A crop growth drought stress detection system and method

Abstract

The invention discloses a crop growth drought stress detection system and method, belonging to the technical field of facility agriculture biological information detection. The system comprises an acquisition platform, a multi-spectral polarization imaging system, a power supply system, a control module, a whiteboard automatic correction device, a USB signal receiver, a signal reading circuit, a storage module and a processing module. By developing a portable multi-spectral polarization imaging system, optimizing data processing and fusion algorithms, the state of plant drought stress can be obtained in time, and accurate data support can be provided for agricultural production.

Description

Technical Field

[0001] The present invention relates to the technical field of facility agriculture biological information detection, and more particularly to a crop growth drought stress detection system and method. Background Art

[0002] With the rapid development of modern agriculture and the intensification of global climate change, plant drought stress has become an important factor affecting crop yield and quality. In order to timely detect and effectively respond to plant drought stress, the agricultural field continues to explore and innovate detection methods to provide more accurate water management solutions. In the detection of drought stress in plants, the main methods currently used include physiological and biochemical indicator detection methods, soil moisture detection methods, remote sensing technology, spectroscopy and visual observation, etc., but these methods have certain limitations:

[0003] The physiological and biochemical index detection method evaluates the moisture status of crops by detecting chlorophyll content, transpiration rate, stomatal conductance, relative water content, etc. However, the cost of professional equipment and technicians is high and the detection process is complicated. The soil moisture detection method uses soil moisture sensors, soil water potential meters and other equipment to detect the moisture content in the soil in real time. However, due to the difference between the soil moisture status and the actual moisture absorbed and utilized by crops, it cannot directly reflect the moisture demand of crops. Remote sensing technology and spectral visual observation methods are the main methods for crop drought stress in recent years. However, due to the relatively high spatial resolution requirements of remote sensing data, it is impossible to detect plants in a small area. The spectral point source method and the visual reliance on plant color texture and other methods cannot fully characterize the characteristic information of plant drought stress, thus limiting the further development and application of plant drought stress detection.

[0004] Multispectral imaging technology overcomes the limitations of point source detection of spectral technology and provides richer plant information. At the same time, polarization imaging technology uses the polarization characteristics of light to obtain microstructural information that traditional imaging methods cannot provide. Multispectral polarization imaging technology provides a new solution for crop drought stress detection, which can achieve non-destructive detection of crops and improve the accuracy and timeliness of detection, providing strong support for precision agricultural production.

[0005] At present, there are some relevant applications in non-destructive testing of plants based on multispectral and polarization image technology in China. When acquiring polarization spectrum information, the polarization optical system of the Chinese patent "CN 103018179A" is fixedly connected to the near-infrared detection system and fixed at a specific angle. Therefore, the information collected by the device only contains a specific angle of polarization, and the information provided by the polarization optical system is not fully utilized. The Chinese patent "CN 103048266A" is relatively more complete in function and data processing, and uses more information. However, the biggest disadvantage of this system is that it cannot realize outdoor real-time detection, because the system involves laboratory instruments such as scanning electron microscopes and microscopes, which need to be used in conjunction with laboratory operations. And the system does not have the function of drought stress detection. The investigation found that although spectral and polarization technologies have played an important role in plant stress, the methods and settings of polarization and spectral fusion detection are not common in the application of plant drought or water stress. In summary, these technologies still have certain limitations, such as incomplete information acquisition, high cost, high technical requirements, and inability to realize outdoor real-time detection. The detection technology of plant drought stress is constantly developing, and multispectral and polarization imaging technologies provide new possibilities for plant drought stress.

[0006] Therefore, how to provide a crop growth drought stress detection system and method that can solve the above problems is an issue that technical personnel in this field urgently need to solve. Summary of the invention

[0007] In view of this, the present invention provides a crop growth drought stress detection system and method, which are used to solve the technical problems existing in the above-mentioned prior art.

[0008] In order to achieve the above object, the present invention provides the following technical solutions:

[0009] A crop growth drought stress detection system, comprising:

[0010] Acquisition platform, multi-spectral polarization imaging system, power supply system, control module, whiteboard automatic correction device, USB signal receiver, signal reading circuit, storage module and processing module;

[0011] The acquisition platform is an adjustable movable door-type truss structure with a fixed pulley, and the width and height are adjustable. The geometric center position of the upper end of the truss is used to fix the multi-spectral polarization imaging system, and the two side positions are used to fix the power supply system and the whiteboard automatic correction device respectively;

[0012] The multi-spectral polarization imaging system comprises, from top to bottom, a CMOS image sensor, a liquid crystal tunable filter, a polarizer, and a lens group;

[0013] The polarizer includes a polarization drive circuit, a small gear, a large gear and a polarizer, and the polarizer and the liquid crystal tunable filter respectively adjust the polarization state and the spectral range of light;

[0014] The CMOS image sensor is used to convert the received optical signal into an electrical signal and store it in the form of a digital image;

[0015] The CMOS image sensor works together with the liquid crystal tunable filter and the polarizer to capture light information of different wavelengths and record light information of different polarization angles, thereby generating image data containing rich spectral and polarization data;

[0016] The power supply system supplies power to the multi-spectral polarization imaging system and the whiteboard automatic correction device;

[0017] The control module is located on the mobile client, allowing the user to set remote commands through the mobile client, which are received by the USB signal receiver and used to adjust the parameters and working status of the crop growth drought stress detection system. The mobile client provides a user-friendly user interface for the control module, allowing the user to remotely set system parameters, monitor system status and receive real-time feedback;

[0018] The whiteboard automatic correction device has a built-in signal receiver, and the whiteboard automatic correction device can rotate between 0° and 90°. When the control module remote command is in the whiteboard correction mode, it is unfolded to the 0° position for correction, and when the control module remote command is in the data collection mode, it is retracted to the 90° position;

[0019] The USB signal receiver receives the instruction of the control module and adjusts the relevant parameters;

[0020] Specifically, the built-in signal receiver of the automatic whiteboard correction integrates a signal reading circuit and is considered as a built-in embedded body, which is not the same device as the USB signal receiver mentioned in the context.

[0021] The signal reading circuit reads the image data captured by the CMOS image sensor and transmits it to the storage module and the processing module;

[0022] The storage module is a USB flash drive, which stores image data acquired from the CMOS image sensor, the image data including all digital images stored by the CMOS sensor that are converted from light signals of all bands and different polarization angles into electrical signals;

[0023] The processing module performs data preprocessing and analysis modeling on the digital images stored in the storage module. The data preprocessing includes extracting the band information and reflectivity of the image, generating an image including image background segmentation, polarization angle, polarization degree and intensity. The analysis modeling includes analyzing the spectral characteristics and polarization information in the multispectral image, establishing a crop growth index prediction model, establishing a correspondence between the image feature information of the drought stress gradient and the polarization angle, and based on the correspondence, constructing a crop drought stress multispectral polarization detection model.

[0024] Preferably, the power supply system is a mobile power supply or a solar power generation system.

[0025] Preferably, the control module is located in the mobile client, allowing the user to set remote commands through the mobile client, which are received by the USB signal receiver and used to adjust the parameters and working status of the crop growth drought stress detection system. The mobile client provides a user-friendly user interface for the control module, allowing the user to remotely set system parameters, monitor system status and receive real-time feedback.

[0026] Preferably, the polarizer is an electrically controlled polarization rotator driven by a large gear and a small gear;

[0027] Wherein, the pinion is a driving wheel with a motor and a driving circuit, and has an infrared receiver for receiving infrared control signals;

[0028] The large gear is a driven wheel with a hollow geometric center. The geometric center is equipped with a full-band polarizer. The polarizer rotates 0-360° around the geometric center axis of the CMOS image sensor to adjust the polarization angle, thereby changing the polarization state of the incident light. Through this adjustment, the polarizer can enhance the light signal at a specific wavelength and receive light signals with different specific polarization directions, so that the CMOS image sensor generates an image containing different spectral and polarization data.

[0029] Preferably, the liquid crystal tunable filter contains an electronic unit and a liquid crystal layer, and the USB signal receiver receives the instruction of the control module, and the electronic unit adjusts the electric field of the liquid crystal layer, thereby changing the refractive index of the liquid crystal. This change in refractive index will cause the transmittance of the liquid crystal tunable filter to light of different wavelengths to change.

[0030] Preferably, the data preprocessing includes extracting the band information and reflectivity of the image, generating an image including image background segmentation, polarization angle, polarization degree and intensity, and the analysis and modeling includes analyzing the spectral characteristics and polarization information in the multispectral image, establishing a crop growth index prediction model, establishing a correspondence between the image feature information of the drought stress gradient and the polarization angle, and based on the correspondence, constructing a multispectral polarization detection model for crop drought stress.

[0031] The processing module performs data preprocessing and analysis modeling on the image data stored in the storage module to generate an image containing rich spectral and polarization data. The data preprocessing includes band channel alignment, crop growth index extraction, image background segmentation, and polarization angle, polarization intensity and intensity image synthesis. Based on the image obtained by data preprocessing and the data analysis, the spectral characteristics and polarization information in the multispectral image are established, and the corresponding relationship between the drought stress gradient and the image feature information of the polarization angle is established. Based on the corresponding relationship, a crop drought stress multispectral polarization detection model is constructed, thereby obtaining the relationship between the degree of drought stress and the spectral and polarization characteristics.

[0032] Preferably, the band channel registration adopts a maximum likelihood estimation method to align the images in spatial position by estimating translation, rotation and scale transformation parameters between the images.

[0033] Preferably, the crop growth index extraction includes: Normalized Difference Vegetation Index NDVI and Normalized Difference Water Index NDWI.

[0034] Preferably, the polarization angle, polarization degree and intensity image include intensity images of four different polarization angles, as well as Stokes parameters and polarization degree;

[0035] The Stokes parameter calculation formula is:

[0036] ;

[0037] The formula for calculating the degree of polarization is:

[0038] ;

[0039] The formula for calculating the polarization angle is:

[0040] ;

[0041] In the formula, Represent the horizontal and vertical directions respectively. Represents the total intensity of the light field, which is the total intensity of 0° polarized light Perpendicular to the direction of propagation The sum of the squares of the electric field strength on the plane; express and Direction intensity difference of linear polarized light; express and Direction linear polarized light intensity difference.

[0042] Preferably, the analysis modeling is support vector machine (SVM) modeling, and the data set is a multispectral polarization image and its reflectivity screened and processed by an image processing system. The multispectral image and reflectivity are associated with the degree of drought stress, and an independent validation data set is used to evaluate the predictive ability of the model. The parameters of the model are adjusted according to the validation results to improve the accuracy and generalization ability of the model.

[0043] A method for detecting drought stress of crop growth, comprising:

[0044] S1: Plant crops on a plot with a width comparable to the adjustable truss structure;

[0045] S2: subject crops to a gradient of drought stress;

[0046] S3: Move the acquisition platform equipped with the multi-spectral polarization imaging system to the center of a corn plant plot under a certain stress treatment and fix the fixed pulley;

[0047] S4: Connect the multi-spectral polarization imaging system to the control module on the mobile client through the WIFI hotspot, set the data acquisition mode, and set the data capture method of the multi-spectral polarization imaging system;

[0048] S5: When the data acquisition mode is in the whiteboard calibration mode, the whiteboard automatic calibration device is rotated to place the whiteboard within the field of view, and the multispectral polarization imaging system automatically focuses and collects whiteboard data under the coordinated action of various components;

[0049] When the data acquisition mode is in the data acquisition mode, the whiteboard automatic correction device is retracted, and the multispectral polarization imaging system automatically focuses and collects crop plant data;

[0050] S6: The USB signal receiver receives the remote command and adjusts the parameters of the liquid crystal tunable filter;

[0051] S7: setting and changing the required polarization angle of the polarizer, and using the multispectral polarization imaging system to collect multispectral polarization data of different drought stress samples, and saving them in the storage module in the form of image data;

[0052] S8: Data preprocessing of the multispectral image is performed by transmitting the data of the storage module to the processing module of the computer, including band channel registration, crop growth index extraction, image background segmentation, and polarization angle, polarization intensity and intensity image synthesis, to generate an image containing rich spectral and polarization data;

[0053] S9: In the processing module, based on the spectrum and image data obtained by the processing in S8, the spectral characteristics and polarization information in the multispectral image are analyzed, a crop growth index prediction model is established according to the reflectance of different wavelengths, and a correspondence between the image feature information of the drought stress gradient and the polarization angle is established according to the polarization image information at different wavelengths. Based on the correspondence, a crop drought stress multispectral polarization detection model is constructed, and the relationship between the degree of drought stress and the spectrum and polarization characteristics is obtained through the crop drought stress multispectral polarization detection model;

[0054] S10: Analyze and evaluate the stress state of crops based on the crop drought stress multispectral polarization detection model obtained by the processing module.

[0055] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a crop growth drought stress detection system that can simultaneously collect multi-spectral information and polarization information of crops. The device supports real-time detection in the field without relying on a laboratory environment. The use of an adjustable polarizer can capture polarization information at different angles, thereby more comprehensively analyzing the spectral characteristics and moisture status of crop leaves. The specific beneficial effects are as follows:

[0056] (1) Comprehensive information acquisition: The present invention can simultaneously capture the reflectance information of crops in different spectral bands and the image features at different polarization angles. This multi-dimensional information acquisition method can more comprehensively reflect the changes in the physiological state of crops under drought stress compared to single spectral or polarization imaging methods.

[0057] (2) Improved diagnostic accuracy: The present invention can more accurately identify and quantify changes in plant physiological and ecological processes in crop drought stress diagnosis. Spectral information can reveal changes in the chemical composition of leaves, while polarization information can reflect the microstructure, water content, and cell arrangement of the leaf surface. The combination of the two makes the diagnostic results more reliable and reduces the possibility of misjudgment.

[0058] (3) Nondestructive testing technology: The present invention adopts a non-contact, non-destructive testing method and does not cause any physical damage to the crops.

[0059] (4) Improving agricultural management efficiency: By detecting crop drought stress in real time, farmers and agricultural managers can adjust irrigation strategies in a timely manner and optimize water resource allocation, thereby improving agricultural production efficiency and crop yields. In addition, accurate drought stress diagnosis can also help predict pest and disease risks, take preventive measures, reduce pesticide use, and promote sustainable agricultural development.

[0060] (5) Promote the progress of agricultural science and technology: The application of this invention is not limited to the diagnosis of crop drought stress, but can also be extended to the monitoring of other crops and ecosystems. Its innovative technical ideas and methodology provide new research tools and means for the fields of agricultural remote sensing, precision agriculture, ecological monitoring, etc., and promote the progress and development of agricultural science and technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0062] Figure 1 A schematic diagram of the structure of a crop growth drought stress detection system provided by the present invention;

[0063] Figure 2 This is a schematic diagram of the image processing and modeling process provided by the present invention.

[0064] In the attached figure:

[0065] 1-acquisition platform, 2-power system, 3-whiteboard automatic correction device, 4-small gear, 5-polarizer drive circuit, 6-CMOS image sensor, 7-polarizer, 8-large gear, 9-liquid crystal tunable filter, 10-USB signal receiver, 11-lens group, 12-light shield, 13-mobile client, 14-field of view of the crop to be measured, 15-storage module, 16-processing module, 17-signal reading circuit. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] The purpose of the present invention is to provide a crop growth drought stress detection system, comprising: an acquisition platform, a multispectral polarization imaging system, a power supply system, a control module, a whiteboard automatic correction device, a USB signal receiver, a signal reading circuit, a storage module and a processing module; by developing a portable multispectral polarization imaging system, optimizing data processing and fusion algorithms, timely acquiring the state of plant drought stress, and providing accurate data support for agricultural production.

[0068] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0069] See also Figure 1 The embodiment of the present invention discloses a crop growth drought stress detection system including an acquisition platform 1, a multi-spectral polarization imaging system, a power supply system 2, a control module 13, a whiteboard automatic correction device 3, a USB signal receiver 10, a signal reading circuit 17, a storage module 15 and a processing module 16.

[0070] The multi-spectral polarization imaging system includes, from top to bottom, a CMOS image sensor 6, a liquid crystal tunable filter 9, a polarizer, and a lens group 11;

[0071] The polarizer includes a polarization drive circuit, a small gear 4, a large gear 8, and a polarizer 7, and the polarizer and the liquid crystal tunable filter adjust the polarization state and the spectral range of light respectively;

[0072] The CMOS image sensor is used to convert the received optical signal into an electrical signal and store it in the form of a digital image;

[0073] The CMOS image sensor works together with the liquid crystal tunable filter and the polarizer to capture light information of different wavelengths and record light information of different polarization angles, thereby generating image data containing rich spectral and polarization data;

[0074] The collection platform 1 is an adjustable movable door-type truss with a stopper fixed pulley, with a minimum size of 0.5m×2.0m×1.5m, and the left and right widths can be adjusted by 0.6m each, and the height can be adjusted by 0.6m. The stopper fixed pulley is used for the movement and stop of the collection platform.

[0075] The acquisition platform 1 is an adjustable truss structure with adjustable width and height. The geometric center position of the upper end of the truss is used to fix the multispectral imaging system, which is composed of a CMOS image sensor 6, a polarizer driven by a polarization driver 5, a large gear 8, and a small gear 4, a liquid crystal tunable filter 9, and a lens group 11 from top to bottom. The present invention can be used to control multispectral polarization imaging according to current, and can be used to capture spectral imaging information of any band.

[0076] The left and right sides of the acquisition platform 1 are respectively used to fix the power supply system 2 and the whiteboard automatic correction device 3, which are respectively used to supply power to the crop growth drought stress detection system and to correct the reflectivity of the multi-spectral polarization imaging system.

[0077] The collection platform 1 is an adjustable movable door-type truss with a stopper fixed pulley, with a minimum size of 0.5m×2.0m×1.5m, and the left and right widths can be adjusted by 0.6m each, and the height can be adjusted by 0.6m. The stopper fixed pulley is used for the movement and stop of the collection platform.

[0078] Specifically, the light source used in the crop growth drought stress detection system is full-band sunlight or halogen light source, which can provide parallel line light source.

[0079] The control module is located in the mobile client, which connects the multispectral polarization imaging system to the mobile client through a WIFI hotspot. The control module can adjust the state of the multispectral polarization imaging system, including: data acquisition mode, wavelength range, capture time, camera height, image storage format, etc.

[0080] The CMOS image sensor 6 is a component for capturing images. It converts the received optical signal into an electrical signal and stores it in digital form. The CMOS image sensor 6 cooperates with the liquid crystal tunable filter 9 and the polarizer driven by the polarization driver 5, the large gear 8, and the small gear 4 to capture the light information of different wavelengths and record the light information of different polarization angles, thereby generating an image containing rich spectral and polarization data.

[0081] The polarizer is a polarization electrically controlled rotator driven by a polarization driver 5, a large gear 8, and a small gear 4. The small gear 4 is a driving wheel with a motor and a driving circuit, and has an infrared receiver for receiving infrared control signals. The large gear 8 is a driven wheel with a hollow geometric center. The geometric center is equipped with a full-band polarizer, and the polarizer 7 can rotate 0-360° around the geometric center axis of the CMOS image sensor 6 to adjust the polarization angle, adjust the polarization state of the light incident on the system, and enhance the signal at a specific wavelength, thereby collecting light information in different polarization states.

[0082] Specifically, the specific wavelength refers to the wavelength of light that can be enhanced by wavelength screening through a liquid crystal tunable filter and adjusting the polarization angle. Different substances have different absorption and reflection characteristics for light of different wavelengths.

[0083] The liquid crystal tunable filter 9 has a USB signal receiver, which can adjust the external current signal by receiving remote commands to change its transmission characteristics, thereby realizing tuning and filtering of light of a specific wavelength. It is mainly used to adjust the spectral range of the sensor, and the adjustment range can be the full band, especially the visible light to near infrared band, specifically 400-1700nm, including any band or band combination within this band range, so as to realize the spectral analysis task of different wavelengths from visible to near infrared.

[0084] The whiteboard automatic correction device 3 is a correction platform that can rotate left and right or flip up and down, fixed at a fixed height on one side of the acquisition platform 1, and the longitudinal movable angle of the fixed point is 0-90°. When the multispectral polarization imaging system is in the whiteboard correction mode, the automatic correction device receives instructions and automatically rotates the correction whiteboard to the 0° position, facing the multispectral polarization imaging system; when the multispectral polarization imaging system is in the data acquisition mode, the automatic correction device receives instructions and automatically rotates the correction whiteboard to the 90° position.

[0085] The power supply system is a mobile power supply or a solar power generation system, which supplies power to the multi-spectral polarization imaging system and the whiteboard automatic correction device.

[0086] The storage module 15 is a USB flash drive, which is used to store all data information collected during the operation of the multi-spectral polarization imaging system.

[0087] The processing module 16 includes image preprocessing and modeling analysis. This module can process the multispectral image data collected by the CMOS image sensor 6, including band channel registration, crop growth index extraction, image background segmentation, and polarization angle, polarization intensity and intensity image synthesis. This module can also be used for analysis and modeling, including analyzing the spectral characteristics and polarization information in the multispectral image, establishing a crop growth index prediction model, establishing a corresponding relationship between the image feature information of the drought stress gradient and the polarization angle, and based on the corresponding relationship, constructing a crop drought stress multispectral polarization detection model.

[0088] The band channel registration adopts the maximum likelihood estimation method, which estimates the translation, rotation and scale transformation parameters between images to align the images in space.

[0089] The crop growth index extraction includes the normalized difference vegetation index (NDVI) and the normalized difference water index (NDWI).

[0090] The U-Net segmentation model is used for image background segmentation, and the background information is set to black to extract the foreground information and reduce the interference of the background on the foreground.

[0091] The polarization angle, polarization degree and intensity images include intensity images of four different polarization angles, as well as Stokes parameters and polarization degree.

[0092] The Stokes parameter calculation formula is:

[0093]

[0094] The polarization degree calculation formula is:

[0095]

[0096] The polarization angle calculation formula is:

[0097]

[0098] In a specific embodiment, a crop growth drought stress detection method based on a crop growth drought stress detection system provided by the present invention comprises:

[0099] S1: Plant crops on a plot with a width comparable to the adjustable truss structure;

[0100] S2: subject crops to a gradient of drought stress;

[0101] S3: Move the crop growth drought stress detection system with the power supply system, automatic correction module and multi-spectral polarization imaging system installed to the center of a corn plant plot with a certain stress treatment and fix the fixed pulley;

[0102] S4: Connect the multi-spectral polarization imaging system to the control module on the mobile client through the WIFI hotspot, set the data acquisition mode, and set the data capture method of the multi-spectral polarization imaging system;

[0103] S5: When the data acquisition mode is in the whiteboard calibration mode, the whiteboard automatic calibration device is rotated to place the whiteboard within the field of view, and the multispectral polarization imaging system automatically focuses and collects whiteboard data;

[0104] When the data acquisition mode is in the data acquisition mode, the whiteboard automatic correction device is retracted, and the multispectral polarization imaging system automatically focuses and collects crop plant data;

[0105] S6: The USB signal receiver receives the remote command and adjusts the parameters of the liquid crystal tunable filter;

[0106] S7: setting and changing the required polarization angle of the polarizer, and collecting multispectral polarization data of different drought stress samples using a multispectral polarization imaging system;

[0107] S8: Preprocess the multispectral image through the processing module, including denoising, correction and other steps. Analyze the spectral features and polarization information in the multispectral image, establish the corresponding relationship between the image feature information of the drought stress gradient and the polarization angle, and build a crop drought stress multispectral polarization detection model based on the corresponding relationship. For details, see Figure 2 The image processing and modeling process shown in the figure obtains the relationship between the degree of drought stress and the spectral and polarization characteristics through the crop drought stress multi-spectral polarization detection model;

[0108] S9: Analyze and evaluate the stress status of crops based on the crop drought stress multispectral polarization detection model.

[0109] In a specific embodiment, taking corn crops as an example, the following method steps are performed:

[0110] S1: Plant crops on a plot with a width comparable to the adjustable truss structure;

[0111] S2: subject crops to a gradient of drought stress;

[0112] S3: Move the acquisition platform equipped with the multi-spectral polarization imaging system to the center of a corn plant plot under a certain stress treatment and fix the fixed pulley;

[0113] S4: Connect the multi-spectral polarization imaging system to the control module on the mobile client through the WIFI hotspot, set the data acquisition mode, and set the data capture method of the multi-spectral polarization imaging system;

[0114] S5: When the data acquisition mode is in the whiteboard calibration mode, the whiteboard automatic calibration device is rotated to place the whiteboard within the field of view, and the multispectral polarization imaging system automatically focuses and collects whiteboard data under the coordinated action of various components;

[0115] When the data acquisition mode is in the data acquisition mode, the whiteboard automatic correction device is retracted, and the multispectral polarization imaging system automatically focuses and collects crop plant data;

[0116] S6: The USB signal receiver receives the remote command and adjusts the parameters of the liquid crystal tunable filter;

[0117] S7: setting and changing the required polarization angle of the polarizer, and using the multispectral polarization imaging system to collect multispectral polarization data of different drought stress samples, and saving them in the storage module in the form of image data;

[0118] S8: Data preprocessing of the multispectral image is performed by transmitting the data of the storage module to the processing module of the computer, including band channel registration, crop growth index extraction, image background segmentation, and polarization angle, polarization intensity and intensity image synthesis, to generate an image containing rich spectral and polarization data;

[0119] S9: In the processing module, based on the spectrum and image data obtained by the processing in S8, the spectral characteristics and polarization information in the multispectral image are analyzed, a crop growth index prediction model is established according to the reflectance of different wavelengths, and a correspondence between the image feature information of the drought stress gradient and the polarization angle is established according to the polarization image information at different wavelengths. Based on the correspondence, a crop drought stress multispectral polarization detection model is constructed, and the relationship between the degree of drought stress and the spectrum and polarization characteristics is obtained through the crop drought stress multispectral polarization detection model;

[0120] S10: Analyze and evaluate the stress state of crops based on the crop drought stress multispectral polarization detection model obtained by the processing module.

[0121] The corn canopy is at a fixed height from the multispectral camera, which is adjusted as the corn grows. The width is adjusted as the width of the planting plot, so that the multispectral polarization camera is always on the geometric center axis of the rectangular plot.

[0122] The information of collecting different drought stress samples includes: collecting multispectral polarization images at four polarization angles of 0°, 45°, 90° and 135° at the same point; collecting any fixed point of a corn field with different drought stress gradients, which is within the imaging range of the camera and is a corn plant treated with the same stress gradient.

[0123] The processing of the captured multi-spectral polarization images refers to using the image processing system to perform channel registration on the spectral polarization multi-images, extract the reflectance, polarization angle, polarization degree, and intensity images of corn, fuse and analyze these images, and select the characteristic image that can reflect the greatest stress change of corn caused by drought stress. Multi-information fusion analysis is performed on the characteristic image, and a multi-information fusion model of multi-spectral polarization-image is constructed to establish a detection model for the drought stress state of corn.

[0124] The corn drought stress gradient refers to the national farmland moisture content standard. The evaluation of corn drought stress status is guided by the modeling results and the performance of the detection model is evaluated with reference to the set stress gradient.

[0125] The multispectral polarization detection model for corn drought stress is a support vector machine (SVM) model. The dataset used is the multispectral polarization image and its reflectivity after being screened and processed by the image processing system. The multispectral image and reflectivity are associated with the degree of drought stress, and an independent validation dataset is used to evaluate the prediction ability of the model. The parameters of the model are adjusted according to the validation results to improve the accuracy and generalization ability of the model.

[0126] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0127] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A crop growth drought stress detection system, characterized in that: include: Acquisition platform, multi-spectral polarization imaging system, power supply system, control module, whiteboard automatic correction device, USB signal receiver, signal reading circuit, storage module and processing module; The acquisition platform is an adjustable movable door-type truss structure with a fixed pulley, and the width and height are adjustable. The geometric center position of the upper end of the truss is used to fix the multi-spectral polarization imaging system, and the two side positions are used to fix the power supply system and the whiteboard automatic correction device respectively; The multi-spectral polarization imaging system comprises, from top to bottom, a CMOS image sensor, a liquid crystal tunable filter, a polarizer, and a lens group; The polarizer includes a polarization drive circuit, a small gear, a large gear and a polarizer, and the polarizer and the liquid crystal tunable filter respectively adjust the polarization state and the spectral range of light; The CMOS image sensor is used to convert the received optical signal into an electrical signal and store it in the form of a digital image; The CMOS image sensor works together with the liquid crystal tunable filter and the polarizer to capture light information of different wavelengths and record light information of different polarization angles, thereby generating image data containing rich spectral and polarization data; The power supply system supplies power to the multi-spectral polarization imaging system and the whiteboard automatic correction device; The control module is located on the mobile client, allowing the user to set remote commands through the mobile client, which are received by the USB signal receiver and used to adjust the parameters and working state of the crop growth drought stress detection system; The whiteboard automatic correction device has a built-in signal receiver, and the whiteboard automatic correction device rotates between 0° and 90°. When the control module remote command is in the whiteboard correction mode, it is unfolded to the 0° position for correction, and when the control module remote command is in the data collection mode, it is retracted to the 90° position; The USB signal receiver receives instructions from the control module; The signal reading circuit reads the image data captured by the CMOS image sensor and transmits it to the storage module and the processing module; The storage module stores image data acquired from the CMOS image sensor, the image data including all digital images stored by the CMOS sensor that are converted from optical signals of all wavelength bands and different polarization angles into electrical signals; The processing module performs data preprocessing and analysis modeling on the digital images stored in the storage module.

2. A crop growth drought stress detection system according to claim 1, characterized in that: The control module is located on the mobile client, allowing the user to set remote commands through the mobile client, and being received by the USB signal receiver, for adjusting the parameters and working state of the crop growth drought stress detection system.

3. A crop growth drought stress detection system according to claim 1, characterized in that: The polarizer is a polarization electrically controlled rotator driven by a large gear and a small gear; Wherein, the pinion is a driving wheel with a motor and a driving circuit, and has an infrared receiver for receiving infrared control signals; The large gear is a driven wheel with a hollow geometric center. The geometric center is equipped with a full-band polarizer. The polarizer rotates 0-360° around the geometric center axis of the CMOS image sensor to adjust the polarization angle, thereby changing the polarization state of the incident light. Through this adjustment, the polarizer enhances the light signal at a specific wavelength and receives light signals with different specific polarization directions, so that the CMOS image sensor generates an image containing different spectral and polarization data.

4. A crop growth drought stress detection system according to claim 1, characterized in that: The liquid crystal tunable filter contains an electronic unit and a liquid crystal layer. The USB signal receiver receives instructions from the control module, and the electronic unit adjusts the electric field of the liquid crystal layer, thereby changing the refractive index of the liquid crystal.

5. A crop growth drought stress detection system according to claim 1, characterized in that: The data preprocessing includes band channel alignment, crop growth index extraction, image background segmentation, and polarization angle, polarization intensity and intensity image synthesis to generate an image containing rich spectral and polarization data. The analysis and modeling includes analyzing the spectral characteristics and polarization information in the multispectral image, establishing a crop growth index prediction model, establishing a correspondence between the drought stress gradient and the image feature information of the polarization angle, and based on the correspondence, constructing a crop drought stress multispectral polarization detection model.

6. A crop growth drought stress detection system according to claim 5, characterized in that: The band channel registration adopts the maximum likelihood estimation method to align the images in spatial position by estimating the translation, rotation and scale transformation parameters between the images.

7. A crop growth drought stress detection system according to claim 5, characterized in that: The extracted crop growth index includes the normalized difference vegetation index NDVI and the normalized difference water index NDWI.

8. A crop growth drought stress detection system according to claim 5, characterized in that: The polarization angle, polarization degree and intensity images include intensity images of four different polarization angles, as well as Stokes parameters and polarization degree; The Stokes parameter calculation formula is: ； The formula for calculating the degree of polarization is: ； The formula for calculating the polarization angle is: ； In the formula, Represent the horizontal and vertical directions respectively. Represents the total intensity of the light field, which is the total intensity of 0° polarized light Perpendicular to the direction of propagation The sum of the squares of the electric field strength on the plane; express and Direction intensity difference of linear polarized light; express and Direction linear polarized light intensity difference.

9. A crop growth drought stress detection system according to claim 6, characterized in that: The analysis modeling is support vector machine (SVM) modeling, and the data set is a multispectral polarization image and its reflectivity that has been screened and processed by an image processing system. The multispectral image and reflectivity are associated with the degree of drought stress, and an independent validation data set is used to evaluate the prediction ability of the model. The parameters of the model are adjusted according to the validation results to improve the accuracy and generalization ability of the model.

10. A method for detecting crop growth drought stress using the crop growth drought stress detection system according to any one of claims 1 to 9, characterized in that: include S1: Plant crops on a plot with a width comparable to the adjustable truss structure; S2: subject crops to a gradient of drought stress; S3: Move the acquisition platform equipped with the multi-spectral polarization imaging system to the center of a corn plant plot under a certain stress treatment and fix the fixed pulley; S4: Connect the multi-spectral polarization imaging system to the control module on the mobile client through the WIFI hotspot, set the data acquisition mode, and set the data capture method of the multi-spectral polarization imaging system; S5: When the data acquisition mode is in the whiteboard calibration mode, the whiteboard automatic calibration device is rotated to place the whiteboard within the field of view, and the multispectral polarization imaging system automatically focuses and collects whiteboard data under the coordinated action of various components; When the data acquisition mode is in the data acquisition mode, the whiteboard automatic correction device is retracted, and the multispectral polarization imaging system automatically focuses and collects crop plant data; S6: The USB signal receiver receives the remote command and adjusts the parameters of the liquid crystal tunable filter; S7: setting and changing the required polarization angle of the polarizer, and using the multispectral polarization imaging system to collect multispectral polarization data of different drought stress samples, and saving them in the storage module in the form of image data; S8: Data preprocessing of the multispectral image is performed by transmitting the data of the storage module to the processing module of the computer, including band channel registration, crop growth index extraction, image background segmentation, and polarization angle, polarization intensity and intensity image synthesis, to generate an image containing rich spectral and polarization data; S9: In the processing module, based on the spectrum and image data obtained by the processing in S8, the spectral characteristics and polarization information in the multispectral image are analyzed, a crop growth index prediction model is established according to the reflectance of different wavelengths, and a correspondence between the image feature information of the drought stress gradient and the polarization angle is established according to the polarization image information at different wavelengths. Based on the correspondence, a crop drought stress multispectral polarization detection model is constructed, and the relationship between the degree of drought stress and the spectrum and polarization characteristics is obtained through the crop drought stress multispectral polarization detection model; S10: Analyze and evaluate the stress state of crops based on the crop drought stress multispectral polarization detection model obtained by the processing module.

Images