Big data analysis method and system based on image processing

Abstract

The invention discloses a big data analysis method and system based on image processing, and relates to the technical field of crop image processing. According to the method, through a preset big data algorithm, crop plant height, leaf area and leaf pest area parameters are comprehensively analyzed, a cylinder model containing crop growth indexes and pest indexes is constructed and compared with a theoretical healthy cylinder, and through grooving and calculation of the area and height difference value of a filling area, a growth score and a disease score are obtained through conversion. Finally, the crop state index is obtained, the comprehensive state of the crop in the current growth stage is quantified comprehensively, visually and accurately, and the growth level can be judged more accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of crop image processing, and in particular to a big data analysis method and system based on image processing. Background Art

[0002] With the rapid development of information technology, image processing and big data analysis technologies have provided new ways to solve these problems in agricultural production.

[0003] However, a big data analysis method based on image processing in the prior art still has the following shortcomings in practical application: The extraction and comprehensive analysis of crop growth parameters are still not perfect, and cannot truly reflect the actual growth status of crops. In addition, the identification of the pest area of ​​crop leaves only relies on simple image feature analysis, which makes it difficult to accurately distinguish different types of lesions and their severity, resulting in low accuracy of data analysis. In addition, when evaluating the growth conditions of farmland, there is a lack of in-depth analysis of the relationship between meteorological factors and crop growth, and it is impossible to effectively distinguish whether abnormal growth is caused by meteorological conditions or other factors.

[0004] Therefore, a big data analysis method and system based on image processing are introduced. Summary of the invention

[0005] The purpose of the present invention is to solve the problems pointed out in the background technology and to propose a big data analysis method and system based on image processing.

[0006] The purpose of the present invention can be achieved by the following technical solution: A big data analysis method based on image processing, comprising: Data analysis: Extract crop growth parameters from farmland crop image information, and use the preset big data algorithm to conduct comprehensive analysis to determine the crop state index of farmland crops at the current growth stage; based on the crop state index of farmland crops at the current growth stage, output the growth level of farmland at the current growth stage; the growth parameters include crop plant height, crop leaf area, and crop leaf pest area; the growth level includes abnormal growth level, normal growth level, and supernormal growth level; Farmland assessment: If the farmland growth level output at a certain growth stage is an abnormal growth level, the farmland in the current growth stage and the past growth stage will be evaluated, and a risk assessment report of the farmland at the current time point will be pushed to the farmer based on the analysis results.

[0007] As a preferred embodiment of the present invention, the plant height and the leaf area of ​​the crop are extracted, and a comprehensive analysis is performed using a preset big data algorithm, specifically: For crop image information, identify the top and bottom positions of the crop in the image, and combine the image scale and camera parameters to obtain the actual plant height of the crop at the current growth stage; Count the number of pixels of crop leaves in the image, and convert the actual area according to the scale of the image to obtain the leaf area of ​​the crop at the current growth stage; Identify the current growth stage of the crop in the field, and preset the normal plant height and normal leaf area corresponding to the actual plant height and leaf area at different growth stages, which are recorded as t1 and t2; After marking the actual plant height and leaf area of ​​the crop at the current growth stage as h1 and h2 respectively, substitute them into the formula A weighted calculation is performed to determine the crop growth index p of the farmland crops at the current growth stage; wherein a1 and a2 are the influencing weight factors of the actual plant height h1 and the leaf area h2, respectively.

[0008] As a preferred embodiment of the present invention, the crop leaf pest area of ​​the crop is extracted and a comprehensive analysis is performed using a preset big data algorithm, specifically: Identify the disease spot features in the crop leaf location area and segment them from the image. After segmentation, they will be matched with the disease spot color features pre-stored in the database; set different disease spot color features to correspond to an influence weight coefficient; After matching, the number of pixels of each disease spot color feature is counted, and the actual area is converted according to the scale of the image to obtain the characteristic area of ​​each disease spot in the current growth stage of the crop; The characteristic area of ​​each disease spot of the crop at the current growth stage is multiplied by the corresponding impact weight coefficient, and then the sum is obtained to obtain the disease spot impact index of the crop at the current growth stage; The allowable disease spot impact index corresponding to different growth stages is preset, and the disease spot impact index of the crop at the current growth stage is used as the numerator, and the corresponding allowable disease spot impact index is used as the denominator. The crop pest index of the crop at the current growth stage is obtained after the ratio calculation.

[0009] As a preferred embodiment of the present invention, the crop state index of the farmland crops at the current growth stage is determined as follows: The crop growth index p is used as the radius of the bottom of the cylinder, and the crop pest index is used as the vertical height of the cylinder. The cylinder is constructed and the pre-constructed theoretical healthy cylinder is extracted. The constructed cylinder is grooved on the theoretical healthy cylinder, that is, after the centers of the two groups of cylinders are aligned, the constructed cylinder is grooved from the bottom of the theoretical healthy cylinder to obtain a combined model diagram of farmland crops at the current growth stage. The remaining bottom circle area in the digging direction in the combined model diagram is filled, and the area of ​​the filled area is calculated to obtain the growth assessment value of the farmland crops in the current growth stage; the height difference between the constructed cylinder and the theoretical healthy cylinder is calculated and recorded as the height difference; The intervals of each group of assessment values ​​corresponding to the preset growth assessment values ​​correspond to a growth score; similarly, the intervals of each group of difference values ​​corresponding to the preset high difference values ​​correspond to a disease score; The growth score and disease score of the farmland crops in the current growth stage are multiplied by the corresponding set weight coefficients, and then the sum is obtained to obtain the crop state index of the farmland crops.

[0010] As a preferred embodiment of the present invention, the growth level of the farmland in the current growth stage is output, specifically: The crop state index of each farmland crop in the current growth stage is averaged to obtain the crop comprehensive index of the farmland in the current growth stage; The crop comprehensive index is compared with the preset reference index range. If the crop comprehensive index is higher than the preset reference index range, the extraordinary growth level is output; if the crop comprehensive index is lower than the preset reference index range, the abnormal growth level is output; if the crop comprehensive index is within the preset reference index range, the normal growth level is output.

[0011] As a preferred embodiment of the present invention, the farmland in the current growth stage and the past growth stage is evaluated, specifically: Extract the farmland growth levels corresponding to the farmland in the current growth stage and the past growth stages, and identify the growth stages with abnormal growth levels as the output results; The growth stage of abnormal growth level is divided into meteorological influence growth stage and abnormal growth stage; The number of abnormal growth stages is counted and recorded as the number of abnormalities; the crop comprehensive index corresponding to the farmland in the abnormal production stage is extracted, and the difference is calculated with the lowest value in the reference index range respectively, and the mean of the difference calculated for each abnormal production stage is calculated to obtain the growth risk value of the farmland at the current time point; The growth risk value and the number of anomalies at the current time point are multiplied by the corresponding set weight coefficients, and then the sum is calculated to obtain the comprehensive risk index of the farmland at the current time point.

[0012] As a preferred embodiment of the present invention, the basis for dividing the growth stage of the abnormal growth level into the meteorological influence growth stage is: Get the temperature, light duration and precipitation corresponding to each growth stage, recorded as q1, q2 and q3, and set the optimal temperature, optimal light duration and optimal precipitation for each growth stage according to the type of crops currently planted in the farmland, recorded as f1, f2 and f3; According to the formula , weighted calculations were performed on the temperature, sunshine duration, and precipitation corresponding to each growth stage to obtain the meteorological impact index corresponding to each growth stage. ; Where z1, z2 and z3 represent the allowable temperature difference, allowable light duration difference and allowable precipitation difference; where n1, n2 and n3 are the weight factors of temperature, light duration and precipitation respectively; The meteorological impact index of each growth stage The growth stages above the meteorological impact threshold are compared with the preset meteorological impact threshold, and the growth stages above the meteorological impact threshold are marked as meteorologically affected growth stages, and the rest are marked as abnormal growth stages.

[0013] As a preferred implementation of the present invention, a risk assessment report at the current time point is generated, specifically: The growth stage and number of occurrences of the meteorological impact growth stage, the growth stage and number of occurrences of the abnormal growth stage, and the growth risk value at the current time point are extracted and filled into the preset report template to generate a risk assessment report at the current time point.

[0014] A big data analysis system based on image processing, comprising: Image processing module: collects image information of farmland crops at different growth stages and pre-processes the collected image information of farmland crops; Image analysis module: extracts crop growth parameters from farmland crop image information, and analyzes and determines the crop state index of farmland crops at the current growth stage; based on the crop state index of farmland crops at the current growth stage, outputs the growth grade of the farmland at the current growth stage; Risk assessment module: When the farmland growth level is abnormal at a certain growth stage, the farmland in the current growth stage and the past growth stage is assessed, and a risk assessment report of the farmland at the current time point is pushed to the farmer based on the analysis results. Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention uses a preset big data algorithm to comprehensively analyze the parameters of crop plant height, leaf area and leaf pest area, construct a cylindrical model containing crop growth index and pest index, and compare it with the theoretical healthy cylinder. By digging grooves, calculating the filling area and height difference, it is converted into growth score and disease score, and finally the crop status index is obtained. It comprehensively, intuitively and accurately quantifies the comprehensive status of crops in the current growth stage, and can more accurately judge the growth level; 2. When the farmland growth level is abnormal, the present invention analyzes the farmland growth level in the current and past growth stages in detail, calculates the meteorological impact index in combination with meteorological data, divides the abnormal growth stage into the meteorological impact growth stage and the growth abnormality stage, and determines the risk degree and generates a risk assessment report by counting the number of abnormal stages, calculating the growth risk value and the comprehensive risk index, so as to help farmers make accurate decisions and take effective measures to deal with the problem of abnormal crop growth. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0016] Figure 1 is a flow chart of the present invention; Figure 2 It is a principle block diagram of the present invention; Figure 3 It is a schematic diagram of the combined model diagram in the present invention. DETAILED DESCRIPTION

[0017] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0018] Example 1

[0019] See also Figure 1 and Figure 3 As shown, a big data analysis method based on image processing includes: Image processing: drones equipped with high-definition cameras are used to fly over farmland at a predetermined route and altitude to collect crop image information at different growth stages. The time interval for image collection is set according to the crop growth cycle and monitoring requirements to ensure that key changes in the crop growth process can be captured. The collected crop image information is preprocessed. Preprocessing includes denoising the collected images, using algorithms such as Gaussian filtering to remove noise interference in the image and improve image clarity. Image enhancement is then performed, using methods such as histogram equalization to enhance the image contrast and make the crop features more obvious. Image segmentation is then performed, using threshold segmentation, edge detection and other algorithms to separate the crop from the background for subsequent analysis of crop features. It should be noted that precise flight routes are developed according to the shape, area and key monitoring areas of the farmland. For example, for rectangular farmland, parallel round-trip routes are used to ensure that every corner of the farmland can be photographed. The flight altitude is determined by the camera resolution and the required image accuracy, generally between 30-100 meters. For example, if a high-definition camera with a resolution of 20 million pixels is used, in order to clearly capture crop details, the flight altitude is set to 50 meters. At this time, the ground resolution can reach 2-3 cm / pixel, which is sufficient to meet the needs of monitoring crop growth conditions; The camera's exposure mode is set to automatic exposure to adapt to shooting needs under different lighting conditions. The shutter speed is set between 1 / 1000-1 / 2000 seconds to avoid image blur caused by the drone's flight speed. The sensitivity (ISO) is adjusted according to the light intensity. When the sun is bright, the ISO is set to 100-200; on cloudy days or in low light conditions, the ISO is appropriately increased to 400-800, but it should be noted that too high an ISO may introduce noise. The image format is selected as JPEG or RAW. The RAW format can retain more image details, but the file size is larger and is suitable for monitoring tasks that require extremely high image quality; the JPEG format file is smaller and easier to store and transmit, and can generally meet analysis needs. Taking corn as an example, in the seedling stage, the image acquisition time interval can be set to 7-10 days because the crop grows relatively slowly; in the jointing stage and the trumpet stage, the crop grows rapidly, and in order to capture the growth changes in time, the time interval is shortened to 3-5 days; in the tasseling and silking stage and the grain filling stage, the time interval can be flexibly adjusted to 2-4 days according to the needs of pest and disease monitoring and the key nodes of crop growth. With such settings, the morphological changes of corn during growth can be fully recorded.

[0020] Data analysis: Extract crop growth parameters from farmland crop image information, and use the preset big data algorithm to conduct comprehensive analysis to determine the crop state index of farmland crops at the current growth stage; based on the crop state index of farmland crops at the current growth stage, output the growth level of farmland at the current growth stage; the growth parameters include crop plant height, crop leaf area, and crop leaf pest area; the growth level includes abnormal growth level, normal growth level, and supernormal growth level; Specifically: For crop image information, identify the top and bottom positions of the crop in the image, and combine the image scale and camera parameters to obtain the actual plant height of the crop at the current growth stage; It should be noted that a reference object of known length (such as a ruler of 1 meter in length) is placed in the farmland, and the scale of the image is calculated based on the pixel length of the reference object in the image and the actual length; for example, if the pixel length of the ruler in the image is 100 pixels and the actual length is 1 meter, the image scale is 100 pixels / meter; After obtaining the crop outline through image segmentation, the image processing algorithm is used to identify the pixel positions of the top and bottom of the crop in the image. Assume that the pixel position of the top of the crop is (x1, y1) and the pixel position of the bottom is (x2, y2); The actual height of the crop is calculated using trigonometric functions based on the image scale and camera parameters (such as focal length, shooting angle, etc.). For example, if the camera focal length is 50 mm and the shooting angle is vertically downward, the vertical pixel distance calculated from the pixel position is 200 pixels. Combined with the image scale of 100 pixels / meter, the crop height can be calculated to be 2 meters.

[0021] Count the number of pixels of crop leaves in the image, and convert the actual area according to the scale of the image to obtain the leaf area of ​​the crop at the current growth stage; It should be noted that for the segmented crop leaf images, the number of pixels occupied by the leaves is counted, and a linear regression model between the leaf area and the number of pixels is established by measuring and counting the number of pixels of a large number of leaf samples of different areas. For example, after measuring and analyzing 100 leaf samples, the regression equation between the leaf area S (square centimeters) and the number of pixels N is obtained as S=0.1N+5, where 0.1 is the regression coefficient and 5 is the constant term; Identify the current growth stage of the crop in the field, and preset the normal plant height and normal leaf area corresponding to the actual plant height and leaf area at different growth stages, which are recorded as t1 and t2; After marking the actual plant height and leaf area of ​​the crop at the current growth stage as h1 and h2 respectively, substitute them into the formula Perform weighted calculation to determine the crop growth index p of the farmland crops at the current growth stage; where a1 and a2 are the influencing weight factors of the actual plant height h1 and leaf area h2 respectively; It should be noted that the crop growth index can intuitively and quantitatively present the crop growth status. By comparing the actual plant height and leaf area with the preset normal plant height and normal leaf area, and combining weighted calculation to obtain the crop growth index, farmers or agricultural researchers can quickly understand the degree of deviation of the current crop growth from the normal state.

[0022] Identify the disease spot features in the crop leaf location area and segment them from the image. After segmentation, they will be matched with the disease spot color features pre-stored in the database; for example, the initial color is light yellow and the color becomes reddish brown in the later stage; set different disease spot color features to correspond to an influence weight coefficient; the more serious the disease spot color, the higher the corresponding influence weight coefficient, which is set in the range of 1.138-1.285; After matching, the number of pixels of each disease spot color feature is counted, and the actual area is converted according to the scale of the image to obtain the characteristic area of ​​each disease spot in the current growth stage of the crop; The characteristic area of ​​each disease spot of the crop at the current growth stage is multiplied by the corresponding impact weight coefficient, and then the sum is obtained to obtain the disease spot impact index of the crop at the current growth stage; The permissible disease spot impact index corresponding to different growth stages is preset, and the disease spot impact index of the crop at the current growth stage is used as the numerator, and the corresponding permissible disease spot impact index is used as the denominator. The crop pest index of the crop at the current growth stage is obtained after the ratio calculation; It should be noted that by identifying and segmenting the characteristics of lesions and matching them with the database, the type and severity of the lesions can be accurately determined. In the past, farmers relied on experience to judge pests and diseases, which was highly subjective and prone to misjudgment. This method uses quantified lesion color characteristics and corresponding weight coefficients to greatly improve the accuracy of judgment.

[0023] The crop growth index p is used as the radius of the bottom of the cylinder, and the crop pest index is used as the vertical height of the cylinder. The cylinder is constructed and the pre-constructed theoretical healthy cylinder is extracted. The bottom size and vertical height of the theoretical healthy cone are both higher than the maximum cone that can be constructed. This means that in the actual monitoring process, any actually constructed cone (representing the current crop growth status) can clearly show the degree of deviation compared with the theoretical healthy cone. The constructed cylinder is used to dig a groove on the theoretical healthy cylinder, that is, after aligning the centers of the two groups of cylinders, the constructed cylinder is grooved from the bottom of the theoretical healthy cylinder to obtain a combined model diagram of farmland crops at the current growth stage. The remaining bottom circle area in the digging direction in the combined model diagram is filled, and the area of ​​the filled area is calculated to obtain the growth assessment value of the farmland crops in the current growth stage; the height difference between the constructed cylinder and the theoretical healthy cylinder is calculated and recorded as the height difference; The intervals of each group of evaluation values ​​corresponding to the preset growth evaluation values ​​correspond to a growth score; the growth score range is set at 1-10, and the smaller the growth evaluation value, the higher the corresponding growth score; similarly, the intervals of each group of difference values ​​corresponding to the preset high difference values ​​correspond to a disease score; the disease score range is set at 1-10, and the smaller the high difference value, the higher the corresponding disease score; The growth score and disease score of the farmland crops in the current growth stage are multiplied by the corresponding set weight coefficients, and then the sum is obtained to obtain the crop state index of the farmland crops; It should be noted that the crop growth index and crop pest index are converted into the geometric parameters of the cylinder to construct the model. By digging grooves on the theoretical healthy cylinder and calculating the area of ​​the filling area and the height difference, the growth status of the crop and the degree of influence of pests are converted into specific values, namely the growth assessment value and the height difference. Furthermore, through the preset intervals and scoring rules, these values ​​are converted into growth scores and disease scores, and finally the crop status index is obtained. This method can comprehensively, intuitively and accurately quantify the comprehensive status of the crop at the current growth stage.

[0024] The crop state index of each farmland crop in the current growth stage is averaged to obtain the crop comprehensive index of the farmland in the current growth stage; Compare the crop comprehensive index with the preset reference index range. If the crop comprehensive index is higher than the preset reference index range, output an abnormal growth level; if the crop comprehensive index is lower than the preset reference index range, output an abnormal growth level; if the crop comprehensive index is within the preset reference index range, output a normal growth level; Farmland assessment: If the farmland growth level output at a certain growth stage is an abnormal growth level, the farmland in the current growth stage and the past growth stage will be assessed, and a risk assessment report of the farmland at the current time point will be pushed to the farmer based on the analysis results; Specifically: Extract the farmland growth levels corresponding to the farmland in the current growth stage and the past growth stages, and identify the growth stages with abnormal growth levels as the output results; The temperature, duration of sunshine and precipitation corresponding to each growth stage are obtained; the temperature is the average of the temperatures at each time point corresponding to each growth stage, which are recorded as q1, q2 and q3. According to the types of crops currently planted in the farmland, the optimal temperature, optimal duration of sunshine and optimal precipitation of each growth stage are set, which are recorded as f1, f2 and f3. According to the formula , weighted calculations were performed on the temperature, sunshine duration, and precipitation corresponding to each growth stage to obtain the meteorological impact index corresponding to each growth stage. ; Where z1, z2 and z3 represent the allowable temperature difference, allowable light duration difference and allowable precipitation difference; where n1, n2 and n3 are the weight factors of temperature, light duration and precipitation respectively; The meteorological impact index of each growth stage Compare with the preset meteorological impact threshold, mark the growth stage above the meteorological impact threshold as the meteorological impact growth stage, and mark the rest as the abnormal growth stage; The number of abnormal growth stages is counted and recorded as the number of abnormalities; the crop comprehensive index corresponding to the farmland in the abnormal production stage is extracted, and the difference is calculated with the lowest value in the reference index range respectively, and the mean of the difference calculated for each abnormal production stage is calculated to obtain the growth risk value of the farmland at the current time point; Multiply the growth risk value and the number of anomalies at the current time point by the corresponding set weight coefficients, and then sum them up to obtain the comprehensive risk index of the farmland at the current time point; Extract the growth stage and number of occurrences of the meteorological impact growth stage, the growth stage and number of occurrences of the abnormal growth stage, and the growth risk value at the current time point, and fill them into the preset report template to generate a risk assessment report at the current time point; It should be noted that by comparing the growth levels of farmland in the current and past growth stages in detail, and in-depth analysis of meteorological data in each stage, including the differences in temperature, sunshine duration, precipitation and the optimal conditions required for crops, and performing weighted calculations to derive the meteorological impact index, it is possible to accurately determine whether abnormal growth is dominated by meteorological factors or caused by other unknown factors. For example, if the meteorological impact index of a certain stage far exceeds the threshold and is marked as a meteorologically affected growth stage, it can be clearly seen that meteorological conditions are the main cause of abnormal growth in that stage; conversely, if it is marked as an abnormal growth stage, it indicates that it is necessary to explore problems from other aspects such as soil quality, pests and diseases, etc., and point out the key direction for farmers to solve the problem.

[0025] The growth risk value and the comprehensive risk index are calculated to transform the complex farmland growth conditions into intuitive and quantitative data. The growth risk value is obtained by averaging the difference between the comprehensive index of crops in the abnormal growth stage and the minimum value of the reference index, reflecting the degree to which the crops deviate from the normal growth level in the abnormal stage; the comprehensive risk index further combines the growth risk value with the number of abnormalities, and considers their respective weight coefficients to comprehensively quantify the overall risk level of the farmland at the current time point, which enables farmers to clearly understand the severity of the risks faced by the farmland.

[0026] Example 2

[0027] See also Figure 2 As shown, based on the big data analysis method based on image processing provided in Example 1 of the present application, Example 2 of the present application proposes a big data analysis system based on image processing. Example 2 is only a preferred method of Example 1, and the implementation of Example 2 will not affect the independent implementation of Example 1.

[0028] Specifically, the big data analysis method based on image processing provided in Embodiment 2 of the present application is different in that it includes: The image processing module is used to collect image information of farmland crops at different growth stages and pre-process the collected image information of farmland crops; The image analysis module is used to extract crop growth parameters from the farmland crop image information, and analyze and determine the crop state index of the farmland crops at the current growth stage; based on the crop state index of the farmland crops at the current growth stage, output the growth grade of the farmland at the current growth stage; The risk assessment module is used to evaluate the farmland in the current growth stage and the past growth stage when the farmland growth level is output as an abnormal growth level at a certain growth stage, and push the risk assessment report of the farmland at the current time point to the farmer based on the analysis results; The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A big data analysis method based on image processing, characterized in that: include: Data analysis: Extract crop growth parameters from farmland crop image information, and use the preset big data algorithm to conduct comprehensive analysis to determine the crop state index of farmland crops at the current growth stage; based on the crop state index of farmland crops at the current growth stage, output the growth level of farmland at the current growth stage; the growth parameters include crop plant height, crop leaf area, and crop leaf pest area; the growth level includes abnormal growth level, normal growth level, and supernormal growth level; Farmland assessment: If the farmland growth level output at a certain growth stage is an abnormal growth level, the farmland in the current growth stage and the past growth stage will be assessed, and a risk assessment report for the farmland at the current time point will be pushed to the farmer based on the analysis results.

2. The big data analysis method based on image processing according to claim 1, characterized in that: Extract the plant height and leaf area of ​​crops, and use the preset big data algorithm to conduct comprehensive analysis, specifically: For crop image information, identify the top and bottom positions of the crop in the image, and combine the image scale and camera parameters to obtain the actual plant height of the crop at the current growth stage; Count the number of pixels of crop leaves in the image, and convert the actual area according to the scale of the image to obtain the leaf area of ​​the crop at the current growth stage; Identify the current growth stage of the crop in the field, and preset the normal plant height and normal leaf area corresponding to the actual plant height and leaf area at different growth stages, which are recorded as t1 and t2; After marking the actual plant height and leaf area of ​​the crop at the current growth stage as h1 and h2 respectively, substitute them into the formula A weighted calculation is performed to determine the crop growth index p of the farmland crops at the current growth stage; wherein a1 and a2 are the influencing weight factors of the actual plant height h1 and the leaf area h2, respectively.

3. The big data analysis method based on image processing according to claim 2, characterized in that: Extract the crop leaf pest area of ​​the crop and use the preset big data algorithm for comprehensive analysis, specifically: Identify the disease spot features in the crop leaf location area and segment them from the image. After segmentation, they will be matched with the disease spot color features pre-stored in the database; set different disease spot color features to correspond to an influence weight coefficient; After matching, the number of pixels of each disease spot color feature is counted, and the actual area is converted according to the scale of the image to obtain the characteristic area of ​​each disease spot in the current growth stage of the crop; The characteristic area of ​​each disease spot of the crop at the current growth stage is multiplied by the corresponding impact weight coefficient, and then the sum is obtained to obtain the disease spot impact index of the crop at the current growth stage; The allowable disease spot impact index corresponding to different growth stages is preset, and the disease spot impact index of the crop at the current growth stage is used as the numerator, and the corresponding allowable disease spot impact index is used as the denominator. The crop pest index of the crop at the current growth stage is obtained after the ratio calculation.

4. The big data analysis method based on image processing according to claim 3 is characterized in that: Determine the crop status index of the crop in the field at the current growth stage, specifically: The crop growth index p is used as the radius of the bottom of the cylinder, and the crop pest index is used as the vertical height of the cylinder. The cylinder is constructed and the pre-constructed theoretical healthy cylinder is extracted. The constructed cylinder is grooved on the theoretical healthy cylinder, that is, after the centers of the two groups of cylinders are aligned, the constructed cylinder is grooved from the bottom of the theoretical healthy cylinder to obtain a combined model diagram of farmland crops at the current growth stage. The remaining bottom circle area in the digging direction in the combined model diagram is filled, and the area of ​​the filled area is calculated to obtain the growth assessment value of the farmland crops in the current growth stage; the height difference between the constructed cylinder and the theoretical healthy cylinder is calculated and recorded as the height difference; The intervals of each group of assessment values ​​corresponding to the preset growth assessment values ​​correspond to a growth score; similarly, the intervals of each group of difference values ​​corresponding to the preset high difference values ​​correspond to a disease score; The growth score and disease score of the farmland crops in the current growth stage are multiplied by the corresponding set weight coefficients, and then the sum is obtained to obtain the crop state index of the farmland crops.

5. The big data analysis method based on image processing according to claim 4 is characterized in that: Output the growth level of the farmland in the current growth stage, specifically: The crop state index of each farmland crop in the current growth stage is averaged to obtain the crop comprehensive index of the farmland in the current growth stage; The crop comprehensive index is compared with the preset reference index range. If the crop comprehensive index is higher than the preset reference index range, the extraordinary growth level is output; if the crop comprehensive index is lower than the preset reference index range, the abnormal growth level is output; if the crop comprehensive index is within the preset reference index range, the normal growth level is output.

6. The big data analysis method based on image processing according to claim 5, characterized in that: Evaluate fields in current and past growth stages, specifically: Extract the farmland growth levels corresponding to the farmland in the current growth stage and the past growth stages, and identify the growth stages with abnormal growth levels as the output results; The growth stage of abnormal growth level is divided into meteorological influence growth stage and abnormal growth stage; The number of abnormal growth stages is counted and recorded as the number of abnormalities; the crop comprehensive index corresponding to the farmland in the abnormal production stage is extracted, and the difference is calculated with the lowest value in the reference index range respectively, and the mean of the difference calculated for each abnormal production stage is calculated to obtain the growth risk value of the farmland at the current time point; The growth risk value and the number of anomalies at the current time point are multiplied by the corresponding set weight coefficients, and then the sum is calculated to obtain the comprehensive risk index of the farmland at the current time point.

7. The big data analysis method based on image processing according to claim 6, characterized in that: The basis for dividing the growth stage of abnormal growth level into meteorological influence growth stage is: Get the temperature, light duration and precipitation corresponding to each growth stage, recorded as q1, q2 and q3, and set the optimal temperature, optimal light duration and optimal precipitation for each growth stage according to the type of crops currently planted in the farmland, recorded as f1, f2 and f3; According to the formula , weighted calculations were performed on the temperature, sunshine duration, and precipitation corresponding to each growth stage to obtain the meteorological impact index corresponding to each growth stage. ; Where z1, z2 and z3 represent the allowable temperature difference, allowable light duration difference and allowable precipitation difference; where n1, n2 and n3 are the weight factors of temperature, light duration and precipitation respectively; The meteorological impact index of each growth stage The growth stages above the meteorological impact threshold are compared with the preset meteorological impact threshold, and the growth stages above the meteorological impact threshold are marked as meteorologically affected growth stages, and the rest are marked as abnormal growth stages.

8. The big data analysis method based on image processing according to claim 7, characterized in that: Generate a risk assessment report at the current time point, specifically: The growth stage and number of occurrences of the meteorological impact growth stage, the growth stage and number of occurrences of the abnormal growth stage, and the growth risk value at the current time point are extracted and filled into the preset report template to generate a risk assessment report at the current time point.

9. A big data analysis system based on image processing, applied to a big data analysis method based on image processing as claimed in any one of claims 1 to 8, characterized in that: include: Image processing module: collects image information of farmland crops at different growth stages and pre-processes the collected image information of farmland crops; Image analysis module: extracts crop growth parameters from farmland crop image information, and analyzes and determines the crop state index of farmland crops at the current growth stage; based on the crop state index of farmland crops at the current growth stage, outputs the growth grade of the farmland at the current growth stage; Risk assessment module: When the farmland growth level is output as an abnormal growth level at a certain growth stage, the farmland in the current growth stage and the past growth stage is evaluated, and a risk assessment report of the farmland at the current time point is pushed to the farmer based on the analysis results.

Images