CN104173038A

CN104173038A - Frequency domain laser speckle imaging based blood flow velocity measuring method

Info

Publication number: CN104173038A
Application number: CN201410438659.5A
Authority: CN
Inventors: 童善保; 李皓; 刘祺; 卢洪阳; 李瑶
Original assignee: Shanghai Jiao Tong University
Current assignee: Shanghai Shenhua Smart Medical Technology Co ltd
Priority date: 2014-08-29
Filing date: 2014-08-29
Publication date: 2014-12-03
Anticipated expiration: 2034-08-29
Also published as: CN104173038B

Abstract

The invention discloses a blood flow velocity measurement method based on frequency-domain laser speckle imaging, which comprises the following steps: irradiating a laser beam on a measured object, then using an imaging system to image the measured object, and collecting the measured object through an image sensor. The original speckle image of the object, and then convert the dynamic speckle intensity of a single pixel in the time domain in the collected original speckle image to the frequency domain, and calculate the power spectral density, and perform polynomial fitting on the power spectral density to obtain a smooth Then transform the smooth curve to the time domain through Fourier transform, calculate the autocovariance function of the pixel points and perform normalization, then establish the measurement model of blood flow velocity, and obtain the relationship between the autocovariance function and the blood flow velocity Finally, the blood flow velocity value is obtained by fitting; the invention not only eliminates static noise, improves the measurement accuracy of blood flow velocity, but also avoids the influence of imaging environmental factors such as light source intensity, irradiation angle, etc., and improves measurement stability. sex.

Description

Measurement method of blood flow velocity based on frequency domain laser speckle imaging

技术领域technical field

本发明涉及生物组织血流成像领域，具体涉及一种基于频域激光散斑成像的血流速度测量方法。The invention relates to the field of biological tissue blood flow imaging, in particular to a blood flow velocity measurement method based on frequency-domain laser speckle imaging.

背景技术Background technique

激光散斑衬比成像是一种利用光学成像系统传输激光散斑图像以实现对全场血流进行成像的技术。激光散斑衬比成像系统主要由相干光源和图像采集设备组成，相干光经生物组织散射后随机叠加，并通过图像采集设备采集随机散斑图样进行空间衬比分析，最终估算出血流速度。激光散斑衬比成像方法具有操作简单、实用性强等特点，因此在生物医学研究、临床诊断、外科引导、皮肤、牙科、眼科和神经科学等领域获得了广泛的应用。Laser speckle contrast imaging is a technology that uses an optical imaging system to transmit laser speckle images to image blood flow in the whole field. The laser speckle contrast imaging system is mainly composed of a coherent light source and an image acquisition device. The coherent light is randomly superimposed after being scattered by biological tissue, and the random speckle pattern is collected by the image acquisition device for spatial contrast analysis, and finally the blood flow velocity is estimated. The laser speckle contrast imaging method has the characteristics of simple operation and strong practicability, so it has been widely used in the fields of biomedical research, clinical diagnosis, surgical guidance, skin, dentistry, ophthalmology and neuroscience.

然而采用激光散斑衬比成像方法进行血流速度测量时，由于受环境条件(如光源强度、照射角度、成像物和相机参数)以及生物组织中的散射过程(如速度分布、静态散斑和多重散斑)的影响，使得已有的激光散斑衬比成像方法测得的血流速度存在较大误差。However, when laser speckle contrast imaging is used to measure blood flow velocity, due to environmental conditions (such as light source intensity, illumination angle, imaging object and camera parameters) and scattering processes in biological tissues (such as velocity distribution, static speckle and Due to the influence of multiple speckles), there is a large error in the blood flow velocity measured by the existing laser speckle contrast imaging method.

针对以上问题，Parthasarathy等人利用多曝光时间散斑衬比成像方法获得自相关函数来代替使用单个衬比值获得单点的自相关函数值，以减少静态散射带来的影响，同时可以获得速度分布类型等有效信息。之后，Thompson和Andrews又在此基础上指出，将自协方差曲线转换为多普勒频谱的形式，便可使用激光多普勒测定中的算法估算出血流速度。通过以上方法虽然已经在很大程度上提高了散斑衬比度值的有效性，但是由于散斑图像依然受多种成像环境因素影响，因此利用激光散斑衬比成像方法直接测量血流速度中存在的误差问题仍然没有得到有效解决。In response to the above problems, Parthasarathy et al. used the multi-exposure time speckle contrast imaging method to obtain the autocorrelation function instead of using a single contrast value to obtain the autocorrelation function value of a single point, so as to reduce the influence of static scattering and obtain the velocity distribution Valid information such as type. Later, Thompson and Andrews pointed out on this basis that by converting the autocovariance curve into the form of Doppler spectrum, the algorithm in laser Doppler measurement can be used to estimate the blood flow velocity. Although the effectiveness of the speckle contrast value has been greatly improved through the above methods, since the speckle image is still affected by various imaging environmental factors, the laser speckle contrast imaging method is used to directly measure the blood flow velocity The problem of errors in the system has not been effectively resolved.

发明内容Contents of the invention

本发明为了克服以上不足，提供一种可以消除环境因数影响，提高测量准确度的基于频域激光散斑成像的血流速度测量方法。In order to overcome the above shortcomings, the present invention provides a blood flow velocity measurement method based on frequency-domain laser speckle imaging that can eliminate the influence of environmental factors and improve measurement accuracy.

为了解决上述技术问题，本发明的技术方案是：一种基于频域激光散斑成像的血流速度测量方法，包括以下步骤：In order to solve the above technical problems, the technical solution of the present invention is: a blood flow velocity measurement method based on frequency-domain laser speckle imaging, comprising the following steps:

S1：将激光光束照射在被测物体上；S1: irradiate the laser beam on the object to be measured;

S2：利用成像系统对被测物体成像；S2: Using the imaging system to image the object under test;

S3：利用图像传感器采集被测物体的原始散斑图像；S3: using the image sensor to collect the original speckle image of the measured object;

S4：对原始散斑图像中的单个像素点进行计算，以获得单个像素点的自协方差函数，对单个像素点的自协方差函数进行归一化处理，包括以下步骤：S4: Calculate a single pixel in the original speckle image to obtain an autocovariance function of a single pixel, and normalize the autocovariance function of a single pixel, including the following steps:

S41：利用公式(Ⅰ)对采集的原始散斑图像中单个像素点(x,y)处于时域中的动态散斑强度进行傅里叶变换，转换到频域：S41: Use the formula (I) to perform Fourier transform on the dynamic speckle intensity of a single pixel point (x, y) in the time domain in the collected original speckle image, and convert it to the frequency domain:

$\overset{~ ~}{I I} ((ω ω)) = = \frac{11}{22 π π} &Integral; &Integral; I I ((t t)) {e e}^{- - iωt iωt} dt dt - - - - - - ((I I))$

其中I(t)表示时域中像素点(x,y)处光强序列，表示频域中像素点(x,y)处光强序列，x和y分别表示像素点的横坐标和纵坐标；Where I(t) represents the light intensity sequence at the pixel point (x, y) in the time domain, Represents the light intensity sequence at the pixel point (x, y) in the frequency domain, where x and y represent the abscissa and ordinate of the pixel, respectively;

S42：计算功率谱密度并对进行多项式拟合得到平滑曲线；S42: Calculate the power spectral density and to Perform polynomial fitting to obtain a smooth curve;

S43：利用公式(Ⅱ)将步骤S42中的平滑曲线通过傅里叶变换转换到时域，计算像素点(x,y)的自协方差函数：S43: Transform the smooth curve in step S42 into the time domain by Fourier transform using formula (II), and calculate the autocovariance function of the pixel point (x, y):

${C C}_{t t} ((τ τ)) = = {&Integral; &Integral; | | \overset{~ ~}{I I} ((ω ω)) | |}^{22} {e e}^{iωτ iωτ} dω dω - - {&lang; &lang; \overset{~ ~}{I I} ((ω ω)) &rang; &rang;}^{22} - - - - - - ((II II))$

其中τ代表时间间隔；where τ represents the time interval;

并对C_t(τ)进行归一化处理；And normalize C _t (τ);

S5：建立血流速度的测量模型，获得自协方差函数与绝对血流速度之间的关系：S5: Establish a measurement model of blood flow velocity, and obtain the relationship between the autocovariance function and the absolute blood flow velocity:

${C C}_{t t} ((τ τ)) = = {e e}^{- - \frac{{M m}^{22} {v v}_{00}^{22} {τ τ}^{22}}{{l l}_{00}^{22} + + {M m}^{22} {\overset{~ ~}{v v}}^{22} {τ τ}^{22}}} [[\frac{{l l}_{00}^{33}}{{(({l l}_{00}^{22} + + {M m}^{22} {\overset{~ ~}{v v}}^{22} {τ τ}^{22}))}^{\frac{33}{22}}} + + \frac{22 {M m}^{44} {v v}_{00}^{22} {\overset{~ ~}{v v}}^{22} {l l}_{00} {τ τ}^{44}}{{(({l l}_{00}^{22} + + {M m}^{22} {\overset{~ ~}{v v}}^{22} {τ τ}^{22}))}^{\frac{55}{22}}}]] - - - - - - ((III III))$

其中M是成像系统的放大倍数，其中τ代表时间间隔；l₀＝0.41Mλ/NA，λ代表照射光波长，NA是数值孔径，v₀是像素点(x,y)的平均速度，是像素点(x,y)的均方根速度。Where M is the magnification of the imaging system, where τ represents the time interval; l ₀ =0.41Mλ/NA, λ represents the wavelength of the illuminating light, NA is the numerical aperture, v ₀ is the average velocity of the pixel point (x, y), is the root mean square velocity of the pixel point (x, y).

S6：将步骤S43中经归一化处理的自协方差函数C_t(τ)代入公式(Ⅲ)进行拟合，得到像素点(x,y)的血流速度v₀；S6: Substituting the normalized autocovariance function C _t (τ) in step S43 into the formula (Ⅲ) for fitting to obtain the blood flow velocity v ₀ of the pixel point (x, y);

S7：对原始散斑图像中的每个像素点重复步骤S4-S6，对生物组织特定区域或病灶区域血流速度的动态监测和分析。S7: Repeating steps S4-S6 for each pixel in the original speckle image, dynamic monitoring and analysis of blood flow velocity in a specific area of biological tissue or lesion area.

进一步的，所述的成像系统为透镜组成像系统。Further, the imaging system is a lens group imaging system.

本发明提供的基于频域激光散斑成像的血流速度测量方法，首先将图像传感器采集的原始散斑图像转换到频域中进行处理，得到自协方差函数，并进行归一化处理；接着建立血流速度的测量模型，将归一化后的自协方差函数与血流速度联系起来，通过拟合得到最终的血流速度，通过此方法不仅消除了静态噪声，提高了血流速度的测量准确度，而且避免了成像环境因素如光源强度、照射角度等的影响，提高了测量稳定性。The blood flow velocity measurement method based on frequency-domain laser speckle imaging provided by the present invention first converts the original speckle image collected by the image sensor into the frequency domain for processing, obtains the autocovariance function, and performs normalization processing; then Establish a measurement model of blood flow velocity, link the normalized autocovariance function with blood flow velocity, and obtain the final blood flow velocity through fitting. This method not only eliminates static noise, but also improves the accuracy of blood flow velocity. The measurement accuracy is high, and the influence of imaging environmental factors such as light source intensity and irradiation angle is avoided, and the measurement stability is improved.

附图说明Description of drawings

图1是本发明提出的基于频域激光散斑成像的血流速度测量方法的流程图；Fig. 1 is the flowchart of the method for measuring blood flow velocity based on frequency-domain laser speckle imaging proposed by the present invention;

图2是本发明透镜组成像系统结构示意图；Fig. 2 is a schematic structural diagram of the imaging system of the lens group of the present invention;

图3a～3c是本发明方法和传统方法在流体模拟实验中的测量结果对比图；Fig. 3a～3c are the comparison diagrams of the measurement results of the method of the present invention and the traditional method in the fluid simulation experiment;

图4a～4c是本发明方法和传统方法对大鼠右耳血管内血流速度的测量结果对比图。Figures 4a to 4c are comparison diagrams of the measurement results of the blood flow velocity in the rat's right ear blood vessel by the method of the present invention and the traditional method.

图中所示：1、激光器；2、被测物体；3、透镜组；4、图像传感器；5、PC机。As shown in the figure: 1. Laser; 2. Object to be measured; 3. Lens group; 4. Image sensor; 5. PC.

具体实施方式Detailed ways

下面结合附图对本发明作详细描述：The present invention is described in detail below in conjunction with accompanying drawing:

如图1所示，本发明提供一种基于频域激光散斑成像的血流速度测量方法，包括以下步骤：As shown in Figure 1, the present invention provides a blood flow velocity measurement method based on frequency-domain laser speckle imaging, including the following steps:

其中τ代表时间间隔；where τ represents the time interval;

并对C_t(τ)进行归一化处理；And normalize C _t (τ);

假设t₀＝0时物平面上的点(x₀,y₀)处只有一个颗粒，运动速度为v，在激光光束的照射下，(x₀,y₀)相对应的像平面上的点(x,y)处散射的电场振幅分布可表示为δ(x-x₀-vτ,y-y₀)，而对于数值孔径固定的成像系统，点(x,y)处的电场振幅可表示为：Suppose there is only one particle at point (x ₀ , y ₀ ) on the object plane when t ₀ = 0, and the moving speed is v, under the irradiation of the laser beam, the point on the image plane corresponding to (x ₀ , y ₀ ) The distribution of electric field amplitude scattered at (x,y) can be expressed as δ(xx ₀ -vτ,yy ₀ ), and for an imaging system with fixed numerical aperture, the electric field amplitude at point (x,y) can be expressed as:

U(x,y)＝δ(x-Mx₀-Mvτ,y-My₀)*h(x,y)U(x,y)＝δ(x-Mx ₀ -Mvτ,y-My ₀ )*h(x,y)

(Ⅲ-1)(Ⅲ-1)

＝h(x-Mx₀-Mvτ,y-My₀)＝h(x-Mx ₀ -Mvτ,y-My ₀ )

其中，M是成像系统的放大倍数，h(x,y)是成像系统的光学传递函数，*为卷积运算，τ代表时间间隔。假设图像畸变可以忽略，则h(x,y)可以表示为系统的点传递函数：Among them, M is the magnification of the imaging system, h(x,y) is the optical transfer function of the imaging system, * is the convolution operation, and τ represents the time interval. Assuming that the image distortion can be ignored, h(x,y) can be expressed as the point transfer function of the system:

其中，J₁是一阶第一类贝塞尔函数，数，λ是照射光的波长，从上式中看出h(x,y)为艾里斑图样，半径r₀为：Among them, _J1 is the first-order Bessel function of the first kind, λ is the wavelength of the irradiated light. It can be seen from the above formula that h(x, y) is an Airy disk pattern, and the radius r ₀ is:

${r r}_{00} = = 0.61 0.61 \frac{Mλ Mλ}{NA NA} - - - - - - ((III III - - 33))$

假设成像的颗粒有单一的速度v，经过时间间隔τ，(x₀,y₀)相对应的像平面上的点(x₁,y₁)的自协方差函数C_t(v,τ)归一化处理后可表示为：Assuming that the imaged particle has a single velocity v, after a time interval τ, the autocovariance function C _t (v,τ) of the point (x ₁ ,y ₁ ) on the image plane corresponding to (x ₀ ,y ₀ ) is normalized to After normalization, it can be expressed as:

${C C}_{t t} ((v v,, τ τ)) = = &lang; &lang; \frac{| | h h {(({x x}_{11} - - M m {x x}_{00},, {y the y}_{11} - - M m {y the y}_{00}))}^{* *} h h (({x x}_{11} - - M m {x x}_{00} - - Mvτ Mvτ,, {y the y}_{11} - - M m {y the y}_{00})) | |}{{| | h h (({x x}_{11} - - M m {x x}_{00},, {y the y}_{11} - - M m {y the y}_{00})) | |}^{22}} &rang; &rang; - - - - - - ((III III - - 44))$

假定t₀＝0时物平面上任意初始位置上均有足够多的颗粒，运动速度都为v，则针对时间间隔τ，将每个初始位置上所有颗粒对应像平面内的自协方差的时间平均值可以转换为空间平均值：Assuming that there are enough particles at any initial position on the object plane when t ₀ =0, and the moving speed is v, then for the time interval τ, the time of all particles corresponding to the autocovariance in the image plane at each initial position The mean can be converted to a spatial mean:

${C C}_{t t} ((v v,, τ τ)) = = \frac{&Integral; &Integral; &Integral; &Integral; {| | h h {((x x,, y the y))}^{* *} h h ((x x - - Mvτ Mvτ,, y the y)) | |}^{22} dxdy dxdy}{&Integral; &Integral; {| | h h ((x x,, y the y)) | |}^{22} dxdy dxdy} - - - - - - ((III III - - 55))$

这里对整个x-y平面进行积分，x₁-Mx₀和y₁-My₀用x和y代替。Here the integral is performed over the entire xy plane, x ₁ -Mx ₀ and y ₁ -My ₀ are replaced by x and y.

根据式(Ⅲ-5)，C_t(v,τ)看做两个相距M_vτ的艾里斑的重叠区域。则C_t(v,τ)可以近似表示为如下高斯函数：According to formula (Ⅲ-5), C _t (v,τ) can be regarded as the overlapping area of two Airy disks with a distance of M _vτ . Then C _t (v,τ) can be approximately expressed as the following Gaussian function:

${C C}_{t t} ((v v,, τ τ)) = = {e e}^{\frac{- - {((Mvτ Mvτ))}^{22}}{{l l}_{00}^{22}}} - - - - - - ((III III - - 66))$

其中l₀＝Mvτ₀是去相关长度，代表两个艾里斑重叠区域是原始重叠区域的e^-1，其中原始重叠区域是指两个艾里斑图样完全重叠的区域。Where l ₀ =Mvτ ₀ is the decorrelation length, which means that the overlapping area of the two Airy disks is e ^-1 of the original overlapping area, where the original overlapping area refers to the area where the two Airy disk patterns completely overlap.

对于给定的速度分布P(v)，自协方差函数可以表示为：For a given velocity distribution P(v), the autocovariance function can be expressed as:

${C C}_{t t} ((τ τ)) = = {&Integral; &Integral;}_{00}^{+ + \infty \infty} P P ((v v)) {C C}_{t t} ((v v,, τ τ)) dv dv - - - - - - ((III III - - 77))$

血流速度分布被认为是布朗运动和高斯分布共同组成的。因此，假设血流速度分布P(v)由两种速度类型组成，可表示为:The distribution of blood flow velocity is considered to be composed of Brownian motion and Gaussian distribution. Therefore, it is assumed that the blood flow velocity distribution P(v) consists of two velocity types, which can be expressed as:

$P P ((v v)) = = \frac{22}{\sqrt{π π} {\overset{~ ~}{v v}}^{33}} {((v v - - {v v}_{00}))}^{22} {e e}^{- - \frac{{((v v - - {v v}_{00}))}^{22}}{{\overset{~ ~}{v v}}^{22}}} - - - - - - ((III III - - 88))$

由(Ⅲ-7)和(Ⅲ-8)式可以解得C_t(τ)的表达式为：From formulas (Ⅲ-7) and (Ⅲ-8), the expression of C _t (τ) can be solved as follows:

其中M是成像系统的放大倍数，l₀＝0.41Mλ/NA，λ代表光波长，NA是数值孔径，v₀是像素点(x,y)的平均速度，是像素点(x,y)的均方根速度。Where M is the magnification of the imaging system, l ₀ =0.41Mλ/NA, λ represents the wavelength of light, NA is the numerical aperture, v ₀ is the average velocity of the pixel point (x, y), is the root mean square velocity of the pixel point (x, y).

S6：将步骤S43中经归一化处理的自协方差函数C_t(τ)代入公式(Ⅲ)进行拟合，得到像素点(x,y)的血流速度v₀。S6: Substituting the normalized autocovariance function C _t (τ) in step S43 into the formula (III) for fitting to obtain the blood flow velocity v ₀ of the pixel point (x, y).

S7：对原始散斑图像中的每个像素点重复步骤S4-S8，对生物组织特定区域或病灶区域血流速度的动态监测和分析。S7: Repeating steps S4-S8 for each pixel in the original speckle image, dynamic monitoring and analysis of blood flow velocity in a specific area of biological tissue or lesion area.

如图2所示，搭建透镜组成像系统，激光器1发射激光照射于被测物体2上，散射激光经过透镜组3收集，并通过高帧速率的图像传感器4进行成像，最终将图像信号传入PC机5对血流速度进行监测和分析。As shown in Figure 2, a lens group imaging system is built. The laser 1 emits laser light and irradiates the measured object 2. The scattered laser light is collected by the lens group 3, and is imaged by the high frame rate image sensor 4, and finally the image signal is transmitted to the The PC 5 monitors and analyzes the blood flow velocity.

以下为流体模拟实验：The following is the fluid simulation experiment:

被测对象2为匀速流动的聚苯乙烯颗粒(0.1wt％,平均直径为3.2μm)，通过注射泵来控制聚苯乙烯颗粒以0.1mm/s,0.5mm/s,1mm/s的速度匀速流动，用780nm的激光器进行照射，然后通过固定在立体显微镜(NA＝0.2)上的高速图像传感器成像以1000帧每秒的速率采集原始激光散斑图像，并进行测量。The measured object 2 is polystyrene particles (0.1wt%, with an average diameter of 3.2μm) flowing at a constant speed, and the polystyrene particles are controlled by a syringe pump at a constant speed of 0.1mm/s, 0.5mm/s, and 1mm/s. flow, irradiated with a 780nm laser, and then imaged by a high-speed image sensor fixed on a stereo microscope (NA=0.2) to collect original laser speckle images at a rate of 1000 frames per second and measure them.

图像传感器采集的激光散斑图像如图3a所示，其中实线所选区域是采集原始散斑的区域；通过计算得到液体不同流速下所选定区域的速度分布如图3b所示，其中x轴0.0点处为所选区域x方向的中心，虚线是测得的散点值，实线是对虚线进行拟合后的抛物线曲线，由下至上分别代表流速为0.1mm/s,0.5mm/s,1mm/s的测量结果；测量得到的液体速度与实际液体速度之间的关系如图3c所示，其中散点值为实验值，直线代表理论值，从图中可以看出，对于不同流速的液体，采用本实施例的方法测得的结果与理论值非常接近，因此可以看出本实施例的方法具有较高的准确度。The laser speckle image collected by the image sensor is shown in Figure 3a, where the area selected by the solid line is the area where the original speckle was collected; the velocity distribution of the selected area under different flow rates of the liquid obtained through calculation is shown in Figure 3b, where x The 0.0 point of the axis is the center of the selected area in the x direction, the dotted line is the measured scatter point value, and the solid line is the parabolic curve after fitting the dotted line, representing the flow velocity of 0.1mm/s and 0.5mm/ s, 1mm/s measurement results; the relationship between the measured liquid velocity and the actual liquid velocity is shown in Figure 3c, where the scattered points are the experimental values, and the straight line represents the theoretical value. It can be seen from the figure that for different For the flow rate of the liquid, the result measured by the method of this embodiment is very close to the theoretical value, so it can be seen that the method of this embodiment has higher accuracy.

以下为动物活体的血流速度测量实验：The following is the blood flow velocity measurement experiment in living animals:

被测对象是一只Sprague Dawley大鼠，使用水合氯醛对其进行麻醉，并用脱毛膏除去大鼠右耳的毛，将其固定在立体显微镜下。将四个相同规格的780nm的激光器放置在四个角度不同的位置上(分别标记为角度1，2，3，4)，从不同的角度照射右耳，然后采用传统方法和本文方法分别对右耳血管内的血流速度进行测量。由于大鼠在短时间(5分钟)内处于稳定状态，因此在此期间，大鼠的血流速度处于稳定状态。为了进一步说明本发明的性能，采用传统方法和本发明方法两种方法进行测量。The test object was a Sprague Dawley rat, which was anesthetized with chloral hydrate, and the hair of the rat's right ear was removed with depilatory cream, and it was fixed under a stereomicroscope. Place four 780nm lasers of the same specification at four positions with different angles (respectively marked as angles 1, 2, 3, 4), irradiate the right ear from different angles, and then use the traditional method and the method in this paper to treat the right ear respectively. The blood flow velocity in the blood vessels of the ear is measured. Since the rat is in a steady state for a short period of time (5 minutes), the blood flow velocity of the rat is in a steady state during this period. In order to further illustrate the performance of the present invention, two methods, the traditional method and the method of the present invention, are used for measurement.

图像传感器采集的激光散斑图像如图4a所示，图中实线为两种方法采集原始散斑的区域。采用传统方法和本发明方法在不同测量角度下测得血管中心的血流速度值如图4b所示，其中左侧代表采用本发明方法测得的血流速度值，右侧代表采用传统方法测得的血流速度值，从图中可以看出采用传统方法进行测量时，在不同的测量角度下测得的血流速度偏差较大，说明照射条件的变化对传统方法的影响比较大。而针对不同的照射条件，通过本发明方法测得的血流速度只有很小的偏差，因此本发明方法具有更高的稳定性。图4c显示了两种方法在角度2和角度3下对同一采集区域测得的血流速度曲线，可以看出采用本发明的方法在两种角度下测得的值非常接近，而采用传统方法在两种角度下测得的值存在较大差距，说明本发明方法具有更好的稳定性。The laser speckle image collected by the image sensor is shown in Fig. 4a, and the solid line in the figure is the original speckle area collected by the two methods. The blood flow velocity values of the blood vessel center measured at different measurement angles using the traditional method and the method of the present invention are shown in Figure 4b. It can be seen from the figure that when the traditional method is used for measurement, the blood flow velocity measured at different measurement angles has a large deviation, indicating that changes in irradiation conditions have a greater impact on the traditional method. However, for different irradiation conditions, the blood flow velocity measured by the method of the present invention has only a small deviation, so the method of the present invention has higher stability. Figure 4c shows the blood flow velocity curves measured by the two methods at the same acquisition area at angle 2 and angle 3, it can be seen that the values measured by the method of the present invention are very close to the two angles, while the traditional method There is a large gap between the values measured under the two angles, which shows that the method of the present invention has better stability.

综上所述，本发明提供的基于频域激光散斑成像的血流速度测量方法，首先将图像传感器采集的原始散斑图像转换到频域中进行处理，得到自协方差函数，并进行归一化处理；接着建立血流速度的测量模型，将归一化后的自协方差函数与血流速度联系起来，通过拟合得到最终的血流速度，不仅消除了静态噪声，提高了血流速度的测量准确度，而且避免了成像环境因素如光源强度、照射角度等的影响，提高了测量稳定性。In summary, the blood flow velocity measurement method based on frequency-domain laser speckle imaging provided by the present invention first converts the original speckle image collected by the image sensor into the frequency domain for processing, obtains the autocovariance function, and performs regression Normalization processing; then establish a measurement model of blood flow velocity, link the normalized autocovariance function with blood flow velocity, and obtain the final blood flow velocity through fitting, which not only eliminates static noise, but also improves blood flow velocity. The measurement accuracy of the speed is high, and the influence of imaging environmental factors such as light source intensity and irradiation angle is avoided, and the measurement stability is improved.

虽然说明书中对本发明的实施方式进行了说明，但这些实施方式只是作为提示，不应限定本发明的保护范围。在不脱离本发明宗旨的范围内进行各种省略、置换和变更均应包含在本发明的保护范围内。Although the embodiments of the present invention have been described in the specification, these embodiments are only used as hints and should not limit the protection scope of the present invention. Various omissions, substitutions and changes within the scope not departing from the gist of the present invention shall be included in the protection scope of the present invention.

Claims

1. A blood flow velocity measurement method based on frequency-domain laser speckle imaging, characterized in that, comprising the following steps:

S1: irradiate the laser beam on the object to be measured;

S2: Using the imaging system to image the object under test;

S3: using the image sensor to collect the original speckle image of the measured object;

S4: Calculate a single pixel in the original speckle image to obtain an autocovariance function of a single pixel, and normalize the autocovariance function of a single pixel;

S5: Establish a measurement model of blood flow velocity, and obtain the relationship between the autocovariance function and blood flow velocity;

S6: Fitting the normalized autocovariance function and the blood flow velocity measurement model to obtain the blood flow velocity of a single pixel;

S7: Steps S4-S6 are repeated to obtain the blood flow velocity of each pixel in the original speckle image, and then dynamically monitor and analyze the blood flow velocity in a specific area of biological tissue or a lesion area.

2. The blood flow velocity measurement method based on frequency-domain laser speckle imaging according to claim 1, wherein the imaging system is a lens group imaging system.

3. The blood flow velocity measurement method based on frequency-domain laser speckle imaging according to claim 1, wherein the step S4 comprises:

S41: Use formula (1) to perform Fourier transform on the dynamic speckle intensity of a single pixel point (x, y) in the collected original speckle image in the time domain, and convert it to the frequency domain:

\overset{~ ~}{I I} ((ω ω)) = = \frac{11}{22 π π} &Integral; &Integral; I I ((t t)) {e e}^{- - iωt iωt} dt dt - - - - - - ((I I))

Where I(t) represents the light intensity sequence at the pixel point (x, y) in the time domain, Represents the light intensity sequence at the pixel point (x, y) in the frequency domain, where x and y represent the abscissa and ordinate of the pixel, respectively;

S42: Calculate the power spectral density and to Perform polynomial fitting to obtain a smooth curve;

S43: Transform the smooth curve in step S42 into the time domain by Fourier transform using formula (II), and calculate the autocovariance function of the pixel point (x, y):

{C C}_{t t} ((τ τ)) = = {&Integral; &Integral; | | \overset{~ ~}{I I} ((ω ω)) | |}^{22} {e e}^{iωτ iωτ} dω dω - - {&lang; &lang; \overset{~ ~}{I I} ((ω ω)) &rang; &rang;}^{22} - - - - - - ((II II))

where τ represents the time interval;

Normalize C _t (τ).

4. The blood flow velocity measurement method based on frequency-domain laser speckle imaging according to claim 3, characterized in that, in the step S5, the relationship between the autocovariance function and the absolute blood velocity is:

{C C}_{t t} ((τ τ)) = = {e e}^{- - \frac{{M m}^{22} {v v}_{00}^{22} {τ τ}^{22}}{{l l}_{00}^{22} + + {M m}^{22} {\overset{~ ~}{v v}}^{22} {τ τ}^{22}}} [[\frac{{l l}_{00}^{33}}{{(({l l}_{00}^{22} + + {M m}^{22} {\overset{~ ~}{v v}}^{22} {τ τ}^{22}))}^{\frac{33}{22}}} + + \frac{22 {M m}^{44} {v v}_{00}^{22} {\overset{~ ~}{v v}}^{22} {l l}_{00} {τ τ}^{44}}{{(({l l}_{00}^{22} + + {M m}^{22} {\overset{~ ~}{v v}}^{22} {τ τ}^{22}))}^{\frac{55}{22}}}]] - - - - - - ((III III))

Where M is the magnification of the imaging system, τ represents the time interval, l ₀ =0.41Mλ/NA, λ is the wavelength of the illuminating light, NA is the fixed numerical aperture of the imaging system, v ₀ is the average velocity of the pixel (x, y) , is the root mean square velocity of the pixel point (x, y).

5. The method for measuring blood flow velocity based on frequency-domain laser speckle imaging according to claim 4, wherein the normalized autocovariance function C _t (τ) in step S43 is substituted into formula (Ⅲ) Fitting is performed to obtain the blood flow velocity v ₀ of the pixel point (x, y).