Intervertebral foramen endoscope image calibration method and system

Abstract

The invention relates to the technical field of data identification, in particular to an intervertebral foramen mirror image calibration method and system. The method comprises the following steps: obtaining a standard image of an intervertebral foramen endoscope operation area, and carrying out intervertebral foramen key feature point set matching and position change feature sampling processing to obtain an intervertebral foramen position change feature point set; performing three-dimensional coordinate system matching positioning and calibration plate projection conversion on the standard image of the foramen intervertebrale endoscope operation area to obtain a foramen intervertebrale calibration plate space coordinate point set; calibration plate angular point detection and angular point posture feature point set detection are carried out on the space coordinate point set of the foramen intervertebrale calibration plate to obtain an angular point position feature point set of the foramen intervertebrale calibration plate and an angular point angle feature point set of the foramen intervertebrale calibration plate; and performing calibration mapping transformation matrix estimation calculation and image calibration correction on the intervertebral foramen position change three-dimensional matching coordinate system point set to obtain an intervertebral foramen image calibration correction coordinate system point set. According to the invention, accurate positioning and precision of foramen intervertebrale mirror image calibration can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data recognition, and in particular to a method and system for calibrating a percutaneous endoscopic image. Background Art

[0002] Transforaminal endoscopic lumbar discectomy is a minimally invasive surgical technique. In this surgery, it is necessary to accurately understand the structure and position of the intervertebral foramen to ensure the safety and effectiveness of the surgery. Image calibration is one of the key steps to obtain accurate spatial information. At present, traditional transforaminal endoscopic image calibration methods mainly rely on traditional image processing technology. However, these methods have the problems of low calibration accuracy and inaccurate positioning. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a percutaneous endoscopic image calibration method and system to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for calibrating a transforaminal endoscopic image comprises the following steps:

[0005] Step S1: Obtain a standard image of the percutaneous endoscopic surgery area, and match the key feature point set of the intervertebral foramen on the standard image of the percutaneous endoscopic surgery area to obtain a feature point set of the intervertebral foramen edge texture and a feature point set of the intervertebral foramen contour structure; perform position change feature sampling processing on the feature point set of the intervertebral foramen contour structure based on the feature point set of the intervertebral foramen edge texture to obtain a feature point set of the intervertebral foramen position change;

[0006] Step S2: performing three-dimensional coordinate system matching and positioning of the standard image of the intervertebral foraminal endoscopic surgery area based on the intervertebral foraminal position change feature point set to obtain the intervertebral foraminal position change three-dimensional matching coordinate system point set; performing calibration plate projection transformation on the intervertebral foraminal position change three-dimensional matching coordinate system point set to obtain the intervertebral foraminal calibration plate space coordinate point set;

[0007] Step S3: performing calibration plate corner point detection on the intervertebral foramen calibration plate spatial coordinate point set to obtain an intervertebral foramen calibration plate corner point set; performing corner point posture feature point set detection on the intervertebral foramen calibration plate corner point set to obtain an intervertebral foramen calibration plate corner point position feature point set and an intervertebral foramen calibration plate corner point angle feature point set;

[0008] Step S4: Based on the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set, a calibration mapping transformation matrix is ​​estimated and calculated for the intervertebral foramen position change three-dimensional matching coordinate system point set to obtain the intervertebral foramen calibration mapping transformation estimation matrix; the intervertebral foramen calibration mapping transformation estimation matrix is ​​used to perform image calibration correction on the intervertebral foramen calibration plate space coordinate point set to obtain the intervertebral foramen image calibration correction coordinate system point set.

[0009] The present invention first obtains the image of the intervertebral foraminal endoscopic surgery process, and performs image preprocessing on the acquired intervertebral foraminal endoscopic surgery area image, so that it can accurately analyze the edge contour structure of the intervertebral foramen, thereby providing a data basis guarantee for the subsequent processing process. By matching the intervertebral foraminal key feature point set with the standard image of the intervertebral foraminal endoscopic surgery area, the edge texture feature point set and the contour structure feature point set of the intervertebral foramen are obtained. The implementation of this step can effectively extract the key feature points of the intervertebral foraminal area, and provide reliable data support for the subsequent position change analysis and processing process. At the same time, by sampling the position change feature of the intervertebral foraminal contour structure feature point set based on the intervertebral foraminal edge texture feature point set, the dynamic change of the position of the intervertebral foramen can be understood in real time, which provides a basis for the subsequent spatial three-dimensional coordinate system positioning, and helps to improve the accuracy and safety of image calibration. Secondly, by combining the intervertebral foraminal position change feature point set to match and locate the standard image of the intervertebral foraminal endoscopic surgery area with the three-dimensional coordinate system, the feature points of the intervertebral foraminal position change in the image can be accurately matched to the actual three-dimensional spatial position. This step can achieve the accurate correspondence between the image and the actual anatomical structure, and provide reliable data support for the subsequent processing process. By performing calibration plate projection transformation on the three-dimensional matching coordinate system point set of the intervertebral foramen position change, the matched three-dimensional space coordinate point set can be projected into the calibration plate space, thereby obtaining the intervertebral foramen calibration plate space coordinate point set. This process can realize the space coordinate transformation and calibration of the intervertebral foramen image, provide more accurate positioning and navigation information for the subsequent image calibration process, and help improve the success rate and accuracy of intervertebral foraminal endoscope image calibration. Then, by performing calibration plate corner point detection on the intervertebral foramen calibration plate space coordinate point set, the corner point position on the intervertebral foramen calibration plate can be accurately identified. The main function of this step is to determine the geometric shape of the calibration plate, which provides important basic data for subsequent positioning and navigation. Subsequently, by analyzing the corner point posture feature of the intervertebral foraminal calibration plate, rich corner point feature information can be extracted from the calibration plate image to discover the changing rules and patterns of the corner points under different conditions. By detecting the feature point set according to the position and angle feature changes of the corner points, the representative position feature points and angle feature points on the calibration plate can be accurately extracted. These feature points can describe the position relationship and rotation angle of the calibration plate under different postures, thus providing an important reference basis for the subsequent processing process. By accurately detecting the position feature points and angle feature points, the accurate positioning and navigation of the intervertebral foraminal endoscope image calibration can be achieved, providing more reliable data support for the subsequent processing process. Finally, by mapping the position feature point set and angle feature point set extracted from the intervertebral foraminal calibration plate to the three-dimensional matching coordinate system point set of the intervertebral foraminal position change, the key to this step is to convert the feature point information on the intervertebral foraminal calibration plate into the matching coordinate system, thereby achieving accurate description and matching of the intervertebral foraminal position change, providing an important data basis for subsequent image calibration and positioning.By establishing the matrix form of the calibration mapping transformation matrix according to the calibration mapping transformation relationship of the calibration mapping process, the transformation relationship of the position and angle can be uniformly expressed. In addition, by estimating each parameter matrix of the calibration mapping transformation and calculating the intervertebral foramen calibration mapping transformation estimation matrix, this step is based on the aforementioned error estimation result, and the final calibration mapping transformation estimation matrix is ​​estimated through mathematical calculation. The matrix can accurately describe the position and angle transformation relationship between the intervertebral foramen calibration plate and the matching coordinate system. In addition, by using the intervertebral foramen calibration mapping transformation estimation matrix to perform image calibration correction on the intervertebral foramen calibration plate space coordinate point set, the intervertebral foramen image can be corrected according to the estimated mapping transformation matrix, so that the feature points in the image can be accurately calibrated and mapped to the matching coordinate system, thereby improving the accuracy of intervertebral foraminal endoscope image calibration and the accuracy of calibration positioning.

[0010] Preferably, the present invention further provides a percutaneous foraminal endoscopy image calibration system for executing the percutaneous foraminal endoscopy image calibration method as described above, the percutaneous foraminal endoscopy image calibration system comprising:

[0011] The intervertebral foramen position feature point set analysis module is used to obtain the standard image of the intervertebral foraminal endoscopic surgery area, and match the intervertebral foraminal key feature point set on the standard image of the intervertebral foraminal endoscopic surgery area to obtain the intervertebral foraminal edge texture feature point set and the intervertebral foraminal contour structure feature point set; based on the intervertebral foraminal edge texture feature point set, the intervertebral foraminal contour structure feature point set is sampled for position change features to obtain the intervertebral foraminal position change feature point set;

[0012] The intervertebral foramen calibration plate projection conversion module is used to perform three-dimensional coordinate system matching and positioning of the standard image of the intervertebral foraminal endoscopic surgery area based on the intervertebral foraminal position change feature point set to obtain the intervertebral foraminal position change three-dimensional matching coordinate system point set; the intervertebral foraminal position change three-dimensional matching coordinate system point set is subjected to calibration plate projection conversion to obtain the intervertebral foraminal calibration plate space coordinate point set;

[0013] The calibration plate corner point posture feature analysis module is used to detect the calibration plate corner points on the intervertebral foramen calibration plate spatial coordinate point set to obtain the intervertebral foramen calibration plate corner point set; perform corner point posture feature point set detection on the intervertebral foramen calibration plate corner point set to obtain the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set;

[0014] The intervertebral foramen image calibration and correction module is used to estimate and calculate the calibration mapping transformation matrix of the three-dimensional matching coordinate system point set of the intervertebral foramen position change based on the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set to obtain the intervertebral foramen calibration mapping transformation estimation matrix; use the intervertebral foramen calibration mapping transformation estimation matrix to perform image calibration and correction on the intervertebral foramen calibration plate space coordinate point set, thereby obtaining the intervertebral foramen image calibration and correction coordinate system point set.

[0015] In summary, the present invention provides a percutaneous foraminal endoscopic image calibration system, which consists of an intervertebral foraminal position feature point set analysis module, an intervertebral foraminal calibration plate projection conversion module, a calibration plate corner point posture feature analysis module and an intervertebral foraminal image calibration correction module. It can implement any one of the percutaneous foraminal endoscopic image calibration methods described in the present invention, and is used to combine the operations between computer programs running on each module to implement any one of the percutaneous foraminal endoscopic image calibration methods. The internal structures of the system cooperate with each other, which can greatly reduce duplication of work and manpower investment, and can quickly and effectively provide a more accurate and efficient percutaneous foraminal endoscopic image calibration process, thereby simplifying the operation process of the percutaneous foraminal endoscopic image calibration system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0017] Figure 1 It is a schematic diagram of the steps of the percutaneous foraminal endoscope image calibration method of the present invention;

[0018] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0019] Figure 3 for Figure 2 Detailed step flow chart of step S14 in FIG. DETAILED DESCRIPTION

[0020] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0022] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0023] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for calibrating a transforaminal endoscopic image, the method comprising the following steps:

[0024] Step S1: Obtain a standard image of the percutaneous endoscopic surgery area, and match the key feature point set of the intervertebral foramen on the standard image of the percutaneous endoscopic surgery area to obtain a feature point set of the intervertebral foramen edge texture and a feature point set of the intervertebral foramen contour structure; perform position change feature sampling processing on the feature point set of the intervertebral foramen contour structure based on the feature point set of the intervertebral foramen edge texture to obtain a feature point set of the intervertebral foramen position change;

[0025] Step S2: performing three-dimensional coordinate system matching and positioning of the standard image of the intervertebral foraminal endoscopic surgery area based on the intervertebral foraminal position change feature point set to obtain the intervertebral foraminal position change three-dimensional matching coordinate system point set; performing calibration plate projection transformation on the intervertebral foraminal position change three-dimensional matching coordinate system point set to obtain the intervertebral foraminal calibration plate space coordinate point set;

[0026] Step S3: performing calibration plate corner point detection on the intervertebral foramen calibration plate spatial coordinate point set to obtain an intervertebral foramen calibration plate corner point set; performing corner point posture feature point set detection on the intervertebral foramen calibration plate corner point set to obtain an intervertebral foramen calibration plate corner point position feature point set and an intervertebral foramen calibration plate corner point angle feature point set;

[0027] Step S4: Based on the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set, a calibration mapping transformation matrix is ​​estimated and calculated for the intervertebral foramen position change three-dimensional matching coordinate system point set to obtain the intervertebral foramen calibration mapping transformation estimation matrix; the intervertebral foramen calibration mapping transformation estimation matrix is ​​used to perform image calibration correction on the intervertebral foramen calibration plate space coordinate point set to obtain the intervertebral foramen image calibration correction coordinate system point set.

[0028] In the embodiment of the present invention, please refer to Figure 1 FIG. 1 is a schematic diagram of the steps of the percutaneous foraminal endoscopy image calibration method of the present invention. In this example, the steps of the percutaneous foraminal endoscopy image calibration method include:

[0029] Step S1: Obtain a standard image of the percutaneous endoscopic surgery area, and match the key feature point set of the intervertebral foramen on the standard image of the percutaneous endoscopic surgery area to obtain a feature point set of the intervertebral foramen edge texture and a feature point set of the intervertebral foramen contour structure; perform position change feature sampling processing on the feature point set of the intervertebral foramen contour structure based on the feature point set of the intervertebral foramen edge texture to obtain a feature point set of the intervertebral foramen position change;

[0030] In an embodiment of the present invention, an optical lens is installed on a percutaneous endoscopic lens to capture images of the percutaneous endoscopic surgery process, so as to obtain the initial image of the percutaneous endoscopic surgery process in real time, and the collected initial image of the percutaneous endoscopic surgery area is pre-processed by using an appropriate image denoising algorithm (such as Gaussian filtering, median filtering, etc.), an image enhancement algorithm, and a standardization method, so as to effectively reduce the noise level in the initial image of the percutaneous endoscopic surgery area, improve the quality of the initial image of the percutaneous endoscopic surgery area, enhance the contrast of the image so that the image parameters such as brightness and color can meet the standards, thereby obtaining a standard image of the percutaneous endoscopic surgery area. Secondly, by using appropriate image processing techniques (including image texture feature analysis and image contour structure analysis methods), feature analysis is performed on the standard image of the perforaminal endoscopic surgery area to deeply analyze the edge texture and contour structure in the standard image of the perforaminal endoscopic surgery area, and the corresponding key feature points are extracted, including texture feature points such as edge texture roughness and direction, and structural feature points such as the shape, size and position of the intervertebral foramen. The standard image of the perforaminal endoscopic surgery area is analyzed by using a position matching analysis method to match the feature points in the image with the intervertebral foramen position in the actual surgical scene, so as to more accurately understand the edge texture feature point set of the actual intervertebral foramen area (such as the actual texture distribution and texture changes during intervertebral foramen surgery) and the contour structure feature point set (such as the actual intervertebral foramen contour shape and size during intervertebral foramen surgery), thereby obtaining the intervertebral foramen edge texture feature point set and the intervertebral foramen contour structure feature point set. Finally, the feature point set of the intervertebral foramen edge texture is combined with the feature point set of the intervertebral foramen edge contour structure to detect and identify the deformation of the intervertebral foramen edge contour structure, and understand the change trend, morphological evolution process and other information of the intervertebral foramen edge contour structure area. At the same time, according to the change trend, morphological evolution process and other information of the intervertebral foramen edge contour structure area, the position movement change information of the intervertebral foramen contour is accurately captured and tracked, and a representative feature point set is selected from it, and finally the intervertebral foramen position change feature point set is obtained.

[0031] Step S2: performing three-dimensional coordinate system matching and positioning of the standard image of the intervertebral foraminal endoscopic surgery area based on the intervertebral foraminal position change feature point set to obtain the intervertebral foraminal position change three-dimensional matching coordinate system point set; performing calibration plate projection transformation on the intervertebral foraminal position change three-dimensional matching coordinate system point set to obtain the intervertebral foraminal calibration plate space coordinate point set;

[0032] The embodiment of the present invention processes the standard image of the intervertebral foraminal endoscopic surgery area by using a three-dimensional coordinate system modeling method to map the standard image of the intervertebral foraminal endoscopic surgery area from a two-dimensional plane to a three-dimensional space, and annotates and samples the feature points of the mapped three-dimensional coordinate system space to effectively extract a representative feature point set in the intervertebral foraminal image. At the same time, the mapping matching degree between the feature point set of the intervertebral foraminal position change and the feature point set in the three-dimensional space feature point set of the intervertebral foraminal image is calculated and quantitatively evaluated, which is used for matching and positioning of the three-dimensional matching coordinate system point set of the intervertebral foraminal position change, so as to accurately correspond the feature points of the intervertebral foraminal position change in the image to the actual three-dimensional space position, and realize the accurate correspondence between the image coordinate space and the actual intervertebral foraminal anatomical structure, thereby obtaining the three-dimensional matching coordinate system point set of the intervertebral foraminal position change. Then, the selected three-dimensional matching coordinate system point set of the intervertebral foraminal position change is projected and transformed by installing a calibration plate, so as to project and transform the matched three-dimensional space coordinate point set into the calibration plate space, and finally obtain the intervertebral foraminal calibration plate space coordinate point set.

[0033] Step S3: performing calibration plate corner point detection on the intervertebral foramen calibration plate spatial coordinate point set to obtain an intervertebral foramen calibration plate corner point set; performing corner point posture feature point set detection on the intervertebral foramen calibration plate corner point set to obtain an intervertebral foramen calibration plate corner point position feature point set and an intervertebral foramen calibration plate corner point angle feature point set;

[0034] The embodiment of the present invention detects the spatial coordinate point set of the intervertebral foramen calibration plate by using a suitable corner point detection algorithm (such as Harris corner point detection, Shi-Tomasi corner point detection, etc.) to accurately identify the corner point positions on the intervertebral foramen calibration plate, thereby obtaining the intervertebral foramen calibration plate corner point set. Secondly, by inputting the intervertebral foramen calibration plate corner point set into the pyramid convolutional neural network for feature extraction processing, rich corner point feature information is extracted from the intervertebral foramen calibration plate corner point set to effectively capture the geometric and visual features of the intervertebral foramen calibration plate corner points, and the extracted intervertebral foramen calibration plate corner point feature information data is analyzed by using the feature pattern analysis method to extract the change law and pattern of the corner points under different conditions. Then, the corner point posture (including corner point position change and corner point angle change) is identified and analyzed for the analyzed angle change pattern to identify and understand the change of the calibration plate corner points under different postures, and according to the identified different posture changes, the corner points with significant position change characteristics and angle change characteristics are screened out from the intervertebral foramen calibration plate corner point set, so that it can describe the position relationship and rotation angle of the calibration plate under different postures, and finally the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set are obtained.

[0035] Step S4: Based on the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set, a calibration mapping transformation matrix is ​​estimated and calculated for the intervertebral foramen position change three-dimensional matching coordinate system point set to obtain the intervertebral foramen calibration mapping transformation estimation matrix; the intervertebral foramen calibration mapping transformation estimation matrix is ​​used to perform image calibration correction on the intervertebral foramen calibration plate space coordinate point set to obtain the intervertebral foramen image calibration correction coordinate system point set.

[0036] The embodiment of the present invention forms a suitable calibration mapping transformation calculation formula by combining the intervertebral foramen calibration plate corner point position feature point set, the intervertebral foramen calibration plate corner point angle feature point set and the homogeneous coordinates of the calibration mapping points in the intervertebral foramen position change three-dimensional matching coordinate system point set and the calibration mapping transformation matrix, transforms the coordinate system mapping of the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set to the coordinate system of the intervertebral foramen position change three-dimensional matching coordinate system point set, so as to convert the feature point information on the intervertebral foramen calibration plate to the matching coordinate system, and by analyzing the transformation characteristics of the coordinate form of the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set to the coordinate form of the intervertebral foramen position change three-dimensional matching coordinate system point set, constructs the matrix form of the corresponding calibration mapping transformation matrix, and at the same time, estimates the constructed calibration mapping transformation matrix to estimate the matrix form of uniformly transforming the intervertebral foramen position and angle changes, thereby obtaining the intervertebral foramen calibration mapping transformation estimation matrix. Then, the image is calibrated by using the estimated intervertebral foramen calibration mapping transformation estimation matrix to calibrate the corresponding position and angle changes in the intervertebral foramen calibration plate spatial coordinate point set, so that the feature points in the image can be accurately calibrated and mapped to the matching coordinate system for correction to correct the position distortion in the intervertebral foraminal endoscopic surgery area image, and finally the intervertebral foramen image calibration correction coordinate system point set is obtained.

[0037] The present invention first obtains the image of the intervertebral foraminal endoscopic surgery process, and performs image preprocessing on the acquired intervertebral foraminal endoscopic surgery area image, so that it can accurately analyze the edge contour structure of the intervertebral foramen, thereby providing a data basis guarantee for the subsequent processing process. By matching the intervertebral foraminal key feature point set with the standard image of the intervertebral foraminal endoscopic surgery area, the edge texture feature point set and the contour structure feature point set of the intervertebral foramen are obtained. The implementation of this step can effectively extract the key feature points of the intervertebral foraminal area, and provide reliable data support for the subsequent position change analysis and processing process. At the same time, by sampling the position change feature of the intervertebral foraminal contour structure feature point set based on the intervertebral foraminal edge texture feature point set, the dynamic change of the position of the intervertebral foramen can be understood in real time, which provides a basis for the subsequent spatial three-dimensional coordinate system positioning, and helps to improve the accuracy and safety of image calibration. Secondly, by combining the intervertebral foraminal position change feature point set to match and locate the standard image of the intervertebral foraminal endoscopic surgery area with the three-dimensional coordinate system, the feature points of the intervertebral foraminal position change in the image can be accurately matched to the actual three-dimensional spatial position. This step can achieve the accurate correspondence between the image and the actual anatomical structure, and provide reliable data support for the subsequent processing process. By performing calibration plate projection transformation on the three-dimensional matching coordinate system point set of the intervertebral foramen position change, the matched three-dimensional space coordinate point set can be projected into the calibration plate space, thereby obtaining the intervertebral foramen calibration plate space coordinate point set. This process can realize the space coordinate transformation and calibration of the intervertebral foramen image, provide more accurate positioning and navigation information for the subsequent image calibration process, and help improve the success rate and accuracy of intervertebral foraminal endoscope image calibration. Then, by performing calibration plate corner point detection on the intervertebral foramen calibration plate space coordinate point set, the corner point position on the intervertebral foramen calibration plate can be accurately identified. The main function of this step is to determine the geometric shape of the calibration plate, which provides important basic data for subsequent positioning and navigation. Subsequently, by analyzing the corner point posture feature of the intervertebral foraminal calibration plate, rich corner point feature information can be extracted from the calibration plate image to discover the changing rules and patterns of the corner points under different conditions. By detecting the feature point set according to the position and angle feature changes of the corner points, the representative position feature points and angle feature points on the calibration plate can be accurately extracted. These feature points can describe the position relationship and rotation angle of the calibration plate under different postures, thus providing an important reference basis for the subsequent processing process. By accurately detecting the position feature points and angle feature points, the accurate positioning and navigation of the intervertebral foraminal endoscope image calibration can be achieved, providing more reliable data support for the subsequent processing process. Finally, by mapping the position feature point set and angle feature point set extracted from the intervertebral foraminal calibration plate to the three-dimensional matching coordinate system point set of the intervertebral foraminal position change, the key to this step is to convert the feature point information on the intervertebral foraminal calibration plate into the matching coordinate system, thereby achieving accurate description and matching of the intervertebral foraminal position change, providing an important data basis for subsequent image calibration and positioning.By establishing the matrix form of the calibration mapping transformation matrix according to the calibration mapping transformation relationship of the calibration mapping process, the transformation relationship of the position and angle can be uniformly expressed. In addition, by estimating each parameter matrix of the calibration mapping transformation and calculating the intervertebral foramen calibration mapping transformation estimation matrix, this step is based on the aforementioned error estimation result, and the final calibration mapping transformation estimation matrix is ​​estimated through mathematical calculation. The matrix can accurately describe the position and angle transformation relationship between the intervertebral foramen calibration plate and the matching coordinate system. In addition, by using the intervertebral foramen calibration mapping transformation estimation matrix to perform image calibration correction on the intervertebral foramen calibration plate space coordinate point set, the intervertebral foramen image can be corrected according to the estimated mapping transformation matrix, so that the feature points in the image can be accurately calibrated and mapped to the matching coordinate system, thereby improving the accuracy of intervertebral foraminal endoscope image calibration and the accuracy of calibration positioning.

[0038] Preferably, step S1 comprises the following steps:

[0039] Step S11: performing image acquisition and processing of the transforaminal endoscopic surgery process through an optical lens to obtain an initial image of the transforaminal endoscopic surgery area;

[0040] Step S12: performing image denoising processing on the initial image of the transforaminal endoscopic surgery area to obtain a denoised image of the transforaminal endoscopic surgery area;

[0041] Step S13: performing contrast enhancement processing on the denoised image of the transforaminal endoscopic surgery area to obtain a contrast-enhanced image of the transforaminal endoscopic surgery area; performing image standardization on the contrast-enhanced image of the transforaminal endoscopic surgery area to obtain a standard image of the transforaminal endoscopic surgery area;

[0042] Step S14: matching the intervertebral foramen key feature point set on the standard image of the intervertebral foraminal endoscopic surgery area to obtain the intervertebral foramen edge texture feature point set and the intervertebral foramen contour structure feature point set;

[0043] Step S15: performing position change feature sampling processing on the intervertebral foramen contour structure feature point set based on the intervertebral foramen edge texture feature point set to obtain the intervertebral foramen position change feature point set.

[0044] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in FIG. 1 , in this embodiment, step S1 includes the following steps:

[0045] Step S11: performing image acquisition and processing of the transforaminal endoscopic surgery process through an optical lens to obtain an initial image of the transforaminal endoscopic surgery area;

[0046] In the embodiment of the present invention, an optical lens is installed on the perforaminal endoscope to capture images of the perforaminal endoscope during the surgery, so as to obtain the initial image of the perforaminal endoscope during the surgery in real time, and finally obtain the initial image of the perforaminal endoscope surgery area.

[0047] Step S12: performing image denoising processing on the initial image of the transforaminal endoscopic surgery area to obtain a denoised image of the transforaminal endoscopic surgery area;

[0048] The embodiment of the present invention processes the collected initial image of the perforaminal endoscopic surgery area by using an appropriate image denoising algorithm (such as Gaussian filtering, median filtering, etc.) to effectively reduce the noise level in the initial image of the perforaminal endoscopic surgery area and improve the quality of the initial image of the perforaminal endoscopic surgery area, and finally obtains a denoised image of the perforaminal endoscopic surgery area.

[0049] Step S13: performing contrast enhancement processing on the denoised image of the transforaminal endoscopic surgery area to obtain a contrast-enhanced image of the transforaminal endoscopic surgery area; performing image standardization on the contrast-enhanced image of the transforaminal endoscopic surgery area to obtain a standard image of the transforaminal endoscopic surgery area;

[0050] The embodiment of the present invention processes the denoised image of the perforaminal endoscopic surgery area by using an image enhancement algorithm to enhance the contrast of the image, so that the image can be clearer, thereby obtaining a contrast-enhanced image of the perforaminal endoscopic surgery area. Then, the contrast-enhanced image of the perforaminal endoscopic surgery area is processed by using a standardized method so that the image parameters such as brightness and color can meet the standards, and finally a standard image of the perforaminal endoscopic surgery area is obtained.

[0051] Step S14: matching the intervertebral foramen key feature point set on the standard image of the intervertebral foraminal endoscopic surgery area to obtain the intervertebral foramen edge texture feature point set and the intervertebral foramen contour structure feature point set;

[0052] In the embodiment of the present invention, the standard image of the perforaminal endoscopic surgery area is subjected to feature analysis by using appropriate image processing technology (including methods such as image texture feature analysis and image contour structure analysis), so as to deeply analyze the edge texture and contour structure in the standard image of the perforaminal endoscopic surgery area, and extract corresponding key feature points, including texture feature points such as roughness and direction of the edge texture, and structural feature points such as shape, size and position of the intervertebral foramen, and the standard image of the perforaminal endoscopic surgery area is analyzed by using a position matching analysis method to associate the feature points in the image with the intervertebral foramen position in the actual surgical scene, thereby analyzing the matching relationship between the intervertebral foramen image and the actual position, and then, by matching the feature sampling points in the standard image of the perforaminal endoscopic surgery area according to the matching relationship between the intervertebral foramen image and the actual position, the edge texture feature point set (such as the actual texture distribution and texture change during the intervertebral foramen surgery) and the contour structure feature point set (such as the actual intervertebral foramen contour shape and size during the intervertebral foramen surgery) of the actual intervertebral foramen area are more accurately understood, and finally the intervertebral foramen edge texture feature point set and the intervertebral foramen contour structure feature point set are obtained.

[0053] Step S15: performing position change feature sampling processing on the intervertebral foramen contour structure feature point set based on the intervertebral foramen edge texture feature point set to obtain the intervertebral foramen position change feature point set.

[0054] The embodiment of the present invention detects the intervertebral foramen contour structure feature point set in combination with the intervertebral foramen edge texture feature point set to detect and identify the deformation of the intervertebral foramen edge contour structure, and understand the change trend, morphological evolution process and other information of the intervertebral foramen edge contour structure area. At the same time, position change tracking analysis is performed based on the change trend, morphological evolution process and other information of the intervertebral foramen edge contour structure area to accurately capture and track the position movement change information of the intervertebral foramen contour, and select a representative feature point set therefrom, and finally obtain the intervertebral foramen position change feature point set.

[0055] The present invention firstly uses an optical lens to collect and process the image of the perforaminal endoscopic surgery process, which can provide clear and high-resolution images, so that it can accurately analyze the edge contour structure of the intervertebral foramen, thereby providing a data basis guarantee for the subsequent processing process. Secondly, by performing image denoising on the initial image of the perforaminal endoscopic surgery area, the noise level in the initial image of the perforaminal endoscopic surgery area can be reduced, and the quality of the initial image of the perforaminal endoscopic surgery area can be effectively improved. This step helps to remove image noise caused by lighting changes or equipment interference, and provides a more reliable data basis for the subsequent image processing process, ensuring that image calibration can be performed under clear image data. Then, by performing contrast enhancement processing and image standardization on the denoised image of the perforaminal endoscopic surgery area, the image of the perforaminal endoscopic surgery area can be made more consistent in contrast and brightness, and the perception and recognition of image details in the subsequent image processing process can be improved, which also helps to reduce visual fatigue and improve the accurate cognition of the intervertebral foramen structure. Next, the key feature point set of the intervertebral foramen is matched with the standard image of the intervertebral foraminal endoscopic surgery area to obtain the edge texture feature point set and contour structure feature point set of the intervertebral foramen. The implementation of this step can effectively extract the key feature points of the intervertebral foramen area and provide reliable data support for the subsequent position change analysis and processing process. Finally, by sampling the position change feature of the intervertebral foramen contour structure feature point set based on the intervertebral foramen edge texture feature point set, the dynamic change of the position of the intervertebral foramen can be understood in real time, which provides a basis for the subsequent spatial three-dimensional coordinate system positioning and helps to improve the accuracy and safety of image calibration.

[0056] Preferably, step S14 comprises the following steps:

[0057] Step S141: performing edge texture and contour structure key feature analysis on the standard image of the transforaminal endoscopic surgery area to obtain key feature data of the edge texture of the transforaminal endoscopic lens and key feature data of the contour structure of the transforaminal endoscopic lens;

[0058] Step S142: performing feature point sampling processing on the key feature data of the edge texture of the percutaneous foraminal endoscope and the key feature data of the contour structure of the percutaneous foraminal endoscope to obtain feature sampling points of the edge texture of the percutaneous foraminal endoscope and feature sampling points of the contour structure of the percutaneous foraminal endoscope;

[0059] Step S143: performing intervertebral foramen position matching analysis on the standard image of the intervertebral foraminal endoscope surgical area to obtain intervertebral foraminal endoscope actual position matching relationship data;

[0060] Step S144: performing texture feature point set matching on the texture feature sampling points at the edge of the intervertebral foraminal endoscope according to the actual position matching relationship data of the intervertebral foraminal endoscope to obtain a texture feature point set at the edge of the intervertebral foraminal endoscope;

[0061] Step S145: performing structural feature point set matching on the structural feature sampling points of the intervertebral foraminal endoscope contour according to the actual position matching relationship data of the intervertebral foraminal endoscope, so as to obtain the structural feature point set of the intervertebral foraminal endoscope contour.

[0062] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 2 Detailed step flow diagram of step S14 in the embodiment, step S14 includes the following steps:

[0063] Step S141: performing edge texture and contour structure key feature analysis on the standard image of the transforaminal endoscopic surgery area to obtain key feature data of the edge texture of the transforaminal endoscopic lens and key feature data of the contour structure of the transforaminal endoscopic lens;

[0064] The embodiment of the present invention uses appropriate image processing technology (including image texture feature analysis and image contour structure analysis) to perform feature analysis on the standard image of the perforaminal endoscopic surgery area, so as to deeply analyze the edge texture and contour structure in the standard image of the perforaminal endoscopic surgery area, and extract corresponding key feature information, including the roughness, direction and other features of the edge texture, and the shape, size and position of the intervertebral foramen and other structural features, and finally obtains the key feature data of the edge texture of the perforaminal endoscopic lens and the key feature data of the contour structure of the perforaminal endoscopic lens.

[0065] Step S142: performing feature point sampling processing on the key feature data of the edge texture of the percutaneous foraminal endoscope and the key feature data of the contour structure of the percutaneous foraminal endoscope to obtain feature sampling points of the edge texture of the percutaneous foraminal endoscope and feature sampling points of the contour structure of the percutaneous foraminal endoscope;

[0066] The embodiment of the present invention processes the key feature data of the edge texture of the percutaneous foraminal endoscope and the key feature data of the contour structure of the percutaneous foraminal endoscope respectively by using a feature point sampling method, so as to effectively extract the most representative and decisive feature information points in the image, and finally obtain the feature sampling points of the edge texture of the percutaneous foraminal endoscope and the feature sampling points of the contour structure of the percutaneous foraminal endoscope.

[0067] Step S143: performing intervertebral foramen position matching analysis on the standard image of the intervertebral foraminal endoscope surgical area to obtain intervertebral foraminal endoscope actual position matching relationship data;

[0068] The embodiment of the present invention analyzes the standard image of the percutaneous endoscopic surgery area by using a position matching analysis method to associate the feature points in the image with the position of the intervertebral foramen in the actual surgical scene, and establish an accurate correspondence between the two, thereby analyzing the matching relationship between the intervertebral foramen image and the actual position, and finally obtaining the percutaneous endoscopic actual position matching relationship data.

[0069] Step S144: performing texture feature point set matching on the texture feature sampling points at the edge of the intervertebral foraminal endoscope according to the actual position matching relationship data of the intervertebral foraminal endoscope to obtain a texture feature point set at the edge of the intervertebral foraminal endoscope;

[0070] The embodiment of the present invention matches the edge texture feature sampling points of the percutaneous foraminal endoscope according to the matching relationship between the intervertebral foraminal image feature points and the actual position in the actual position matching relationship data of the percutaneous foraminal endoscope, so as to more accurately understand the edge texture feature point set of the actual intervertebral foraminal area, such as the actual texture distribution and texture changes during the intervertebral foraminal surgery, and finally obtain the edge texture feature point set of the intervertebral foraminal endoscope.

[0071] Step S145: performing structural feature point set matching on the structural feature sampling points of the intervertebral foraminal endoscope contour according to the actual position matching relationship data of the intervertebral foraminal endoscope, so as to obtain the structural feature point set of the intervertebral foraminal endoscope contour.

[0072] The embodiment of the present invention matches the percutaneous foraminal endoscope contour structure feature sampling points according to the matching relationship between the intervertebral foraminal image feature points and the actual position in the percutaneous foraminal endoscope actual position matching relationship data, so as to more accurately understand the contour structure feature point set of the actual intervertebral foraminal area, such as the actual intervertebral foraminal contour shape and size during the intervertebral foraminal surgery, and finally obtain the intervertebral foraminal contour structure feature point set.

[0073] The present invention firstly obtains the key feature data of the edge texture of the perforaminal mirror and the key feature data of the contour structure of the perforaminal mirror by analyzing the key features of the edge texture and contour structure of the standard image of the perforaminal mirror surgery area. By deeply analyzing the edge texture and contour structure in the image, this step can reveal the detailed information of the key features in the surgical area image, including the roughness, direction and other features of the edge texture, as well as the structural features such as the shape, size and position of the intervertebral foramen, providing a strong basis for subsequent processing. Secondly, by sampling the key feature data of the edge texture of the perforaminal mirror and the key feature data of the contour structure of the perforaminal mirror, the key of this step is that by sampling the key feature points, the most representative and decisive feature information in the image can be effectively extracted, providing efficient data support for further matching and analysis. Then, by performing the intervertebral foramen position matching analysis on the standard image of the perforaminal mirror surgery area, the actual position matching relationship of the intervertebral foramen can be analyzed, and the feature points in the image can be associated with the intervertebral foramen position in the actual surgical scene, and an accurate corresponding relationship between the two can be established. This step helps to ensure that the subsequent feature point matching and analysis are more accurate and reliable. Next, by matching the texture feature point set of the edge texture feature sampling points of the foraminal mirror according to the actual position matching relationship data of the foraminal mirror, the edge texture features of the foraminal region, such as texture distribution and texture change, can be more accurately understood, providing an important basis for accurate positioning and operation in the subsequent processing. Finally, by matching the structure feature point set of the contour structure feature sampling points of the foraminal mirror according to the actual position matching relationship data of the foraminal mirror, important information such as the shape, size and contour features of the foramina can be understood, providing accurate spatial reference for positioning and operation in the subsequent processing. The results of these point set matching help improve the accuracy and success rate of image calibration.

[0074] Preferably, step S15 comprises the following steps:

[0075] Step S151: performing local feature description processing on the texture feature point set of the edge of the intervertebral foramen to obtain local texture structure information data of the edge of the intervertebral foramen;

[0076] The embodiment of the present invention describes each edge texture feature point in the intervertebral foramen edge texture feature point set by using a local feature descriptor (such as SIFT, SURF or ORB), so as to describe the texture information around each feature point in detail, including the density, direction, size and other features of the texture, and ensure that the texture characteristics of the intervertebral foramen area can be more accurately reflected, and finally the local texture structure information data of the intervertebral foramen edge is obtained.

[0077] Step S152: performing contour deformation detection on the intervertebral foramen contour structure feature point set according to the intervertebral foramen edge local texture structure information data to obtain intervertebral foramen edge contour structure deformation information data;

[0078] The embodiment of the present invention detects the intervertebral foramen edge contour structure feature point set by using the described local texture structure information data of the intervertebral foramen edge (including the change information of the texture structure characteristics of the intervertebral foramen edge), so as to detect and identify the deformation of the intervertebral foramen edge contour structure, and understand the change trend, morphological evolution process and other information of the intervertebral foramen edge contour structure area, and finally obtain the intervertebral foramen edge contour structure deformation information data.

[0079] Step S153: performing intervertebral foramen motion estimation on the intervertebral foramen contour structure feature point set to obtain intervertebral foramen contour motion trajectory information data; performing position change tracking analysis on the intervertebral foramen contour motion trajectory information data based on the intervertebral foramen edge contour structure deformation information data to obtain intervertebral foramen contour position change information data;

[0080] The embodiment of the present invention processes the feature point set of the intervertebral foramen contour structure by using a motion estimation method to estimate the overall motion trajectory of the intervertebral foramen, thereby obtaining intervertebral foramen contour motion trajectory information data. Then, the intervertebral foramen contour motion trajectory information data is analyzed by combining the morphological evolution process of the intervertebral foramen edge contour structure in the intervertebral foramen edge contour structure deformation information data to accurately capture and track the position motion change information of the intervertebral foramen contour, and finally obtain the intervertebral foramen contour position change information data.

[0081] Step S154: performing data smoothing processing on the intervertebral foramen contour position change information data to obtain intervertebral foramen contour position change standard data; performing dynamic reconstruction processing on the intervertebral foramen contour position change standard data using three-dimensional reconstruction technology to obtain intervertebral foramen position change reconstruction structure space;

[0082] The embodiment of the present invention uses data smoothing filtering technology (such as Kalman filtering or extended Kalman filtering algorithms) to smooth the intervertebral foramen contour position change in the intervertebral foramen contour position change information data to eliminate the instability caused by image noise or motion estimation errors, thereby obtaining intervertebral foramen contour position change standard data. Then, by using three-dimensional reconstruction technology, the position change of the intervertebral foramen in the intervertebral foramen contour position change standard data is presented in a dynamic manner, and a more intuitive and three-dimensional intervertebral foramen position change scene perception space is provided, and finally the intervertebral foramen position change reconstruction structure space is obtained.

[0083] Step S155: performing feature point set sampling processing on the intervertebral foramen position change reconstruction structure space to obtain an intervertebral foramen position change feature point set.

[0084] The embodiment of the present invention processes the reconstructed structure space of the intervertebral foramen position change obtained after reconstruction by using a feature point set sampling method to sample and select the most representative feature points and ensure that they can effectively summarize the key feature conditions of the intervertebral foramen position change, and finally obtain the intervertebral foramen position change feature point set.

[0085] The present invention firstly performs local feature description processing on the texture feature point set of the intervertebral foramen edge, and can describe the texture information around each feature point in detail, including the density, direction, size and other features of the texture. The local texture structure information data of the intervertebral foramen edge obtained in this way can more accurately reflect the texture characteristics of the intervertebral foramen area, and provide strong support for the subsequent contour deformation detection. Secondly, by performing contour deformation detection on the intervertebral foramen contour structure feature point set according to the local texture structure information data of the intervertebral foramen edge, the deformation of the intervertebral foramen edge contour structure can be identified. These deformation information data are of great significance for understanding the change trend and morphological evolution of the intervertebral foramen area, and provide real-time morphological reference for the subsequent processing process. Then, by performing intervertebral foramen motion estimation on the intervertebral foramen contour structure feature point set, and performing position change tracking analysis on the intervertebral foramen contour motion trajectory information data based on the intervertebral foramen edge contour structure deformation information data, this series of steps can accurately capture the motion change of the intervertebral foramen contour, and provide timely position guidance for the subsequent three-dimensional coordinate system positioning operation process. Next, by smoothing the data of the position change information of the intervertebral foramen contour, the instability and noise that may be caused by the position and angle of the camera during the surgical operation or image acquisition process can be effectively eliminated, providing a smoother and more reliable data basis for subsequent analysis and visualization, thereby obtaining more stable and reliable standard data of the position change of the intervertebral foramen contour. In addition, by using three-dimensional reconstruction technology to dynamically reconstruct the standard data of the position change of the intervertebral foramen contour, the key to this step is to be able to present the position change of the intervertebral foramen in a three-dimensional form, providing a more intuitive and three-dimensional scene perception of the position change of the intervertebral foramen for subsequent processing, which is helpful for fine image calibration. Finally, by sampling the feature point set of the reconstructed structure space of the position change of the intervertebral foramen, these feature point sets can effectively summarize the key features of the position change of the intervertebral foramen, provide more intuitive and concise visual information for the subsequent processing, help better understand the changes in the intervertebral foramen area, and provide accurate guidance and data support for the subsequent spatial three-dimensional coordinate system positioning.

[0086] Preferably, step S2 comprises the following steps:

[0087] Step S21: performing a three-dimensional coordinate system conversion on the standard image of the transforaminal endoscopic surgery area to obtain a three-dimensional coordinate system space of the transforaminal image;

[0088] The embodiment of the present invention processes the standard image of the perforaminal endoscopic surgery area by using a three-dimensional coordinate system modeling method to map the standard image of the perforaminal endoscopic surgery area from a two-dimensional plane to a three-dimensional space, and make it accurately describe the spatial structure and geometric shape of the perforaminal surgery area, and finally obtain the three-dimensional coordinate system space of the perforaminal image.

[0089] Step S22: performing feature point labeling sampling processing on the three-dimensional coordinate system space of the intervertebral foramen image to obtain a three-dimensional space feature point set of the intervertebral foramen image;

[0090] The embodiment of the present invention labels and samples the feature points of the three-dimensional coordinate system space of the intervertebral foramen image to effectively extract a representative feature point set in the intervertebral foramen image, so that it can better describe the morphology and structure of the intervertebral foramen image, and finally obtain a three-dimensional spatial feature point set of the intervertebral foramen image.

[0091] Step S23: performing mapping matching calculation on the three-dimensional space feature point set of the intervertebral foramen image based on the intervertebral foramen position change feature point set using the position change mapping matching calculation formula to obtain the intervertebral foramen position change mapping matching degree;

[0092] The embodiment of the present invention forms a suitable position change mapping matching calculation formula by combining the matching influence weight parameter of the feature point, the matching Gaussian kernel standard deviation, the matching influence control coefficient, the spatial horizontal coordinate parameter, the spatial vertical coordinate parameter, the spatial vertical coordinate parameter and related parameters to perform mapping matching calculation on the feature point set of the intervertebral foramen position change and the feature points corresponding to the feature point set of the intervertebral foramen image three-dimensional space feature point set, so as to quantitatively evaluate the similarity and correspondence between the feature point set of the intervertebral foramen position change and the feature point set of the intervertebral foramen image three-dimensional space feature point set, and finally obtain the intervertebral foramen position change mapping matching degree. In addition, the position change mapping matching calculation formula can also use any position mapping matching method in the art to replace the mapping matching calculation process, and is not limited to the position change mapping matching calculation formula.

[0093] Step S24: mapping and matching the three-dimensional spatial feature point set of the intervertebral foramen image according to the mapping matching degree of the intervertebral foramen position change to obtain a three-dimensional matching coordinate system point set of the intervertebral foramen position change;

[0094] The embodiment of the present invention processes the calculated intervertebral foramen position change mapping matching degree to screen out feature points with higher matching degree, and uses them for matching and positioning of the intervertebral foramen position change three-dimensional matching coordinate system point set, so as to accurately correspond the feature points of the intervertebral foramen position change in the image to the actual three-dimensional space position, thereby achieving accurate correspondence between the image coordinate space and the actual intervertebral foramen anatomical structure, and finally obtaining the intervertebral foramen position change three-dimensional matching coordinate system point set.

[0095] Step S25: performing calibration plate projection transformation on the three-dimensional matching coordinate system point set of the intervertebral foramen position change to obtain the intervertebral foramen calibration plate space coordinate point set.

[0096] In the embodiment of the present invention, a calibration plate is installed to project and transform the selected three-dimensional matching coordinate system point set of intervertebral foramen position changes, so as to project and transform the matched three-dimensional space coordinate point set into the calibration plate space, and finally obtain the intervertebral foramen calibration plate space coordinate point set.

[0097] The present invention firstly transforms the standard image of the intervertebral foraminal endoscopic surgery area into a three-dimensional coordinate system, and can map the image information from the two-dimensional plane to the three-dimensional space, so as to accurately describe the spatial structure and geometric shape of the surgical area. This conversion process can provide more information dimensions, so that the intervertebral foraminal anatomical structure of the surgical area can be better understood in a three-dimensional environment, and accurate positioning and navigation can be provided for the subsequent image calibration process. Secondly, by performing feature point labeling sampling processing on the three-dimensional coordinate system space of the intervertebral foraminal image, a representative feature point set in the intervertebral foraminal image can be effectively extracted. These feature points can not only describe the morphology and structure of the intervertebral foraminal image, but also serve as the basis for subsequent matching and positioning, providing important support for the accuracy and safety of the image calibration operation. Then, by combining the intervertebral foraminal position change feature point set with a suitable position change mapping matching calculation formula, the three-dimensional space feature point set of the intervertebral foraminal image is mapped and matched, and the similarity and correspondence between the feature point set of the intervertebral foraminal position change and the feature point set of the three-dimensional space feature point set of the intervertebral foraminal image can be quantitatively evaluated. Such mapping matching calculation provides a basis for subsequent positioning and navigation, thereby ensuring the accuracy and stability of the subsequent processing process. Next, by mapping and matching the three-dimensional spatial feature point set of the intervertebral foramen image according to the calculated mapping matching degree of the intervertebral foramen position change, the feature points of the intervertebral foramen position change in the image can be accurately matched to the actual three-dimensional spatial position. This step can achieve accurate correspondence between the image and the actual anatomical structure, providing reliable data support for subsequent processing. Finally, by performing calibration plate projection transformation on the three-dimensional matching coordinate system point set of the intervertebral foramen position change, the matched three-dimensional spatial coordinate point set can be projected into the calibration plate space, thereby obtaining the intervertebral foramen calibration plate spatial coordinate point set. This process can achieve spatial coordinate transformation and calibration of the intervertebral foramen image, providing more accurate positioning and navigation information for the subsequent image calibration process, and helping to improve the success rate and accuracy of intervertebral foraminal endoscope image calibration.

[0098] Preferably, the position change mapping matching calculation formula in step S23 is specifically:

[0099]

[0100] Where M is the matching degree of the intervertebral foramen position change mapping, N is the number of feature points in the feature point set, j is the index item of the feature point in the feature point set, and w is the j is the matching influence weight parameter of the jth feature point, σ j is the standard deviation of the matching Gaussian kernel of the jth feature point, d j is the matching influence control coefficient of the jth feature point, x aj is the spatial horizontal coordinate parameter of the jth feature point in the intervertebral foramen position change feature point set, x bj is the spatial horizontal coordinate parameter of the jth feature point in the three-dimensional spatial feature point set of the intervertebral foramen image, y aj is the spatial ordinate parameter of the jth feature point in the intervertebral foramen position change feature point set, y bj is the spatial ordinate parameter of the jth feature point in the three-dimensional spatial feature point set of the intervertebral foramen image, z aj is the spatial vertical coordinate parameter of the jth feature point in the intervertebral foramen position change feature point set, z bj is the spatial vertical coordinate parameter of the jth feature point in the three-dimensional spatial feature point set of the intervertebral foramen image, and η is the correction value of the matching degree of the intervertebral foramen position change mapping.

[0101] The present invention obtains a position change mapping matching calculation formula by using a specific mathematical model and after verification, which is used to perform mapping matching calculation on the three-dimensional spatial feature point set of the intervertebral foramen image. The position change mapping matching calculation formula combines the number of feature points in the feature point set, which means that it can handle the matching of multiple feature points, which is very important for intervertebral foramen image calibration, because the positions and changes of multiple feature points may need to be considered in intervertebral foramen image calibration. Secondly, by using the Gaussian kernel function The distance between the feature point set of intervertebral foramen position change and the feature point set of intervertebral foramen image three-dimensional space is weighted, which means that the feature points with closer distance contribute more to the matching degree, and the feature points with farther distance contribute less to the matching degree. The matching influence weight parameter is also used to adjust the contribution of each feature point to the matching degree. If a feature point should contribute more to the matching degree, its weight parameter can be increased; vice versa. The width of the Gaussian kernel function is adjusted by using the matching Gaussian kernel standard deviation, thereby affecting the degree of influence of the distance between feature points on the matching degree. In addition, the shape of the Gaussian kernel function is adjusted by using the matching influence control coefficient, thereby affecting the degree of influence of the distance between feature points on the matching degree. In addition, the correction value is introduced to adjust the overall matching degree, thereby affecting the matching result of the feature point set. Therefore, the formula can effectively quantify the matching degree between feature points through the weighted calculation of the Gaussian kernel function, realize the mapping from the three-dimensional space of the intervertebral foramen image to the three-dimensional matching coordinate system of the intervertebral foramen position change, and thus provide key information for subsequent image calibration navigation and positioning. In summary, this formula fully considers the mapping matching degree M of the intervertebral foramen position change, the number of feature points in the feature point set N, the index item j of the feature point in the feature point set, and the matching influence weight parameter w of the jth feature point j , the matching Gaussian kernel standard deviation σ of the jth feature point j , the matching influence control coefficient d of the jth feature point j , the spatial horizontal coordinate parameter x of the jth feature point in the intervertebral foramen position change feature point set aj , the spatial horizontal coordinate parameter x of the jth feature point in the three-dimensional spatial feature point set of the intervertebral foramen image bj , the spatial ordinate parameter y of the jth feature point in the intervertebral foramen position change feature point set aj , the spatial ordinate parameter y of the jth feature point in the three-dimensional spatial feature point set of the intervertebral foramen image bj , the spatial vertical coordinate parameter z of the jth feature point in the intervertebral foramen position change feature point set aj , the spatial vertical coordinate parameter z of the jth feature point in the three-dimensional spatial feature point set of the intervertebral foramen image bj , the correction value η of the intervertebral foramen position change mapping matching degree, according to the intervertebral foramen position change mapping matching degree M and the mutual correlation between the above parameters constitutes a functional relationship The formula can realize the mapping and matching calculation process of the three-dimensional spatial feature point set of the intervertebral foramen image. At the same time, the introduction of the correction value η of the mapping matching degree of the intervertebral foramen position change can be adjusted according to the error situation occurring in the calculation process, thereby improving the accuracy and applicability of the position change mapping matching calculation formula.

[0102] Preferably, step S3 comprises the following steps:

[0103] Step S31: performing calibration plate corner point detection on the intervertebral foramen calibration plate spatial coordinate point set to obtain an intervertebral foramen calibration plate corner point set;

[0104] The embodiment of the present invention detects the spatial coordinate point set of the intervertebral foramen calibration plate by using a suitable corner point detection algorithm (such as Harris corner point detection, Shi-Tomasi corner point detection, etc.) to accurately identify the corner point positions on the intervertebral foramen calibration plate, and finally obtains the intervertebral foramen calibration plate corner point set.

[0105] Step S32: using a pyramid convolutional neural network to perform corner point feature extraction processing on the intervertebral foramen calibration plate corner point set to obtain intervertebral foramen calibration plate corner point feature information data;

[0106] The embodiment of the present invention inputs the intervertebral foramen calibration plate corner point set into a pyramid convolutional neural network for feature extraction processing, extracts rich corner point feature information from the intervertebral foramen calibration plate corner point set, so as to effectively capture the geometric and visual features of the intervertebral foramen calibration plate corner points, and can better describe the morphology and structure of the calibration plate, and finally obtain the intervertebral foramen calibration plate corner point feature information data.

[0107] Step S33: performing feature change pattern analysis on the feature information data of the corner points of the intervertebral foramen calibration plate to obtain feature change pattern data of the corner points of the intervertebral foramen calibration plate; performing position and angle feature change recognition analysis on the feature change pattern data of the corner points of the intervertebral foramen calibration plate to obtain position change pattern data of the corner points of the intervertebral foramen calibration plate and angle change pattern data of the corner points of the intervertebral foramen calibration plate;

[0108] The embodiment of the present invention uses a characteristic pattern analysis method to analyze the extracted intervertebral foramen calibration plate corner point characteristic information data to extract the change rules and patterns of the corner points under different conditions, thereby obtaining the intervertebral foramen calibration plate corner point characteristic change pattern data. Then, by performing corner point posture recognition analysis (including corner point position change and corner point angle change) on the intervertebral foramen calibration plate corner point characteristic change pattern data, the change of the calibration plate corner points under different postures is recognized and understood, and finally the intervertebral foramen calibration plate corner point position change pattern data and the intervertebral foramen calibration plate corner point angle change pattern data are obtained.

[0109] Step S34: performing position feature point set detection on the intervertebral foramen calibration plate corner point set according to the position change pattern data of the intervertebral foramen calibration plate corner points to obtain the position feature point set of the intervertebral foramen calibration plate corner points;

[0110] The embodiment of the present invention processes the intervertebral foramen calibration plate corner point set according to the corner point position change situation in the intervertebral foramen calibration plate corner point position change pattern data, so as to screen out corner points with significant position change characteristics from the intervertebral foramen calibration plate corner point set, and enable them to describe the position relationship of the calibration plate in different postures, and finally obtain the intervertebral foramen calibration plate corner point position feature point set.

[0111] Step S35: performing angle feature point set detection on the intervertebral foramen calibration plate corner point set according to the angle variation pattern data of the intervertebral foramen calibration plate corner point to obtain the angle feature point set of the intervertebral foramen calibration plate corner point.

[0112] The embodiment of the present invention processes the intervertebral foramen calibration plate corner point set according to the corner point angle change in the intervertebral foramen calibration plate corner point angle change pattern data, so as to screen out corner points with significant angle change characteristics from the intervertebral foramen calibration plate corner point set, and enable them to describe the rotation angle of the calibration plate in different postures, and finally obtain the intervertebral foramen calibration plate corner point angle feature point set.

[0113] The present invention firstly detects the corner points of the calibration plate on the spatial coordinate point set of the intervertebral foramen calibration plate, and can accurately identify the corner point positions on the intervertebral foramen calibration plate. The main function of this step is to determine the geometric shape of the calibration plate, and provide important basic data for subsequent positioning and navigation. By accurately detecting the corner points, the accuracy and stability of the subsequent analysis can be ensured, and data support can be provided for the accuracy of the subsequent processing process. Secondly, by using the pyramid convolutional neural network to perform corner point feature extraction processing on the corner point set of the intervertebral foramen calibration plate, rich corner point feature information can be extracted from the calibration plate image. The neural network has a strong representation ability in processing complex image data, can automatically learn and extract features, and effectively capture the geometric and visual features of the corner points. The corner point feature information data obtained in this way can better describe the morphology and structure of the calibration plate, and provide more reliable feature representation for subsequent pattern analysis and matching. Then, by performing feature change pattern analysis on the corner point feature information data of the intervertebral foramen calibration plate, the change law and pattern of the corner points under different conditions can be found. By identifying the position and angle feature changes of the corner points, we can understand the changes of the calibration plate in different postures, which provides an important reference for subsequent posture estimation and matching. Such analysis can help understand the deformation of the calibration plate and provide more accurate information support for subsequent image calibration and positioning. Next, by detecting the position feature point set of the corner point set of the intervertebral foraminal calibration plate according to the position change pattern data of the corner point of the intervertebral foraminal calibration plate, the representative position feature points on the calibration plate can be accurately extracted. These feature points can describe the position relationship of the calibration plate in different postures, providing an important reference for subsequent positioning and navigation. By accurately detecting the position feature points, accurate positioning and navigation of intervertebral foraminal endoscope image calibration can be achieved, thereby improving the success rate of image calibration. Finally, by detecting the angle feature point set of the corner point set of the intervertebral foraminal calibration plate according to the angle change pattern data of the corner point of the intervertebral foraminal calibration plate, the representative angle feature points on the calibration plate can be accurately extracted. These feature points can describe the rotation angle of the calibration plate in different postures, providing an important reference for subsequent processing. By accurately detecting the angle feature points, precise positioning and navigation of the percutaneous endoscopic image calibration can be achieved, providing more reliable data support for subsequent processing.

[0114] Preferably, step S4 comprises the following steps:

[0115] Step S41: mapping and transforming the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set to the intervertebral foramen position change three-dimensional matching coordinate system point set using the calibration mapping transformation calculation formula;

[0116] The embodiment of the present invention forms a suitable calibration mapping transformation calculation formula by combining the intervertebral foramen calibration plate corner point position feature point set, the intervertebral foramen calibration plate corner point angle feature point set and the homogeneous coordinates of the calibration mapping points in the intervertebral foramen position change three-dimensional matching coordinate system point set and the calibration mapping transformation matrix to transform the coordinate system mapping of the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set to the coordinate system of the intervertebral foramen position change three-dimensional matching coordinate system point set, so as to convert the feature point information on the intervertebral foramen calibration plate into the matching coordinate system, thereby achieving accurate description and matching of the intervertebral foramen position change. In addition, the calibration mapping transformation calculation formula can also use any mapping transformation matching method in the field to replace the mapping transformation process, and is not limited to the calibration mapping transformation calculation formula.

[0117] The calibration mapping transformation calculation formula is as follows:

[0118]

[0119] In the formula, C i is the homogeneous coordinate of the i-th calibration mapping point in the three-dimensional matching coordinate system point set of the intervertebral foramen position change, A i is the homogeneous coordinate of the i-th calibration mapping point in the feature point set of the intervertebral foramen calibration plate corner point position, B i is the homogeneous coordinate of the i-th calibration mapping point in the angle feature point set of the intervertebral foramen calibration plate, T is the transposition symbol, · is the matrix dot multiplication symbol, i is the index item of the calibration mapping point in the feature point set, n is the number of calibration mapping points in the feature point set, and G is the calibration mapping transformation matrix;

[0120] The present invention obtains a calibration mapping transformation calculation formula by using a specific mathematical model and verifying it, which is used to realize the coordinate system mapping transformation of the feature point set of the corner point position of the intervertebral foramen calibration plate and the feature point set of the corner point angle of the intervertebral foramen calibration plate to the coordinate system of the three-dimensional matching coordinate system point set of the intervertebral foramen position change. The calibration mapping transformation calculation formula combines the feature point set of the corner point position of the intervertebral foramen calibration plate and the feature point set of the angle to comprehensively consider the three-dimensional information of the intervertebral foramen position change. The advantage of this is that the position and posture changes of the intervertebral foramen can be captured more accurately, thereby improving the precision and accuracy of subsequent matching. The use of homogeneous coordinates can simplify the calculation, and the rotation, translation and scaling operation process can be well described. This method makes the calibration mapping transformation calculation more flexible and efficient. Secondly, by expressing the calibration mapping transformation matrix G as a combination of rotation, translation and scaling matrices, the calculation is more intuitive and clear, and this expression method facilitates the understanding and adjustment of the transformation parameters. By comprehensively considering the information of multiple feature points and accurately estimating the transformation parameters, the accuracy of the calibration mapping transformation is improved, which helps to ensure the effect of the intervertebral foramen image calibration and correction, and improves the success rate of the subsequent matching process. Therefore, the calibration mapping transformation calculation formula plays an important role in the intervertebral foramen image processing process. By comprehensively considering the information of multiple feature points and combining the matrix calculation method, it realizes the accurate description and effective mapping of the intervertebral foramen position change, thereby providing a reliable foundation for subsequent image calibration and correction. In summary, this formula fully considers the homogeneous coordinates C of the i-th calibration mapping point in the three-dimensional matching coordinate system point set of the intervertebral foramen position change. i , the homogeneous coordinates A of the i-th calibration mapping point in the feature point set of the intervertebral foramen calibration plate corner point position i , the homogeneous coordinates B of the i-th calibration mapping point in the angle feature point set of the intervertebral foramen calibration plate i , transposition symbol T, matrix dot multiplication symbol ·, index item i of the calibration mapping point in the feature point set, number n of calibration mapping points in the feature point set, calibration mapping transformation matrix G, in order to realize the calibration mapping transformation process of the calibration mapping transformation calculation formula, a calibration mapping transformation matrix composed of rotation matrix X, translation matrix P, scaling matrix S and zero matrix is ​​constructed by fully considering the calibration mapping transformation relationship According to the change of intervertebral foramen position, the homogeneous coordinate C of the i-th calibration mapping point in the three-dimensional matching coordinate system point set is obtained. i The correlation between the above sub-parameters constitutes a functional relationship Thereby, the coordinate system mapping transformation of the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set can be achieved to the coordinate system of the intervertebral foramen position change three-dimensional matching coordinate system point set.

[0121] Step S42: construct the matrix form of the calibration mapping transformation matrix G according to the calibration mapping transformation relationship of the calibration mapping transformation calculation formula:

[0122]

[0123] Among them, X is the rotation matrix, P is the translation matrix, and S is the scaling matrix;

[0124] The embodiment of the present invention analyzes the transformation characteristics of the coordinate form of the feature point set of the corner point position of the intervertebral foramen calibration plate and the feature point set of the corner point angle of the intervertebral foramen calibration plate into the coordinate form of the point set of the three-dimensional matching coordinate system of the intervertebral foramen position change, that is, how to convert the transformation feature information (including rotation, translation, scaling, etc.) from one point set matrix form to another point set matrix form, and expresses the calibration mapping transformation relationship in matrix form according to the analysis results, and finally constructs the matrix form of the corresponding calibration mapping transformation matrix G.

[0125] Step S43: performing matrix estimation calculation on the rotation matrix X, the translation matrix P and the scaling matrix S in the calibration mapping transformation matrix G to obtain a rotation estimation matrix, a translation estimation matrix and a scaling estimation matrix;

[0126] The embodiment of the present invention gradually derives the matrix estimation calculation formulas of the rotation matrix, translation matrix and scaling matrix by combining the homogeneous coordinates of the calibration mapping points of the three-dimensional matching coordinate system point set of the intervertebral foramen position change, the homogeneous coordinates of the calibration mapping points of the intervertebral foramen calibration plate corner point position feature point set, the homogeneous coordinates of the calibration mapping points of the intervertebral foramen calibration plate corner point angle feature point set, transposition symbols, matrix point multiplication symbols and related parameters to perform estimation calculations, thereby estimating the estimated matrix forms of the rotation matrix X, translation matrix P and scaling matrix S in the calibration mapping transformation matrix G, and finally obtaining the rotation estimation matrix, translation estimation matrix and scaling estimation matrix. In addition, the matrix estimation calculation method can also use any matrix estimation method in the art to replace the matrix estimation calculation process of the present invention, and is not limited to the matrix estimation calculation method.

[0127] Step S44: performing calibration mapping transformation matrix estimation calculation through the rotation estimation matrix, the translation estimation matrix and the scaling estimation matrix to obtain the intervertebral foramen calibration mapping transformation estimation matrix;

[0128] The embodiment of the present invention performs estimation calculation by substituting the rotation estimation matrix, translation estimation matrix and scaling estimation matrix obtained by estimation calculation into the matrix form equation of the calibration mapping transformation matrix G, so as to estimate the matrix form that uniformly transforms the position and angle changes of the intervertebral foramen, and finally obtains the intervertebral foramen calibration mapping transformation estimation matrix.

[0129] Step S45: performing image calibration correction on the intervertebral foramen calibration plate space coordinate point set using the intervertebral foramen calibration mapping transformation estimation matrix to obtain an intervertebral foramen image calibration correction coordinate system point set.

[0130] The embodiment of the present invention performs image calibration on the corresponding position and angle changes in the spatial coordinate point set of the intervertebral foramen calibration plate by using the estimated intervertebral foramen calibration mapping transformation estimation matrix, so that the feature points in the image can be accurately calibrated and mapped to the matching coordinate system for correction, so as to correct the position distortion in the image of the intervertebral foraminal endoscopic surgery area, and ensure that the image data can be represented in an accurate three-dimensional space, and finally obtain the intervertebral foramen image calibration correction coordinate system point set.

[0131] The present invention first maps the position feature point set and angle feature point set extracted from the intervertebral foramen calibration plate to the point set of the three-dimensional matching coordinate system of the intervertebral foramen position change by using a suitable calibration mapping transformation calculation formula. The key to this step is to convert the feature point information on the intervertebral foramen calibration plate into the matching coordinate system, thereby realizing the accurate description and matching of the intervertebral foramen position change, and providing an important data basis for subsequent image calibration and positioning. Secondly, the matrix form of the calibration mapping transformation matrix G is established by the calibration mapping transformation relationship according to the calibration mapping transformation calculation formula. The purpose of this step is to express the calibration mapping transformation relationship in a matrix form, so as to facilitate subsequent calculation and processing. By constructing the calibration mapping transformation matrix, the transformation relationship of the position and angle can be uniformly expressed, which facilitates the subsequent error estimation and correction. Then, by performing matrix estimation calculation on the rotation matrix X, translation matrix P and scaling matrix S in the calibration mapping transformation matrix G, the rotation estimation matrix, translation estimation matrix and scaling estimation matrix are obtained. The purpose of this step is to estimate the various parameter matrices of the calibration mapping transformation and obtain their accurate estimation values, thereby improving the accuracy and reliability of image calibration. Next, the calibration mapping transformation matrix is ​​estimated and calculated by using the rotation estimation matrix, the translation estimation matrix, and the scaling estimation matrix, that is, the intervertebral foramen calibration mapping transformation estimation matrix is ​​calculated. This step is based on the aforementioned error estimation results. The final calibration mapping transformation matrix is ​​estimated by mathematical calculation. This matrix can accurately describe the position and angle transformation relationship between the intervertebral foramen calibration plate and the matching coordinate system. Finally, the intervertebral foramen calibration plate space coordinate point set is image calibrated and corrected by using the intervertebral foramen calibration mapping transformation estimation matrix. The key to this step is to be able to correct the intervertebral foramen image according to the estimated mapping transformation matrix, so that the feature points in the image can be accurately calibrated and mapped to the matching coordinate system, providing accurate basic data support for subsequent intervertebral foraminal endoscopic surgery navigation and positioning.

[0132] Preferably, the matrix estimation calculation in step S43 is solved by the following matrix estimation calculation formula:

[0133]

[0134] In the formula, is the rotation estimation matrix, is the translation estimation matrix, is the scaling estimate matrix.

[0135] The present invention obtains a suitable matrix estimation calculation formula by combining the calibration mapping transformation calculation formula and the calibration mapping transformation matrix verification and derivation, which is used to perform matrix estimation calculation on the rotation matrix X, translation matrix P and scaling matrix S in the calibration mapping transformation matrix G. The formula fully considers the rotation estimation matrix Translation Estimation Matrix Scaling the Estimation Matrix The homogeneous coordinate C of the i-th calibration mapping point in the three-dimensional matching coordinate system point set of the intervertebral foramen position change i , the homogeneous coordinates A of the i-th calibration mapping point in the feature point set of the intervertebral foramen calibration plate corner point position i , the homogeneous coordinates B of the i-th calibration mapping point in the angle feature point set of the intervertebral foramen calibration plate i , transposition symbol T, matrix multiplication symbol ·, index item i of the calibration mapping point in the feature point set, number n of calibration mapping points in the feature point set, firstly, the product of the position change 3D matching coordinates of each calibration mapping point and the position feature point coordinate transposition is accumulated, and the product of the angle feature point coordinates and the position feature point coordinate transposition is divided, so as to obtain an estimated function relationship of the estimated rotation matrix Secondly, the three-dimensional matching coordinates of the position change of each calibration mapping point minus the product of the rotation estimation matrix and the corner point position feature point coordinates are accumulated, and the product of the transposed position feature point coordinates is divided by the product of the angle feature point coordinates and the transposed position feature point coordinates to obtain an estimated translation matrix. Then, the three-dimensional matching coordinates of the position change of each calibration mapping point are subtracted from the product of the rotation estimation matrix and the corner point position feature point coordinates, and then the product of the translation estimation matrix and the corner point position feature point coordinates are subtracted, and then the product of the position feature point coordinates transposed is accumulated, and finally the angle feature point coordinates and the angle feature point coordinates transposed product accumulation result are divided, so as to obtain an estimated function relationship of the estimated scaling matrix. In summary, these matrix estimation calculation formulas can ensure the accurate estimation of the calibration mapping transformation matrix in the entire matrix estimation calculation process, so that the three-dimensional matching coordinate system of the intervertebral foramen position change can be successfully mapped to the calibration plate, thereby realizing the accurate correction of the intervertebral foramen image. This matrix calculation method is crucial to ensure the precision and accuracy of matching.

[0136] Preferably, the present invention further provides a percutaneous foraminal endoscopy image calibration system for executing the percutaneous foraminal endoscopy image calibration method as described above, the percutaneous foraminal endoscopy image calibration system comprising:

[0137] The intervertebral foramen position feature point set analysis module is used to obtain the standard image of the intervertebral foraminal endoscopic surgery area, and match the intervertebral foraminal key feature point set on the standard image of the intervertebral foraminal endoscopic surgery area to obtain the intervertebral foraminal edge texture feature point set and the intervertebral foraminal contour structure feature point set; based on the intervertebral foraminal edge texture feature point set, the intervertebral foraminal contour structure feature point set is sampled for position change features to obtain the intervertebral foraminal position change feature point set;

[0138] The intervertebral foramen calibration plate projection conversion module is used to perform three-dimensional coordinate system matching and positioning of the standard image of the intervertebral foraminal endoscopic surgery area based on the intervertebral foraminal position change feature point set to obtain the intervertebral foraminal position change three-dimensional matching coordinate system point set; the intervertebral foraminal position change three-dimensional matching coordinate system point set is subjected to calibration plate projection conversion to obtain the intervertebral foraminal calibration plate space coordinate point set;

[0139] The calibration plate corner point posture feature analysis module is used to detect the calibration plate corner points on the intervertebral foramen calibration plate spatial coordinate point set to obtain the intervertebral foramen calibration plate corner point set; perform corner point posture feature point set detection on the intervertebral foramen calibration plate corner point set to obtain the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set;

[0140] The intervertebral foramen image calibration and correction module is used to estimate and calculate the calibration mapping transformation matrix of the three-dimensional matching coordinate system point set of the intervertebral foramen position change based on the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set to obtain the intervertebral foramen calibration mapping transformation estimation matrix; use the intervertebral foramen calibration mapping transformation estimation matrix to perform image calibration and correction on the intervertebral foramen calibration plate space coordinate point set, thereby obtaining the intervertebral foramen image calibration and correction coordinate system point set.

[0141] In summary, the present invention provides a percutaneous foraminal endoscopic image calibration system, which consists of an intervertebral foraminal position feature point set analysis module, an intervertebral foraminal calibration plate projection conversion module, a calibration plate corner point posture feature analysis module and an intervertebral foraminal image calibration correction module. It can implement any one of the percutaneous foraminal endoscopic image calibration methods described in the present invention, and is used to combine the operations between computer programs running on each module to implement any one of the percutaneous foraminal endoscopic image calibration methods. The internal structures of the system cooperate with each other, which can greatly reduce duplication of work and manpower investment, and can quickly and effectively provide a more accurate and efficient percutaneous foraminal endoscopic image calibration process, thereby simplifying the operation process of the percutaneous foraminal endoscopic image calibration system.

[0142] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0143] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for calibrating a transforaminal endoscopic image, characterized in that: The following steps are involved: Step S1: Obtain a standard image of the percutaneous endoscopic surgery area, and match the key feature point set of the intervertebral foramen on the standard image of the percutaneous endoscopic surgery area to obtain a feature point set of the intervertebral foramen edge texture and a feature point set of the intervertebral foramen contour structure; perform position change feature sampling processing on the feature point set of the intervertebral foramen contour structure based on the feature point set of the intervertebral foramen edge texture to obtain a feature point set of the intervertebral foramen position change; Step S2: performing three-dimensional coordinate system matching and positioning of the standard image of the intervertebral foraminal endoscopic surgery area based on the intervertebral foraminal position change feature point set to obtain the intervertebral foraminal position change three-dimensional matching coordinate system point set; performing calibration plate projection transformation on the intervertebral foraminal position change three-dimensional matching coordinate system point set to obtain the intervertebral foraminal calibration plate space coordinate point set; Step S3: performing calibration plate corner point detection on the intervertebral foramen calibration plate spatial coordinate point set to obtain an intervertebral foramen calibration plate corner point set; performing corner point posture feature point set detection on the intervertebral foramen calibration plate corner point set to obtain an intervertebral foramen calibration plate corner point position feature point set and an intervertebral foramen calibration plate corner point angle feature point set; Step S4: Based on the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set, a calibration mapping transformation matrix is ​​estimated and calculated for the intervertebral foramen position change three-dimensional matching coordinate system point set to obtain the intervertebral foramen calibration mapping transformation estimation matrix; the intervertebral foramen calibration mapping transformation estimation matrix is ​​used to perform image calibration correction on the intervertebral foramen calibration plate space coordinate point set to obtain the intervertebral foramen image calibration correction coordinate system point set.

2. The method for calibrating intervertebral foraminal endoscopy images according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: performing image acquisition and processing of the transforaminal endoscopic surgery process through an optical lens to obtain an initial image of the transforaminal endoscopic surgery area; Step S12: performing image denoising processing on the initial image of the transforaminal endoscopic surgery area to obtain a denoised image of the transforaminal endoscopic surgery area; Step S13: performing contrast enhancement processing on the denoised image of the transforaminal endoscopic surgery area to obtain a contrast-enhanced image of the transforaminal endoscopic surgery area; performing image standardization on the contrast-enhanced image of the transforaminal endoscopic surgery area to obtain a standard image of the transforaminal endoscopic surgery area; Step S14: matching the intervertebral foramen key feature point set on the standard image of the intervertebral foraminal endoscopic surgery area to obtain the intervertebral foramen edge texture feature point set and the intervertebral foramen contour structure feature point set; Step S15: performing position change feature sampling processing on the intervertebral foramen contour structure feature point set based on the intervertebral foramen edge texture feature point set to obtain the intervertebral foramen position change feature point set.

3. The method for calibrating intervertebral foraminal endoscopy images according to claim 2, characterized in that: Step S14 includes the following steps: Step S141: performing edge texture and contour structure key feature analysis on the standard image of the transforaminal endoscopic surgery area to obtain key feature data of the edge texture of the transforaminal endoscopic lens and key feature data of the contour structure of the transforaminal endoscopic lens; Step S142: performing feature point sampling processing on the key feature data of the edge texture of the percutaneous foraminal endoscope and the key feature data of the contour structure of the percutaneous foraminal endoscope to obtain feature sampling points of the edge texture of the percutaneous foraminal endoscope and feature sampling points of the contour structure of the percutaneous foraminal endoscope; Step S143: performing intervertebral foramen position matching analysis on the standard image of the intervertebral foraminal endoscope surgical area to obtain intervertebral foraminal endoscope actual position matching relationship data; Step S144: performing texture feature point set matching on the texture feature sampling points at the edge of the intervertebral foraminal endoscope according to the actual position matching relationship data of the intervertebral foraminal endoscope to obtain a texture feature point set at the edge of the intervertebral foraminal endoscope; Step S145: performing structural feature point set matching on the structural feature sampling points of the intervertebral foraminal endoscope contour according to the actual position matching relationship data of the intervertebral foraminal endoscope, so as to obtain the structural feature point set of the intervertebral foraminal endoscope contour.

4. The method for calibrating intervertebral foraminal endoscopy images according to claim 2, characterized in that: Step S15 includes the following steps: Step S151: performing local feature description processing on the texture feature point set of the edge of the intervertebral foramen to obtain local texture structure information data of the edge of the intervertebral foramen; Step S152: performing contour deformation detection on the intervertebral foramen contour structure feature point set according to the intervertebral foramen edge local texture structure information data to obtain intervertebral foramen edge contour structure deformation information data; Step S153: performing intervertebral foramen motion estimation on the intervertebral foramen contour structure feature point set to obtain intervertebral foramen contour motion trajectory information data; performing position change tracking analysis on the intervertebral foramen contour motion trajectory information data based on the intervertebral foramen edge contour structure deformation information data to obtain intervertebral foramen contour position change information data; Step S154: performing data smoothing processing on the intervertebral foramen contour position change information data to obtain intervertebral foramen contour position change standard data; performing dynamic reconstruction processing on the intervertebral foramen contour position change standard data using three-dimensional reconstruction technology to obtain intervertebral foramen position change reconstruction structure space; Step S155: performing feature point set sampling processing on the intervertebral foramen position change reconstruction structure space to obtain an intervertebral foramen position change feature point set.

5. The method for calibrating intervertebral foraminal endoscopy images according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: performing a three-dimensional coordinate system conversion on the standard image of the transforaminal endoscopic surgery area to obtain a three-dimensional coordinate system space of the transforaminal image; Step S22: performing feature point labeling sampling processing on the three-dimensional coordinate system space of the intervertebral foramen image to obtain a three-dimensional space feature point set of the intervertebral foramen image; Step S23: performing mapping matching calculation on the three-dimensional space feature point set of the intervertebral foramen image based on the intervertebral foramen position change feature point set using the position change mapping matching calculation formula to obtain the intervertebral foramen position change mapping matching degree; Step S24: mapping and matching the three-dimensional spatial feature point set of the intervertebral foramen image according to the mapping matching degree of the intervertebral foramen position change to obtain a three-dimensional matching coordinate system point set of the intervertebral foramen position change; Step S25: performing calibration plate projection transformation on the three-dimensional matching coordinate system point set of the intervertebral foramen position change to obtain the intervertebral foramen calibration plate space coordinate point set.

6. The method for calibrating intervertebral foraminal endoscopy images according to claim 5, characterized in that: The position change mapping matching calculation formula in step S23 is specifically: Where M is the matching degree of the intervertebral foramen position change mapping, N is the number of feature points in the feature point set, j is the index item of the feature point in the feature point set, and w is the j is the matching influence weight parameter of the jth feature point, σ j is the standard deviation of the matching Gaussian kernel of the jth feature point, d j is the matching influence control coefficient of the jth feature point, x aj is the spatial horizontal coordinate parameter of the jth feature point in the intervertebral foramen position change feature point set, x bj is the spatial horizontal coordinate parameter of the jth feature point in the three-dimensional spatial feature point set of the intervertebral foramen image, y aj is the spatial ordinate parameter of the jth feature point in the intervertebral foramen position change feature point set, y bj is the spatial ordinate parameter of the jth feature point in the three-dimensional spatial feature point set of the intervertebral foramen image, z aj is the spatial vertical coordinate parameter of the jth feature point in the intervertebral foramen position change feature point set, z bj is the spatial vertical coordinate parameter of the jth feature point in the three-dimensional spatial feature point set of the intervertebral foramen image, and η is the correction value of the matching degree of the intervertebral foramen position change mapping.

7. The method for calibrating transforaminal endoscopic images according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: performing calibration plate corner point detection on the intervertebral foramen calibration plate spatial coordinate point set to obtain an intervertebral foramen calibration plate corner point set; Step S32: using a pyramid convolutional neural network to perform corner point feature extraction processing on the intervertebral foramen calibration plate corner point set to obtain intervertebral foramen calibration plate corner point feature information data; Step S33: performing feature change pattern analysis on the feature information data of the corner points of the intervertebral foramen calibration plate to obtain feature change pattern data of the corner points of the intervertebral foramen calibration plate; performing position and angle feature change recognition analysis on the feature change pattern data of the corner points of the intervertebral foramen calibration plate to obtain position change pattern data of the corner points of the intervertebral foramen calibration plate and angle change pattern data of the corner points of the intervertebral foramen calibration plate; Step S34: performing position feature point set detection on the intervertebral foramen calibration plate corner point set according to the position change pattern data of the intervertebral foramen calibration plate corner points to obtain the position feature point set of the intervertebral foramen calibration plate corner points; Step S35: performing angle feature point set detection on the intervertebral foramen calibration plate corner point set according to the angle variation pattern data of the intervertebral foramen calibration plate corner point to obtain the angle feature point set of the intervertebral foramen calibration plate corner point.

8. The method for calibrating transforaminal endoscopic images according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: mapping and transforming the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set to the intervertebral foramen position change three-dimensional matching coordinate system point set using the calibration mapping transformation calculation formula; The calibration mapping transformation calculation formula is as follows: In the formula, C i is the homogeneous coordinate of the i-th calibration mapping point in the three-dimensional matching coordinate system point set of the intervertebral foramen position change, A i is the homogeneous coordinate of the i-th calibration mapping point in the feature point set of the intervertebral foramen calibration plate corner point position, B i is the homogeneous coordinate of the i-th calibration mapping point in the angle feature point set of the intervertebral foramen calibration plate, T is the transposition symbol, · is the matrix dot multiplication symbol, i is the index item of the calibration mapping point in the feature point set, n is the number of calibration mapping points in the feature point set, and G is the calibration mapping transformation matrix; Step S42: construct the matrix form of the calibration mapping transformation matrix G according to the calibration mapping transformation relationship of the calibration mapping transformation calculation formula: Among them, X is the rotation matrix, P is the translation matrix, and S is the scaling matrix; Step S43: performing matrix estimation calculation on the rotation matrix X, the translation matrix P and the scaling matrix S in the calibration mapping transformation matrix G to obtain a rotation estimation matrix, a translation estimation matrix and a scaling estimation matrix; Step S44: performing calibration mapping transformation matrix estimation calculation through the rotation estimation matrix, the translation estimation matrix and the scaling estimation matrix to obtain the intervertebral foramen calibration mapping transformation estimation matrix; Step S45: performing image calibration correction on the intervertebral foramen calibration plate space coordinate point set using the intervertebral foramen calibration mapping transformation estimation matrix to obtain an intervertebral foramen image calibration correction coordinate system point set.

9. The method for calibrating intervertebral foraminal endoscopy images according to claim 8, characterized in that: The matrix estimation calculation in step S43 is solved by the following matrix estimation calculation formula: In the formula, is the rotation estimation matrix, is the translation estimation matrix, is the scaling estimate matrix.

10. A transforaminal endoscopic image calibration system, characterized in that: Used to execute the percutaneous foraminal endoscopy image calibration method as claimed in claim 1, the percutaneous foraminal endoscopy image calibration system comprises: The intervertebral foramen position feature point set analysis module is used to obtain the standard image of the intervertebral foraminal endoscopic surgery area, and match the intervertebral foraminal key feature point set on the standard image of the intervertebral foraminal endoscopic surgery area to obtain the intervertebral foraminal edge texture feature point set and the intervertebral foraminal contour structure feature point set; based on the intervertebral foraminal edge texture feature point set, the intervertebral foraminal contour structure feature point set is sampled for position change features to obtain the intervertebral foraminal position change feature point set; The intervertebral foramen calibration plate projection conversion module is used to perform three-dimensional coordinate system matching and positioning of the standard image of the intervertebral foraminal endoscopic surgery area based on the intervertebral foraminal position change feature point set to obtain the intervertebral foraminal position change three-dimensional matching coordinate system point set; the intervertebral foraminal position change three-dimensional matching coordinate system point set is subjected to calibration plate projection conversion to obtain the intervertebral foraminal calibration plate space coordinate point set; The calibration plate corner point posture feature analysis module is used to detect the calibration plate corner points on the intervertebral foramen calibration plate spatial coordinate point set to obtain the intervertebral foramen calibration plate corner point set; perform corner point posture feature point set detection on the intervertebral foramen calibration plate corner point set to obtain the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set; The intervertebral foramen image calibration and correction module is used to estimate and calculate the calibration mapping transformation matrix of the three-dimensional matching coordinate system point set of the intervertebral foramen position change based on the intervertebral foramen calibration plate corner point position feature point set and the intervertebral foramen calibration plate corner point angle feature point set to obtain the intervertebral foramen calibration mapping transformation estimation matrix; use the intervertebral foramen calibration mapping transformation estimation matrix to perform image calibration and correction on the intervertebral foramen calibration plate space coordinate point set, thereby obtaining the intervertebral foramen image calibration and correction coordinate system point set.

Images