Medical image registration method, control system, electronic device and storage medium

Abstract

The present invention relates to a medical image registration method, a control system, an electronic device and a storage medium, comprising the steps of acquiring a white light image and a fluorescence image; performing image feature extraction on the white light image and the fluorescence image respectively to acquire pixel points with image feature information; aligning the pixel points with image feature information on the white light image and the fluorescence image in response to the acquired pixel points with image feature information to obtain a target image; updating the image feature information in response to the pixel points with no image feature information acquired, and performing image feature extraction on the white light image and the fluorescence image using the updated image feature information until the pixel points with image feature information are acquired to align the white light image and the fluorescence image to obtain the target image, thereby improving the success rate and efficiency of image registration and reducing the risk of doctors misjudging and interfering with real lesions or diagnostic results.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to a medical image registration method, a control system, an electronic device and a storage medium. Background Art

[0002] During medical diagnosis, the basis for determining the condition of lesions mostly comes from medical images captured by medical imaging equipment. Due to noise in the shooting environment or the equipment itself, the captured images can only express the conditions of the lesion and its surroundings from a certain angle (such as the outline, structure, texture, etc. of the lesion). For example, an unstable shooting environment causes fluctuations in the brightness level of the lesion image, resulting in noise such as spots and stripes in the image, which obscures the texture details of the lesion and can only be observed from the single angle of the outline. Therefore, doctors cannot obtain effective lesion information based on these incomplete images and cannot formulate a diagnostic plan. Medical image registration technology has been proposed to align target images of the same lesion from multiple devices or the same device at different times and environments, providing a complementary description of the lesion information from multiple angles, displaying its complete information, and assisting in the accurate diagnosis of the disease. Medical image registration is one of the basic steps in medical image processing and is of great significance in the processes of medical image fusion, intelligent diagnosis of lesions, image-guided surgery, etc.

[0003] According to interactivity, medical image registration devices can be divided into manual registration and automatic registration; according to the position of the registration control points, they can be divided into external control point-based registration and internal control point-based registration; according to the registration process, they can be divided into feature-based registration and grayscale-based registration. Current registration devices are unable to maintain high registration accuracy while maintaining high-speed real-time performance. For example, completely manual registration is prone to unstable accuracy, and although automatic registration has stable registration accuracy, it is limited in speed. Feature-based registration often uses deep learning methods and has high accuracy, but the process is too complicated, resulting in poor real-time performance. Grayscale-based registration has high real-time performance due to its simple processing method, but the registration accuracy cannot meet clinical needs.

[0004] At present, when performing image registration on medical images, it is necessary to extract image features of the medical images and obtain multiple pixel points with image feature information (such as color features, texture features, shape features, spatial relationship features, geometric features, contour features, regional features, etc.), and then further match the pixel points at corresponding positions on different medical images. The successfully matched pixel points are spatially superimposed to obtain a new matching image to complete the image registration. Therefore, obtaining pixel points with image feature information is the key to performing image registration and obtaining the registered target image. However, in the existing technology, due to the low accuracy of the calculation results of the image processing algorithm when extracting image feature information of medical images, it is difficult to ensure that the pixel points with image feature information can be accurately obtained on the medical images to be registered, thereby making it impossible to perform image registration on the medical images, reducing the success rate and efficiency of image registration, and easily causing doctors to misjudge and interfere with the real lesions or diagnostic results. Summary of the Invention

[0005] (1) Technical problems to be solved

[0006] The present invention provides a medical image registration method, control system, electronic device and storage medium, which improve the accuracy of the entire calculation result by repeatedly updating the calculation result of the image gradient function, thereby avoiding the problem of pixels with image feature information being missed due to low calculation accuracy, improving the success rate and efficiency of image registration, and reducing the risk of doctors misjudging and interfering with the actual lesions or diagnostic results.

[0007] (2) Technical solution

[0008] In a first aspect, an embodiment of the present invention proposes a medical image registration method, wherein the medical image includes a white light image and a fluorescence image, including step S1: acquiring the white light image and the fluorescence image; step S2: performing image feature extraction on the white light image and the fluorescence image respectively to obtain pixel points with image feature information, including step S21: selecting any pixel point on the white light image and the fluorescence image, and planning the center pixel and neighborhood pixels of the pixel point; step S22: using an image gradient function to calculate the absolute value of the gradient difference between the center pixel and the neighborhood pixel as a first difference, and calculating the absolute value of the gradient difference between each of the neighborhood pixels as a second difference, and calculating a first ratio of the first difference to a first threshold and a second ratio of the second difference to a second threshold; step S23: comparing the first ratio with a first critical value to obtain a first difference. and the size of the second ratio and the second critical value; step S24: in response to the first ratio being greater than the first critical value and the second ratio being less than or equal to the second critical value, it is judged that the pixel point has image feature information; otherwise, it is judged that the pixel point does not have image feature information; step S3: in response to the pixel point with image feature information being obtained, step S5 is entered; in response to the pixel point with image feature information not being obtained, step S4 is entered; step S4: the calculation result of the image gradient function is updated, and the updated calculation result is used to re-compare the corresponding first critical value and the second critical value with the ratio of the first threshold and the second threshold until the pixel point with image feature information is obtained, and step S5 is entered; step S5: the pixel point with image feature information on the white light image is aligned with the pixel point with image feature information on the fluorescence image to obtain the target image.

[0009] Furthermore, the calculation result of the image gradient function is updated, and the corresponding first critical value and the second critical value are re-compared with the ratio of the updated calculation result to the first threshold and the second threshold until the pixel point with image feature information is obtained, including step S41: using the image gradient function to calculate the first gradient value of the center pixel and the neighborhood pixel and the second gradient value when the neighborhood pixel is used as the center pixel and the center pixel is used as the corresponding neighborhood pixel, and at the same time calculating the absolute value of the difference between the Nth power of the first gradient value and the Nth power of the second gradient value as the third difference; step S42: using the image gradient function to calculate the third gradient value of any two of the neighborhood pixels and the second gradient value of the two neighborhood pixels The fourth gradient value after the interchange of positions, and the absolute value of the difference between the Nth power of the third gradient value and the Nth power of the third gradient value is calculated as the fourth difference; step S43: calculate the third ratio of the third difference to the first threshold and the fourth ratio of the fourth difference to the second threshold, and compare the size of the third ratio with the first critical value and the size of the fourth ratio with the second critical value; step S44: in response to the third ratio being greater than the first critical value and the fourth ratio being less than or equal to the second critical value, it is judged that the pixel point has image feature information, and then enter step S5; otherwise, it is judged that the pixel point does not have image feature information, and then enter step S41, calculate the absolute value of the difference between the N+1 power of the first gradient value and the N+1 power of the second gradient value, until it is judged that the pixel point has image feature information.

[0010] Furthermore, the method further includes step S6: evaluating the image quality of the target image, and adjusting the light source intensity in response to the quality evaluation result of the target image.

[0011] Furthermore, the image quality of the target image is evaluated, and the light source intensity is adjusted in response to the quality evaluation result of the target image, including step S61: calculating the mean of the grayscale values ​​of all pixels on the white light image as a first mean; and the mean of the grayscale values ​​of all pixels on the fluorescence image as a second mean; and the mean of the grayscale values ​​of all pixels on the target image as a third mean; step S62: calculating the absolute value of the difference between the first mean and the third mean and the absolute value of the difference between the second mean and the third mean, and averaging the absolute value of the difference between the first mean and the third mean and the absolute value of the difference between the second mean and the third mean to obtain a fourth mean; step S63: calculating the ratio of the absolute value of the difference between the fourth mean and the first mean or the second mean and the first mean or the second mean as a fifth ratio; step S64: comparing the fifth ratio with the third critical value, and adjusting the light source intensity in response to the fifth ratio being greater than the third critical value.

[0012] Furthermore, comparing the fifth ratio with the third critical value, and adjusting the light source intensity in response to the fifth ratio being greater than the third critical value, also includes increasing the light source intensity in response to a positive difference between the third mean and the first mean or the second mean, and decreasing the light source intensity in response to a negative difference between the third mean and the first mean or the second mean.

[0013] In the second aspect, an embodiment of the present invention also proposes a control system for implementing the medical image registration method of the first aspect, comprising a first camera unit for acquiring the white light image; a second camera unit for acquiring the fluorescence image; an image processing module electrically connected to the first camera unit and the second camera unit, for receiving the white light image and the fluorescence image, performing feature extraction on the white light image and the fluorescence image to obtain pixel points with image feature information, and performing image registration on the pixel points with image feature information on the white light image and the pixel points with image feature information on the fluorescence image.

[0014] Furthermore, the image processing module includes an image data receiving module, which is electrically connected to the first camera unit and the second camera unit, and is used to receive the white light image and the fluorescence image; and an image feature extraction module, which is electrically connected to the image data receiving module, and is used to extract image features from the white light image and the fluorescence image and obtain pixel points with image feature information; and an information feature updating module, which is electrically connected to the image feature extraction module, and is used to update the calculation result of the image gradient function in response to a second control instruction, and transmit the calculation result to the image feature extraction module; and an image registration module, which is electrically connected to the image feature extraction module, and is used to align the pixel points with image feature information on the white light image with the pixel points with image feature information on the fluorescence image in response to the first control instruction to obtain the target image; and an FPGA chip unit, which is electrically connected to the image feature extraction module, the image feature updating module and the image registration module, and is used to generate a first control instruction according to the pixel points with image feature information obtained by the image feature extraction module, and to generate a second control instruction according to the pixel points with image feature information not obtained by the image feature extraction module.

[0015] Furthermore, it also includes a light source module, which is electrically connected to the first camera unit and the second camera unit, and is used to provide light source energy to the first camera unit and the second camera unit. The image processing module also includes the image quality assessment module electrically connected to the image registration module and the FPGA chip unit, and is used to assess the image quality of the target image. The FPGA chip unit generates a third control instruction for adjusting the light source intensity of the light source module based on the image quality assessment result of the target image.

[0016] In a third aspect, an embodiment of the present invention further proposes an electronic device comprising a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to implement the medical image registration method described in the first aspect when executing the instructions stored in the memory.

[0017] In a fourth aspect, an embodiment of the present invention further proposes a non-volatile computer-readable storage medium having computer program instructions stored thereon, characterized in that when the computer program instructions are executed by a processor, the medical image registration method described in the first aspect is implemented.

[0018] (3) Beneficial effects

[0019] In summary, the present invention obtains pixel points with image feature information by updating the features of pixel points on the medical image that do not have image feature information, which can ensure that all pixel points on the medical image have image feature information, improve the success rate and efficiency of image registration, and reduce the risk of doctors misjudging and interfering with real lesions or diagnostic results. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 This is a control flow chart of the medical image registration method of the present invention.

[0022] Figure 2 It is a control flow chart of the method for obtaining pixel points with image feature information of the present invention.

[0023] Figure 3 is another control flow chart of the medical image registration method of the present invention.

[0024] Figure 4 This is a control flow chart of the information feature updating method of the present invention.

[0025] Figure 5 It is a control flow chart of the method for adjusting the light source intensity of the present invention.

[0026] Figure 6 This is a structural distribution diagram of pixel points of the present invention.

[0027] Figure 7 It is a structural principle diagram of the control system of the present invention.

[0028] Figure 8It is a structural principle diagram of the image processing module of the present invention.

[0029] Figure 9 It is another structural principle diagram of the control system of the present invention.

[0030] Figure 10 Schematic diagram of the image registration process of the present invention.

[0031] In the picture:

[0032] 10-first camera unit; 20-second camera unit; 30-image processing module; 40-image data receiving module; 50-image feature extraction module; 60-information feature updating module; 70-image registration module; 80-FPGA chip unit; 90-image quality assessment module; 100-light source module. DETAILED DESCRIPTION

[0033] The following detailed description of the embodiments of the present invention is provided in conjunction with the accompanying drawings and examples. The detailed description of the following embodiments and the accompanying drawings are intended to illustrate the principles of the present invention and are not intended to limit the scope of the present invention. That is, the present invention is not limited to the described embodiments and covers any modifications, replacements, and improvements to the parts, components, and connection methods without departing from the spirit of the present invention.

[0034] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0035] like Figure 1 and Figure 2 As shown, in a first aspect, the present invention relates to a medical image registration method, wherein the medical image includes a white light image and a fluorescence image, and the registration method includes the following steps: step S1: acquiring a white light image and a fluorescence image; step S2: performing image feature extraction on the white light image and the fluorescence image respectively to obtain pixel points with image feature information, wherein step S2 specifically includes step S21: selecting any pixel point on the white light image and the fluorescence image, and planning the central pixel and neighboring pixels of the pixel point, such as Figure 6As shown, taking an 8×8 pixel matrix of pixel points as an example, the central pixel is planned to be 4 pixels at the center of the image, and the remaining 60 pixels are all neighborhood pixels; step S22: using the image gradient function to calculate the absolute value of the gradient difference between the central pixel and the neighborhood pixel as the first difference, and calculating the absolute value of the gradient difference between each neighborhood pixel as the second difference, and calculating a first ratio of the first difference to the first threshold and a second ratio of the second difference to the second threshold, the first difference is the absolute value of the gradient difference between the first gradient value of the central pixel and the neighborhood pixel calculated by the image gradient function and the second gradient value calculated when the neighborhood pixel is the central pixel and the central pixel is the corresponding neighborhood pixel, for example, a is the central pixel and b is the neighborhood pixel, that is, the image gradient function is used to calculate the gradient between the central pixel a and the neighborhood pixel. The first gradient value of pixel b is denoted as ab, and the second gradient value of the central pixel b and the neighboring pixel a calculated using the image gradient function after swapping the positions of the central pixel a and the neighboring pixel b is denoted as ba. Therefore, the first difference is the absolute value of ab-ba; similarly, the second difference is the absolute value of the gradient difference between the third gradient value of the two neighboring pixels calculated using the image gradient function and the fourth gradient value calculated when the two neighboring pixels are swapped. For example, c is the first neighboring pixel and d is the second neighboring pixel, that is, the gradient value of the first neighboring pixel c and the second neighboring pixel d calculated using the image gradient function is denoted as cd, and the gradient value of the first neighboring pixel d and the second neighboring pixel c calculated using the image gradient function after swapping the positions of the first neighboring pixel c and the second neighboring pixel d is dc. Therefore, the second difference is the absolute value of cd-dc.The first threshold is selected as the larger of the gradient value ab calculated by the image gradient function for the center pixel a and the neighboring pixel b, or the gradient value ba calculated by the image gradient function after the positions of the center pixel a and the neighboring pixel b are swapped, and the second threshold is selected as the smaller of the gradient value cd calculated by the image gradient function for the first neighboring pixel c and the second neighboring pixel d, or the gradient value dc calculated by the image gradient function after the positions of the first neighboring pixel c and the second neighboring pixel d are swapped, and the first ratio is compared with the first critical value, and the second ratio is compared with the second critical value, the first critical value is selected as 10%, and the second critical value is selected as 2%; step S24: in response to the first ratio, If the pixel point has image feature information and the second ratio is greater than the first critical value and is less than or equal to the second critical value, it is judged that the pixel point has image feature information; otherwise, it is judged that the pixel point does not have image feature information; step S3: in response to the pixel point with image feature information being obtained, enter step S5; in response to the pixel point with image feature information not being obtained, enter step S4; step S4: update the calculation result of the image gradient function, and use the updated calculation result and the ratio of the above-mentioned first threshold and the above-mentioned second threshold to re-compare the corresponding first critical value and the above-mentioned second critical value until the pixel point with image feature information is obtained, and enter step S5; step S5: align the pixel point with image feature information on the white light image with the pixel point with image feature information on the fluorescence image to obtain the target image, such as. Figure 10 As shown, specifically, the pixel points with image feature information on the white light image and the pixel points with image feature information on the fluorescence image correspond to each other in spatial position and their coordinate positions are the same.

[0036] The present invention repeatedly updates the calculation results of the image gradient function, and uses the updated results to re-compare the corresponding first critical value and the second critical value until it is identified that the pixel points on the medical image that were initially judged not to have image feature information are actually pixels with image feature information. This ensures that all pixels on the medical image have image feature information, and uses the repeatedly updated calculation results to improve its accuracy, thereby avoiding the problem of pixels with image feature information being omitted due to low calculation accuracy, improving the success rate and efficiency of image registration, and reducing the risk of doctors misjudging and interfering with real lesions or diagnostic results.

[0037] As a preferred embodiment, Figure 4As shown, the calculation result of the image gradient function is updated, and the updated calculation result is used to re-compare the corresponding first critical value and the second critical value with the ratio of the first threshold and the second threshold until the pixel point with image feature information is obtained, including step S41: using the image gradient function to calculate the first gradient value of the center pixel and the neighborhood pixel and the second gradient value when the neighborhood pixel is used as the center pixel and the center pixel is used as the corresponding neighborhood pixel, and at the same time calculating the absolute value of the difference between the Nth power of the first gradient value and the Nth power of the second gradient value as the third difference; step S42: using the image gradient function to calculate the third gradient value of any two neighborhood pixels and the fourth gradient value after the two neighborhood pixels are swapped, and at the same time calculating the Nth power of the third gradient value The fourth difference is calculated by calculating the absolute value of the difference between the Nth power of the first gradient value and the Nth power of the third gradient value, where N is a natural number greater than or equal to 2. Step S43: Calculating a third ratio of the third difference value to the first threshold value and a fourth ratio of the fourth difference value to the second threshold value, and comparing the third ratio with the first critical value and the fourth ratio with the second critical value. Step S44: In response to the third ratio being greater than the first critical value and the fourth ratio being less than or equal to the second critical value, it is determined that the pixel point has image feature information, and the process proceeds to step S5. Otherwise, it is determined that the pixel point does not have image feature information, and the process proceeds to step S41, calculating the absolute value of the difference between the N+1th power of the first gradient value and the N+1th power of the second gradient value, and repeating steps S41 to S44 until it is determined that the pixel point has image feature information. By using the Nth power to update the gradient values ​​between the central pixel and the neighboring pixels, and between the neighboring pixels, the computational complexity of the image gradient function is reduced, the output of the calculation results is accelerated, and the acquisition rate of pixels with image feature information is improved.

[0038] As another preferred embodiment, Figure 3 As shown, the medical image registration method further includes step S6: evaluating the image quality of the target image and adjusting the light source intensity in response to the quality evaluation result of the target image. By adjusting the light source intensity, the light source intensity of the white light image and the fluorescence image is adjusted to meet the quality requirements of the target image.

[0039] As another optional implementation.

[0040] Preferably, if Figure 5As shown, the image quality of the target image is evaluated, and the light source intensity is adjusted in response to the quality evaluation result of the target image, including step S61: calculating the mean of the grayscale values ​​of all pixels on the white light image as a first mean; and the mean of the grayscale values ​​of all pixels on the fluorescence image as a second mean; and the mean of the grayscale values ​​of all pixels on the target image as a third mean; step S62: calculating the absolute value of the difference between the first mean and the third mean and the absolute value of the difference between the second mean and the third mean, and averaging the absolute value of the difference between the first mean and the third mean and the absolute value of the difference between the second mean and the third mean to obtain a fourth mean; step S63: calculating the absolute value of the difference between the fourth mean and the first mean or the second mean and the absolute value of the difference between the first mean and the third mean and the absolute value of the difference between the second mean and the third mean to obtain a fourth mean. The ratio of the second mean is the fifth ratio, that is, the ratio of the absolute value of the difference between the fourth mean and the first mean to the first mean and the ratio of the absolute value of the difference between the fourth mean and the second mean to the second mean are both the fifth ratio; step S64: comparing the fifth ratio with the third critical value, and adjusting the light source intensity in response to the fifth ratio being greater than the third critical value, that is, adjusting the white light source intensity when the ratio of the absolute value of the difference between the fourth mean and the first mean to the first mean is greater than the third critical value, and adjusting the fluorescence light source intensity when the ratio of the absolute value of the difference between the fourth mean and the second mean to the second mean is greater than the third critical value; if the fifth ratio is less than or equal to the third critical value, the light source intensity is not adjusted, and the target image, white light image, and fluorescence image are directly output. The third critical value is selected as 10%.

[0041] Preferably, if Figure 5 As shown, the method of comparing the fifth ratio with the third critical value and adjusting the light source intensity in response to the fifth ratio being greater than the third critical value further includes increasing the light source intensity in response to a positive difference between the third mean and the first mean or the second mean, that is, increasing the white light source intensity when the difference between the third mean and the first mean is positive or increasing the fluorescent light source intensity when the difference between the third mean and the second mean is positive, and decreasing the light source intensity in response to a negative difference between the third mean and the first mean or the second mean, that is, decreasing the white light source intensity when the difference between the third mean and the first mean is negative or decreasing the fluorescent light source intensity when the difference between the third mean and the second mean is negative.

[0042] Second, as Figure 7As shown, the present invention relates to a control system for implementing the medical image registration method of the first aspect, the control system comprising a first camera unit 10 for acquiring the white light image; and a second camera unit 20 for acquiring the fluorescence image; and an image processing module 30 electrically connected to the first camera unit 10 and the second camera unit 20, for receiving the white light image and the fluorescence image, performing feature extraction on the white light image and the fluorescence image to obtain pixel points having image feature information, and performing image registration on the pixel points having image feature information on the white light image and the pixel points having image feature information on the fluorescence image.

[0043] Preferably, if Figure 8 As shown, the image processing module 30 includes an image data receiving module 40, which is electrically connected to the first camera unit 10 and the second camera unit 20 and is used to receive the white light image and the fluorescence image; and an image feature extraction module 50, which is electrically connected to the image data receiving module 40 and is used to extract image features from the white light image and the fluorescence image and obtain pixel points with image feature information; and an information feature updating module 60, which is electrically connected to the image feature extraction module 50 and is used to update the calculation result of the image gradient function in response to the second control instruction and transmit the calculation result to the image feature extraction module 50; and image registration Module 70 is electrically connected to the image feature extraction module 50, and is used to align the pixel points with image feature information on the white light image with the pixel points with image feature information on the fluorescence image in response to the first control instruction to obtain the target image; and the FPGA chip unit 80 is electrically connected to the image feature extraction module 50, the image feature update module 60 and the image registration module 70, and is used to generate a first control instruction based on the pixel points with image feature information obtained by the image feature extraction module 50, and to generate a second control instruction based on the pixel points with image feature information not obtained by the image feature extraction module 50.

[0044] Preferably, if Figure 9 As shown, the control system also includes a light source module 100, which is electrically connected to the first camera unit 10 and the second camera unit 20, and is used to provide light source energy to the first camera unit 10 and the second camera unit 20. The image processing module 30 also includes an image quality assessment module 90 electrically connected to the image registration module 70 and the FPGA chip unit 80, and is used to assess the image quality of the target image. The FPGA chip unit 80 generates a third control instruction for adjusting the light source intensity of the light source module 100 based on the image quality assessment result of the target image, and is used to adjust the light source energy output by the light source module 100 to the first camera unit 10 and the second camera unit 20, that is, the light source intensity.

[0045] In a third aspect, the present invention further provides an electronic device comprising a processor and a memory for storing processor executable instructions, wherein the processor is configured to implement the medical image registration method of the first aspect when executing the instructions stored in the memory.

[0046] In a fourth aspect, the present invention further provides a non-volatile computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the medical image registration method of the first aspect.

[0047] It should be noted that the various embodiments in this specification are described in a progressive manner. References to the same or similar parts between the various embodiments are sufficient. Each embodiment focuses on the differences from the other embodiments. The present invention is not limited to the specific steps and structures described above and shown in the figures. Furthermore, for the sake of brevity, detailed descriptions of known methods and technologies are omitted here.

[0048] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art without departing from the scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included within the scope of the claims of the present application.

Claims

1. A medical image registration method, wherein the medical image includes a white light image and a fluorescence image, characterized in that: include: Step S1: Acquire the white light image and the fluorescence image; Step S2: performing image feature extraction on the white light image and the fluorescence image respectively to obtain pixel points having image feature information, including: Step S21: selecting any pixel point on the white light image and the fluorescence image, and planning the central pixel and neighboring pixels of the pixel point; Step S22: using an image gradient function to calculate the absolute value of the gradient difference between the central pixel and the neighboring pixels as a first difference, and calculating the absolute value of the gradient difference between each of the neighboring pixels as a second difference, and calculating a first ratio of the first difference to a first threshold and a second ratio of the second difference to a second threshold; Step S23: comparing the first ratio with the first critical value and the second ratio with the second critical value; Step S24: In response to the first ratio being greater than a first threshold and the second ratio being less than or equal to a second threshold, determining that the pixel has image feature information; otherwise, determining that the pixel does not have image feature information; Step S3: In response to the pixel point with image feature information being obtained, proceed to step S5; in response to the pixel point with image feature information not being obtained, proceed to step S4; Step S4: updating the calculation result of the image gradient function, and re-comparing the first critical value and the second critical value corresponding to the updated calculation result with the ratio of the first threshold value to the second threshold value until a pixel point with image feature information is obtained, and then proceeding to step S5; Step S5: registering the pixel points with image feature information on the white light image and the pixel points with image feature information on the fluorescence image to obtain a target image.

2. The medical image registration method according to claim 1, characterized in that: The updating of the calculation result of the image gradient function and re-comparing the first critical value and the second critical value corresponding to the updated calculation result with the ratio of the first threshold value to the second threshold value until a pixel point with image feature information is obtained includes: Step S41: using the image gradient function to calculate a first gradient value between the central pixel and the neighboring pixels, and a second gradient value when the neighboring pixel is used as the central pixel and the central pixel is used as the corresponding neighboring pixel, and calculating an absolute value of a difference between the Nth power of the first gradient value and the Nth power of the second gradient value as a third difference; Step S42: using the image gradient function to calculate a third gradient value of any two neighboring pixels and a fourth gradient value obtained by swapping the positions of the two neighboring pixels, and calculating an absolute value of a difference between the Nth power of the third gradient value and the Nth power of the third gradient value as a fourth difference; Step S43: calculating a third ratio of the third difference to the first threshold and a fourth ratio of the fourth difference to the second threshold, and comparing the third ratio to the first critical value and the fourth ratio to the second critical value; Step S44: In response to the third ratio being greater than the first critical value and the fourth ratio being less than or equal to the second critical value, it is determined that the pixel point has image feature information, and the process proceeds to step S5; otherwise, it is determined that the pixel point does not have image feature information, and the process proceeds to step S41, calculating the absolute value of the difference between the N+1 power of the first gradient value and the N+1 power of the second gradient value, until it is determined that the pixel point has image feature information.

3. The medical image registration method according to claim 1, wherein: The method further includes step S6: evaluating the image quality of the target image, and adjusting the light source intensity in response to the quality evaluation result of the target image.

4. The medical image registration method according to claim 3, wherein: The evaluating the image quality of the target image and adjusting the light source intensity in response to the quality evaluation result of the target image includes: Step S61: calculating the mean of the grayscale values ​​of all pixels on the white light image as a first mean; the mean of the grayscale values ​​of all pixels on the fluorescence image as a second mean; and the mean of the grayscale values ​​of all pixels on the target image as a third mean; Step S62: calculating the absolute value of the difference between the first mean and the third mean and the absolute value of the difference between the second mean and the third mean, and averaging the absolute value of the difference between the first mean and the third mean and the absolute value of the difference between the second mean and the third mean to obtain a fourth mean; Step S63: Calculate the ratio of the absolute value of the difference between the fourth mean and the first mean or the second mean to the first mean or the second mean as a fifth ratio; Step S64 : comparing the fifth ratio with the third critical value, and adjusting the light source intensity in response to the fifth ratio being greater than the third critical value.

5. The medical image registration method according to claim 4, characterized in that: Comparing the fifth ratio with the third critical value, and adjusting the light source intensity in response to the fifth ratio being greater than the third critical value, also includes increasing the light source intensity in response to a positive difference between the third mean and the first mean or the second mean, and decreasing the light source intensity in response to a negative difference between the third mean and the first mean or the second mean.

6. A control system for implementing the medical image registration method of claims 1-5, comprising: A first camera unit (10) for acquiring the white light image; a second imaging unit (20), configured to acquire the fluorescent image; An image processing module (30) is electrically connected to the first camera unit (10) and the second camera unit (20), and is used to receive the white light image and the fluorescence image, perform feature extraction on the white light image and the fluorescence image to obtain pixel points with image feature information, and perform image registration on the pixel points with image feature information on the white light image and the pixel points with image feature information on the fluorescence image.

7. The control system according to claim 6, characterized in that: The image processing module (30) comprises: an image data receiving module (40), electrically connected to the first camera unit (10) and the second camera unit (20), and configured to receive the white light image and the fluorescent image; An image feature extraction module (50) is electrically connected to the image data receiving module (40) and is used to extract image features from the white light image and the fluorescent image and obtain pixel points having image feature information; an information feature updating module (60) electrically connected to the image feature extraction module (50) for updating a calculation result of the image gradient function in response to a second control instruction and transmitting the calculation result to the image feature extraction module (50); An image registration module (70) is electrically connected to the image feature extraction module (50) and is used to register the pixel points having image feature information on the white light image with the pixel points having image feature information on the fluorescence image in response to a first control instruction to obtain the target image; and an FPGA chip unit (80), electrically connected to the image feature extraction module (50), the information feature update module (60), and the image registration module (70), for generating a first control instruction based on pixel points with image feature information acquired by the image feature extraction module (50), and generating a second control instruction based on pixel points with no image feature information acquired by the image feature extraction module (50).

8. The control system according to claim 7, characterized in that: The system further includes a light source module (100) electrically connected to the first camera unit (10) and the second camera unit (20) for providing light source energy to the first camera unit (10) and the second camera unit (20); the image processing module (30) further includes an image quality assessment module (90) electrically connected to the image registration module (70) and the FPGA chip unit (80) for assessing the image quality of the target image; the FPGA chip unit (80) generates a third control instruction for adjusting the light source intensity of the light source module (100) based on the image quality assessment result of the target image.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to implement the method according to any one of claims 1 to 5 when executing the instructions stored in the memory.

10. A non-volatile computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images