CN103323533A - System and method for detecting plant diseases based on near field acoustic holography technology - Google Patents

Abstract

The invention relates to a system and method for detecting plant diseases based on near-field acoustic holography technology. Based on the acoustic emission phenomenon of plants, the cylindrical-spherical non-conformal surface is established based on the near-field acoustic holography technology of distributed source boundary point method. Acoustic holography experimental model, the simulation experiment is carried out for single sound source and double sound source respectively. Feasible experimental parameters are obtained by changing the radius of the holographic cylinder, the number of holographic measuring points, the number of nodes on the reconstructed sphere, the distance from the fictional solution source to the nodes on the reconstructed sphere, etc. in the model. The simulation results show that the method can accurately identify and locate the location of the acoustic emission source. According to the experimental parameters, a signal acquisition acoustic holographic experimental system with DSP as the core is established, and the upper computer uses Labview as the software platform to analyze and process the data. In the present invention, plant acoustic emission signals are used as important sensitive information to understand their growth status and disease conditions, and through the research of acoustic emission signals based on the distributed source boundary point method near-field acoustic holography technology, timely detection and identification of diseases are provided to detect plant diseases. And prevention and control provides a new method and new way.

Description

System and method for detecting plant diseases based on near-field acoustic holography technology

technical field

The invention belongs to the field of agricultural biotechnology, in particular to a system and method for detecting plant diseases based on near-field acoustic holography technology.

Background technique

Plants are often subjected to various environmental stresses during their growth and development, among which disease stress and water deficit reduce crop yield more than the sum of all other stresses. The total area of pests and diseases in the country is constantly increasing, and there are problems of excessive pesticide residues. Therefore, how to quickly, accurately and reliably judge the distribution area and degree of crop diseases in a timely manner is the key to effective prevention and control measures and precise Pesticide application is the key to disease control.

Plants affected by disease stress lead to changes in plant transpiration rate and increased tension; when it exceeds a limit value, the continuum of the water column can no longer be maintained due to the failure of the cohesive force between water molecules or the failure of the adhesion to the vessel wall. Breakage or evacuation occurs thereby, which is the "cavitation" phenomenon of plant vessels. When air pockets appear in the plant xylem, the tension will be released suddenly to generate a shock wave. Accompanied by the shock wave, an acoustic emission signal is generated. The plant transmits its own disease information to us through this signal. According to the distribution law of the acoustic emission signal The degree of disease stress of plants can be judged.

Near-field acoustic holography is an advanced noise measurement and analysis cutting-edge technology for sound source identification, localization, and spatial sound field visualization. The spatial transformation algorithm of the sound field is the core of near-field acoustic holography, and many reconstruction algorithms have been formed after years of development. Based on the NAH technology of Fourier transform, the algorithm and experiment are easy to implement and the calculation speed is fast, but the window effect and winding error will be introduced in the reconstruction process, and the holographic surface and the source surface are required to be conformal regular shapes, such as plane, Cylindrical or spherical, etc., to a certain extent limit its scope of application. Based on the NAH technology of the boundary element method (BEM), this method effectively overcomes the restrictions on the shape of the source surface and the holographic surface. The low efficiency of its numerical calculation affects its popularization and application in engineering. On this basis, the distributed source boundary point method is improved. The core of this method is to use the special solution generated by a series of special solution sources distributed in the normal direction of the boundary node of the vibrating body (deviating from the analysis domain) to indirectly construct the vibration The transfer matrix between the source and the sound field effectively avoids the problems of variable interpolation, singular integral processing, non-uniqueness of the solution, and large amount of calculation when the boundary element method is used to solve the sound radiation problem. The advantage of being suitable for the analysis of sound sources of arbitrary shapes, and improving the calculation speed, it has been widely used.

The location of plant acoustic emission sources through near-field acoustic holography technology can provide theoretical basis and experimental parameters for the precise prevention and control of plant diseases, so as to effectively and targetedly spray pesticides and reduce the amount of pesticides used. In addition, plant diseases have a long or short incubation period. During the incubation period, it is possible that the internal physiological structure changes but the external characteristics do not change. Therefore, the disease can be detected in time when the disease does not appear on the plant. And prevention is necessary.

Contents of the invention

The purpose of the present invention is to overcome the deficiencies of the prior art and provide a system and method for detecting plant diseases based on near-field acoustic holography. Using holographic technology as a theoretical basis, an acoustic holographic experimental model is established to identify the acoustic emission source signal of plants and analyze the acoustic field, locate the location of the acoustic emission source, determine the status of plant diseases, and implement precise spraying.

The present invention is achieved through the following technical solutions:

A system for detecting plant diseases based on near-field acoustic holography technology, including sequentially connected acoustic emission sensor arrays, broadband acoustic emission preamplifiers, AD converters, DSP processors, host computers and graphic displays, running acoustic emissions in the host computer Source identification algorithm analysis and processing software.

Moreover, the acoustic emission sensor array collects acoustic emission signals, which are amplified and filtered by the broadband acoustic emission preamplifier and then enter the AD converter for multi-channel signal conversion. The AD converter converts the data and transmits it to the DSP processor. The data is uploaded to the upper computer, and the upper computer uses Labview as the software platform to analyze and process the data, complete the reconstruction of the sound source, and display the sound field information in a graphical way to locate the location of the sound emission source and judge the degree of plant disease stress .

Moreover, the acoustic emission sensor and the broadband acoustic emission preamplifier select a resonant frequency of 70 kHz or above, and use a high-pass filter circuit to separate power and signals.

Moreover, the AD converter is AD7656 from American Analog Devices.

Moreover, the DSP processor adopts TMS320VC5509, the highest main frequency can reach 144MHz, and can complete three data read operations and two data write operations in one cycle.

And, described upper computer sound source recognition algorithm analysis processing software is to be software platform with Labview, has included a large amount of tools and functions for data acquisition, analysis, display and storage, utilizes MATLAB Script node, realizes in Labview to Matlab language call.

A method for detecting plant diseases based on near-field acoustic holography, the steps are:

⑴According to the acoustic emission phenomenon of plants, based on the near-field acoustic holography technology of the distributed source boundary point method, a cylindrical-spherical non-conformal surface acoustic holographic experimental model is established;

(2) Set the number of holographic measurement points and the number of reconstructed spherical nodes for a single sound source on the experimental model, and conduct simulation experiment analysis to select parameters;

(3) According to the parameter selection results of the single sound source simulation, the dual sound sources are reconstructed, and the validity of the method is verified;

(4) Through the analysis of the experimental data, the radius of the holographic cylinder, the applicable range of the sound source frequency, the selection rule of the distance from the fictitious special solution point source to the node on the reconstruction sphere, and the main parameters affecting the reconstruction accuracy are given;

⑸According to the best experimental parameters of the simulation, the acoustic holographic experimental system is designed, and the acoustic emission signal is collected by the sensor array, and then enters the AD sampling circuit after amplification and filtering, and the DSP provides the AD sampling control signal to realize synchronous sampling by multiple ADCs. The A/D conversion data is transmitted to the DSP in parallel, and uploaded to the host computer after processing. The host computer uses Labview as the software platform to analyze and process the data, complete the reconstruction of the sound source, and display the sound field information in a graphical way. Position the sound The location of the emission source can determine the degree of plant disease stress.

Moreover, the cylindrical-spherical non-conformal surface acoustic holography experimental model is to establish a spatial Cartesian coordinate system with the axial center as the origin, where r _h is the radius of the holographic cylindrical surface, _rs is the radius of the reconstructed spherical surface, and L is the holographic The height of the measurement surface along the axial direction, h is the distance from the imaginary special solution point source to the node on the reconstruction sphere, the special solution point source is distributed in the normal direction conformal to the reconstruction surface and away from the node, and let the special solution The number of solution sources is equal to the number of nodes on the reconstruction surface.

Moreover, the method for setting the number of holographic measurement points and the number of reconstruction spherical nodes on the experimental model for a single sound source is: for a single sound source, the number of holographic measurement points is 24 and the number of nodes on the reconstruction sphere is 14, or the number of holographic measurement points is 16 and the number of reconstruction spheres is 14. The number of nodes on the sphere is 14, or the number of holographic measurement points is 16 and the number of nodes on the reconstruction sphere is 10.

Advantage and positive effect of the present invention are:

1. The present invention applies the near-field acoustic holography technology to the detection of plant diseases, which can detect diseases in the early stage of plant diseases without disease spots, improve the control effect, reduce the cost of control, improve economic benefits, and provide a basis for realizing precise control.

2. The present invention expands the distributed source boundary point method from the research of low-frequency and steady-state signals to the research of high-frequency and instantaneous and unsteady-state signals under the condition that the number of holographic surface measurement points is small. The key to its realization is the special solution The structure of the source, the selected special solution source is a point source, assuming that the special solution source is distributed at a certain distance in the normal direction away from the node and conformal to the reconstruction surface, and the number of special solution sources is related to the knot on the reconstruction surface Points are equal.

3. In the present invention, the acoustic emission signal sent by plants under disease stress can be regarded as a point sound source, which radiates evenly to the surroundings, so the reconstruction surface can be selected as a spherical surface, and the main stem of the plant is cylindrical as the holographic measurement surface , established a cylindrical-spherical non-conformal surface acoustic holographic experimental model, which is more in line with the characteristics of plant acoustic emission signals.

4. Since the acoustic emission signal collected in this system is an instantaneous and unsteady signal, it is necessary to use the snapshot method to obtain the complex sound pressure on the holographic surface.

5. The present invention builds an acoustic holographic experimental system that uses high-speed AD and DSP for signal acquisition and processing, and organically combines Labview and Matlab to perform sound field analysis and identify sound sources; Labview can provide rich graphic controls, using graphical The programming method is simple, and the upper computer uses Labview as the software platform to collect and store data and calculate the complex sound pressure of the holographic surface. Matlab contains a variety of matrix operations, rich algorithm operations and powerful drawing functions, which are used for some complex numerical calculations, programming of various algorithms and complex graphics drawing. Using the drawing function of matlab, the reconstructed and analyzed sound The source sound pressure distribution results are displayed in two-dimensional, three-dimensional, and four-dimensional graphs to improve the visualization effect.

6. The present invention uses plant acoustic emission signals as important sensitive information to understand their growth status and disease status, and through the study of acoustic emission signals based on the distributed source boundary point method near-field acoustic holography technology, discovers and identifies diseases in time, for detection Plant diseases and their prevention and control provide a new method and new approach.

Description of drawings

Fig. 1 is the acoustic holography experimental model of the present invention;

Fig. 2 is the distribution figure of reconstruction spherical surface node of the present invention; Wherein, Fig. 2 (a) is the distribution figure of 14 nodes on the reconstruction sphere, Fig. 2 (b) is the distribution figure of 10 nodes on the reconstruction sphere;

Fig. 3 is the reconstruction effect diagram under the experimental parameters that the single sound source has 24 measuring points on the holographic surface and the number of nodes on the reconstruction spherical surface is 14; wherein, Fig. 3 (a) is the reconstruction sound pressure and theoretical sound pressure of the single sound source Pressure comparison chart, Fig. 3(b) is the 3D reconstruction sound pressure map of a single sound source, Fig. 3(c) is the rotation plane view of the 3D reconstruction sound pressure map of a single sound source, Fig. 3(d) is a 4D reconstruction sound pressure map of a single sound source picture;

Fig. 4 is the reconstruction effect diagram under the experimental parameters that the measurement points of the dual sound source on the holographic surface are 16 and the number of nodes on the reconstruction spherical surface is 10; wherein, Fig. 4 (a) is a comparison between the reconstruction sound pressure of the dual sound source and the theoretical sound pressure Fig. 4(b) is a three-dimensional reconstructed sound pressure map of two sound sources, Fig. 4(c) is a rotation plan view of a three-dimensional reconstructed sound pressure map of two sound sources, and Fig. 4(d) is a four-dimensional reconstructed sound pressure map of two sound sources;

Fig. 5 is a block diagram of the acoustic holography experimental system of the present invention.

Detailed ways

The present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings. It should be noted that this embodiment is illustrative, not restrictive, and cannot limit the protection scope of the present invention.

A method for detecting plant diseases based on near-field acoustic holography, the steps are:

⑴According to the acoustic emission phenomenon of plants, based on the near-field acoustic holography technology of the distributed source boundary point method, an experimental model of cylindrical-spherical non-conformal surface acoustic holography is established; this experiment is based on the "acoustic emission" phenomenon of plants The boundary point method near-field acoustic holography technology is used as the theoretical basis to establish a cylindrical-spherical non-conformal surface acoustic holography experimental model, as shown in Figure 1. A space Cartesian coordinate system is established with the axial center as the origin, where r _h is the radius of the holographic cylinder, _rs is the radius of the reconstructed spherical surface, L is the height of the holographic measurement surface along the axial direction, and h is the point source to the reconstruction of the fictitious special solution The distance of the nodes on the spherical surface distributes the special solution point sources in the conformal shape with the reconstruction surface and in the normal direction away from the nodes, and makes the number of special solution sources equal to the number of nodes on the reconstruction surface.

(2) Set the number of holographic measurement points and the number of reconstructed spherical nodes for a single sound source on the experimental model, and conduct simulation experiment analysis to select parameters; first, set 24 measurement points distributed on the holographic surface, and the angular distance of holographic points in the circumferential direction △ θ=π/4, the height distance in the z direction △Z=0.45m, and the holographic measurement surface is 0.9m high in the axial direction. The reconstructed sphere is evenly divided into 4 equal parts in the latitude and longitude directions, with a total of 14 nodes. The node distribution diagram of the reconstructed spherical surface is shown in Fig. 2(a). According to the size of the general plant radius, the radius of the holographic cylindrical surface is selected as 0.01m, the radius of the reconstructed spherical surface is 0.009m, f=80kHz, and the sound source is located at the position of 0.007m from the center of the sphere (at the origin) in the direction of the fourth node Carry out the simulation, and the reconstruction effect is shown in Figure 3 (a), (b), (c), and (d).

It can be seen from Figure 3(a) that the sound pressure value of node 4 is the highest, and the sound pressure values of other nodes conform to the law of sound pressure attenuation with distance, so it can be determined that there is a sound source. Combining Figures 3(b), 3(c), and 3(d), it can be seen that the angle between the sound source (the darkest position) and the x-axis is pi/2, and the angle between the sound source and the z-axis is pi/4. The height is 0.105m, which is consistent with the assumption.

The frequency of the sound source is taken at intervals of 20kHz within the range of 10-300kHz, and the radius of the holographic cylinder and the radius of the reconstruction sphere are respectively selected: 0.01m, 0.009m; 0.02m, 0.019m; 0.03m, 0.029m. Under different holographic cylindrical surface radii and corresponding reconstruction sphere radii, the distance from the sound source to the center of the sphere in different node directions is changed for multiple simulations. The results show that: for each sound source frequency, by adjusting the fictitious special solution point source The distance to the nodes on the reconstruction sphere can reconstruct any position of the sound source and has high reconstruction accuracy. At the same time, the distance h between the special solution point source and the nodes on the reconstruction sphere can be obtained when the best reconstruction effect is achieved. Selection rule: the sum of the distance from the sound source to the center of the sphere and h is equal to the radius of the reconstructed sphere.

However, for plants with a small radius, the distance between holographic measuring points in the circumferential direction is △ _max = 7.85 mm, which means that the maximum sensor installed on the holographic surface is about 7-8 mm, which brings inconvenience to the actual operation.

Secondly, in order to solve the problems of actual measurement difficulties and sensor selection, the number of measuring points is reduced to 16, the angular spacing of holographic points in the circumferential direction is △θ=π/2, and the height spacing of z-direction is △Z=0.3m. The axial height is 0.9m, and there are still 14 nodes on the reconstructed spherical surface. Also select different holographic cylindrical surface radii and corresponding reconstruction sphere radii, the range and value of the sound source frequency remain unchanged, the distance from the sound source to the center of the sphere is changed accordingly, and the simulation is carried out according to the selection rule of the parameter h, it can be seen that : Only when the ratio of the distance from the sound source to the center of the sphere to the radius of the corresponding holographic cylinder is greater than or equal to 0.4 can the reconstruction effect be better, and some individual frequencies can be accurately reconstructed only when the ratio is larger.

Again, without changing the number of measuring points, the number of measuring points is still 16, and the number of nodes on the reconstructed spherical surface is changed to 10, which are divided into 3 equal parts and 4 equal parts in the latitude and longitude directions respectively. The distribution of nodes on the reconstructed spherical surface is as follows Figure 2(b) shows. Under these experimental parameters, the radius r _h of the holographic cylinder, the frequency f of the sound source, and the distance d from the sound source to the center of the sphere are changed to conduct a simulation experiment. Some statistics of the simulation results are shown in Table 1:

Table 1 Statistics table of simulation results

The simulation results show that: under the experimental parameters, within the frequency range of 10-300kHz, by reasonably selecting h, any position of the sound source can be reconstructed, with high reconstruction accuracy, and the selection rule of the parameter h is the same as the above , and the radius of the holographic cylinder can be increased to 0.05m.

In addition, at this time, the number of holographic circumferential measuring points is reduced, and the measuring point spacing is relatively large. For plants with a small radius, the circumferential measuring point spacing △ _max = 15.7mm. At this time, the maximum sensor installed on the holographic surface is about 15-16mm , which improves the practical operability.

It can be seen from the simulation experiments that the reconstruction accuracy of the near-field acoustic holography of the distributed source boundary point method is affected by many factors, among which the key parameters include: the distance between the circumferential and axial measuring points on the holographic measurement surface, The number of upper nodes is the distance from the imaginary special solution point source to the nodes on the reconstructed sphere.

In the case of ensuring sufficient accuracy, the number of measuring points should be as few as possible to reduce the difficulty of actual operation and measurement cost, and considering the selection of sensors and hardware data processing, the number of measuring points on the holographic surface is finally selected to be 16. The directional angular spacing △θ=π/2, the height spacing △z=0.3m in the z direction, the holographic measurement surface is 0.9m high in the axial direction, and the experimental parameters with 10 nodes on the reconstruction surface are the reference parameters of the actual measurement system. Under the parameters, the sound source localization can be accurately realized in the frequency range of 10 ~ 300kHz.

(3) Based on the parameter selection results of the single sound source simulation, reconstruct the double sound source and verify the effectiveness of the method; since the acoustic emission signal does not necessarily occur uniquely, multiple frequency signals can be used to express different sound sources, and each frequency Represents the signal after each acoustic emission signal passes through a narrowband filter. According to the parameter selection results of single sound source simulation in this experiment, multi-sound source reconstruction is carried out for f ₁ =100kHz and f ₂ =150kHz, and the reconstruction effects are shown in Figure 4 (a), (b), (c), (d) shown.

From Figure 4(a), it can be seen that the sound pressure values of nodes 7 and 9 are higher than those of other nodes, and nodes 6, 7, 8, and 9 are on the same section. If there is a sound source point, it must be around the middle of nodes 7 and 9. The sound pressure generated by the sound source at nodes 6 and 8 at this position is similar to that at nodes 7 and 9. The simulation results show that their The sound pressure values are very different, indicating that there may be two sound sources, which are at nodes 7 and 9 respectively. In addition, since the sound pressure values of other nodes are lower than those of nodes 7 and 9, and conform to the law of sound pressure attenuation with distance, it can be determined that there are two sound sources. Combining Figures 4(b), 4(c), and 4(d), it can be seen that the height direction of the two sound sources is around 0.29m, and the angle between one sound source and the x-axis is 0, and the angle between the sound source and the z-axis The angle is 2/3*pi, the angle between another sound source and the x-axis is pi, and the angle between the other sound source and the z-axis is 2/3*pi, which can accurately identify the sound source. The simulation results of other parameters with two sound sources are similar, which verifies the effectiveness of the method.

⑷Comprehensive simulation results show that under the experimental parameters of holographic cylinder radius (0.01m~0.05m) and sound source frequency (10~300kHz), the method can accurately identify and locate the location of the acoustic emission source, and draw The selection rule of the distance from the fictitious special solution point source to the node on the reconstructed sphere: the sum of the distance from the sound source to the center of the sphere and the distance h from the fictitious special solution point source to the node on the reconstructed sphere is equal to the radius of the reconstructed sphere, which clearly affects the reconstruction The main parameters of accuracy: the distance between the circumferential and axial measuring points on the holographic measurement surface, the number of nodes on the reconstructed sphere, and the distance from the fictional special solution point source to the nodes on the reconstructed sphere.

⑸According to the best experimental parameters of the simulation, the acoustic holographic experimental system is designed, and the acoustic emission signal is collected by the sensor array, and then enters the AD sampling circuit after amplification and filtering, and the DSP provides the AD sampling control signal to realize synchronous sampling by multiple ADCs. The A/D conversion data is transmitted to the DSP in parallel, and uploaded to the host computer after processing. The host computer uses Labview as the software platform to analyze and process the data, complete the reconstruction of the sound source, and display the sound field information in a graphical way. Position the sound The location of the emission source can determine the degree of plant disease stress.

According to the experimental parameters obtained from the simulation results, considering the reasonable arrangement of experimental parameters and actual operation and other factors, the best experimental parameters are selected and the acoustic holographic experimental system is designed, as shown in Figure 5.

A plant disease detection system based on near-field acoustic holography technology, mainly including acoustic emission sensor array, sensor bracket, preamplifier, A/D converter, DSP processor, host computer sound source identification algorithm analysis and processing software, etc.

Since the acoustic emission signal is an instantaneous and non-stationary signal, a snapshot method is needed to obtain the complex sound pressure on the holographic surface. According to the parameter selection results of the above-mentioned simulation experiments, according to the requirement that the angular spacing of 16 holographic points in the circumferential direction △θ=π/2, the height spacing in the z direction △z=0.3m, and the height of the holographic measurement surface along the axial direction is 0.9m, it is reasonable Array of sensors.

Considering the characteristics of the acoustic emission signal and the influence of environmental noise, the ultrasonic frequency band above 20kHz is generally selected for analysis, combined with the requirements of the sampling rate, the acoustic emission sensor and the broadband acoustic emission preamplifier with a resonant frequency of 70kHz and above can be selected. The signal output and power input of the preamplifier are collinear, and a high-pass filter circuit needs to be designed to separate the power and signal. The AD converter can use the AD7656 of the American ADI company. Shorting the three pairs of conversion start pins can realize 6 channels of simultaneous sampling with a sampling rate of up to 250ksps and a resolution of 16 bits. Since 16 measuring points are required, three piece. All the AD conversion start pins of the three AD7656s share one GPIO port of the DSP to control 16 channels of synchronous sampling and conversion. The three address lines of the DSP can be used as the chip selection signal of the AD. It has two interface modes of parallel and serial with the DSP. , the platform uses the parallel interface mode, the 16-bit data line of AD7656 is directly connected with the 16-bit data line of DSP, after all conversions are completed, the DSP is triggered to enter an external interrupt to read data.

The main stem of the plant is selected as the research object in the experiment, and the sensor array is fixed on the bracket and close to the surface of the main stem of the plant. The multi-channel signal collected by the sensor array is amplified and filtered before entering the AD sampling circuit, and the DSP provides the AD sampling signal. The control signal realizes synchronous sampling by multiple ADCs, and the A/D conversion data is transmitted to the DSP in parallel, and uploaded to the host computer after processing. The upper computer uses Labview as the software platform, which contains a large number of tools and functions for data acquisition, analysis, display and storage, and uses the MATLAB Script node to realize the call of Matlab language in Labview; the software platform uses graphical programming language to process data Carry out auto-spectrum and cross-spectrum calculations, taking the first sensor as a reference point, cross-spectrum the signals collected by each measurement sensor and the signal collected by the first sensor to obtain the sound pressure phase at each measurement point, and the signals collected by each measurement sensor The sound pressure amplitude at each measuring point is obtained by making an autospectrum, and the complex sound pressure of the holographic surface is obtained according to the sound pressure calculation formula. Use the MATLAB Script node in Labview, and use the holographic complex sound pressure as the input in the algorithm to analyze and process the data, complete the reconstruction of the sound source, and use the powerful drawing function of matlab to reconstruct the sound pressure distribution of the sound source after analysis The results are displayed in two-dimensional, three-dimensional and four-dimensional graphs to improve the visualization effect.

Claims (9)

1. A system for detecting plant diseases based on near-field acoustic holography, characterized in that: comprise an acoustic emission sensor array, a broadband acoustic emission preamplifier, an AD converter, a DSP processor, an upper computer and a graphic display connected in sequence, Run the sound source identification algorithm analysis and processing software in the host computer. 2. The system for detecting plant diseases based on near-field acoustic holography technology according to claim 1, characterized in that: the acoustic emission sensor array collects acoustic emission signals, amplifies and filters them through the broadband acoustic emission preamplifier and enters the AD The converter performs multi-channel signal conversion, the AD converter converts the data and transmits it to the DSP processor, and the processed data is uploaded to the host computer. The host computer uses Labview as the software platform to perform data analysis and processing, and complete the reconstruction of the sound source. The sound field information is displayed graphically to locate the source of the sound emission and determine the degree of plant disease stress. 3. The system for detecting plant diseases based on near-field acoustic holography according to claim 1, characterized in that: the acoustic emission sensor and the broadband acoustic emission preamplifier select a resonant frequency of 70kHz and above, and use a high-pass filter circuit to perform Power and signal separation. 4. The system for detecting plant diseases based on near-field acoustic holography technology according to claim 1, characterized in that: the AD converter is AD7656 from American Analog Devices. 5. The system for detecting plant diseases based on near-field acoustic holography technology according to claim 1, characterized in that: the DSP processor adopts TMS320VC5509, the highest main frequency can reach 144MHz, and data reading can be completed three times in one cycle operation and two data write operations. 6. The system for detecting plant diseases based on near-field acoustic holography according to claim 1 is characterized in that: the sound source recognition algorithm analysis and processing software of the host computer is based on Labview as a software platform, including a large number of tools and The function is used for data acquisition, analysis, display and storage, and the MATLAB Script node is used to realize the call of Matlab language in Labview. 7. A method for detecting plant diseases based on near-field acoustic holography, characterized in that: the steps are: ⑴According to the acoustic emission phenomenon of plants, based on the near-field acoustic holography technology of the distributed source boundary point method, a cylindrical-spherical non-conformal surface acoustic holographic experimental model is established; (2) Set the number of holographic measurement points and the number of reconstructed spherical nodes for a single sound source on the experimental model, and conduct simulation experiment analysis to select parameters; (3) According to the parameter selection results of the single sound source simulation, the dual sound sources are reconstructed, and the validity of the method is verified; (4) Through the analysis of the experimental data, the radius of the holographic cylinder, the applicable range of the sound source frequency, the selection rule of the distance from the fictitious special solution point source to the node on the reconstruction sphere, and the main parameters affecting the reconstruction accuracy are given; ⑸According to the best experimental parameters of the simulation, the acoustic holographic experimental system is designed, and the acoustic emission signal is collected by the sensor array, and then enters the AD sampling circuit after amplification and filtering, and the DSP provides the AD sampling control signal to realize synchronous sampling by multiple ADCs. The A/D conversion data is transmitted to the DSP in parallel, and uploaded to the host computer after processing. The host computer uses Labview as the software platform to analyze and process the data, complete the reconstruction of the sound source, and display the sound field information in a graphical way. Position the sound The location of the emission source can determine the degree of plant disease stress. 8. The method for detecting plant diseases based on near-field acoustic holography according to claim 7, characterized in that: the cylindrical-spherical non-conformal surface acoustic holography experimental model is to establish a spatial rectangular coordinate with the axial center as the origin system, where r _h is the radius of the holographic cylindrical surface, _rs is the radius of the reconstructed sphere, L is the height of the holographic measurement surface along the axial direction, h is the distance from the imaginary special solution point source to the node on the reconstructed sphere, and the special solution The point sources are distributed conformally to the reconstruction surface and in the direction away from the normal of the nodes, and the number of specific solution sources is equal to the number of nodes on the reconstruction surface. 9. The method for detecting plant diseases based on near-field acoustic holography according to claim 7, characterized in that: the method for setting the number of holographic measuring points and the number of reconstructed spherical nodes on the experimental model for a single sound source is: A single sound source has 24 holographic measuring points and 14 nodes on the reconstructed sphere, or 16 holographic measuring points and 14 nodes on the reconstructed sphere, or 16 holographic measuring points and 10 nodes on the reconstructed sphere.

Images