Blood vessel feature extraction method and device and blood vessel imaging instrument

Abstract

The embodiment of the present application provides a kind of blood vessel feature extraction method, device and blood vessel imaging appearance, wherein the method is applied to the processing module of blood vessel imaging appearance, blood vessel imaging appearance includes processing module and the optical system connected with processing module, method includes: obtaining the blood vessel original image information and the blood vessel feature information of patient that blood vessel imaging appearance is collected;Determine convolution window based on the optical system parameters of blood vessel imaging appearance and blood vessel feature information;Different direction blood vessel distribution characteristics in original image information are extracted based on convolution window;The blood vessel feature of blood vessel original image is obtained by accumulating the blood vessel distribution characteristics of different directions.It can be based on the method of windowed fourier transform to extract venous blood vessel original image to obtain more blood vessel information, and the feature image of different directions can be convolved using image discretization method to achieve the purpose of time division multiplexing, which can reduce the use of memory and improve processing efficiency.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, specifically to a method, apparatus, and vascular imaging instrument for extracting vascular features. Background Technology

[0002] Existing methods for segmenting vein images all employ directional valley-shaped images to identify the acquired veins. Because the grayscale level of the vein region is low, while the background grayscale level is relatively high, it exhibits a valley-shaped characteristic in the longitudinal section. Therefore, the valley-shaped areas with lower grayscale levels in the original image can be extracted as vein regions. However, in practical applications, low-quality vein images are often acquired during actual injection punctures, often exhibiting low contrast, abrupt changes in local illumination, and uneven overall brightness. In such cases, the valley-shaped feature is almost indistinct, and accurate extraction of the vein image is not possible.

[0003] Because the process of acquiring vein images using a vein acquisition device is affected by various environmental factors and the subject's own characteristics (such as differences in blood vessel thickness, blood flow velocity, uneven skin thickness distribution, and skin color), low-quality vein images are often acquired during actual injection punctures. These images typically exhibit low contrast, abrupt changes in local illumination, and uneven overall brightness. In such raw images, the valley-shaped feature is almost indistinct. Therefore, basic methods based on the principle of directional valley-shaped detection suffer from the technical problem of difficulty in extracting blood vessels, leading to missed vascular imaging and insufficient vascular reconstruction. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method, apparatus and vascular imaging instrument for extracting vascular features, so as to solve the above-mentioned technical problems.

[0005] In a first aspect, embodiments of the present invention provide a method for extracting vascular features. This method is applied to the processing module of a vascular imaging instrument, which includes a processing module and an optical system connected to the processing module. The method includes: acquiring raw vascular image information collected by the vascular imaging instrument and vascular feature information of a patient; determining a convolution window based on the optical system parameters of the vascular imaging instrument and the vascular feature information; extracting vascular distribution characteristics in different directions from the raw image information based on the convolution window; and accumulating the vascular distribution characteristics in different directions to obtain the vascular features of the raw vascular image.

[0006] Furthermore, the steps for obtaining the patient's vascular feature information include: pre-collecting the patient's vascular feature information using ultrasound equipment; wherein the vascular feature information includes: vascular diameter and vascular direction; associating and storing the patient's information with the vascular feature information, and generating a corresponding vascular feature information table; searching the vascular feature information table to obtain the patient's corresponding vascular feature information.

[0007] Furthermore, the step of determining the convolution window based on the optical system parameters and vascular feature information of the vascular imaging instrument includes: acquiring the optical system parameters of the vascular imaging instrument; wherein the optical system parameters include the lens focal length of the vascular imaging instrument; determining the number of pixels in the vascular image based on the lens focal length, the vascular diameter and pixel size in the vascular feature information, and a first conversion formula; wherein the first conversion formula is expressed as follows:

[0008] N = fc × H C / D c / X

[0009] Where N is the number of pixels, fc is the lens focal length, and h c For pixel size, D c H represents the physical distance. C X represents the diameter of the blood vessel; X represents the pixel size.

[0010] The convolution window is determined based on the number of pixels and the second transformation formula.

[0011] Furthermore, the second conversion formula is expressed as follows:

[0012] The side length of the convolution window = 2N+1

[0013] Where N is the number of pixels.

[0014] Furthermore, the step of extracting the blood vessel distribution characteristics in different directions from the original image information based on the convolution window includes: obtaining the number of blood vessel directions pre-analyzed; determining the convolution sine function based on the number of blood vessel directions; and performing convolution processing on the original image information within the convolution window based on the convolution sine function to obtain the blood vessel distribution characteristics in different directions.

[0015] Furthermore, the step of performing convolution processing on the original image information within the convolution window includes: performing cross-clock domain discretization processing on the original image to output the original image as multiple discrete image data streams; wherein the number of image data streams is the same as the number of blood vessel directions; convolving the image data streams based on the convolution sine function to obtain multiple different convolution values; and summing the convolution values ​​to make the image data streams continuous.

[0016] Furthermore, before performing convolution processing on the original image information within the convolution window based on the convolution sine function, the method also includes: normalizing different original image information to ensure that the original image information does not exceed the range of the convolution window.

[0017] Secondly, embodiments of the present invention also provide a vascular feature extraction device, wherein a processing module is applied to a vascular imaging instrument. The vascular imaging instrument includes a processing module and an optical system connected to the processing module, for performing any of the methods described above; including: an acquisition module for acquiring raw vascular image information collected by the vascular imaging instrument and vascular feature information of a patient; a determination module for determining a convolution window based on the optical system parameters of the vascular imaging instrument and the vascular feature information; an extraction module for extracting vascular distribution characteristics in different directions from the raw image information based on the convolution window; and an accumulation module for accumulating the vascular distribution characteristics in different directions to obtain the vascular features of the raw vascular image.

[0018] In a third aspect, embodiments of the present invention also provide a vascular imaging device, comprising a processing module and an optical system connected to the processing module; wherein the processing module is configured with the vascular feature extraction device described above, for performing any of the methods described above; and the optical system is used to acquire raw vascular image information of a patient.

[0019] Furthermore, the optical system includes: a camera module and a light-emitting module; wherein the camera module is connected to the processing module; the light-emitting module is used to emit infrared light to the patient's body; the camera module is used to receive the infrared light reflected from the patient's body, and to collect the reflected infrared light signal as the patient's original vascular image information and send the original vascular image information to the processing module.

[0020] The embodiments of the present invention bring the following beneficial effects:

[0021] This invention provides a method for extracting vascular features. The method is applied to the processing module of a vascular imaging instrument, which includes a processing module and an optical system connected to the processing module. The method includes: acquiring raw vascular image information and patient vascular feature information collected by the vascular imaging instrument; determining a convolution window based on the optical system parameters of the vascular imaging instrument and the vascular feature information; extracting vascular distribution characteristics in different directions from the raw image information based on the convolution window; and accumulating the vascular distribution characteristics in different directions to obtain the vascular features of the raw vascular image. This method can acquire the raw venous image of a patient, determine the size of the venous feature window based on the patient's vascular thickness information, analyze the distribution characteristics of the acquired raw venous image to obtain the feature coefficients of a certain direction, convolve them with a sine function, design Gabor filters in multiple directions for vascular feature matching, and finally process the data to obtain the vascular image. This method can extract more vascular information from the raw venous image based on windowed Fourier transform, and can use image discretization to convolve the feature images in different directions to achieve time-division multiplexing, reducing memory usage and improving processing efficiency.

[0022] Other features and advantages of the invention will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the techniques described above.

[0023] To make the above-mentioned objects, features and advantages of this disclosure more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0024] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0025] Figure 1 A flowchart of a method for extracting blood vessel features provided by the present invention;

[0026] Figure 2 This is a schematic diagram of an imaging device for a blood vessel imaging system provided by the present invention;

[0027] Figure 3 A flowchart of another method for extracting blood vessel features provided by the present invention;

[0028] Figure 4 A schematic diagram of a raw blood vessel image provided by the present invention;

[0029] Figure 5 A schematic diagram of an extracted blood vessel image provided by the present invention;

[0030] Figure 6 This is a schematic diagram of the structure of a vascular feature extraction device provided by the present invention;

[0031] Figure 7 This is a schematic diagram of the structure of a vascular imaging device provided by the present invention;

[0032] Figure 8 This is a practical application diagram of a vascular imaging device provided by the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] Currently, most vein segmentation algorithms employ directional valley detection or optimized versions thereof. The basic principle of directional valley detection is based on the analysis of the grayscale values ​​of veins. It is known that vein regions have lower grayscale values, while the background grayscale is relatively higher, and the longitudinal section generally exhibits a valley-shaped characteristic. Veins are located within these valleys. Since several valleys appear in the longitudinal section of the image, and the locations of these valleys are precisely where the veins are situated, extracting the vein region can be transformed into extracting the valley-shaped regions in the image.

[0035] Because the process of acquiring vein images using a vein acquisition device is affected by various environmental factors and the subject's own characteristics (such as differences in blood vessel thickness, blood flow velocity, uneven skin thickness distribution, and skin color), low-quality vein images are often acquired during actual injection punctures. These images typically exhibit low contrast, abrupt changes in local illumination, and uneven overall brightness. In such raw images, the valley-shaped features are almost indistinct. Therefore, basic methods based on the principle of directional valley-shaped detection are difficult to extract blood vessels, leading to missed vascular imaging and insufficient vascular reconstruction.

[0036] Based on this, embodiments of the present invention provide a method for extracting blood vessel features. Figure 1 A flowchart of a method for extracting vascular features is shown, wherein the method is applied to the processing module of a vascular imaging instrument, which includes a processing module and an optical system connected to the processing module, such as... Figure 1 As shown, the method includes:

[0037] Step S101: Obtain the original vascular image information and the patient's vascular feature information acquired by the vascular imaging instrument;

[0038] Specifically, the process of obtaining a patient's vascular characteristic information can be achieved through the following steps A1-A3, including:

[0039] Step A1: Collect vascular feature information of the patient in advance using ultrasound equipment; the vascular feature information includes: vascular diameter and vascular direction;

[0040] Step A2: Associate and store the patient's information with the vascular feature information, and generate the corresponding vascular feature information table;

[0041] Step A3: Locate the vascular feature information table to obtain the patient's corresponding vascular feature information.

[0042] In practical applications, we can cooperate with hospitals in advance to obtain the patient's basic information, and collect a series of information about the patient through ultrasound equipment, and obtain information on the thickness of blood vessels. According to the statistical data, the blood vessels of adult women are about 3 mm and the blood vessels of adult men are about 4 mm.

[0043] Step S102: Determine the convolution window based on the optical system parameters and vascular feature information of the vascular imaging instrument;

[0044] Specifically, the process of determining the convolution window based on the optical system parameters and vascular feature information of the angiography system can be implemented by the following steps B1-B3, including:

[0045] Step B1: Obtain the optical system parameters of the angiography system; wherein, the optical system parameters include the lens focal length of the angiography system.

[0046] Step B2: Determine the number of pixels in the blood vessel image based on the lens focal length, blood vessel diameter and pixel size from the blood vessel feature information, and the first conversion formula; wherein, the first conversion formula is expressed as follows:

[0047] N = fc × H C / D c / X

[0048] Where N is the number of pixels, fc is the lens focal length, and D c H represents the physical distance. C X represents the diameter of the blood vessel; X represents the pixel size.

[0049] Step B3: Determine the convolution window based on the number of pixels and the second transformation formula.

[0050] Specifically, the second conversion formula mentioned above is expressed as follows:

[0051] The side length of the convolution window = 2N+1

[0052] Where N is the number of pixels.

[0053] In practical applications, a conversion formula between world coordinate size and pixel count can be established based on the optical system parameters of the angiography system to obtain its filter window size. Specifically, for example... Figure 2 The schematic diagram of an angiography system shows that the formula for the lens focal length and distance is:

[0054]

[0055] Where fc is the lens focal length, and D c H represents the physical distance. C This refers to the diameter of the blood vessel.

[0056] The formula for pixel size versus world physical size can be obtained through conversion:

[0057]

[0058] Then, using the formula pixel size = number of pixels × pixel size, we can obtain the final conversion formula:

[0059] N = fc × H C / D c / X

[0060] Specifically, in practical applications, the optical parameters of the vascular imaging instrument are fc = 6mm, Hc is about 3mm-4mm, the pixel size is 6um, and the Dc distance is 26cm. Therefore, the number of pixels is about 7-9. If the middle value of 8 is selected, the size of the filtering window is 8×2+1=17, that is, the filtering coefficient is a two-dimensional 17x17 matrix.

[0061] Step S103: Extract the blood vessel distribution characteristics in different directions from the original image information based on the convolution window;

[0062] In practical applications, the high-speed I / O performance and parallel processing capabilities of FPGAs can be leveraged to facilitate real-time image acquisition and computation. Furthermore, the field-programmable nature of FPGAs allows for different algorithm optimizations based on the specific imaging conditions, providing significant flexibility.

[0063] Therefore, in practical applications, a near-infrared image acquisition system can be built based on an FPGA platform to store raw near-infrared images. Under the influence of near-infrared light, the FPGA reads the light signal data image captured by the sensor and returned by human skin through absorption, scattering, and reflection—that is, the raw near-infrared data—and caches it in an external SDRAM memory chip before saving the image to a flash chip. Finally, the system can interact with the FPGA's Nios2 CPU via host computer software, communicating through the UART protocol to upload the image data to a PC.

[0064] Step S104: Accumulate the vascular distribution characteristics in different directions to obtain the vascular features of the original vascular image.

[0065] This invention provides a method for extracting vascular features. The method is applied to the processing module of a vascular imaging instrument, which includes a processing module and an optical system connected to the processing module. The method includes: acquiring raw vascular image information and vascular feature information of the patient obtained by the vascular imaging instrument; determining a convolution window based on the optical system parameters of the vascular imaging instrument and the vascular feature information; extracting vascular distribution characteristics in different directions from the raw image information based on the convolution window; and accumulating the vascular distribution characteristics in different directions to obtain the vascular features of the raw vascular image. This method can acquire raw images of the patient's veins, determine the size of the venous feature window based on the patient's vascular thickness information, analyze the distribution characteristics of the acquired raw images to obtain feature coefficients for a certain direction, convolve these coefficients with a sine function, design Gabor filters in multiple directions for vascular feature matching, and finally process the data to obtain the vascular image. The original image of veins can be extracted based on the windowed Fourier transform method, which can obtain more vascular information. Furthermore, the image discretization method can be used to convolve the feature images in different directions to achieve time-division multiplexing, which can reduce memory usage and improve processing efficiency.

[0066] Based on the above embodiments, Figure 3 A flowchart of another method for extracting vascular features is shown, such as... Figure 3 As shown, the method specifically includes the following steps:

[0067] Step S301: Obtain the original vascular image information acquired by the vascular imaging instrument and the patient's vascular feature information;

[0068] Step S302: Determine the convolution window based on the optical system parameters and vascular feature information of the vascular imaging instrument;

[0069] Step S303: Obtain the number of blood vessel directions obtained from the pre-statistical analysis;

[0070] Specifically, statistical analysis reveals that there are approximately six directions of blood vessel orientation. By convolving the sine function with a rotation angle of 30°, the Gabor filter coefficients for each direction are obtained. Then, the near-infrared blood vessel image is convolved in all six directions, and the results are summed to obtain the final result.

[0071] Step S304: Determine the convolution sine function based on the number of blood vessel directions;

[0072] Step S305: Perform convolution processing on the original image information within the convolution window based on the convolution sine function to obtain the blood vessel distribution characteristics in different directions.

[0073] Specifically, the process of convolving the original image information within the convolution window can be implemented by the following steps C1-C3, including:

[0074] Step C1: Perform cross-clock domain discretization processing on the original image, and output the original image as multiple discrete image data streams; wherein the number of image data streams is the same as the number of blood vessel directions;

[0075] Step C2 involves convolving the image data stream using a convolution sine function to obtain multiple different convolution values.

[0076] Step C3 sums the convolution values ​​to make the image data stream continuous.

[0077] In practical applications, the convolution window size is NxN. The following explanation uses N=17 as an example, but it is not limited to 17. It only needs to be an odd number and greater than 3.

[0078] Here, time-division multiplexing can be achieved by combining image discretization with a method to improve video stream clocking. This not only trades speed for area (using a single convolutional module to perform six convolutions), but also keeps the latency in the nanosecond range, without requiring additional memory storage overhead.

[0079] Specifically, the angiography sensor input in this embodiment uses a 24MHz on-path clock, with 845 cycles per line and 752 effective data pixels. However, in practical applications, the maximum usable width of the target area is only 320 pixels. Therefore, the time cycle of one line is 35.21ns. When using a 120MHz clock to process the 320 pixels of the target, it only takes 2.67ns. If this is repeated 6 times, the accumulated time is still less than the cycle of one line. Therefore, it is entirely possible to complete 6 processing cycles by increasing the video stream clock and utilizing the idle time of ineffective data, thus not wasting idle time, keeping the latency in the nanosecond range, and achieving time-division multiplexing, trading speed for resources, and reducing redundant overhead of DSP and other resources.

[0080] First, the `video_bridge_s2t` module can be used. This module uses two RAM ping-pong operations for cross-clock domain discretization processing, i.e., a 24M continuous data stream input and a 120M discrete data stream output (groups of 6). The image width changes from `width` to `width*6`, `o_de_flag` represents the start flag of each group, and the `Data_ctrl_17x17` module is used to generate image data of the size of the filtered window. Sixteen RAM blocks are used for row buffering, simultaneously outputting 17 rows of data. Each row of data is delayed by 16 clock cycles using the DFF register. The `de_flag` signal output from the previous module controls read / write enable, ultimately resulting in a 17x17 window of image data, which remains synchronously unchanged for 6 cycles. For example, the image showing the window data corresponding to each point is horizontally copied 6 times.

[0081] Subsequently, the filter_gauss_17x17 module can be used to expand a single convolution operation in a certain direction into a matrix during parallel processing of the FPGA, as follows:

[0082]

[0083] The convolution value is obtained by multiplying and adding the data within the window. The parallel processing of the FPGA allows the counting result to be completed within a few cycles. Simultaneously, counting begins based on the `de_flag` signal, and six fixed Gabor filter coefficients obtained by switching between six directions are used to pipeline six convolution processes. The pipelined convolution results are retained, using `de_flag` as a marker; six values ​​represent a group, and these six data points are summed to obtain the extraction result for that point.

[0084] Finally, the results obtained from the video_bridge_t2s module, with blood vessel results for one direction available every 6 cycles, are stored in RAM and then read all at once, completing the image discretization to continuous processing conversion, resulting in the extracted blood vessel map, as shown below. Figures 4-5 As shown, where, Figure 4 This is a schematic diagram of a raw image of a blood vessel. Figure 5 This is a schematic diagram of an extracted blood vessel image.

[0085] Step S306: Accumulate the vascular distribution characteristics in different directions to obtain the vascular features of the original vascular image.

[0086] In practical applications, before performing convolution processing on the original image information within the convolution window based on the convolution sine function, the method also includes:

[0087] Different original image information is normalized to ensure that the original image information does not exceed the range of the convolution window.

[0088] Specifically, since the vascular imager can utilize the characteristic that hemoglobin in human blood absorbs near-infrared light more strongly than other tissues to image blood vessels, specifically in imaging, the imaging at the vein will be darker than other tissues. By selecting and analyzing the distribution characteristics of multiple near-infrared original images in a 17x17 window, it can be obtained that its distribution conforms to a two-dimensional normal distribution curve. Select an appropriate standard deviation σ to obtain the weight distribution. In order to ensure that the value after weighted summation does not exceed the original maximum range of the pixel value, the above formula needs to be normalized to make 0 < w(i,j) < 1. As shown in the following formula:

[0089]

[0090]

[0091]

[0092] Corresponding to the above method embodiments, the embodiments of the present invention provide a vascular feature extraction device, as Figure 6 shown. This device is applied to the processing module of the vascular imager, as Figure 7 shown in a schematic structural diagram of a vascular imager. The vascular imager 700 includes a processing module 701 and an optical system 702 connected to the processing module, which is used to execute any one of the above methods; it includes:

[0093] An acquisition module 601, configured to acquire the vascular original image information collected by the vascular imager and the vascular feature information of the patient;

[0094] A determination module 602, configured to determine a convolution window based on the optical system parameters of the vascular imager and the vascular feature information;

[0095] An extraction module 603, configured to extract the vascular distribution characteristics in different directions from the original image information based on the convolution window;

[0096] An accumulation module 604, configured to accumulate the vascular distribution characteristics in different directions to obtain the vascular features of the vascular original image.

[0097] Specifically, the above optical system is used to collect the vascular original image information of the patient.

[0098] Among them, the optical system 702 includes: a camera module 721 and a light-emitting module 722; among them, the camera module 721 is connected to the processing module 701;

[0099] The light-emitting module is configured to emit infrared light to the patient's body;

[0100] The camera module is used to receive infrared light reflected from the patient's body and collect the reflected infrared light signal as the patient's raw vascular image information and send the raw vascular image information to the processing module.

[0101] Figure 8 This diagram illustrates a practical application of a blood vessel imaging system. In practical applications, the aforementioned processing module can employ methods such as... Figure 8 The sensor board shown includes a camera module with a lens, a light-emitting module with an optical engine and LEDs for light emission, and a dichroic mirror and a light-diffusing sheet for color adjustment.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

[0103] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0104] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0105] Finally, it should be noted that the above embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method of extracting a blood vessel feature, characterized by, The application discloses a processing module applied to a blood vessel imaging instrument, and relates to the technical field of blood vessel imaging instruments. The method comprises the following steps: acquiring blood vessel original image information collected by a blood vessel imaging instrument and blood vessel feature information of a patient; determining a convolution window based on optical system parameters of the blood vessel imaging instrument and the blood vessel feature information; extracting blood vessel distribution characteristics in different directions from the blood vessel original image information based on the convolution window; accumulating the blood vessel distribution characteristics in different directions to obtain blood vessel features of the blood vessel original image; wherein the step of determining the convolution window based on the optical system parameters of the blood vessel imaging instrument and the blood vessel feature information comprises: acquiring optical system parameters of the blood vessel imaging instrument; wherein the optical system parameters comprise a lens focal length of the blood vessel imaging instrument; N= / / X Wherein, N is the number of said pixels, fc is the focal length of said lens is the physical distance, is the diameter of said blood vessel; X is the pixel size; determining the number of image elements of the blood vessel original image based on the lens focal length, a blood vessel diameter in the blood vessel feature information, an image element size and a first conversion formula; wherein the first conversion formula is expressed as the following formula: determining the convolution window based on the number of image elements and a second conversion formula; wherein the second conversion formula is expressed as the following formula: the side length of the convolution window = 2N+1; 2. The method of claim 1, wherein, wherein N is the number of image elements. The step of acquiring the blood vessel feature information of the patient comprises: previously collecting the blood vessel feature information of the patient by using an ultrasonic device; wherein the blood vessel feature information comprises a blood vessel diameter and a blood vessel direction; storing the patient information of the patient and the blood vessel feature information in association and generating a corresponding blood vessel feature information table; 3. The method of claim 1, wherein, looking up the blood vessel feature information table to acquire the blood vessel feature information corresponding to the patient. The step of extracting the blood vessel distribution characteristics in different directions from the original image information based on the convolution window comprises: acquiring the number of blood vessel directions previously statistically analyzed; determining a convolution sine function based on the number of blood vessel directions; 4. The method of claim 3, wherein, performing convolution processing on the original image information in the convolution window based on the convolution sine function to obtain the blood vessel distribution characteristics in different directions. The step of performing convolution processing on the original image information in the convolution window comprises: performing cross-clock domain discretization processing on the original image to output the blood vessel original image as a plurality of discrete image data streams; wherein the number of the image data streams is the same as the number of blood vessel directions; performing convolution on the image data streams based on the convolution sine function and obtaining a plurality of different convolution values; 5. The method of claim 3, wherein, summing the convolution values to continuously process the image data streams. Before performing the convolution processing on the original image information in the convolution window based on the convolution sine function, the method further comprises:

6. A blood vessel feature extraction apparatus characterized by comprising: performing normalization processing on different original image information to ensure that the original image information does not exceed the range of the convolution window. The application discloses a processing module applied to a blood vessel imaging instrument, and relates to the technical field of blood vessel imaging instruments. The method comprises the following steps: acquiring blood vessel original image information collected by a blood vessel imaging instrument and blood vessel feature information of a patient; A determining module is configured to determine a convolution window based on optical system parameters of the blood vessel imaging device and the blood vessel feature information; An extracting module is configured to extract blood vessel distribution characteristics in different directions in the original image information based on the convolution window; An accumulating module is configured to accumulate the blood vessel distribution characteristics in different directions to obtain blood vessel features of the blood vessel original image.

7. An angiographic imaging system, characterized by The blood vessel feature extraction device comprises a processing module and an optical system connected to the processing module; The processing module is configured with the blood vessel feature extraction device in claim 6 to execute the method in any one of claims 1-5; The optical system is configured to collect blood vessel original image information of a patient.

8. The blood vessel visualizer of claim 7, wherein, The optical system comprises a camera module and a light-emitting module; the camera module is connected to the processing module; The light-emitting module is configured to emit infrared light to a patient's body; The camera module is configured to receive infrared light reflected by the patient's body, collect the reflected infrared light signal, and send the blood vessel original image information of the patient to the processing module.

Images