CN113128554B - A target positioning method, system, device and medium based on template matching - Google Patents

Abstract

The invention discloses a target positioning method, a system, a device and a medium based on template matching, wherein the method comprises the following steps: constructing a template image pyramid and constructing a first grid; selecting a first grid with the average gradient amplitude being greater than or equal to a preset first threshold value as a second grid; determining first gradient amplitude and first gradient direction entropy of all pixel points in a second grid, determining a first local feature point according to the first gradient amplitude and the first gradient direction entropy, and establishing a first LBP histogram feature vector according to the first local feature point; and establishing a second LBP histogram feature vector of the target image pyramid, and determining the position and the rotation angle of the target to be positioned according to the similarity matching of the first LBP histogram feature vector and the second LBP histogram feature vector. The invention reduces the requirement on system computing power, overcomes the defect of large influence of rotation in the prior art, improves the accuracy of target positioning, and can be widely applied to the technical field of computer vision.

Description

A target positioning method, system, device and medium based on template matching

technical field

The present invention relates to the technical field of computer vision, in particular to a target positioning method, system, device and medium based on template matching.

Background technique

Template matching is a classic method in the field of computer vision and image processing. By giving a template image and a target image, the feature vector of the image is extracted, the similarity between the template image and the candidate window in the target image is calculated, and the candidate window most similar to the template image is calculated. The window is the matching result.

The LBP (Local Binary Patterns) algorithm has been proposed by Timo Ojala et al since 2002. By comparing the gray value of pixels and eight neighborhoods, a total of 36 sets of binary patterns are used to represent the rotation invariance of pixels, among which uniform The 9 binary modes ≤ 2 are called equivalent modes, and the remaining binary modes are called mixed modes. Variants based on LBP are also constantly innovating, but the core of the LBP-based algorithm is the binary pattern of local pixels. The MRELBP algorithm proposed by Li Liu et al. includes the ELBP_CI method based on the center strength, the ELBP_NI method based on the neighborhood strength, the ELBP_RD method based on the radial difference and the ELBP_AD method based on the angular difference, and on the basis of ELBP, the median value is used. The response of the filter replaces the value of a single pixel, and finally 5 modes with symmetry are selected from the blending mode and merged with 9 modes of the equivalent mode, but the rotation invariant stability in the blending mode is not enough, affected by the rotation. Great impact. The LBP-NMF algorithm proposed by Ioan Buciu et al. is aimed at facial expressions and extracts the LBP histogram features of the four parts of the eyes, nose and mouth for template matching, because the LBP histogram information of each point in the region needs to be It is used for calculation, so the calculation time must be more, and the accuracy of the result is also related to the location division of these four parts.

SUMMARY OF THE INVENTION

The purpose of the present invention is to solve one of the technical problems existing in the prior art at least to a certain extent.

To this end, an object of the embodiments of the present invention is to provide a target localization method based on template matching. On the one hand, the method establishes an LBP histogram feature vector according to the selected local feature points, without performing the detection of each pixel in the image. On the other hand, the stable pixel points are determined according to the similarity matching of the feature vector of the LBP histogram, which overcomes the shortcomings of the prior art that are greatly affected by rotation, and improves the accuracy of target positioning.

Another object of the embodiments of the present invention is to provide a target positioning system based on template matching.

In order to achieve the above technical purpose, the technical solutions adopted in the embodiments of the present invention include:

In a first aspect, an embodiment of the present invention provides a template matching-based target positioning method, including the following steps:

obtaining a template image, constructing a template image pyramid according to the template image, and constructing a first grid according to the template image pyramid;

Determine the average gradient magnitude of all pixel points in the first grid, and select the first grid whose average gradient magnitude is greater than or equal to a preset first threshold as the second grid;

Determine the first gradient magnitude and the first gradient direction entropy value of all the pixel points in the second grid, and determine the first local feature point according to the first gradient magnitude and the first gradient direction entropy value, and according to The first local feature point establishes a first LBP histogram feature vector;

Acquiring a target image, constructing a target image pyramid according to the target image, and determining a second local feature point of the target image pyramid, and then establishing a second LBP histogram feature vector according to the second local feature point, and then according to the The similarity between the first LBP histogram feature vector and the second LBP histogram feature vector determines the position and rotation angle of the target to be located.

Further, in an embodiment of the present invention, the step of acquiring a template image, constructing a template image pyramid according to the template image, and constructing a first grid according to the template image pyramid specifically includes:

Obtain a template image, and expand the width and height of the template image to obtain an expanded template image;

down-sampling the expanded template image to generate a template image pyramid;

Extracting the first pyramid image of each layer in the template image pyramid, and adaptively constructing a first grid by quadtree segmentation according to the first pyramid image;

Wherein, the first pyramid image is covered by the corresponding first grid.

Further, in an embodiment of the present invention, in the step of adaptively constructing a first grid through quadtree segmentation according to the first pyramid image, when the width or height of the first grid is less than When the preset sixth threshold is used, no quadtree segmentation is performed, and the sixth threshold satisfies the following conditions:

Among them, μ _s represents the sixth threshold, W _TL represents the width of the top image of the template image pyramid, and H _TL represents the height of the top image of the template image pyramid.

Further, in an embodiment of the present invention, the average gradient magnitude of all pixels in the first grid is determined, and the first grid whose average gradient magnitude is greater than or equal to a preset first threshold is selected as the first grid. The second grid step, which specifically includes:

Calculate the sum of the gradient magnitudes of all pixels in the first grid, and determine the average gradient magnitude according to the number of pixels in the first grid and the sum of the gradient magnitudes;

If it is determined that the average gradient magnitude is greater than or equal to the first threshold, the corresponding first grid is acquired as the second grid.

Further, in an embodiment of the present invention, the first gradient magnitude and the first gradient direction entropy value of all the pixel points in the second grid are determined, and the first gradient magnitude and the first gradient direction entropy value are determined according to the first gradient magnitude and the A gradient direction entropy value determines a first local feature point, and the step of establishing a first LBP histogram feature vector according to the first local feature point specifically includes:

determining the first gradient magnitude and the first gradient direction entropy value of all pixels in the second grid;

Obtain a preset second threshold and a third threshold, and select pixels whose first gradient magnitude is greater than or equal to the second threshold and whose first gradient direction entropy value is greater than or equal to the third threshold as the first local feature point;

The first LBP histogram feature vector is extracted by taking the first local feature point as the central pixel point.

Further, in an embodiment of the present invention, the step of extracting the first LBP histogram feature vector by using the first local feature point as a central pixel point is specifically:

The first LBP histogram feature vector of the first local feature point is established by taking the first local feature point as the central pixel point and combining the nine modes of uniform≤2 in the LBP equivalent mode.

Further, in an embodiment of the present invention, the acquiring a target image, constructing a target image pyramid according to the target image, and determining a second local feature point of the target image pyramid, and then according to the second local feature point to establish the second LBP histogram feature vector, and then according to the similarity matching of the first LBP histogram feature vector and the second LBP histogram feature vector to determine the position and rotation angle of the target to be located. include:

Obtaining a target image, constructing a target image pyramid according to the target image, and constructing a third grid according to the target image pyramid;

Determine the average gradient magnitude of all pixel points in the third grid, and select the third grid whose average gradient magnitude is greater than or equal to a preset fourth threshold as the fourth grid;

Determine the second gradient magnitude and the second gradient direction entropy value of all the pixel points in the fourth grid, and determine the second local feature point according to the second gradient magnitude and the second gradient direction entropy value, and according to The second local feature point establishes a second LBP histogram feature vector;

Calculate the similarity of the first LBP histogram feature vector of each layer in the template image pyramid and the target image pyramid and the second LBP histogram feature vector of the corresponding position, and perform similarity matching from the top layer to the bottom layer in turn, and The second local feature points whose similarity is less than the preset fifth threshold are discarded, and the second local feature points whose similarity is greater than or equal to the fifth threshold are screened out at the bottom of the target image pyramid, so as to determine the to-be-located feature. The position and rotation angle of the target.

In a second aspect, an embodiment of the present invention provides a target positioning system based on template matching, including:

a grid construction module for acquiring a template image, constructing a template image pyramid according to the template image, and constructing a first grid according to the template image pyramid;

a grid selection module, configured to determine the average gradient amplitude of all the pixels in the first grid, and select the first grid whose average gradient amplitude is greater than or equal to a preset first threshold as the second grid;

The LBP histogram feature vector establishment module is used to determine the first gradient magnitude and the first gradient direction entropy value of all the pixels in the second grid, and according to the first gradient magnitude and the first gradient direction entropy value determines a first local feature point, and establishes a first LBP histogram feature vector according to the first local feature point;

The similarity matching module is used to obtain a target image, construct a target image pyramid according to the target image, and determine a second local feature point of the target image pyramid, and then establish a second LBP histogram according to the second local feature point feature vector, and then determine the position and rotation angle of the target to be located according to similarity matching between the first LBP histogram feature vector and the second LBP histogram feature vector.

In a third aspect, an embodiment of the present invention provides a target positioning device based on template matching, including:

at least one processor;

at least one memory for storing at least one program;

When the at least one program is executed by the at least one processor, the at least one processor is caused to implement the above-mentioned method for locating objects based on template matching.

In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, in which a program executable by a processor is stored, and the program executable by the processor is used to execute the above one when executed by the processor Object localization method based on template matching.

The advantages and beneficial effects of the present invention will, in part, be given in the following description, and in part will become apparent from the following description, or be learned by practice of the present invention:

In the embodiment of the present invention, local feature points belonging to edges and textures in the pyramid image are screened based on the gradient magnitude and gradient direction entropy value, and the LBP histogram feature vector is established according to the local feature points, and then the LBP histogram feature vector is determined according to the similarity matching of the LBP histogram feature vectors. Compared with the prior art, the embodiment of the present invention establishes the LBP histogram feature vector according to the selected local feature points on the one hand, and does not need to All pixel points are calculated, which reduces the requirement for system computing power. On the other hand, stable pixel points are determined according to the similarity matching of LBP histogram feature vectors, which overcomes the shortcomings of the prior art that are greatly affected by rotation and improves target positioning. accuracy.

Description of drawings

In order to explain the technical solutions in the embodiments of the present invention more clearly, the following descriptions are given to the accompanying drawings that are used in the embodiments of the present invention. It should be understood that the accompanying drawings in the following introduction are only for the convenience of clearly expressing the technology of the present invention. For some of the embodiments in the solution, for those skilled in the art, other drawings can also be obtained from these drawings without the need for creative work.

1 is a flowchart of steps of a template matching-based target positioning method provided by an embodiment of the present invention;

2 is a structural block diagram of a template matching-based target positioning system provided by an embodiment of the present invention;

FIG. 3 is a structural block diagram of a target positioning apparatus based on template matching according to an embodiment of the present invention.

Detailed ways

The following describes in detail the embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary, only used to explain the present invention, and should not be construed as a limitation of the present invention. The numbers of the steps in the following embodiments are only set for the convenience of description, and the sequence between the steps is not limited in any way, and the execution sequence of each step in the embodiments can be adapted according to the understanding of those skilled in the art Sexual adjustment.

In the description of the present invention, the meaning of multiple is two or more. If the first and second are described, they are only for the purpose of distinguishing technical features, and should not be understood as indicating or implying relative importance or implicit Indicates the number of the indicated technical features or implicitly indicates the order of the indicated technical features. Also, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art.

Referring to FIG. 1 , an embodiment of the present invention provides a target positioning method based on template matching, which specifically includes the following steps:

S101, obtaining a template image, constructing a template image pyramid according to the template image, and constructing a first grid according to the template image pyramid;

Specifically, each layer in the template image pyramid has a corresponding pyramid image, the first grid is used to cover the pyramid image, and the first grid contains all the pixels of the pyramid image, which is convenient for subsequent processing according to the first grid. Screen and identify local feature points belonging to edges and textures. Step S101 includes the following steps:

S1011. Obtain a template image, and expand the width and height of the template image to obtain an expanded template image;

S1012, down-sampling the extended template image to generate a template image pyramid;

S1013, extract the first pyramid image of each layer in the template image pyramid, and construct the first grid adaptively by quadtree segmentation according to the first pyramid image;

Wherein, the first pyramid image is covered by the corresponding first grid.

Specifically, a template image is obtained, and the width and height of the template image are respectively expanded to the nearest power of 2. The expanded image is downsampled with Gaussian smoothing with a scale factor of 2 to generate a template image pyramid.

Rotate the image of each layer in the template image pyramid at the angle of Angle _j , _which is calculated by the following formula:

Among them, Angle _s is the starting angle of the rotation, Angle _E is the ending angle of the rotation, and the reference step size μ _A of the rotation angle is limited by the number of layers of the template image pyramid and the range of the rotation angle, so when the range of the rotation angle is small, μ _A should also decrease, while the value of _μA should increase when the total number of pyramid levels is small. For each layer down from the top layer of the pyramid, the rotation angle step size d _Ai is correspondingly reduced, and the calculation time is reduced while the angle is accurate. Usually μ _A is set to 1.

The smaller the size of the top-level image of the template image pyramid, the shorter the calculation time for matching the target image, but the accuracy of the matching result is also reduced. When the result of the previous layer of the pyramid is not higher than the threshold, it means The termination of the downsampling match with the result.

Further as an optional embodiment, in the step of adaptively constructing the first grid through quadtree segmentation according to the first pyramid image, when the width or height of the first grid is less than the preset sixth threshold, no longer For quadtree segmentation, the sixth threshold satisfies the following conditions:

Among them, μ _s represents the sixth threshold, W _TL represents the width of the top image of the template image pyramid, and H _TL represents the height of the top image of the template image pyramid.

Specifically, all images in the template image pyramid are extracted, including images obtained by rotating each layer of pyramid images at a certain angle, and the first grid is adaptively constructed based on the quadtree principle.

The grid size can be 3*3, while all images need to be completely covered by the corresponding first grid constructed.

The smaller the sixth threshold _μs is, the more grids are filtered, and the range of extracting feature pixels increases. More feature pixels can improve the matching accuracy, but it will waste computing and memory resources. At the same time, more pixels with insufficient information are also included in the feature pixels, which will lead to unstable matching results. The larger the sixth threshold _μs is, the fewer grids pass the screening, the number of feature pixels will be reduced, and the time-consuming matching calculation will be reduced accordingly, but the feature pixels used for matching may not be sufficient, resulting in matching results. Inaccurate. Therefore, the set sixth threshold μ _s should enable at least a three-level quadtree to be established for the top-level image of the pyramid for segmentation.

S102. Determine the average gradient magnitude of all the pixel points in the first grid, and select the first grid whose average gradient magnitude is greater than or equal to a preset first threshold as the second grid.

Specifically, a first grid with an average gradient magnitude greater than or equal to a first threshold can preliminarily determine local feature points including edges and textures, and a second grid can be screened for subsequent determination of local feature points by setting the first threshold. Step S102 specifically includes the following steps:

S1021, calculating the sum of the gradient magnitudes of all pixels in the first grid, and determining the average gradient magnitude according to the number of pixels in the first grid and the sum of the gradient magnitudes;

S1022. It is determined that the average gradient magnitude is greater than or equal to the first threshold, and then acquire the corresponding first grid as the second grid.

Specifically, the sum of the gradient magnitudes of all the pixels in the first grid is calculated, and the average gradient magnitude obtained by dividing the sum of the gradient magnitudes by the number of pixels in the grid, the average gradient magnitude of the selected second grid should be greater than or equal to first threshold.

If the average gradient magnitude is less than the first threshold, it is likely to be the background, or the parts of the image lacking texture and edges, and feature points with sufficient information cannot be extracted from these parts.

In this embodiment of the present invention, the average gradient amplitude is used for screening, which is equivalent to the global filter of the image. Filtering grids with a smaller average gradient amplitude is equivalent to ignoring the areas in the image with few edges and few textures or no edges and no textures. The range of extracting strong feature points in the image.

S103. Determine the first gradient magnitude and the first gradient direction entropy value of all the pixel points in the second grid, and determine the first local feature point according to the first gradient magnitude and the first gradient direction entropy value, and according to the first local feature point to establish the first LBP histogram feature vector;

Specifically, for each screened second grid, the feature information amount of the pixel points in the second grid is evaluated based on the gradient magnitude and gradient direction entropy value, and the local feature points belonging to edges and textures in the image are detected, and then according to This local feature point suggests the LBP histogram feature vector. Step S103 specifically includes the following steps:

S1031, determine the first gradient magnitude and the first gradient direction entropy value of all pixels in the second grid;

S1032, obtaining a preset second threshold and a third threshold, and selecting pixels whose first gradient magnitude is greater than or equal to the second threshold and whose first gradient direction entropy value is greater than or equal to the third threshold as the first local feature point;

S1033 , extracting a first LBP histogram feature vector by using the first local feature point as a central pixel point.

Specifically, for each filtered second grid, a gradient magnitude and gradient direction entropy method is used to detect local feature points belonging to edges and textures in the image. The larger the gradient magnitude of a pixel point, the richer the gradient information contained in this pixel point; the larger the gradient direction entropy value of a certain area of the image, the larger the gradient direction change of the pixel point in this area, the image corners exist in this area. The probability of the region is increased, and the corner point usually means that the target contour has changed, which can provide more feature information for template matching.

Calculate the gradient direction entropy of a pixel point, take the pixel point as the center of the n*n area, and n can be set to 3, that is, calculate the entropy value of the gradient direction of 9 pixel points.

In each second grid, pixels whose gradient magnitude and gradient direction entropy are respectively greater than the corresponding threshold are selected as local feature points.

Further as an optional embodiment, the step of extracting the first LBP histogram feature vector by using the first local feature point as the central pixel point is specifically:

Taking the first local feature point as the center pixel, and combining the nine modes of uniform≤2 in the LBP equivalent mode, the first LBP histogram feature vector of the first local feature point is established.

Specifically, the nine modes with uniform≤2 in the LBP equivalent mode are regarded as stable rotation modes, and there are nine categories in total. The rest that do not belong to the stable rotation mode are classified into another category, so there are 10 types of binary modes used to extract the histogram feature vector of the central pixel in the region.

In a 3*3 area of the image, the gray value of the central pixel is compared with the surrounding 8 pixels. If the gray value of the surrounding pixel is greater than that of the central pixel, it is marked as 1, otherwise it is marked as 0. Therefore, the number of consecutively marked 1 in the 8 pixels will be 0, 1, 2...8, a total of 9 categories, corresponding to the nine modes of uniform≤2 in the LBP equivalent mode, as the stable rotation mode . The rest that do not belong to stable rotation patterns are classified into another category, so there are 10 types of binary patterns used to extract the histogram feature vector of each local feature point.

Similarly, the size of the area with radius r is R _r *R _r , r=1,2,3,..., R _r =2r+1, then each area with radius r has R _r *R _r Pixels, except for the center pixel, there are a total of R _r *R _r -1 pixels in the area, and the histogram feature vector of the center pixel is extracted by 10 types of binary mode.

Local feature points are detected for all images in the template image pyramid, each local feature point is used as the central pixel point of the extracted LBP feature area, and the first LBP histogram feature vector of the central pixel point is extracted.

The larger the radius of the region where the feature vector of the first LBP histogram is obtained, the richer the information contained, but the time required to extract the feature vector increases.

Detect local feature points for all images in the template image pyramid, and extract the improved LBP histogram feature vector with each local feature point as the center of the region.

S104: Acquire a target image, construct a target image pyramid according to the target image, and determine a second local feature point of the target image pyramid, and then establish a second LBP histogram feature vector according to the second local feature point, and then according to the first LBP histogram feature The similarity matching between the vector and the second LBP histogram feature vector determines the position and rotation angle of the target to be located.

Specifically, the processing of the target image is similar to the processing of the template image. The target image pyramid is also constructed first, and then the second local feature points are determined, and then the second LBP histogram feature vector is suggested, and then similarity matching is performed, which is determined according to the matching result. The position and rotation angle of the target to be positioned. Step S104 specifically includes the following steps:

S1041, obtaining a target image, constructing a target image pyramid according to the target image, and constructing a third grid according to the target image pyramid;

S1042, determine the average gradient magnitude of all pixels in the third grid, and select the third grid whose average gradient magnitude is greater than or equal to a preset fourth threshold as the fourth grid;

S1043: Determine the second gradient magnitude and the second gradient direction entropy value of all the pixel points in the fourth grid, and determine the second local feature point according to the second gradient magnitude and the second gradient direction entropy value, and according to the second local feature point to establish the second LBP histogram feature vector;

S1044: Calculate the similarity between the first LBP histogram feature vector of each layer in the template image pyramid and the target image pyramid and the second LBP histogram feature vector at the corresponding position, perform similarity matching from the top layer to the bottom layer in sequence, and compare the similarity The second local feature points whose degree is less than the preset fifth threshold are discarded, and then the second local feature points whose similarity is greater than or equal to the fifth threshold are screened at the bottom of the target image pyramid, thereby determining the position and rotation angle of the target to be located. .

It should be recognized that the process of establishing the second LBP histogram feature vector is basically similar to the foregoing process of establishing the first LBP histogram feature vector, and details are not described herein.

Specifically, after acquiring the target image to be positioned, constructing a target image pyramid with the same number of pyramid layers as the template image pyramid, establishing a second LBP histogram feature vector of the target image pyramid, starting from the top layer of the target image pyramid and the template image pyramid, Sliding the candidate window on the target image, the relative position information of all local feature points in the template image has been recorded in the candidate window, and the eigenvectors of the local feature points of the candidate window are continuously matched with the eigenvectors on the corresponding target image. The similarity is calculated between the histogram feature vectors of all local feature points (ie the first LBP histogram feature vector) and the histogram feature vectors of the feature points at the corresponding positions of the target image (ie the second LBP histogram feature vector).

Each layer may have multiple results, and the results whose similarity is lower than the set fifth threshold are discarded. Finally, after the similarity matching at the bottom of the pyramid is completed, local feature points with high similarity and stable are detected. The local feature points obtain the position and rotation angle of the target to be positioned.

The method steps in the embodiments of the present invention are described above. The embodiment of the present invention provides a target positioning method based on template matching. First, based on the gradient magnitude and gradient direction, the feature points belonging to edges and textures in the target image and the template image are detected; secondly, an improved LBP algorithm is used, The rotation-stable local feature points are screened out from the detected feature points and the LBP histogram feature vector is established; finally, a narrow-range search method is adopted, based on the detected stable local feature points, from the top of the pyramid to the bottom layer by layer Screening to determine the local feature points whose similarity is higher than the preset threshold and stable, so as to obtain the accurate target position and rotation angle. The embodiment of the present invention can accurately calculate the target position and the target rotation angle when the target image has blur, occlusion, noise, complex background and angle rotation.

In the embodiment of the present invention, local feature points belonging to edges and textures in the pyramid image are screened based on the gradient magnitude and gradient direction entropy value, and the LBP histogram feature vector is established according to the local feature points, and then the LBP histogram feature vector is determined according to the similarity matching of the LBP histogram feature vectors. Compared with the prior art, the embodiment of the present invention establishes the LBP histogram feature vector according to the selected local feature points on the one hand, and does not need to All pixel points are calculated, which reduces the requirement for system computing power. On the other hand, stable pixel points are determined according to the similarity matching of LBP histogram feature vectors, which overcomes the shortcomings of the prior art that are greatly affected by rotation and improves target positioning. accuracy.

Referring to FIG. 2, an embodiment of the present invention provides a target positioning system based on template matching, including:

The grid construction module is used for obtaining the template image, constructing the template image pyramid according to the template image, and constructing the first grid according to the template image pyramid;

The grid selection module is used to determine the average gradient magnitude of all the pixels in the first grid, and select the first grid whose average gradient magnitude is greater than or equal to the preset first threshold as the second grid;

The LBP histogram feature vector building module is used to determine the first gradient magnitude and the first gradient direction entropy value of all pixels in the second grid, and determine the first local feature according to the first gradient magnitude and the first gradient direction entropy value point, and establish the first LBP histogram feature vector according to the first local feature point;

The similarity matching module is used to obtain the target image, construct the target image pyramid according to the target image, and determine the second local feature point of the target image pyramid, and then establish the second LBP histogram feature vector according to the second local feature point, and then according to the second local feature point. The similarity matching between the feature vector of the first LBP histogram and the feature vector of the second LBP histogram determines the position and rotation angle of the target to be located.

The contents in the above method embodiments are all applicable to the present system embodiments, the specific functions implemented by the present system embodiments are the same as the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

Referring to FIG. 3, an embodiment of the present invention provides a target positioning device based on template matching, including:

at least one processor;

at least one memory for storing at least one program;

When the above-mentioned at least one program is executed by the above-mentioned at least one processor, the above-mentioned at least one processor is caused to implement the above-mentioned method for target positioning based on template matching.

The contents in the above method embodiments are all applicable to the present device embodiments, the specific functions implemented by the present device embodiments are the same as the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

An embodiment of the present invention further provides a computer-readable storage medium, in which a program executable by a processor is stored, and the program executable by the processor, when executed by the processor, is used to perform the above-mentioned target positioning based on template matching method.

A computer-readable storage medium according to an embodiment of the present invention can execute a template matching-based target positioning method provided by the method embodiment of the present invention, can execute any combination of implementation steps of the method embodiment, and has corresponding functions of the method. and beneficial effects.

The embodiment of the present invention also discloses a computer program product or computer program, where the computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of the computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the method shown in FIG. 1 .

In some alternative implementations, the functions/operations noted in the block diagrams may occur out of the order noted in the operational diagrams. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/operations involved. Furthermore, the embodiments presented and described in the flowcharts of the present invention are provided by way of example in order to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of the various operations are altered and in which sub-operations described as part of larger operations are performed independently.

Furthermore, although the invention is described in the context of functional modules, it is to be understood that, unless stated to the contrary, one or more of the above-described functions and/or features may be integrated in a single physical device and/or In software modules, or one or more functions and/or features may be implemented in separate physical devices or software modules. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary to understand the present invention. Rather, given the attributes, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the modules will be within the routine skill of the engineer. Accordingly, those skilled in the art, using ordinary skill, can implement the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are illustrative only and are not intended to limit the scope of the invention, which is to be determined by the appended claims along with their full scope of equivalents.

If the above functions are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the above-mentioned methods in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

The logic and/or steps represented in flowcharts or otherwise described herein, for example, may be considered an ordered listing of executable instructions for implementing the logical functions, may be embodied in any computer-readable medium, For use with, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other system that can fetch instructions from and execute instructions from an instruction execution system, apparatus, or apparatus) or equipment. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport the program for use by or in connection with an instruction execution system, apparatus, or apparatus.

More specific examples (non-exhaustive list) of computer readable media include the following: electrical connections with one or more wiring (electronic devices), portable computer disk cartridges (magnetic devices), random access memory (RAM), Read Only Memory (ROM), Erasable Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the above-mentioned program can be printed, as it is possible, for example, by optically scanning the paper or other medium, followed by editing, interpretation or other suitable means if necessary Processing is performed to obtain the above program electronically and then stored in computer memory.

It should be understood that various parts of the present invention may be implemented in hardware, software, firmware or a combination thereof. In the above-described embodiments, various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or a combination of the following techniques known in the art: Discrete logic circuits, application specific integrated circuits with suitable combinational logic gates, Programmable Gate Arrays (PGA), Field Programmable Gate Arrays (FPGA), etc.

In the above description of the present specification, reference to the description of the terms "one embodiment/example", "another embodiment/example" or "certain embodiments/examples" etc. means the description in conjunction with the embodiment or example. Particular features, structures, materials, or characteristics are included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, The scope of the invention is defined by the claims and their equivalents.

The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above-mentioned embodiments, and those skilled in the art can also make various equivalent deformations or replacements on the premise of not violating the spirit of the present invention. Equivalent modifications or substitutions are included within the scope defined by the claims of the present application.

Claims (9)

1. A target positioning method based on template matching is characterized by comprising the following steps:

acquiring a template image, constructing a template image pyramid according to the template image, and constructing a first grid according to the template image pyramid;

determining the average gradient amplitude of all pixel points in the first grid, and selecting the first grid with the average gradient amplitude being greater than or equal to a preset first threshold value as a second grid;

determining first gradient amplitude and first gradient direction entropy of all pixel points in the second grid, determining a first local feature point according to the first gradient amplitude and the first gradient direction entropy, and establishing a first LBP histogram feature vector according to the first local feature point;

acquiring a target image, constructing a target image pyramid according to the target image, determining a second local feature point of the target image pyramid, further establishing a second LBP histogram feature vector according to the second local feature point, and determining the position and the rotation angle of a target to be positioned according to the similarity matching of the first LBP histogram feature vector and the second LBP histogram feature vector;

the method comprises the steps of obtaining a target image, constructing a target image pyramid according to the target image, determining a second local feature point of the target image pyramid, further establishing a second LBP histogram feature vector according to the second local feature point, and determining the position and the rotating angle of a target to be positioned according to the similarity matching of the first LBP histogram feature vector and the second LBP histogram feature vector, wherein the steps specifically comprise:

acquiring a target image, constructing a target image pyramid according to the target image, and constructing a third grid according to the target image pyramid;

determining the average gradient amplitude of all pixel points in the third grid, and selecting the third grid with the average gradient amplitude being greater than or equal to a preset fourth threshold value as a fourth grid;

determining second gradient magnitude and second gradient direction entropy of all pixel points in the fourth grid, determining second local feature points according to the second gradient magnitude and the second gradient direction entropy, and establishing second LBP histogram feature vectors according to the second local feature points;

and calculating the similarity of the first LBP histogram feature vector of each layer in the template image pyramid and the target image pyramid and the second LBP histogram feature vector of the corresponding position, sequentially performing similarity matching from the top layer to the bottom layer, eliminating second local feature points of which the similarity is smaller than a preset fifth threshold value, and screening out second local feature points of which the similarity is larger than or equal to the fifth threshold value from the bottom layer of the target image pyramid, thereby determining the position and the rotation angle of the target to be positioned.

2. The target positioning method based on template matching according to claim 1, wherein the step of obtaining a template image, constructing a template image pyramid according to the template image, and constructing a first grid according to the template image pyramid specifically comprises:

acquiring a template image, and expanding the width and the height of the template image to obtain an expanded template image;

downsampling the extended template image to generate a template image pyramid;

extracting a first pyramid image of each layer in the template image pyramid, and constructing a first grid in a self-adaptive manner through quadtree segmentation according to the first pyramid image;

wherein the first pyramid image is covered by a corresponding first grid.

3. The template matching-based target positioning method according to claim 2, wherein in the step of adaptively constructing the first mesh from the first pyramid image through quadtree partitioning, when the width or height of the first mesh is smaller than a preset sixth threshold, the quadtree partitioning is not performed, and the sixth threshold satisfies the following condition:

wherein,μ _sa sixth threshold value is indicated which is,W _TL the width of the top level image representing the template image pyramid,H _TL representing the height of the topmost image of the template image pyramid.

4. The template matching-based target positioning method according to claim 1, wherein the step of determining an average gradient magnitude of all pixel points in the first mesh, and selecting the first mesh having the average gradient magnitude greater than or equal to a preset first threshold as the second mesh specifically includes:

calculating the sum of the gradient amplitudes of all the pixel points in the first grid, and determining the average gradient amplitude according to the number of the pixel points in the first grid and the sum of the gradient amplitudes;

and if the average gradient amplitude is determined to be larger than or equal to the first threshold, acquiring the corresponding first grid as a second grid.

5. The template matching-based target positioning method according to claim 1, wherein the step of determining a first gradient magnitude and a first gradient direction entropy of all pixel points in the second mesh, determining a first local feature point according to the first gradient magnitude and the first gradient direction entropy, and establishing a first LBP histogram feature vector according to the first local feature point specifically comprises:

determining first gradient amplitude and first gradient direction entropy values of all pixel points in the second grid;

acquiring a preset second threshold and a preset third threshold, and selecting pixel points of which the first gradient amplitude is greater than or equal to the second threshold and the first gradient direction entropy is greater than or equal to the third threshold as first local feature points;

and extracting a first LBP histogram feature vector by taking the first local feature point as a central pixel point.

6. The target positioning method based on template matching according to claim 5, wherein the step of extracting the first LBP histogram feature vector by using the first local feature point as a central pixel point specifically comprises:

and establishing a first local feature point LBP histogram feature vector by taking the first local feature point as a central pixel point and combining nine modes of which the uniform is less than or equal to 2 in the LBP equivalent mode.

7. An object positioning system based on template matching, comprising:

the grid construction module is used for acquiring a template image, constructing a template image pyramid according to the template image and constructing a first grid according to the template image pyramid;

the grid selection module is used for determining the average gradient amplitude of all pixel points in the first grid and selecting the first grid with the average gradient amplitude being greater than or equal to a preset first threshold value as a second grid;

an LBP histogram feature vector establishing module, configured to determine first gradient magnitudes and first gradient direction entropy values of all pixel points in the second grid, determine a first local feature point according to the first gradient magnitudes and the first gradient direction entropy values, and establish a first LBP histogram feature vector according to the first local feature point;

the similarity matching module is used for acquiring a target image, constructing a target image pyramid according to the target image, determining a second local feature point of the target image pyramid, further establishing a second LBP histogram feature vector according to the second local feature point, and determining the position and the rotation angle of the target to be positioned according to the similarity matching of the first LBP histogram feature vector and the second LBP histogram feature vector;

the similarity matching module is specifically configured to:

acquiring a target image, constructing a target image pyramid according to the target image, and constructing a third grid according to the target image pyramid;

determining the average gradient amplitude of all pixel points in the third grid, and selecting the third grid with the average gradient amplitude being greater than or equal to a preset fourth threshold value as a fourth grid;

determining second gradient amplitude and second gradient direction entropy values of all pixel points in the fourth grid, determining second local feature points according to the second gradient amplitude and the second gradient direction entropy values, and establishing second LBP histogram feature vectors according to the second local feature points;

and calculating the similarity of the first LBP histogram feature vector of each layer in the template image pyramid and the target image pyramid and the second LBP histogram feature vector of the corresponding position, sequentially performing similarity matching from the top layer to the bottom layer, eliminating second local feature points of which the similarity is smaller than a preset fifth threshold value, and screening out second local feature points of which the similarity is larger than or equal to the fifth threshold value from the bottom layer of the target image pyramid, thereby determining the position and the rotation angle of the target to be positioned.

8. An object positioning device based on template matching, comprising:

at least one processor;

at least one memory for storing at least one program;

when executed by the at least one processor, cause the at least one processor to implement a method of target location based on template matching as claimed in any one of claims 1 to 6.

9. A computer readable storage medium, in which a processor executable program is stored, wherein the processor executable program, when executed by a processor, is adapted to perform a method of template matching based object localization according to any of claims 1 to 6.

Images