A parameterized image correlation matching method, device, equipment and storage medium

Abstract

The application belongs to the technical field of image processing, and relates to a parameterized image correlation matching method, device, equipment and storage medium. The method comprises the following steps: acquiring a reference image and a real-time image, obtaining a first representation of the reference image according to the motion positioning condition of a mobile positioning system, and obtaining an initial representation of a normalized cross-correlation coefficient; performing pixel interpolation in a pixel window to obtain a second representation of the reference image, obtaining a transition representation of the normalized cross-correlation coefficient according to the initial representation of the normalized cross-correlation coefficient, and obtaining a final representation of the normalized cross-correlation coefficient according to the average pixel value of the resampled reference image; and inputting the reference image and the real-time image according to the final representation of the normalized cross-correlation coefficient to obtain an image displacement parameter between the reference image and the real-time image. The application can improve the matching efficiency and reliability of image matching while ensuring high matching accuracy.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a parameterized image correlation matching method, apparatus, device, and storage medium. Background Technology

[0002] With the rise of computer vision, image matching and comparison have found wide applications in numerous fields, including robotics, medical image processing, autonomous driving, security monitoring, and document processing. In the industrial sector, image matching technology has been widely used in precision industrial scenarios, such as the closed-loop servo systems of CNC machine tools, which require detection, measurement, and calibration.

[0003] Image matching refers to comparing one or more images with a known reference image or image library to identify, locate, or describe a target, feature, or object. For example, in two-dimensional image matching, the center point of the window with the highest correlation coefficient in the search region is obtained by comparing the correlation coefficients of windows of the same size in the target region and the search region. Essentially, it is an optimal search problem that applies matching criteria under the condition of primitive similarity.

[0004] In existing technologies, based on matching methods and strategies, image matching algorithms can be divided into four main types: feature point-based, region-based, template-based, and deep learning-based.

[0005] However, existing image matching algorithms have the following problems:

[0006] (1) Feature point matching algorithm. This type of algorithm relies on detecting and matching key feature points in the image, such as SIFT (Scale Invariant Feature Transform), SURF (Speed-Up Robust Feature Transform), ORB (Oriented Fast and Rotated BRIEF), etc. It has problems such as feature points becoming invisible due to changes in viewpoint and occlusion, different descriptors being generated under different lighting conditions, and requiring a large amount of computation.

[0007] (2) Region matching algorithm: The region matching algorithm divides the image into different regions and compares the similarity between these regions, such as image patch matching, HOG (histogram of oriented gradients) matching, etc. Such methods usually focus on local features and may ignore global information.

[0008] (3) Template matching algorithm: Template matching algorithm uses known templates to match images, such as standard cross-correlation, normalized cross-correlation, squared difference matching, etc. It needs to deal with scale and rotation invariance as well as computational complexity and real-time performance.

[0009] (4) Deep learning methods: Deep learning-based image matching algorithms use convolutional neural networks to learn feature representations and matching relationships, such as matching methods based on Siamese or Triplet networks. However, deep learning methods often require a large amount of labeled data to train the model. Collecting and labeling data is time-consuming and laborious. The model is complex and requires a lot of computing resources for training and inference. Summary of the Invention

[0010] Therefore, it is necessary to provide a parameterized image correlation matching method, apparatus, device, and medium to address the above-mentioned technical problems. This method can improve the matching efficiency and reliability of image matching while ensuring high matching accuracy, and is characterized by high accuracy, low complexity, and wide applicability.

[0011] A parameterized image correlation matching method, comprising:

[0012] Acquire reference images and real-time images. Based on the motion positioning of the mobile positioning system, obtain the first representation of the reference image and the initial representation of the normalized cross-correlation coefficient.

[0013] Pixel interpolation is performed within a pixel window to obtain a second representation of the reference image;

[0014] Based on the second representation of the reference image, the vector product representation of the reference image is obtained;

[0015] Based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, the first coefficient, the second coefficient, and the third coefficient are obtained; based on the first coefficient, the second coefficient, and the third coefficient, the transitional representation of the normalized cross-correlation coefficient is obtained, and based on the average pixel value of the resampled reference image, the final representation of the normalized cross-correlation coefficient is obtained.

[0016] Based on the final representation of the normalized cross-correlation coefficient, the reference image and the real-time image are input to obtain the image displacement parameters between the reference image and the real-time image.

[0017] In one embodiment, a reference image and a real-time image are acquired. Based on the motion positioning of the mobile positioning system, a first representation of the reference image is obtained, and an initial representation of the normalized cross-correlation coefficient is obtained, including:

[0018] Acquire reference images and real-time images, and obtain the coordinate transformation relationship between the reference images and real-time images based on the motion positioning of the mobile positioning system;

[0019] Based on the coordinate transformation relationship between the reference image and the real-time image, the first representation of the reference image is obtained;

[0020] Based on the first representation of the reference image, the normalized cross-correlation coefficient is expressed in fractional form, thus obtaining the initial representation of the normalized cross-correlation coefficient.

[0021] In one embodiment, pixel interpolation is performed within a pixel window to obtain a second representation of the reference image, including:

[0022] Pixel interpolation is performed within the pixel window, the nearest integer is selected as the reference point, and a second representation of the reference image is obtained according to the weight function.

[0023] In one embodiment, obtaining the vector product representation of the reference image based on the second representation of the reference image includes:

[0024] Based on the second representation of the reference image, considering multiple row vectors and multiple column vectors, we obtain the vector product representation of the reference image.

[0025] In one embodiment, the first coefficient, the second coefficient, and the third coefficient are obtained based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, including:

[0026] Based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, the numerator and denominator representations of the normalized cross-correlation coefficient are obtained.

[0027] The first coefficient is obtained based on the numerator representation of the normalized cross-correlation coefficient;

[0028] The second and third coefficients are obtained by representing the denominator of the normalized cross-correlation coefficient.

[0029] In one embodiment, a transitional representation of the normalized cross-correlation coefficient is obtained based on the first coefficient, the second coefficient, and the third coefficient, and a final representation of the normalized cross-correlation coefficient is obtained based on the average pixel value of the resampled reference image, including:

[0030] Based on the first, second, and third coefficients, a transitional representation of the normalized cross-correlation coefficient is obtained;

[0031] The final representation of the normalized cross-correlation coefficient is obtained by using the transitional representation of the normalized cross-correlation coefficient and referring to the average pixel value and row vector of the resampled reference image.

[0032] In one embodiment, based on the final representation of the normalized cross-correlation coefficient, a reference image and a real-time image are input to obtain image displacement parameters between the reference image and the real-time image, including:

[0033] Based on the final representation of the normalized cross-correlation coefficient, the maximum value of the normalized cross-correlation coefficient is obtained. Then, the reference image and the real-time image are input, and the gradient descent method is used to obtain the image displacement parameters between the reference image and the real-time image.

[0034] A parameterized image correlation matching device, comprising:

[0035] The acquisition module is used to acquire reference images and real-time images. Based on the motion positioning of the mobile positioning system, it obtains the first representation of the reference image and the initial representation of the normalized cross-correlation coefficient.

[0036] The interpolation module is used to perform pixel interpolation within a pixel window to obtain a second representation of the reference image;

[0037] The calculation module is used to obtain the vector product representation of the reference image based on the second representation of the reference image;

[0038] The transition module is used to obtain the first coefficient, the second coefficient, and the third coefficient based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient; to obtain the transition representation of the normalized cross-correlation coefficient based on the first coefficient, the second coefficient, and the third coefficient; and to obtain the final representation of the normalized cross-correlation coefficient based on the average pixel value of the resampled reference image.

[0039] The output module is used to obtain the image displacement parameters between the reference image and the real-time image based on the final representation of the normalized cross-correlation coefficient.

[0040] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:

[0041] Acquire reference images and real-time images. Based on the motion positioning of the mobile positioning system, obtain the first representation of the reference image and the initial representation of the normalized cross-correlation coefficient.

[0042] Pixel interpolation is performed within a pixel window to obtain a second representation of the reference image;

[0043] Based on the second representation of the reference image, the vector product representation of the reference image is obtained;

[0044] Based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, the first coefficient, the second coefficient, and the third coefficient are obtained; based on the first coefficient, the second coefficient, and the third coefficient, the transitional representation of the normalized cross-correlation coefficient is obtained, and based on the average pixel value of the resampled reference image, the final representation of the normalized cross-correlation coefficient is obtained.

[0045] Based on the final representation of the normalized cross-correlation coefficient, the reference image and the real-time image are input to obtain the image displacement parameters between the reference image and the real-time image.

[0046] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0047] Acquire reference images and real-time images. Based on the motion positioning of the mobile positioning system, obtain the first representation of the reference image and the initial representation of the normalized cross-correlation coefficient.

[0048] Pixel interpolation is performed within a pixel window to obtain a second representation of the reference image;

[0049] Based on the second representation of the reference image, the vector product representation of the reference image is obtained;

[0050] Based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, the first coefficient, the second coefficient, and the third coefficient are obtained; based on the first coefficient, the second coefficient, and the third coefficient, the transitional representation of the normalized cross-correlation coefficient is obtained, and based on the average pixel value of the resampled reference image, the final representation of the normalized cross-correlation coefficient is obtained.

[0051] Based on the final representation of the normalized cross-correlation coefficient, the reference image and the real-time image are input to obtain the image displacement parameters between the reference image and the real-time image.

[0052] The aforementioned parameterized image correlation matching method improves image matching accuracy by explicitly using parameterized image correlation matching. It calculates normalized cross-correlation coefficients using parameterized image displacement to represent the relationships between images, particularly in cases of viewpoint changes, illumination variations, and deformation, enabling more precise matching of image features. This application employs an efficient method to represent the relationships between images, avoiding complex image interpolation or resampling processes. When matching in large-scale image databases, it reduces computational complexity and increases matching speed. This application exhibits strong robustness, is suitable for diverse application scenarios, and demonstrates significant effectiveness in high-precision applications, such as precision industrial manufacturing. Attached Figure Description

[0053] Figure 1 This is an application scenario diagram of a parameterized image correlation matching method in one embodiment;

[0054] Figure 2 This is a flowchart illustrating a parameterized image correlation matching method in one embodiment;

[0055] Figure 3This is a schematic diagram of image interpolation for a 4-pixel × 4-pixel region in one embodiment.

[0056] Figure 4 This is a structural block diagram of a parameterized image correlation matching device in one embodiment;

[0057] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0059] It should be noted that all directional indicators (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicator will also change accordingly.

[0060] Furthermore, the use of terms such as "first" and "second" in this application is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of those features. In the description of this application, "multiple sets" means at least two sets, such as two sets, three sets, etc., unless otherwise explicitly specified.

[0061] In this application, unless otherwise expressly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection, an electrical connection, a physical connection, or a wireless communication connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two elements or the interaction between two elements, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0062] Furthermore, the technical solutions of the various embodiments of this application can be combined with each other, but only if they are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by this application.

[0063] The method provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. Terminal 102 may include, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. Server 104 may be a server corresponding to various portal websites or work system backends.

[0064] This application provides a parameterized image correlation matching method, such as Figure 2 As shown, in one embodiment, the method is applied to Figure 1 Taking the terminal in the example, the explanation includes:

[0065] Step 202: Obtain the reference image and the real-time image. Based on the motion positioning of the mobile positioning system, obtain the first representation of the reference image and the initial representation of the normalized cross-correlation coefficient.

[0066] Specifically:

[0067] Acquire reference images and real-time images, and obtain the coordinate transformation relationship between the reference images and real-time images based on the motion positioning of the mobile positioning system;

[0068] Based on the coordinate transformation relationship between the reference image and the real-time image, the first representation of the reference image is obtained;

[0069] Based on the first representation of the reference image, the normalized cross-correlation coefficient is expressed in fractional form, thus obtaining the initial representation of the normalized cross-correlation coefficient.

[0070] More specifically:

[0071] Obtain reference image Given a real-time image I(x,y), and based on the motion positioning of the Landing System (LSM), the relationship between the reference image and the real-time image involves only 2D image displacement, thus yielding the coordinate transformation relationship between the reference image and the real-time image:

[0072]

[0073] In the formula, x and y represent the real-time image coordinates. and represents the corresponding reference image coordinates, and s and t are two different parameters of image displacement;

[0074] Based on the coordinate transformation relationship between the reference image and the real-time image, the first representation of the reference image is obtained:

[0075]

[0076] In the formula, R represents the expression... and A function for the pixel values ​​of variables;

[0077] Based on the first representation of the reference image, the normalized cross-correlation coefficient (ZNCC) between the reference image and the real-time image is expressed in fractional form, resulting in the initial representation of the normalized cross-correlation coefficient:

[0078]

[0079] In the formula, C(s,t) is the normalized cross-correlation coefficient. Let I(x,y) be the average pixel value of the reference image, and let I(x,y) be the pixel value of the real-time image. This represents the average pixel value of the real-time image.

[0080] In this step, an image transformation model is constructed, where the normalized cross-correlation coefficient is a function of the parameters (s and t), but cannot be obtained directly from the parameters, because R(x+s,y+t) can only be obtained through image interpolation or image resampling based on s and t.

[0081] Step 204: Perform pixel interpolation within the pixel window to obtain a second representation of the reference image.

[0082] Specifically:

[0083] Pixel interpolation is performed within the pixel window, the nearest integer is selected as the reference point, and a second representation of the reference image is obtained according to the weight function.

[0084] More specifically:

[0085] Assuming that the image displacements s and t are both less than one pixel, and pixel interpolation is performed within a 4-pixel × 4-pixel window, such as... Figure 3 As shown, (i,j) represents the coordinates of a pixel.

[0086] When needed When calculating pixel values, the nearest top-left integer is selected as the reference point. The reference point is used as the pixel value g(x,y) of the weight function for that pixel. Then, interpolation is performed, and the result is calculated based on the weight function P. i (u) yields the second representation of the reference image:

[0087]

[0088] Among them, P m+1 (s) and P n+1 (t) are all weight functions P i (u), u is P i The parameters can be replaced by s or t, and P is obtained using bicubic interpolation. i (u) can be represented by the following equation:

[0089]

[0090] Among them, P i In (u), i can be either m or n.

[0091] When P i When i takes the value m in (u), equation (5) is expressed as a vector product as follows:

[0092] P m (u)=[a m0 ,a m1 ,a m2 ,a m3 ][1,u,u 2 ,u 3 ] T (6)

[0093] Let: B = [a] m0 ,a m1 ,a m2 ,a m3 B is a coefficient matrix determined by the interpolation method and is a known quantity.

[0094] Equation (4) can be expressed as follows, which gives the second representation of the reference image:

[0095]

[0096] In the formula, g is the pixel value function, m is the m-th row of the pixel grid, n is the n-th column of the pixel grid, [a m0 ,a m1 ,a m2 ,a m3 ] is the coefficient matrix, and T is the transpose.

[0097] In this step, a reference image representation is prepared, which is written in another form to represent the normalized cross-correlation coefficient through parameters s and t.

[0098] Step 206: Obtain the vector product representation of the reference image based on the second representation of the reference image.

[0099] Specifically:

[0100] Based on the second representation of the reference image, considering multiple row vectors and multiple column vectors, we obtain the vector product representation of the reference image.

[0101] More specifically:

[0102] Based on the second representation of the reference image, i.e., equation (7), and according to the interpolation method, the reference image is represented as a product of row vectors and column vectors, thus obtaining the vector product representation of the reference image:

[0103] R(x+s,y+t)=G(x,y)A(s,t) (8)

[0104] in:

[0105] G(x,y)=[g y-1,x-1 ,g y-1,x ,…,g y+1,x+2 ,g y+2,x+2 (9)

[0106] A(s,t)=[P0(s)P0(y),P0(s)P1(t),…,P3(s)P3(t)] T (10)

[0107] In the formula, G(x,y) is a row vector with 16 terms, and A(s,t) is a column vector with 16 terms, which can be directly calculated based on s and t.

[0108] In this step, the image translation parameters are calculated, and P is determined. i The problem of (u) is transformed into finding A(s,t).

[0109] Step 208: Based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, obtain the first coefficient, the second coefficient, and the third coefficient; based on the first coefficient, the second coefficient, and the third coefficient, obtain the transitional representation of the normalized cross-correlation coefficient; and based on the average pixel value of the resampled reference image, obtain the final representation of the normalized cross-correlation coefficient.

[0110] Specifically:

[0111] Based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, the numerator and denominator representations of the normalized cross-correlation coefficient are obtained; based on the numerator representation of the normalized cross-correlation coefficient, the first coefficient is obtained; based on the denominator representation of the normalized cross-correlation coefficient, the second and third coefficients are obtained.

[0112] Based on the first, second, and third coefficients, a transitional representation of the normalized cross-correlation coefficient is obtained; based on the transitional representation of the normalized cross-correlation coefficient, and referring to the average pixel value and row vector of the resampled reference image, the final representation of the normalized cross-correlation coefficient is obtained.

[0113] More specifically:

[0114] Based on the vector product representation of the reference image (i.e., equation (8)) and the initial representation of the normalized cross-correlation coefficient (i.e., equation (3)), the numerators of the normalized cross-correlation coefficient are directly multiplied apart to obtain:

[0115]

[0116] In the formula, the sum of the pixel values ​​of each point is expressed as the product of the number of points and the average pixel value:

[0117]

[0118] Then, according to equations (11) and (12), the numerator representation of the normalized cross-correlation coefficient is obtained:

[0119]

[0120] Similarly, based on the vector product representation of the reference image (i.e., Equation (8)) and the initial representation of the normalized cross-correlation coefficient (i.e., Equation (3)), the denominators of the normalized cross-correlation coefficient are directly multiplied apart to obtain the denominator representation of the normalized cross-correlation coefficient:

[0121]

[0122] In the formula, N is the number of interpolation points used;

[0123] According to the numerator representation of the normalized cross-correlation coefficient (i.e., equation (13)), we get:

[0124]

[0125] in, Let C be a row vector containing 16 terms, defined by its first coefficient C, then:

[0126]

[0127] Similarly, according to the denominator representation of the normalized cross-correlation coefficient (i.e., equation (14)), we get:

[0128]

[0129] in, Given a 16×16 matrix, defined by the second coefficient D, we get:

[0130]

[0131] Similarly, according to the denominator representation of the normalized cross-correlation coefficient (i.e., equation (14)), Defined as the third coefficient E, we get:

[0132]

[0133] In the formula, G(x,y) is the pixel value at the x and y coordinates;

[0134] Based on the first coefficient C, the second coefficient D, and the third coefficient E, the transitional representation of the normalized cross-correlation coefficient is obtained, that is, equation (3) can be expressed as:

[0135]

[0136] Among them, the average pixel value of the resampled reference image It can be represented as:

[0137]

[0138] Where F is a row vector containing 16 terms, it can be represented as:

[0139]

[0140] Substituting equation (21) into equation (20), we obtain the final expression of the normalized cross-correlation coefficient:

[0141]

[0142] In this step, the cost function is calculated.

[0143] Step 210: Based on the final representation of the normalized cross-correlation coefficient, input the reference image and the real-time image to obtain the image displacement parameters between the reference image and the real-time image.

[0144] Specifically:

[0145] Based on the final representation of the normalized cross-correlation coefficient (i.e., equation (23)), the maximum value of the normalized cross-correlation coefficient is obtained. The reference image and the real-time image are input, and the gradient descent method is used to obtain the image displacement parameters between the reference image and the real-time image.

[0146] In this step, image displacement is calculated. Specifically, in formula (23), the displacement can be calculated based on the reference image. C, F, and D are directly calculated from the real-time image I(x,y). D and F can be prepared in advance based on the reference image, while matrix B is a known matrix determined by the interpolation method. In other words, B, C, D, and F are all known quantities. Therefore, during image correlation matching, the normalized cross-correlation coefficient C(s,t) can be explicitly expressed as a function of A(s,t). Thus, when the normalized cross-correlation coefficient C(s,t) reaches its maximum value, A(s,t) is obtained based on the global maximum value of the normalized cross-correlation coefficient. The corresponding image displacements s and t can then be determined using the gradient descent method, resulting in accurate image displacements. This allows us to obtain the pixel offsets, completing the image correlation matching process without requiring image resampling.

[0147] The aforementioned parameterized image correlation matching method improves image matching accuracy by explicitly using parameterized image correlation matching. It calculates normalized cross-correlation coefficients using parameterized image displacement to represent the relationships between images, particularly in cases of viewpoint changes, illumination variations, and deformation, enabling more precise matching of image features. This application employs an efficient method to represent the relationships between images, avoiding complex image interpolation or resampling processes. When matching in large-scale image databases, it reduces computational complexity and increases matching speed. This application exhibits strong robustness, is suitable for diverse application scenarios, and demonstrates significant effectiveness in high-precision applications, such as precision industrial manufacturing.

[0148] It should be understood that, although Figure 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0149] This application also provides a parameterized image correlation matching device, such as... Figure 4 As shown, in one embodiment, it includes: an acquisition module 402, an interpolation module 404, a calculation module 406, a transition module 408, and an output module 410, wherein:

[0150] The acquisition module 402 is used to acquire a reference image and a real-time image, obtain a first representation of the reference image based on the motion positioning of the mobile positioning system, and obtain an initial representation of the normalized cross-correlation coefficient.

[0151] Interpolation module 404 is used to perform pixel interpolation within a pixel window to obtain a second representation of the reference image;

[0152] The calculation module 406 is used to obtain the vector product representation of the reference image based on the second representation of the reference image;

[0153] The transition module 408 is used to obtain the first coefficient, the second coefficient, and the third coefficient based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient; to obtain the transition representation of the normalized cross-correlation coefficient based on the first coefficient, the second coefficient, and the third coefficient; and to obtain the final representation of the normalized cross-correlation coefficient based on the average pixel value of the resampled reference image.

[0154] The output module 410 is used to obtain the image displacement parameters between the reference image and the real-time image based on the final representation of the normalized cross-correlation coefficient.

[0155] For specific limitations regarding a parameterized image correlation matching device, please refer to the limitations regarding a parameterized image correlation matching method above, which will not be repeated here. Each module in the above device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0156] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When executed by the processor, the computer program implements a parameterized image correlation matching method. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0157] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0158] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method described above.

[0159] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0160] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0161] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0162] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A parameterized image correlation matching method, characterized in that, include: Acquire a reference image and a real-time image. Based on the motion positioning information of the mobile positioning system, obtain the first representation of the reference image: ； In the formula, Indicated by and A function for the pixel values ​​of variables; and Represents real-time image coordinates. and Represents the corresponding reference image coordinates, and These are two different parameters of image displacement; And the initial representation of the normalized cross-correlation coefficient is obtained: ； In the formula, The average pixel value of the reference image. These are the pixel values ​​of the real-time image. The average pixel value of the real-time image; Pixel interpolation is performed within a pixel window, and a second representation of the reference image is obtained using the bicubic interpolation method. ； In the formula, For pixel value functions, For transpose; For weighting functions; Based on the second representation of the reference image, the vector product representation of the reference image is obtained: ； In the formula, It is a row vector. It is a column vector; Based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, the first coefficient, the second coefficient, and the third coefficient are obtained; the first coefficient for: ； The second coefficient for: ； The third coefficient for: ； Based on the first, second, and third coefficients, the transitional representation of the normalized cross-correlation coefficient is obtained: ； And based on the average pixel value of the resampled reference image The final representation of the normalized cross-correlation coefficient is obtained; where, ; The number of interpolation points used; The final representation is as follows: ； Based on the final representation of the normalized cross-correlation coefficient, the reference image and the real-time image are input to obtain the image displacement parameters between the reference image and the real-time image.

2. The parameterized image correlation matching method according to claim 1, characterized in that, Acquire a reference image and a real-time image. Based on the motion positioning of the mobile positioning system, obtain a first representation of the reference image and an initial representation of the normalized cross-correlation coefficient, including: Acquire reference images and real-time images, and obtain the coordinate transformation relationship between the reference images and real-time images based on the motion positioning of the mobile positioning system; Based on the coordinate transformation relationship between the reference image and the real-time image, the first representation of the reference image is obtained; Based on the first representation of the reference image, the normalized cross-correlation coefficient is expressed in fractional form, thus obtaining the initial representation of the normalized cross-correlation coefficient.

3. The parameterized image correlation matching method according to claim 2, characterized in that, Pixel interpolation is performed within a pixel window to obtain a second representation of the reference image, including: Pixel interpolation is performed within the pixel window, the nearest integer is selected as the reference point, and a second representation of the reference image is obtained according to the weight function.

4. The parameterized image correlation matching method according to claim 3, characterized in that, Based on the second representation of the reference image, the vector product representation of the reference image is obtained, including: Based on the second representation of the reference image, considering multiple row vectors and multiple column vectors, we obtain the vector product representation of the reference image.

5. The parameterized image correlation matching method according to claim 4, characterized in that, Based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, the first coefficient, the second coefficient, and the third coefficient are obtained, including: Based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient, the numerator and denominator representations of the normalized cross-correlation coefficient are obtained. The first coefficient is obtained based on the numerator representation of the normalized cross-correlation coefficient; The second and third coefficients are obtained by representing the denominator of the normalized cross-correlation coefficient.

6. The parameterized image correlation matching method according to claim 5, characterized in that, Based on the first, second, and third coefficients, a transitional representation of the normalized cross-correlation coefficient is obtained, and based on the average pixel values ​​of the resampled reference image, the final representation of the normalized cross-correlation coefficient is obtained, including: Based on the first, second, and third coefficients, a transitional representation of the normalized cross-correlation coefficient is obtained; The final representation of the normalized cross-correlation coefficient is obtained by using the transitional representation of the normalized cross-correlation coefficient and referring to the average pixel value and row vector of the resampled reference image.

7. The parameterized image correlation matching method according to claim 6, characterized in that, Based on the final representation of the normalized cross-correlation coefficient, the reference image and the real-time image are input to obtain the image displacement parameters between the reference image and the real-time image, including: Based on the final representation of the normalized cross-correlation coefficient, the maximum value of the normalized cross-correlation coefficient is obtained. Then, the reference image and the real-time image are input, and the gradient descent method is used to obtain the image displacement parameters between the reference image and the real-time image.

8. A parameterized image correlation matching device, characterized in that, The apparatus employs a parameterized image correlation matching method according to any one of claims 1 to 7, the apparatus comprising: The acquisition module is used to acquire reference images and real-time images. Based on the motion positioning of the mobile positioning system, it obtains the first representation of the reference image and the initial representation of the normalized cross-correlation coefficient. The interpolation module is used to perform pixel interpolation within a pixel window to obtain a second representation of the reference image; The calculation module is used to obtain the vector product representation of the reference image based on the second representation of the reference image; The transition module is used to obtain the first coefficient, the second coefficient, and the third coefficient based on the vector product representation of the reference image and the initial representation of the normalized cross-correlation coefficient; to obtain the transition representation of the normalized cross-correlation coefficient based on the first coefficient, the second coefficient, and the third coefficient; and to obtain the final representation of the normalized cross-correlation coefficient based on the average pixel value of the resampled reference image. The output module is used to obtain the image displacement parameters between the reference image and the real-time image based on the final representation of the normalized cross-correlation coefficient, taking the reference image and the real-time image as inputs.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images