WO2020191896A1 - 一种树木纵截面内部缺陷成像方法 - Google Patents

一种树木纵截面内部缺陷成像方法 Download PDF

Info

Publication number
WO2020191896A1
WO2020191896A1 PCT/CN2019/087022 CN2019087022W WO2020191896A1 WO 2020191896 A1 WO2020191896 A1 WO 2020191896A1 CN 2019087022 W CN2019087022 W CN 2019087022W WO 2020191896 A1 WO2020191896 A1 WO 2020191896A1
Authority
WO
WIPO (PCT)
Prior art keywords
grid unit
value
velocity
imaging plane
tree
Prior art date
Application number
PCT/CN2019/087022
Other languages
English (en)
French (fr)
Inventor
李光辉
刘雷
Original Assignee
江南大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 江南大学 filed Critical 江南大学
Priority to US17/005,394 priority Critical patent/US20200393347A1/en
Publication of WO2020191896A1 publication Critical patent/WO2020191896A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N3/00—Investigating strength properties of solid materials by application of mechanical stress
    • G01N3/02—Details
    • G01N3/06—Special adaptations of indicating or recording means
    • G01N3/062—Special adaptations of indicating or recording means with mechanical indicating or recording means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/003—Reconstruction from projections, e.g. tomography
    • G06T11/006—Inverse problem, transformation from projection-space into object-space, e.g. transform methods, back-projection, algebraic methods
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/04—Analysing solids
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/04—Analysing solids
    • G01N29/06—Visualisation of the interior, e.g. acoustic microscopy
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/04—Analysing solids
    • G01N29/07—Analysing solids by measuring propagation velocity or propagation time of acoustic waves
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/40—Filling a planar surface by adding surface attributes, e.g. colour or texture
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
    • G01H5/00—Measuring propagation velocity of ultrasonic, sonic or infrasonic waves, e.g. of pressure waves
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2203/00—Investigating strength properties of solid materials by application of mechanical stress
    • G01N2203/003—Generation of the force
    • G01N2203/0032—Generation of the force using mechanical means
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2203/00—Investigating strength properties of solid materials by application of mechanical stress
    • G01N2203/0058—Kind of property studied
    • G01N2203/006—Crack, flaws, fracture or rupture
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V2210/00—Details of seismic processing or analysis
    • G01V2210/60—Analysis
    • G01V2210/67—Wave propagation modeling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2211/00—Image generation
    • G06T2211/40—Computed tomography
    • G06T2211/424—Iterative

Definitions

  • the invention relates to a method for imaging internal defects in longitudinal sections of trees, belonging to the field of nondestructive inspection of trees.
  • Non-destructive testing also known as non-destructive testing, is the use of different physical and mechanical properties or chemical properties of the material, without destroying the internal and appearance structure and characteristics of the target object, to determine the object-related characteristics (such as shape, displacement, stress, optical characteristics, Fluid properties, mechanical properties, etc.) for testing and inspection, especially the measurement of various defects.
  • object-related characteristics such as shape, displacement, stress, optical characteristics, Fluid properties, mechanical properties, etc.
  • Stress waves refer to elastic mechanical waves that can propagate inside the object under the action of stress after the object is impacted.
  • stress waves were first applied to the detection of properties and defects of rock, soil, concrete, etc., and then forestry scientists applied it to the field of non-destructive testing of trees.
  • the present invention provides a method for imaging the internal defects of the longitudinal section of trees, the method includes:
  • SS1 establishes the corresponding imaging plane based on the measured tree data, divides the imaging plane into grid units of the same size, and assigns each grid unit an initial velocity value; calculates the reference for the propagation velocity of stress waves in various directions inside healthy trees Value, and then get the healthy reference speed value v of each grid unit of the imaging plane;
  • the linear propagation model is used to simulate the propagation of stress waves inside the trees, and the combined iterative reconstruction technology SIRT algorithm is used to adjust the velocity of the imaging plane grid unit;
  • the maximum and minimum velocity values and the fuzzy constraint mechanism based on the grid unit group are used to constrain the velocity of the imaging plane grid unit;
  • the adjusted velocity v'of each grid unit of the imaging plane is obtained, that is, the imaging plane The final velocity distribution; where the fuzzy constraint factor of each grid cell has a value range of [0.5, 1];
  • SS3 compares the adjusted speed v′ of each grid unit with the healthy reference speed value v of each grid unit obtained by S1, if When the predetermined threshold is exceeded, the grid unit corresponding to the mark v ′ ′ is an abnormal grid unit;
  • SS4 performs secondary image smoothing on the marked abnormal grid units, and then determines the imagery of internal defects in the longitudinal section of the tree.
  • the second object of the present invention is to provide a method for imaging internal defects in longitudinal sections of trees, the method further comprising:
  • S1 Establish the corresponding imaging plane based on the measured tree data, divide the imaging plane into grid units with the same size, and assign the same initial velocity value to each grid unit to obtain the initial velocity distribution of the imaging plane ;
  • the linear propagation model is used to simulate the propagation of stress waves inside the trees, and the combined iterative reconstruction technology SIRT algorithm is used to adjust the velocity of the imaging plane grid unit; the maximum and minimum velocity values and based on The fuzzy constraint mechanism of the grid unit group restricts the speed of the grid unit of the imaging plane; obtain the adjusted velocity v'of each grid unit of the imaging plane;
  • the method further includes: calculating reference values for the propagation velocity of stress waves in various directions within the healthy trees, and then obtaining the healthy reference velocity value v of each grid unit of the imaging plane; the S3 is: comparing each of the imaging planes The adjusted velocity v'of the grid unit and the healthy reference velocity v of each grid unit of the imaging plane, calculate Value when When the predetermined threshold is exceeded, the grid cell corresponding to the mark v ′ ′ is an abnormal grid cell.
  • the method further includes: performing secondary image smoothing processing on the abnormal grid unit to obtain a defect map inside the measured tree.
  • the S2 includes:
  • S21 uses the SIRT algorithm to calculate the velocity increment of each grid unit, and applies it to the current velocity value of each grid unit to obtain a new velocity value;
  • S22 imposes maximum and minimum speed limits on the speed value of the grid unit. When the obtained new speed value exceeds the maximum or minimum limit value, the new speed value will be assigned to the speed value exceeded.
  • the limit value
  • a fuzzy constraint based on the grid unit group is applied to the speed value of the grid unit.
  • the value of each grid unit after each iteration is The inversion velocity value is linearly combined with the fully constrained velocity value of each grid unit as the new velocity value of the grid unit;
  • the calculation of the reference value v( ⁇ , ⁇ ) of the propagation speed of the stress wave in each direction inside the healthy tree, and then the healthy reference speed value v of each grid unit of the imaging plane includes:
  • v( ⁇ , ⁇ ) v l ⁇ v R ⁇ (-0.2 ⁇ 2 +1)/[v l ⁇ sin 2 ⁇ +v R ⁇ (-0.2 ⁇ 2 +1) ⁇ cos 2 ⁇ ] (1)
  • v l is the velocity of the stress wave propagating in the longitudinal direction of the tree
  • v R is the velocity value of the stress wave propagating in the radial direction of the tree
  • ⁇ is the angle between the longitudinal section and the radial section corresponding to the propagation direction
  • ⁇ is the corresponding stress wave propagation direction angles
  • reference speed V i represents the health value of the i-th cell
  • v ij is the j-th through the speed reference value of the i-th cell of the propagation path
  • the velocity values by the formula (1) is calculated to give
  • M is the total number of paths through the i-th grid unit
  • N is the number of grid units in the imaging plane.
  • the grid unit corresponding to the mark v' is an abnormal grid unit, and the predetermined threshold is 15%.
  • the fuzzy constraint factor of each grid unit has a value range of [0.5, 1].
  • the value of the fuzzy constraint factor of the grid unit near the center of the tree is greater than the value of the fuzzy constraint factor of the grid unit near the edge of the tree.
  • the method before establishing the corresponding imaging plane according to the measured tree data, the method further includes:
  • the third object of the present invention is to provide an application method of the above method in the field of non-destructive testing, characterized in that, the application method includes: building a non-destructive testing platform, and place the trunk ends of the tree to be measured at random distances along the same longitudinal direction. Arrange a certain number of sensors and connect the sensors to the stress wave signal collector. Each time one of the sensors is hit with a pulse hammer, the sensor at the other end receives the corresponding signal, and the collector records the collected stress wave propagation time , Repeat this way until all the sensors are tapped to obtain the propagation time data between the sensors at both ends; at the same time, use a tape measure to measure the diameter of the tree and the longitudinal section sensor position information for subsequent longitudinal section imaging .
  • each grid unit is assigned an initial velocity value, and the velocity value is greater than zero.
  • the imaging plane is divided into a certain number of grid cells to establish its initial velocity distribution, and then the linear propagation model is used for multiple iterations.
  • the joint Simultaneous iterative reconstruction technique (SIRT) algorithm adjusts the velocity distribution of the imaging plane, uses the maximum and minimum velocity constraints to constrain the velocity of each grid unit of the imaging plane, and uses the blur based on the grid unit group The constraint restricts the speed of each grid unit until the final velocity distribution matches the measured data to end the iteration.
  • SIRT Simultaneous iterative reconstruction technique
  • Figure 1 is an experimental platform for non-destructive testing in the method of the present invention.
  • Fig. 2 is a schematic diagram of a longitudinal imaging plane in the present invention.
  • Fig. 3 is a schematic diagram of the fuzzy constraint matrix in the present invention.
  • Figure 4 is a comparison diagram of longitudinal section imaging of log samples.
  • Figure 5 is the three-dimensional coordinate system of the tree trunk.
  • This embodiment provides a method for imaging defects in the longitudinal section of a tree.
  • the method uses the propagation time of the stress wave inside the tree as input data, divides the imaging plane into a certain number of grid cells to establish its initial velocity distribution, and then uses it.
  • the linear propagation model performs multiple iterations. After each iteration, the SIRT algorithm is used to adjust the velocity distribution of the imaging plane, and the maximum and minimum velocity constraints are used to constrain the velocity of each grid unit of the imaging plane.
  • the fuzzy constraint of the grid cell group limits the speed of each grid cell until the final velocity distribution is more consistent with the measured data to end the iteration, and the speed value of the grid cell at this time is compared with the reference of the tested healthy tree The values are compared to determine whether a certain grid unit is abnormal data or normal data, and then the grid unit imaging is subjected to secondary smoothing processing to obtain the defect position inside the tree.
  • FIG. 1 A certain number of sensors are arranged at random distances along the longitudinal direction at both ends of the trunk of the tree to be measured.
  • the wave signal collector is connected, and each time one of the sensors is struck with a pulse hammer, the sensor at the other end receives the corresponding signal, and the collector records the collected stress wave propagation time. Repeat this way until all the sensors are all After tapping, obtain the propagation time data between the two sensors at both ends; at the same time, use a tape measure to measure the diameter of the tree and the sensor position information of the longitudinal section for subsequent longitudinal section imaging.
  • the subsequent longitudinal section imaging work is started.
  • an imaging plane as shown in Figure 2 is established.
  • the imaging plane is divided into a certain number of grid cells, where each grid cell has the same size.
  • the grid cells are usually divided into smaller sizes, but at the same time, it is necessary to ensure that each grid cell has a propagation path through as much as possible.
  • the initial velocity value is usually any positive value greater than 0, so that the imaging plane is obtained.
  • the reference value v( ⁇ , ⁇ ) of the stress wave propagation in various directions inside the healthy tree is calculated, and then the healthy reference velocity value v of each grid unit of the imaging plane is obtained.
  • the reference value v( ⁇ , ⁇ ) of stress wave propagation in various directions inside healthy trees can be calculated according to the following formula (1)
  • v( ⁇ , ⁇ ) v l ⁇ v R ⁇ (-0.2 ⁇ 2 +1)/[v l ⁇ sin 2 ⁇ +v R ⁇ (-0.2 ⁇ 2 +1) ⁇ cos 2 ⁇ ] (1)
  • v l is the velocity of the stress wave propagating in the longitudinal direction of the tree
  • v R is the velocity value of the stress wave propagating in the radial direction of the tree
  • ⁇ is the angle between the longitudinal section and the radial section corresponding to the propagation direction
  • ⁇ is the corresponding stress wave propagation Direction angle
  • the calculation method of the health reference speed value v of each grid unit can be based on the following formula (2):
  • V i indicates a healthy reference speed value of the i-th cell
  • v ij is the reference velocity passes through the j-th channel of the i-th grid units
  • the speed value obtained from the above formula (1) is calculated to give
  • M is the total number of paths through the i-th grid unit
  • N is the number of grid units in the imaging plane.
  • the linear propagation model is used to simulate the propagation of stress waves inside the trees, and the joint iterative reconstruction technology SIRT algorithm is used to adjust the velocity of the imaging plane grid unit; the maximum and minimum velocity values and based on the grid are used in the adjustment process
  • the fuzzy constraint mechanism of the unit group constrains the speed of the grid unit of the imaging plane; the adjusted velocity v'of each grid unit of the imaging plane is obtained;
  • the SIRT algorithm is used to calculate the velocity increment of each grid unit, and it is applied to the current velocity value of each grid unit to obtain a new velocity value; the SIRT algorithm is used to calculate the velocity increase of each grid unit Please refer to Geophysical Tomography Using Wavefront Migration and Fuzzy Constraints published in 1994.
  • a fuzzy constraint based on the grid unit group is applied to the speed value of the grid unit.
  • the value of each grid unit after each iteration is The inversion velocity value is linearly combined with the fully constrained velocity value of each grid unit as the new velocity value of the grid unit;
  • the integer part of the fuzzy constraint factor of each grid cell represents the type of constraint imposed: a negative value means that the velocity value of the grid cell remains at a fixed value.
  • the algorithm of this application chooses to fix it as the reference velocity value of the grid unit; a positive value represents that the velocity of the grid unit is constrained by the velocity of the grid unit group where it is located, and different integers represent different grid unit groups.
  • the speed of each grid unit group is the average of the reference speeds of all grid units in the same grid unit group.
  • the fractional part of the grid element constraint factor represents the degree of ambiguity of the imposed constraints: 0 means full constraints are used, and greater than 0 means fuzzy constraints are applied, and the larger the fractional part, the higher the degree of ambiguity and uncertainty. Bigger.
  • the algorithm of this application chooses to impose smaller fuzzy constraints on the grid unit group close to the bark to make it conform to the law of stress wave longitudinal propagation as much as possible. For the part closer to the center of the tree, the wood is harder and denser, which makes it easier to There is a high probability that the speed abnormal area will appear, and the uncertainty is large, so a large blur constraint is imposed to make it better adapt to the internal situation of the tree and enhance the realism of the imaging.
  • the above-mentioned iteration end condition for adjusting the velocity of each grid cell using the SIRT algorithm is: when the root mean square error between the measured time data and the time data obtained from the inversion stabilizes, the iteration ends.
  • the above-mentioned stabilization means that the root mean square error fluctuates above and below a certain value in the final stage of the iteration. Generally, about 3 times is considered stable.
  • All the grid cells marked as abnormal grid cells are smoothed using the mean value method to generate the final image of the longitudinal section of the tree, and the health status of the defective part of the tree is judged.
  • Figure 4a is a log image, in which 16 sensors are used for test data; No. 1-8 sensors are arranged longitudinally along end a in Fig. 4a, and No. 9-16 sensors are along end b in Fig. 4a Longitudinal arrangement;
  • Figure 4b is a longitudinal cross-sectional image generated using the Du's method method,
  • Figure 4c is a longitudinal cross-sectional image generated using the LSQR method, and
  • Figure 4d is a longitudinal cross-sectional image obtained using the method provided by this application for detection;
  • LSQR method can refer to An Algorithm for Sparse Linear Equations and Sparse Least Squares published in 1982.
  • the Du’s method detects that the log sample has defects, but there are many false detections, which is quite different from the real situation.
  • the improved LSQR detects the approximate location of the defect, which is more accurate than the Du’s method, but there are still many false detections in the figure.
  • the method proposed in this application detects the defect more accurately, the shape position is closest to the real situation of the defect, and the algorithm has almost no false detection area, and the imaging effect is better.
  • Part of the steps in the embodiments of the present invention can be implemented by software, and the corresponding software program can be stored in a readable storage medium, such as an optical disc or a hard disk.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biochemistry (AREA)
  • Analytical Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Acoustics & Sound (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Optimization (AREA)
  • Algebra (AREA)
  • Mathematical Analysis (AREA)
  • Image Processing (AREA)
  • Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)

Abstract

一种树木纵截面内部缺陷成像方法，属于树木的无损检测领域。方法通过以应力波在树木内部的传播时间作为输入数据，将成像平面划分为一定数量的网格单元来建立成像平面的初始速度分布，之后使用直线传播模型进行多轮迭代，每轮迭代结束之后利用SIRT算法调节成像平面的速度分布，使用最大值以及最小值速度约束以及基于网格单元组的模糊约束对每个网格单元速度进行限制，直到最后的速度分布与所测得的数据较吻合时结束迭代，通过比较此时的网格单元的速度值与被测健康树木的参考值，判断出异常网格单元，之后对网格单元成像进行二次平滑处理，得到树木内部的缺陷位置，能够准确检测出树木缺陷区域，且误检区域少，成像效果好。

Description

一种树木纵截面内部缺陷成像方法 技术领域
本发明涉及一种树木纵截面内部缺陷成像方法，属于树木的无损检测领域。
背景技术
无损检测又称非破坏性检测，是利用材料的不同物理力学性质或化学性质在不破坏目标物体内部及外观结构与特性的前提下，对物体相关特性(如形状、位移、应力、光学特性、流体性质、力学性质等)进行测试与检验，尤其是对各种缺陷的测量。
对于树木的无损检测通常是利用应力波的作用进行检测，应力波是指物体在受到了冲击之后，在应力的作用之下所产生的可以在物体内部中间进行传播的弹性机械波。在我国，应力波首先应用在岩土，混凝土等性能以及缺陷的检测之中，之后林业科技人员才将其应用到了树木的无损检测领域。
目前，国内外对于树木内部缺陷的横截面断层成像检测已经进行了较为广泛的研究，但是树木纵截面成像的结果对于判断树木内部缺陷在纵向上的延伸程度具有重要意义，同时可以为树木内部的三维立体成像提供参考。
发明内容
为了判断树木内部缺陷在纵向上的延伸程度以及为树木内部的三维立体成像提供参考，本发明提供了一种树木纵截面内部缺陷成像方法，所述方法包括：
SS1根据所测量树木的数据建立相应的成像平面，将成像平面划分为具有相同尺寸的网格单元，赋予每个网格单元一个初始的速度值；计算健康树木内部各个方向应力波传播的速度参考值，继而得到成像平面每个网格单元的健康参考速度值v；
SS2每个网格单元有了初始速度值以后，根据成像平面的初始速度分布，使用直线传播模型模拟应力波在树木内部的传播，利用联合迭代重建技术SIRT算法调整成像平面网格单元的速度；调整过程中使用最大与最小速度值和基于网格单元组的模糊约束机制对成像平面网格单元的速度进行约束；得到成像平面每个网格单元调整后的速度v′，即得到成像平面的最终速度分布；其中，每个网格单元的模糊约束因子取值范围为[0.5，1]；
SS3将每个网格单元调整后的速度v′与S1获得的每个网格单元的健康参考速度值v进行比较，若
Figure PCTCN2019087022-appb-000001
超过预定阈值时，标记v ′′对应的网格单元为异常网格单元；
SS4将所标记的异常网格单元进行二次图像平滑处理，进而确定树木纵截面内部缺陷成 像图。
本发明的第二个目的在于提供了一种树木纵截面内部缺陷成像方法，所述方法还包括：
S1：根据所测量树木的数据建立相应的成像平面，将成像平面划分为具有相同的大小尺寸的网格单元，为每个网格单元赋予一个相同的初始速度值，得到成像平面的初始速度分布；
S2：根据成像平面的初始速度分布，使用直线传播模型模拟应力波在树木内部的传播，利用联合迭代重建技术SIRT算法调整成像平面网格单元的速度；调整过程中使用最大与最小速度值和基于网格单元组的模糊约束机制对成像平面网格单元的速度进行约束；得到成像平面每个网格单元调整后的速度v′；
S3：根据成像平面每个网格单元调整后的速度v′确定每个网格单元是否为异常网格单元。
可选的，所述方法还包括：计算健康树木内部各个方向应力波传播的速度参考值，继而得到成像平面每个网格单元的健康参考速度值v；所述S3为：比较成像平面每个网格单元调整后的速度v′和成像平面每个网格单元的健康参考速度值v，计算
Figure PCTCN2019087022-appb-000002
的值，当
Figure PCTCN2019087022-appb-000003
超过预定阈值时，标记v ′′对应的网格单元为异常网格单元。
可选的，所述方法还包括：对异常网格单元进行二次图像平滑处理，得到所测量树木内部的缺陷图。
可选的，所述S2包括：
S21利用SIRT算法计算出每个网格单元的速度增量，将其应用于每个网格单元当前速度值，得到新的速度值；
S22在速度调整的过程中，对网格单元的速度值施加最大与最小速度值限制，当所得到的新的速度值超过最大或最小极限值，则新的速度值就会被赋予为其所超过的极限值；
同时，在速度调整的过程中，对网格单元的速度值施加基于网格单元组的模糊约束，根据每个网格单元的模糊约束因子，将每轮迭代结束之后的每个网格单元的反演速度值与每个网格单元的全约束速度值进行线性组合，作为网格单元的新的速度值；
S23当最后一轮迭代结束后，得到成像平面每个网格单元调整后的速度v′。
可选的，所述计算健康树木内部各个方向应力波传播的速度参考值v(θ，α)，继而得到成像平面每个网格单元的健康参考速度值v，包括：
根据式(1)计算v(θ，α)，根据式(2)计算v；
v(θ，α)＝v l×v R×(-0.2α 2+1)/[v l×sin 2θ+v R×(-0.2α 2+1)×cos 2θ]   (1)
Figure PCTCN2019087022-appb-000004
其中，v l是应力波沿树木纵向传播的速度，v R是应力波沿树木径向传播的速度值，α为传播方向相对应纵截面与径切面的夹角，θ为对应的应力波传播方向角，v i表示第i个网格单元的健康参考速度值，v ij为穿过第i个网格单元的第j条传播路径的速度参考值，该速度值可由公式(1)计算得到，M为穿过第i个网格单元的路径总数，N为成像平面的网格单元数量。
可选的，所述当
Figure PCTCN2019087022-appb-000005
超过预定阈值时，标记v′对应的网格单元为异常网格单元中，预定阈值为15％。
可选的，所述每个网格单元的模糊约束因子取值范围为[0.5，1]。
可选的，靠近树木中心位置处的网格单元的模糊约束因子的取值大于靠近树木边缘位置处的网格单元的模糊约束因子的取值。
可选的，所述根据所测量树木的数据建立相应的成像平面之前，还包括：
在所测量树木的树干两端沿纵向同一方向随机距离地布置预定数量的传感器；将传感器与应力波信号采集仪连接起来，通过脉冲锤敲击的方式获取两端传感器两两之间的传播时间数据；并测量树木的直径以及纵截面的传感器位置信息。
本发明的第三个目的在于提供上述方法在无损检测领域内的应用方法，其特征在于，所述应用方法包括：搭建无损检测平台，在待测量树木的树干两端沿纵向同一方向随机距离地布置一定数量的传感器，将传感器与应力波信号采集仪连接起来，每次使用脉冲锤敲击其中一个传感器，另一端的传感器接收到相应的信号，采集仪将收集到的应力波传播时间进行记录，如此这样重复，直到所有的传感器全部敲击完毕，获得两端传感器两两之间的传播时间数据；同时，使用卷尺测量树木的直径以及纵截面的传感器位置信息，用于之后的纵截面成像。
可选的，所述赋予每个网格单元一个初始的速度值，该速度值大于零。
本发明有益效果是：
通过以应力波在树木内部的传播时间作为输入数据，将成像平面划分为一定数量的网格单元来建立其初始速度分布，之后使用直线传播模型进行多轮迭代，每轮迭代结束之后，利用联合迭代重建技术(Simultaneous iterative reconstruction technique，SIRT)算法调节成像平面的速度分布，使用最大值以及最小值速度约束对成像平面的每个网格单元的速度进行约束，同时使用基于网格单元组的模糊约束对每个网格单元速度进行限制，直到最后的速度分布与所测得的数据较吻合时结束迭代，将此时的网格单元的速度值与被测健康树木的参考值 进行比较，判断出某个网格单元为异常数据还是正常数据，之后对网格单元成像进行二次平滑处理，得到树木内部的缺陷位置，该方法能够准确检测出树木缺陷区域，且误检区域少，成像效果好。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明方法中无损检测的实验平台。
图2是本发明中的纵向成像平面示意图。
图3是本发明中的模糊约束矩阵示意图。
图4是原木试样纵切面成像比较图。
图5是树干的三维坐标系统。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图对本发明实施方式作进一步地详细描述。
实施例一：
本实施例提供一种树木纵截面内部缺陷成像方法，所述方法以应力波在树木内部的传播时间作为输入数据，将成像平面划分为一定数量的网格单元来建立其初始速度分布，之后使用直线传播模型进行多轮迭代，每轮迭代结束之后，利用SIRT算法调节成像平面的速度分布，使用最大值以及最小值速度约束对成像平面的每个网格单元的速度进行约束，同时使用基于网格单元组的模糊约束对每个网格单元速度进行限制，直到最后的速度分布与所测得的数据较吻合时结束迭代，将此时的网格单元的速度值与被测健康树木的参考值进行比较，判断出某个网格单元为异常数据还是正常数据，之后对网格单元成像进行二次平滑处理，得到树木内部的缺陷位置。
具体的，在对树木进行无损检测时，首先搭建无损检测平台，请参考图1，在所测量树木的树干两端沿纵向同一方向随机距离地布置一定数量的传感器，将传感器与匈牙利产FAKOPP应力波信号采集仪连接起来，每次使用脉冲锤敲击其中一个传感器，另一端的传感器接收到相应的信号，采集仪将收集到的应力波传播时间进行记录，如此这样重复，直到所 有的传感器全部敲击完毕，获得两端传感器两两之间的传播时间数据；同时，使用卷尺测量树木的直径以及纵截面的传感器位置信息，用于之后的纵截面成像。
如图2所示，在获取了上述传感器两两之间的传播时间数据、树木的直径以及纵截面的传感器位置信息之后，开始进行后续的纵截面成像工作。
根据所测量的树木直径以及传感器位置信息，建立如图2所示的成像平面。将成像平面划分为一定数量的网格单元，其中每个网格单元具有相同的大小尺寸。为使得成像结果较为精确，通常将网格单元划分为较小尺寸，但是同时需要保证每个网格单元尽量有传播路径穿过。
建立应力波传播的速度模型，为图2所示的成像平面中的每个网格单元赋予一个统一的初始速度值，该初始速度值通常使用大于0的任意正值，如此构建得到成像平面的初始速度分布。
构建得到成像平面的初始速度分布之后，计算健康树木内部各个方向应力波传播的速度参考值v(θ，α)，继而得到成像平面每个网格单元的健康参考速度值v。
健康树木内部各个方向应力波传播的速度参考值v(θ，α)可根据下述式(1)进行计算
v(θ，α)＝v l×v R×(-0.2α 2+1)/[v l×sin 2θ+v R×(-0.2α 2+1)×cos 2θ]   (1)
其中，v l是应力波沿树木纵向传播的速度，v R是应力波沿树木径向传播的速度值，α为传播方向相对应纵截面与径切面的夹角，θ为对应的应力波传播方向角，具体α以及θ如图5中相应位置所示；
每个网格单元的健康参考速度值v的计算方式可根据下述式(2)：
Figure PCTCN2019087022-appb-000006
其中，v i表示第i个网格单元的健康参考速度值，v ij为穿过第i个网格单元的第j条传播路径的速度参考值，该速度值可由上述式(1)计算得到，M为穿过第i个网格单元的路径总数，N为成像平面的网格单元数量。
根据成像平面的初始速度分布，使用直线传播模型模拟应力波在树木内部的传播，利用联合迭代重建技术SIRT算法调整成像平面网格单元的速度；调整过程中使用最大与最小速度值和基于网格单元组的模糊约束机制对成像平面网格单元的速度进行约束；得到成像平面每个网格单元调整后的速度v′；
具体的，利用SIRT算法计算出每个网格单元的速度增量，将其应用于每个网格单元当前速度值，得到新的速度值；利用SIRT算法计算出每个网格单元的速度增量请参考1994年发表的Geophysical Tomography Using Wavefront Migration and Fuzzy Constraints。
在速度调整的过程中，对网格单元的速度值施加最大与最小速度值限制，当所得到的新的速度值超过最大或最小极限值，则新的速度值就会被赋予为其所超过的极限值；
同时，在速度调整的过程中，对网格单元的速度值施加基于网格单元组的模糊约束，根据每个网格单元的模糊约束因子，将每轮迭代结束之后的每个网格单元的反演速度值与每个网格单元的全约束速度值进行线性组合，作为网格单元的新的速度值；
当最后一轮迭代结束后，得到成像平面每个网格单元调整后的速度v′
在上述速度调整的过程中，如图3所示，每个网格单元的模糊约束因子的整数部分代表了所施加的约束类型：负数值代表该网格单元的速度值保持在一个固定值，本申请算法选择将其固定为该网格单元的参考速度值；正数值代表该网格单元的速度受其所在网格单元组的速度约束，不同的整数代表不同的网格单元组。
其中，每个网格单元组的速度即为处于同一网格单元组的所有网格单元参考速度的平均值。
网格单元约束因子的分数部分则代表了所施加约束的模糊程度：0代表使用完全约束，而大于0则代表施加了模糊约束，且小数部分越大，则代表模糊程度越高，不确定性越大。本申请算法选择对靠近树皮部分的网格单元组施加较小的模糊约束，使其尽量符合应力波纵向传播规律，而对于越靠近树木中心的部分，木质较硬，密度较大，容易更大机率地出现速度异常区域，不确定性较大，因此施加较大模糊约束，使其更好地适应树木内部情况，增强成像的真实感。
上述使用SIRT算法调整每个网格单元的速度的迭代结束条件为：当所测时间数据与反演所得时间数据的均方根误差趋于稳定时，迭代结束。上述趋于稳定指迭代最后阶段，均方根误差在某一数值的上下进行波动，一般3次左右即视为稳定。
在得到最终速度分布后，将其与根据上述公式(2)计算得到的每个网格的健康参考速度值v进行比较，计算
Figure PCTCN2019087022-appb-000007
的值，当
Figure PCTCN2019087022-appb-000008
超过预定阈值时，标记v′对应的网格单元为异常网格单元；
具体的，设定当
Figure PCTCN2019087022-appb-000009
时，标记该v′对应的网格单元为异常网格单元；
对标记为异常网格单元的所有网格单元使用均值法进行平滑操作，生成最终的树木纵截面成像图，对树木内部的缺陷部分健康状况进行判断。
为验证本申请方法的检测效果，以下采用通用的成像方法与本申请所公开的方法进行比较：
请参考图4，图4a为原木图像，其中采用16个传感器进行进行测试数据；1-8号传感 器沿着图4a中的a端纵向布置，9-16号传感器沿着图4a中的b端纵向布置；图4b为采用Du’s method方法生成纵截面图像，图4c为采用LSQR方法生成的纵截面图像，图4d为采用本申请所提供的方法进行检测得到的纵截面图像；
Du’s method方法的介绍可参考2015年公开的Stress Wave Tomography of Wood Internal Defects using Ellipse-Based Spatial Interpolation and Velocity Compensation文献中。
LSQR方法的介绍可参考1982年公开的An Algorithm for Sparse Linear Equations and Sparse Least Squares文献中。
由图可知，Du’s method检测出该原木试样存在缺陷，但是误检处比较多，与真实情况相差较大。改进的LSQR检测出了缺陷的大概位置，较Du’s method更为精确，但是图中误检之处依然较多。而本申请所提方法较为精确地检测出了缺陷，形状位置与缺陷真实情况最为接近，并且算法几乎没有误检测区域，成像效果较好。
本发明实施例中的部分步骤，可以利用软件实现，相应的软件程序可以存储在可读取的存储介质中，如光盘或硬盘等。
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (12)

  1. 一种树木纵截面内部缺陷成像方法，其特征在于，所述方法包括：
    SS1根据所测量树木的数据建立相应的成像平面，将成像平面划分为具有相同尺寸的网格单元，赋予每个网格单元一个初始的速度值；计算健康树木内部各个方向应力波传播的速度参考值，继而得到成像平面每个网格单元的健康参考速度值v；
    SS2每个网格单元有了初始速度值以后，根据成像平面的初始速度分布，使用直线传播模型模拟应力波在树木内部的传播，利用联合迭代重建技术SIRT算法调整成像平面网格单元的速度；调整过程中使用最大与最小速度值和基于网格单元组的模糊约束机制对成像平面网格单元的速度进行约束；得到成像平面每个网格单元调整后的速度v′，即得到成像平面的最终速度分布；其中，每个网格单元的模糊约束因子取值范围为[0.5，1]；
    SS3将每个网格单元调整后的速度v′与S1获得的每个网格单元的健康参考速度值v进行比较，若
    Figure PCTCN2019087022-appb-100001
    超过预定阈值时，标记v ′′对应的网格单元为异常网格单元；
    SS4将所标记的异常网格单元进行二次图像平滑处理，进而确定树木纵截面内部缺陷成像图。
  2. 一种树木纵截面内部缺陷成像方法，其特征在于，所述方法还包括：
    S1：根据所测量树木的数据建立相应的成像平面，将成像平面划分为具有相同的大小尺寸的网格单元，为每个网格单元赋予一个相同的初始速度值，得到成像平面的初始速度分布；
    S2：根据成像平面的初始速度分布，使用直线传播模型模拟应力波在树木内部的传播，利用联合迭代重建技术SIRT算法调整成像平面网格单元的速度；调整过程中使用最大与最小速度值和基于网格单元组的模糊约束机制对成像平面网格单元的速度进行约束；得到成像平面每个网格单元调整后的速度v′；
    S3：根据成像平面每个网格单元调整后的速度v′确定每个网格单元是否为异常网格单元。
  3. 根据权利要求2所述的方法，其特征在于，所述方法还包括：计算健康树木内部各个方向应力波传播的速度参考值，继而得到成像平面每个网格单元的健康参考速度值v；所述S3为：比较成像平面每个网格单元调整后的速度v′和成像平面每个网格单元的健康参考速度值v，计算
    Figure PCTCN2019087022-appb-100002
    的值，当
    Figure PCTCN2019087022-appb-100003
    超过预定阈值时，标记v ′′对应的网格单元为异常网格单元。
  4. 根据权利要求3所述的方法，其特征在于，所述方法还包括：对异常网格单元进行二次图像平滑处理，得到所测量树木内部的缺陷图。
  5. 根据权利要求4所述的方法，其特征在于，所述S2包括：
    S21利用SIRT算法计算出每个网格单元的速度增量，将其应用于每个网格单元当前速度值，得到新的速度值；
    S22在速度调整的过程中，对网格单元的速度值施加最大与最小速度值限制，当所得到的新的速度值超过最大或最小极限值，则新的速度值就会被赋予为其所超过的极限值；
    同时，在速度调整的过程中，对网格单元的速度值施加基于网格单元组的模糊约束，根据每个网格单元的模糊约束因子，将每轮迭代结束之后的每个网格单元的反演速度值与每个网格单元的全约束速度值进行线性组合，作为网格单元的新的速度值；
    S23当最后一轮迭代结束后，得到成像平面每个网格单元调整后的速度v′。
  6. 根据权利要求5所述的方法，其特征在于，所述计算健康树木内部各个方向应力波传播的速度参考值v(θ，α)，继而得到成像平面每个网格单元的健康参考速度值v，包括：
    根据式(1)计算v(θ，α)，根据式(2)计算v；
    v(θ，α)＝v l×v R×(-0.2α 2+1)/[v l×sin 2θ+v R×(-0.2α 2+1)×cos 2θ]         (1)
    Figure PCTCN2019087022-appb-100004
    其中，v l是应力波沿树木纵向传播的速度，v R是应力波沿树木径向传播的速度值，α为传播方向相对应纵截面与径切面的夹角，θ为对应的应力波传播方向角，v i表示第i个网格单元的健康参考速度值，v ij为穿过第i个网格单元的第j条传播路径的速度参考值，该速度值可由公式(1)计算得到，M为穿过第i个网格单元的路径总数，N为成像平面的网格单元数量。
  7. 根据权利要求6所述的方法，其特征在于，所述当
    Figure PCTCN2019087022-appb-100005
    超过预定阈值时，标记v′对应的网格单元为异常网格单元中，预定阈值为15％。
  8. 根据权利要求7所述的方法，其特征在于，所述每个网格单元的模糊约束因子取值范围为[0.5，1]。
  9. 根据权利要求8所述的方法，其特征在于，靠近树木中心位置处的网格单元的模糊约束因子的取值大于靠近树木边缘位置处的网格单元的模糊约束因子的取值。
  10. 根据权利要求9所述的方法，其特征在于，所述根据所测量树木的数据建立相应的成像平面之前，还包括：
    在所测量树木的树干两端沿纵向同一方向随机距离地布置预定数量的传感器；将传感器与应力波信号采集仪连接起来，通过脉冲锤敲击的方式获取两端传感器两两之间的传播时间数据；并测量树木的直径以及纵截面的传感器位置信息。
  11. 权利要求1-10任一所述的方法在无损检测领域内的应用方法，其特征在于，所述方法包括：搭建无损检测平台，在待测量树木的树干两端沿纵向同一方向随机距离地布置一定数量的传感器，将传感器与应力波信号采集仪连接起来，每次使用脉冲锤敲击其中一个传感器，另一端的传感器接收到相应的信号，采集仪将收集到的应力波传播时间进行记录，如此这样重复，直到所有的传感器全部敲击完毕，获得两端传感器两两之间的传播时间数据；同时，使用卷尺测量树木的直径以及纵截面的传感器位置信息，用于之后的纵截面成像。
  12. 根据权利要求11所述的方法，其特征在于，所述赋予每个网格单元一个初始的速度值，该速度值大于零。
PCT/CN2019/087022 2019-03-22 2019-05-15 一种树木纵截面内部缺陷成像方法 WO2020191896A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/005,394 US20200393347A1 (en) 2019-03-22 2020-08-28 Imaging Method of Internal Defects in Longitudinal Sections of Trees

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910221698.2 2019-03-22
CN201910221698.2A CN109900789B (zh) 2019-03-22 2019-03-22 一种树木纵截面内部缺陷成像方法

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/005,394 Continuation US20200393347A1 (en) 2019-03-22 2020-08-28 Imaging Method of Internal Defects in Longitudinal Sections of Trees

Publications (1)

Publication Number Publication Date
WO2020191896A1 true WO2020191896A1 (zh) 2020-10-01

Family

ID=66952437

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/087022 WO2020191896A1 (zh) 2019-03-22 2019-05-15 一种树木纵截面内部缺陷成像方法

Country Status (3)

Country Link
US (1) US20200393347A1 (zh)
CN (1) CN109900789B (zh)
WO (1) WO2020191896A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110261850B (zh) * 2019-07-01 2023-05-23 东北林业大学 一种树木内部缺陷检测数据的成像算法
CN110940728B (zh) * 2019-12-17 2022-05-06 湖北民族大学 树木缺陷无损检测方法
CN112485335A (zh) * 2020-11-12 2021-03-12 江南大学 一种基于射线分割的检测树木内部缺陷的断层成像方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104535655A (zh) * 2014-11-24 2015-04-22 清华大学 一种射线追踪式超声Lamb波缺陷层析成像方法
CN105467012A (zh) * 2015-11-23 2016-04-06 江南大学 一种检测树木径切面上缺陷位置的方法
WO2017077486A1 (en) * 2015-11-06 2017-05-11 Peter Lee System and method for assessment of irregularity in a wooden material surrounded by a substrate
CN106885846A (zh) * 2017-01-20 2017-06-23 浙江农林大学 树木缺陷检测装置及检测方法
CN107045003A (zh) * 2016-12-20 2017-08-15 北京航空航天大学 一种基于同步迭代重建算法的铝合金结构腐蚀损伤监测方法
CN107402257A (zh) * 2017-08-14 2017-11-28 浙江农林大学 基于路径分组插值法的木材径切面缺陷成像方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100604415B1 (ko) * 2005-04-04 2006-07-25 재단법인서울대학교산학협력재단 목재 열화 씨티 측정 방법
CN107300587B (zh) * 2017-01-20 2020-08-07 浙江农林大学 树木缺陷检测方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104535655A (zh) * 2014-11-24 2015-04-22 清华大学 一种射线追踪式超声Lamb波缺陷层析成像方法
WO2017077486A1 (en) * 2015-11-06 2017-05-11 Peter Lee System and method for assessment of irregularity in a wooden material surrounded by a substrate
CN105467012A (zh) * 2015-11-23 2016-04-06 江南大学 一种检测树木径切面上缺陷位置的方法
CN107045003A (zh) * 2016-12-20 2017-08-15 北京航空航天大学 一种基于同步迭代重建算法的铝合金结构腐蚀损伤监测方法
CN106885846A (zh) * 2017-01-20 2017-06-23 浙江农林大学 树木缺陷检测装置及检测方法
CN107402257A (zh) * 2017-08-14 2017-11-28 浙江农林大学 基于路径分组插值法的木材径切面缺陷成像方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
JACKSON, M. J. ET AL.: "MIGRATIOM-Geophysical Tomography Using Wavefront Migration and Fuzzy Constraints", REPORT OF INVESTIGATIONS, 31 December 1994 (1994-12-31), XP055737293, ISSN: 1066-5552, DOI: 20191129175839Y *
LIU, LEI ET AL.: "Acoustic tomography based on hybrid wave propagation model for tree decay detection", COMPUTERS AND ELECTRONICS IN AGRICULTURE, vol. 151, 15 June 2018 (2018-06-15), XP085417138, DOI: 20191129175725Y *

Also Published As

Publication number Publication date
US20200393347A1 (en) 2020-12-17
CN109900789A (zh) 2019-06-18
CN109900789B (zh) 2020-05-08

Similar Documents

Publication Publication Date Title
US10877084B2 (en) Nonlinear model transformation solving and optimization method for partial discharge positioning based on multi-ultrasonic sensor
WO2020191896A1 (zh) 一种树木纵截面内部缺陷成像方法
Zhao et al. Ultrasonic Lamb wave tomography in structural health monitoring
Moezi et al. A generally modified cuckoo optimization algorithm for crack detection in cantilever Euler-Bernoulli beams
CN104280455B (zh) 一种用于裂纹方向识别的超声散射系数优化计算方法
MXPA05006835A (es) Metodos para determinar los parametros de yacimiento y pozo de sondeo utilizando tomografia de volumen de fresnel.
Liu et al. Acoustic tomography based on hybrid wave propagation model for tree decay detection
CN104537251B (zh) 一种风机叶片的冲击载荷识别方法
CN113588488B (zh) 电缆的缺陷检测方法、装置、终端设备及存储介质
CN111929549A (zh) 基于局部放电光学信号的gil局部放电源定位方法和系统
Divós et al. Resolution of stress wave based acoustic tomography
CN106770691A (zh) 一种基于压缩感知的汽轮机叶轮超声相控阵信号压缩方法
CN115616674A (zh) 探地雷达训练数据集模拟扩增与道路无损检测方法及系统
US11480550B2 (en) Method and device for mapping components for detecting elongation direction
AU2020271967A1 (en) Method for determining the geometry of a defect on the basis of non-destructive measurement methods using direct inversion
CN107727749A (zh) 一种基于小波包融合特征提取算法的超声定量检测方法
CN104040329A (zh) 用于检测在检查对象内部的缺陷的方法和装置
Wei et al. Propagation velocity model and two-dimensional defect imaging of stress wave in Larch (Larix gmelinii) wood
WO2023240939A1 (zh) 一种树木检测方法、装置及树木检测设备
CN115754007A (zh) 一种基于声发射技术和层析成像技术的损伤检测方法
CN112485335A (zh) 一种基于射线分割的检测树木内部缺陷的断层成像方法
CN114755302A (zh) 一种基于超声相控阵的裂纹表征方法及装置
Visalga et al. Influence of noise on decay predictions in standing trees
IT202100027284A1 (it) Computer-implemented method and a computer system for generating a training dataset for training an artificial neural network
Schmerr, Jr et al. Ultrasonic flaw sizing-An overview

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19920953

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19920953

Country of ref document: EP

Kind code of ref document: A1