Image Registration Method Based on Improved ORB Feature Extraction and Grid Feature Matching

Abstract

The present invention discloses an image registration method based on improved ORB feature extraction and grid feature matching, which includes the following steps: input the image to be extracted; use the improved FAST algorithm to extract the feature points of the image to be stitched; establish a BRIEF descriptor and obtain the BRIEF descriptor with direction; according to the set of feature points of the image to be stitched, establish a Delaunay triangle grid with the feature points as vertices; judge the triangle similarity and obtain the similar triangles composed of feature points; use the Hamming distance to pair the vertices of the similar triangles to obtain the best matching feature point pairs; use the SVD method to obtain the affine matrix parameters; the image registration method proposed by the present invention can improve the feature extraction efficiency and extraction accuracy while ensuring the image quality, reduce the false matching and improve the stability. It not only improves the real-time performance of video stitching, but also reduces the waste of memory resources, lowers the equipment cost, greatly improves the work efficiency, and provides a better method for image stitching.

Description

Technical Field

[0001] The present invention relates to the technical field of image registration methods, and more particularly, to an image registration method based on improved ORB feature extraction and grid feature matching. Background Art

[0002] Image registration is the core and key technology in image stitching technology, and the registration accuracy can directly affect the final stitching effect. Image registration refers to the spatial alignment of two or more images, which is a mapping of images in space and intensity. Accurately finding the position of the overlapping area between two adjacent images, finding a suitable transformation model, and determining the transformation relationship between images are the keys to image registration. Among them, the image feature information is diverse, and thus there are many feature extraction methods: straight line segments, feature structures, feature points (including corner points, high curvature points, etc.), edges, closed regions, and statistical features, etc. At present, there are many image registration methods, but due to different application environments and the diversity of images to be stitched, many factors between images need to be considered. With the development of technology and society, the requirements for the real-time performance and computational efficiency of panoramic image stitching technology are getting higher and higher. Most traditional algorithms use Harris corner operators, SIFT operators, and SURF operators. However, the extraction accuracy and efficiency of traditional algorithms are relatively low, the computational complexity is huge, and they cannot meet the requirements of real-time work, and the image quality is poor. For the feature extraction and matching of some high-resolution digital images, the computational complexity is huge, which seriously affects the stitching efficiency, cannot meet the requirements of real-time work, and occupies a large amount of CPU resources.

[0003] At the ICCV in 2011, Ethan Rublee and Vincent Rabaud proposed the ORB algorithm, which is a new type of efficient local feature descriptor in binary form. The ORB algorithm is divided into two steps. First, the improved FAST algorithm is used to detect feature points, and then the improved BRIEF algorithm is used to describe the feature points. Since both the FAST and BRIEF algorithms have the advantage of fast calculation speed, the ORB algorithm can efficiently and quickly complete the extraction of feature points during the feature extraction process. At the same time, the feature extraction of the ORB algorithm also has rotational invariance and reduces the sensitivity to noise. In view of the problems of slow extraction speed and poor accuracy caused by the huge number of image feature points extracted by traditional methods, further improvement is needed. Therefore, the development direction of image stitching technology should not only pay attention to the accuracy of image matching, but also adopt methods with relatively low computational complexity and high efficiency in extracting features and matching information, greatly reducing the waste of CPU resources and improving the real-time performance of stitching.

[0004] There are several problems in the feature extraction process based on ORB. First, the number of feature points extracted by the FAST algorithm is relatively large, and the phenomenon of distribution clustering is likely to occur. Therefore, the probability of incorrect feature point matching will increase significantly. Second, the FAST feature point detection algorithm can extract a large number of feature points. Therefore, when the ORB algorithm calculates the feature values of local regions of the image, the time consumption will increase abnormally, seriously affecting the feature extraction time. Third, the BRIEF feature description algorithm is sensitive to noise. Therefore, although the integral image is used to compare pixel blocks to a certain extent to solve the problem of noise sensitivity, the problem that the BRIEF algorithm affects feature extraction under uneven illumination has not been well solved. Summary of the Invention

[0005] In view of the above technical problems in the related art, the present invention proposes an image registration method based on improved ORB feature extraction and grid feature matching, which can overcome the above deficiencies of the prior art.

[0006] To achieve the above technical objectives, the technical solution of the present invention is implemented as follows:

[0007] An image registration method based on improved ORB feature extraction and grid feature matching;

[0008] The image registration method based on improved ORB feature extraction and grid feature matching includes the following steps:

[0009] S1. Input the image to be extracted;

[0010] S2. Use the improved FAST algorithm to extract the feature points of the image to be stitched;

[0011] S3. Establish a BRIEF descriptor using the idea of the BRIEF descriptor to obtain a BRIEF descriptor with direction;

[0012] S4. According to the set of feature points of the image to be stitched extracted in S02, establish a Delaunay triangle grid with the feature points as vertices;

[0013] S5. Based on the Delaunay triangle grid established in S04, judge the triangle similarity to obtain similar triangles composed of feature points;

[0014] S6. Use the Hamming distance to pair the vertices of similar triangles to obtain the best matching feature point pairs;

[0015] S7. Use the SVD method to obtain the affine matrix parameters.

[0016] Further, the step S2 uses the improved FAST algorithm to extract the feature points of the image to be stitched, which specifically includes the following steps:

[0017] S201. Let F1 and F2 be the images to be stitched. When detecting feature points of the images to be stitched, a given constant threshold ε is adopted d , so as to obtain a set P of initial corner points, with a total number of N;

[0018] S202. Let (x, y) be any point in the set P of corner points. If there is any point (x′, y′) ∈ P in the 3×3 circular neighborhood centered on this point, and the Harris response function of this point is:

[0019] Cornerness = Det(M) - k(Trace(M)) 2 ,

[0020] and it is greater than the response value of the point (x, y), then remove the point (x, y); perform a loop operation to complete the comparison of all corner points, and form the remaining points in the set P of corner points into a set P′;

[0021] S203. Set a threshold N′. When the number of corner points in P′ is less than N′, all corner points in the set P′ are used as the available feature points to be extracted; when the number of corner points in P′ is greater than N′, select the N′ points with the largest response values in the set P′ as the available feature points to be extracted; among them, the value of N′ is N / 2;

[0022] S204. Adopt the strategies of image region division and adjacent feature point removal to avoid the clustering of feature points and ensure the uniform distribution of feature points; perform uniform region division on the image, and set the total number of feature points in each region to N thd , to ensure the stability of the total number of feature points; then sort the feature points extracted from each image region according to the size of the eigenvalue, and retain N thd points with the largest eigenvalues, and remove other feature points;

[0023] S205. Set a neighbor distance D thd to eliminate the local clustering phenomenon; for different images, set the number of feature points N thd = 300 and the normalized distance D thd = 0.02; when the total number of feature points of the extracted image patches is greater than N thd , sort them automatically according to the eigenvalue size, and screen and remove non-significant feature points to make the feature points evenly distributed, thereby improving the feature point extraction efficiency;

[0024] S206. Complete the feature point extraction work for the F1 and F2 images to be stitched, and obtain the feature point sets of F1 and F2 respectively and The feature points are described by 3 attribute values: The coordinates of the feature points in the images to be stitched are set as (x, y), and (x, y, σ) is the vector description of the feature points.

[0025] Further, in step S3, the BRIEF descriptor is established by adopting the idea of the BRIEF descriptor, which specifically includes the following steps:

[0026] S301. Define the moment of the patch as: m pq = ∑ x，y x p y p I(x, y), where I(x, y) is the image gray value, and x, y are the positions relative to the FAST feature points; the radius of the circular neighborhood is r, and x, y ∈ [-r, r];

[0027] S302. Calculate the centroid of the patch: The angle between the feature point and the centroid is defined as the direction θ of the FAST feature point, then:

[0028] S303. Compare several pairs of pixel blocks by using the integral image. It is set that in the 31×31 pixel neighborhood, randomly select n d pixel blocks, the sub-window is set to 5×5, and compare the sum of the pixels in this window; construct a matrix S from the central coordinates of the randomly selected pixel blocks, and then construct a rotation matrix R according to the direction θ obtained in S302, and rotate S to the corresponding direction: S θ = RS, to obtain a binary string;

[0029] S304. Define the test value τ of the image gray difference value as: where p(x) represents the pixel intensity at x = (u, v) after the patch p is filtered; the binary detection τ is determined by a group of pairs, and the descriptor of the n T -dimensional binary string BRIEF is defined as: d where, nd is set to 256; the descriptor is a binary string, and the Hamming distance is used to pair the feature points; if the corresponding bit values are the same, the Hamming distance is 0, otherwise it is 1; when the number of the same bits is more, the descriptor similarity is greater.

[0030] Further, the method for establishing the Delaunay triangular mesh in S4 specifically includes the following steps:

[0031] S401. For the feature point sets f1 and f2, sort and divide all the points in the set (recursive processing), and divide the point set level by level until the number of feature points in the subset is no more than 3, and all subsets do not intersect;

[0032] ​S402. Construct Delaunay sub - triangulation using the subsets after recursive segmentation;

[0033] S403. Gradually merge the subsets from bottom to top to generate the Delaunay triangular mesh of the entire point set.

[0034] Furthermore, the method for judging the similarity of triangles in S5 specifically includes the following steps:

[0035] S501. Let the triangles in the reference image and the image to be registered be ΔABC and ΔA'B'C' respectively, where A, B, C and A', B', C' are the corresponding point pairs; the triangle similarity is I, and the similarities between the three included angles are I A 、I B 、I C . If the value of angle A is α and the value of angle A' is α', then the similarity between angle A and angle A' is: where σ = α / 6;

[0036] S502. Calculate the similarities I B and I C of the other two pairs of interior angles respectively; the similarity between triangle ΔABC and ΔA'B'C' is Set the similarity threshold ω, which is taken as 0.75 in this method; when I is greater than the threshold ω, it is judged that triangle ΔABC and ΔA'B'C' are similar.

[0037] Advantages of the present invention: The present invention designs and develops an image registration method based on improved ORB feature extraction and grid feature matching. The extracted feature points have scale invariance and rotation invariance, and can combine the geometric information of the feature points for feature extraction, reducing the sensitivity to noise and effectively improving the disadvantage of slow extraction speed of image features, with good stability. This method is different from the traditional method that only considers the geometric correlation between point pairs for point pair matching. By constructing Delaunay triangles and judging triangle similarity, wrong matching points are eliminated. The present invention can improve the video splicing efficiency, accuracy and calculation speed, etc., and solves the problems of poor real - time performance of video splicing effect and large memory resource occupation. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0039] Figure 1It is a flowchart of image registration of the image registration method based on improved ORB feature extraction and grid feature matching according to the embodiments of the present invention. Detailed implementation manners

[0040] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part rather than all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0041] It should be understood that in the description of the embodiments of the present invention, the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the embodiments of the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the embodiments of the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the embodiments of the present invention, the meaning of "a number of" is two or more unless otherwise specifically defined.

[0042] As Figure 1 shown, according to the image registration method based on improved ORB feature extraction and grid feature matching described in the embodiments of the present invention, after studying the key technologies and principles of traditional image feature extraction and matching methods, aiming at the problems of slow feature point extraction speed, low accuracy, and excessive memory resource occupation, the ORB feature extraction method is improved. It mainly achieves the purpose of improving the feature extraction efficiency and extraction accuracy and having better stability by integrating the ORB algorithm with scale clustering analysis, and solves the problems of high redundancy, easy occurrence of mismatching and missing matching in traditional image matching methods. In addition, grid feature matching is used instead of the traditional matching method after the improved ORB algorithm to complete efficient and accurate feature matching.

[0043] The present invention is described in detail below in conjunction with the accompanying drawings. There are several problems with the feature extraction process based on ORB. First, the number of feature points extracted by the FAST algorithm is large, and distribution clustering is prone to occur, so the probability of feature point mismatching will be greatly increased. Second, the FAST feature point detection algorithm can extract a large number of feature points, so when the ORB algorithm calculates the feature values ​​of the local area of ​​the image, the time consumption will be abnormally increased, which seriously affects the feature extraction time. Third, the BRIEF feature description algorithm is more sensitive to noise, so an integral graph is used to compare pixel block pairs. Although the noise sensitivity problem is solved to a certain extent, the problem that the BRIEF algorithm affects feature extraction under the condition of uneven illumination has not been well solved.

[0044] In view of the above problems, the FAST feature point detection algorithm is improved. M is defined as the second-order moment proposed by the Harris operator to describe the gradient distribution in the neighborhood of a pixel point. The Harris corner point response function is: Cornerness = Det(M)-k(Trace(M)) 2 By combining the circular template with the Harris response value of the FAST corner point, the clustering phenomenon of corner point distribution is eliminated, and the corner points with unstable edge responses are eliminated, so as to improve the accuracy and speed of feature extraction and enhance the robustness of subsequent matching.

[0045] The specific implementation process of the improved ORB algorithm is as follows:

[0046] 1. Let F1 and F2 be the images to be stitched. When detecting feature points on the images to be stitched, use the given constant threshold ε d , thus obtaining the point set P of the initial corner points, with a total number of N.

[0047] 2. Let (x, y) be any point in the corner point set P. If, with this point as the center, within a 3×3 circular neighborhood, there is any point (x′, y′)∈P, and the Harris response value of this point is Cornerness=Det(M)-k(Trace(M)) 2 If the response value is greater than that of the (x, y) point, remove the (x, y) point; repeat the operation to complete the comparison of all corner points, and form the remaining points in the corner point set P into a set P′.

[0048] 3. Set the threshold N'. When the number of corner points in P' is less than N', all corner points in the set P' are used as available feature points for extraction. When the number of corner points in P' is greater than N', select the N' points with the largest response value in P' as available feature points for extraction. The value of N' can generally be N / 2.

[0049] 4. Adopt the strategy of image region division and adjacent feature point elimination to avoid clustering of feature points and ensure uniform distribution of feature points. Uniformly divide the image into regions, and set the total number of feature points in each region to N thd , ensuring the stability of the total number of feature points. Then, sort the feature points extracted from each image region according to the magnitude of the eigenvalue, and retain N thd points with the largest eigenvalues, and eliminate other feature points.

[0050] 5. Since N thd significant feature points are retained in each region, the distribution of image feature points is homogenized. However, feature points in local regions of the image may still cluster, so the false matching rate will increase during image matching. Therefore, set the nearest neighbor distance D thd to eliminate local clustering. For different images, set the number of feature points N thd = 300 and the normalized distance D thd = 0.02. The setting of the number of feature points is related to factors such as the image extraction speed, and the threshold can also be appropriately modified; the nearest neighbor distance threshold is set to ensure successful extraction even when the image texture is rich or the distance between feature points is close. When the total number of block feature points extracted from the image is greater than N thd , automatically sort according to the magnitude of the eigenvalue, and screen and remove non-significant feature points to keep the feature points evenly distributed, effectively improving the feature point extraction efficiency and avoiding the problem of increased time consumption.

[0051] 6. Complete the feature point extraction work for the F1 and F2 images to be stitched, and obtain the feature point sets of F1 and F2 and Feature points are described using 3 attribute values: The coordinates of the feature point in the image to be stitched are set as (x, y), and (x, y, σ) is the vector description of the feature point.

[0052] 7. After obtaining the feature points, establish a descriptor using the idea of the BRIEF descriptor.

[0053] To solve the problem that the ORB algorithm does not have rotational invariance, add a direction to the BRIEF descriptor and perform rotation to obtain the oriented BRIEF. First, define the moment of the patch as:

[0054] m pq = ∑ x,y x p y p I(x, y) (1)

[0055] In the formula, I(x, y) is the image grayscale value, where x and y are the positions relative to the FAST feature point, the radius of the circular neighborhood is r, and x, y ∈ [-r, r].

[0056] Then calculate the centroid of the plaque:

[0057]

[0058] The angle between the feature point and the centroid is defined as the direction θ of the FAST feature point:

[0059]

[0060] To address the noise impact caused by uneven illumination and improve real-time performance, integral images are used to compare several pairs of pixel blocks. In a 31×31 pixel neighborhood, nd pixel blocks are randomly selected, the sub-window is set to 5×5, and the pixel sums within this window are compared.

[0061] Construct a matrix S from the central coordinates of the randomly selected pixel blocks, and then construct a rotation matrix R according to the direction θ obtained from equation (3) to rotate S to the corresponding direction:

[0062]

[0063] S θ = RS (5)

[0064] Obtain a binary string, and define the test value τ of the image gray-scale difference value as:

[0065]

[0066] In equation (6), p(x) represents the pixel intensity at x = (u, v) after the image patch p has been filtered. The binary detection τ is determined by a set of pairs. Therefore, the descriptor of the nd-dimensional binary string BRIEF is defined as: T The binary detection τ is determined by a set of pairs. Therefore, the descriptor of the nd-dimensional binary string BRIEF is defined as:

[0067]

[0068] In this method, n d is set to 256. This descriptor is a binary string, and the Hamming distance is used to pair the feature points. If the corresponding bit values are the same, the Hamming distance is 0; otherwise, it is 1. The greater the number of identical bits, the greater the similarity of the descriptors.

[0069] The improved ORB algorithm obtains the BRIEF descriptor with direction according to the above formula.

[0070] 8. Feature matching based on Delaunay triangulation

[0071] a) Using the previously accurately extracted feature point sets f1 and f2, triangulate the convex hull formed by the feature points, and construct the image feature information into a Delaunay triangular mesh. For the convex hull formed by triangulating the plane of the feature point sets f1 and f2, a method of recursively dividing the point set is adopted, and it is divided level by level until the subset only contains three discrete points that can form a triangle, and the final triangular mesh is formed by merging level by level from bottom to top.

[0072] 1) For the feature point sets f1 and f2, sort and divide all the points in the set (recursive processing), divide the point set level by level until the number of feature points in the subset is no more than 3, and all subsets do not intersect.

[0073] 2) Use the subsets after recursive division to construct Delaunay sub-triangular meshes.

[0074] 3) Gradually merge the subsets from bottom to top to generate the Delaunay triangular mesh of the entire point set.

[0075] Since the Delaunay triangulation network is unique, the feature point sets f1 and f2 have a unique Delaunay triangular mesh triangulation.

[0076] b) Judge the triangle similarity, extract the triangle pairs with larger similarity, and obtain the accurate matching point pairs. The method for judging triangle similarity is as follows:

[0077] Let the triangles in the reference image and the image to be registered be ΔABC and ΔA′B′C′ respectively, where A, B, C and A', B', C' are the corresponding point pairs. The triangle similarity is I, and the similarities between the three included angles are I A 、I B 、I C , the value of angle A is α, the value of angle A' is α', then the similarity between angle A and angle A' is:

[0078]

[0079] where σ = α / 6.

[0080] Then calculate the similarities I B and I C of the other two pairs of interior angles respectively. The similarity between triangle ΔABC and ΔA′B′C′ is Set the similarity threshold ω, which is taken as 0.75 in this method. When I is greater than the threshold ω, it is judged that triangle ΔABC and ΔA′B′C′ are similar.

[0081] After obtaining the similar triangle pairs, pair the triangle vertex pairs of the similar triangle pairs using the Hamming distance to obtain the best matching feature point pairs, and finally use the SVD (Singular Value Decomposition) method to obtain the affine matrix parameters.

[0082] In summary, the image registration method based on improved ORB feature extraction and grid feature matching proposed by the present invention can improve the feature extraction efficiency and extraction accuracy while ensuring the image quality, reduce the false matching and improve the stability. It not only improves the real-time performance of video stitching, but also reduces the waste of memory resources, lowers the device cost, greatly improves the work efficiency, and provides a better method for image stitching.

[0083] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An image registration method based on improved ORB feature extraction and grid feature matching, characterized in that: The following steps are involved: S1, input the image to be extracted; S2, using the improved FAST algorithm to extract feature points of the image to be stitched; S3, using the idea of ​​BRIEF descriptor to establish a BRIEF descriptor, and obtaining a BRIEF descriptor with direction; specifically including the following steps: S301. Define the moment of the patch as: m pq =∑ xy x p y p I(x, y), where I(x, y) is the grayscale value of the image, and x, y are the positions relative to the FAST feature points; the radius of the circular neighborhood is r, x, y∈[-r, r]; S302, calculating the center of gravity of the plaque: The angle between the feature point and the center of gravity is defined as the direction θ of the FAST feature point, then: S303, using the integral graph to compare several pairs of pixel blocks, set in a 31×31 pixel neighborhood, randomly select n d The sub-window is set to 5×5, and the pixels and in the window are compared; the center coordinates of the randomly selected pixel blocks are constructed into a matrix S, and then a rotation matrix R is constructed according to the direction θ obtained in S302 to rotate S to the corresponding direction: S θ =RS gets the binary string; S304, define the test value τ of the image grayscale difference value as: Where p(x) represents the image spot p after filtering, at x = (u, v) T The pixel intensity at the binary detection τ is uniquely determined by a set of pairs, n d The binary string BRIEF descriptor for the dimension is defined as: Among them, nd is set to 256; the descriptor is a binary string, and the Hamming distance is used to pair the feature points; if the corresponding bit values ​​are the same, the Hamming distance is 0, otherwise it is 1; the more the number of identical bits, the greater the similarity of the descriptors; S4, establishing a Delaunay triangular mesh with the feature points as vertices according to the feature point set of the image to be stitched extracted in S2; S5, judging the triangle similarity based on the Delaunay triangle mesh established in S4, and obtaining similar triangles composed of feature points; S6. Use the Hamming distance to pair similar triangle vertices to obtain the best matching feature point pair; S7. Use the SVD method to obtain the affine matrix parameters.

2. The image registration method based on improved ORB feature extraction and grid feature matching according to claim 1, characterized in that: The step S2 uses the improved FAST algorithm to extract feature points of the images to be stitched, and specifically includes the following steps: S201, let F1 and F2 be the images to be stitched, and when detecting feature points of the images to be stitched, use a given constant threshold ε d , thus obtaining the point set P of the initial corner points, the total number is N; S202. Let (x, y) be any point in the corner point set P. If, with this point as the center, within a 3×3 circular neighborhood, there is any point (x′, y′)∈P, and the Harris response function of this point is: Cornerness=Det(M)-k(Trace(M)) 2 , If the response value is greater than that of the (x,y) point, remove the (x,y) point; loop the operation to complete the comparison of all corner points, and form the remaining points in the corner point set P into a set P′; S203, setting a threshold value N', when the number of corner points in P' is less than N', all corner points in the set P' are used as available feature points for extraction; and when the number of corner points in P' is greater than N', N' points with the largest response values ​​in the set P' are selected as available feature points for extraction; wherein the value of N' is N / 2; S204, adopting the strategy of image region division and adjacent feature point elimination to avoid clustering of feature points and ensure uniform distribution of feature points; evenly dividing the image into regions and setting the total number of feature points in each region to N thd , to ensure the stability of the total number of feature points; then sort the feature points extracted from each image area according to the size of the feature value, and retain N thd The point with the largest eigenvalue, other feature points are eliminated; S205, setting the neighbor distance D thd To eliminate local clustering; for different images, set the number of feature points N thd =300 and normalized distance D thd =0.02; when the total number of extracted image block feature points is greater than N thd When extracting features, the features are automatically sorted by feature value, and non-significant feature points are filtered and removed to keep the feature points evenly distributed, thereby improving the efficiency of feature point extraction. S206: Extract feature points of the images to be spliced ​​F1 and F2, and obtain feature point sets of F1 and F2 respectively. and Feature points are described using three attribute values: The coordinates of the feature points in the image to be stitched are set to (x, y), and (x, y, σ) is the vector description of the feature points.

3. The image registration method based on improved ORB feature extraction and grid feature matching according to claim 2, characterized in that: The method for establishing the Delaunay triangular mesh in S4 specifically comprises the following steps: S401, for feature point sets f1 and f2, sort and segment all points in the sets, and divide the point sets step by step until the number of feature points in the subsets is no more than 3, and all subsets do not intersect with each other; S402, constructing a Delaunay sub-triangulation network using the subsets after recursive segmentation; S403, gradually merge the subsets from bottom to top to generate a Delaunay triangulated mesh of the entire point set.

4. The image registration method based on improved ORB feature extraction and grid feature matching according to claim 3, characterized in that: The method for determining the triangle similarity in S5 specifically comprises the following steps: S501, suppose the triangles in the reference image and the image to be registered are ΔABC and ΔA′B′C′, where A, B, C and A', B', C' are corresponding point pairs; the triangle similarity is I, and the similarities between the three angles are I A ,I B ,I C , the value of angle A is α, the value of angle A' is α', then the similarity between angle A and angle A' is: Where σ = α / 6; S502, respectively calculate the similarity I of the other two pairs of interior angles B and I C ; The similarity between triangles ΔABC and ΔA′B′C′ is Set the similarity threshold ω, which is 0.75 in this method; when I is greater than the threshold ω, it is determined that triangles ΔABC and ΔA′B′C′ are similar.

Images