour curve area a as Center offset The upper limit value is set as a) Judge e top ≤e max Whether it is established, specifically: If e top ≤e max If established, jump to step 3b); If e top ≤e max If it is not established, the image registration error evaluation ends, the output of the registered two-dimensional image contour curve and the registered three-dimensional model projection contour curve top offset exceeds the tolerance range, the evaluation is unqualified, and the output is e top The value of b) Judge e bot ≤e max Whether it is established, specifically: If e bot ≤e max If established, jump to step 3c); If e bot ≤e max If it is not established, the image registration error evaluation ends, the output of the registered two-dimensional image contour curve and the registered three-dimensional model projection contour curve bottom offset exceeds the tolerance range, the evaluation is unqualified, and the output is e bot The value of c) Let a=1, i=1, α=1, j=1, δ=1, Jump to step 3d); d) Judgment Whether it is established, specifically: like If established, jump to step 3e); like Not true, set i=i+1, jump to step 3f) ； e) Order Will Deposit Point Set In the equation, let α = α + 1 ， i=i+1, jump to step 3f) ； f) Determine whether i≤n is true, specifically: If i≤n holds, jump to step 3d) ； If i≤n is not true, jump to step 3g) ； g) Calculate the center point coordinates of the contour curve area a of the two-dimensional image after registration Skip to step 3h) ； h) Judgment Whether it is established, specifically: like If established, jump to step 3i) ； like Not true, set j = j + 1, jump to step 3) ； i) Order Will Deposit Point Set In the equation, let δ = δ + 1 ， j=j+1, jump to step 3j) ； j) Determine whether j≤m is true, specifically: If j≤m holds, jump to step 3h) ； If j≤m is not true, jump to step 3k) ； k) Calculate the center point coordinates of the projection contour curve area a of the 3D model after registration Skip to step 3) ； l) Calculate the center deviation between the contour curve of the registered two-dimensional image and the contour curve area a of the registered three-dimensional model projection Jump to step 3m); m) Judgment Whether it is established, specifically: like If established, jump to step 3n); like If it is not established, the image registration error evaluation ends, and the center deviation of the output registered 2D image contour curve and the registered 3D model projection contour curve area a exceeds the tolerance range, the evaluation fails, and the output The value of n) Order Jump to step 3o); o) Judgment Whether it is established, specifically: like If yes, set i=1, α=1, j=1, δ=1, a=a+1, and jump to step 3d); like If it does not hold, the image registration error evaluation ends, the output image registration is qualified, the registration accuracy meets the requirements, and the output σ,e top , e bot , value.

Images