Intelligent greenhouse control system and method

Abstract

The invention discloses an intelligent greenhouse control system and method, relates to the technical field of intelligent greenhouses, solves the problem that flowers are easily damaged by an original spraying mode, obtains an image of a spraying area through a high-definition probe, accurately recognizes a contour area through graying processing and a Sobe l algorithm, and improves the spraying efficiency. Further accurately calibrating a florescence region in combination with the gray scale interval of the flower color feature; on the basis, a non-spraying area is reasonably planned, the spraying speed of a spraying opening is accurately regulated and controlled, damage to crops in the flowering period caused by spraying water is effectively avoided, and the growth quality and yield of the crops are greatly improved; compared with traditional greenhouse spraying, the occurrence rate of the problems of flower damage, pollination failure and the like caused by improper spraying is remarkably reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent greenhouses, and in particular to an intelligent greenhouse control system and method. Background Art

[0002] In the field of agricultural greenhouse cultivation, crops such as flowers, fruits and vegetables are extremely sensitive to the environment during the flowering stage. Any improper spraying management will have a serious impact on crop growth and final harvest.

[0003] Traditional greenhouse spraying systems lack effective flowering period identification and precise spraying control mechanisms. On the one hand, it is difficult to accurately judge whether the crops are in the flowering period, and most of the subjective judgments are based on manual experience. This method is not only inefficient, but also inaccurate and prone to misjudgment. On the other hand, during the spraying operation, the spraying range and spraying speed cannot be adjusted according to the actual state of the crops, and excessive or insufficient spraying often occurs. When over-spraying, excessive water may wash away the pollen on the flowers, hinder the pollination process, and may also cause diseases; insufficient spraying cannot meet the water demand for crop growth, which also affects the healthy growth of crops. In addition, the traditional manual inspection method is difficult to monitor the crop growth conditions in every corner of the greenhouse in real time and comprehensively, resulting in the inability to timely discover and deal with flowering-related problems.

[0004] With the continuous improvement of the requirements for the quality and output of agricultural products, as well as the continuous increase in labor costs, the disadvantages of traditional greenhouse spraying methods have become more and more prominent. To solve these problems, the spraying area image is obtained through high-definition probes, and the contour area is accurately identified with the help of grayscale processing and Sobel algorithm. The flowering period area is accurately calibrated in combination with the grayscale interval of flower color characteristics. On this basis, a new type of intelligent greenhouse spraying technology has emerged, which can reasonably plan the non-spraying area and accurately control the spray speed of the spray port. This technology has greatly made up for the shortcomings of the traditional spraying system, provided a more suitable growth environment for crops during the flowering period, and significantly improved the intelligence and refinement level of agricultural production. Summary of the invention

[0005] In view of the deficiencies of the prior art, the present invention provides an intelligent greenhouse control system and method, which solves the problem that the original spraying method can easily damage flowers.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent greenhouse control method, comprising the following steps:

[0007] Step 1: Obtain the regional image associated with each spraying area in the smart greenhouse, and grayscale the acquired regional image to obtain a grayscale image. Then, based on the grayscale value characteristics of different points in the grayscale image, confirm several contour areas associated with the grayscale image. The specific method is as follows:

[0008] Based on the different area images obtained for each sprinkler area, confirm the RGB values associated with different pixel points within the specified area image: Use HD i = 0.299R + 0.587G + 0.114B to confirm the grayscale value HD associated with different pixel points i , where 0.299, 0.587, and 0.114 are all weight factors preset for the corresponding pixel features, where i represents different pixel points associated within the corresponding area image. Based on the different grayscale values HD confirmed for different pixel points i , perform grayscale processing on this area image to generate a grayscale image belonging to this area image;

[0009] Use the Sobel algorithm to confirm the vertical gradient and vertical gradient associated with different pixel points within the grayscale image, and based on the vertical gradient and vertical gradient, confirm the comprehensive gradient ZH associated with the corresponding pixel points i , where X i is the vertical gradient, and its Y i is the vertical gradient;

[0010] Compare the confirmed comprehensive gradient ZH i with the preset value Y1. Those that satisfy: ZH i ≥ Y1 are marked as gradient pixel points, otherwise, no marking is performed. Y1 is the preset determination threshold;

[0011] Mark the continuously appearing gradient pixel points with characteristic contours. Denote the area enclosed by the complete and closed characteristic contour as the contour area and mark it within the grayscale image. Otherwise, no specific marking of the contour area is performed;

[0012] Step2. Based on the different contour areas confirmed within the grayscale image corresponding to the sprinkler area, according to the contour characteristics associated with different contour areas, identify whether the corresponding contour area belongs to the flowering period area and perform marking, and confirm the sprinkling range for the grayscale image with the flowering period area. The specific method is as follows:

[0013] Step21. For several groups of contour areas confirmed within a single group of grayscale images, confirm the grayscale values associated with the pixel points inside the corresponding contour areas, and from them, confirm the minimum grayscale value and the maximum grayscale value to generate the grayscale characteristics of the corresponding contour areas, and sequentially confirm the grayscale characteristics associated with different contour areas;

[0014] Step22. Compare the grayscale characteristics associated with different contour areas with the preset numerical interval. The end point values of the numerical interval are all preset values:

[0015] If the gray - scale feature belongs to a numerical interval, then label this contour area as the flowering - stage area;

[0016] If the gray - scale feature does not belong to the numerical interval, then identify the intersection range between the gray - scale feature and the corresponding numerical interval. If the proportion of the intersection range in the entire range of the gray - scale feature exceeds 90%, then label this contour area as the flowering - stage area; otherwise, do not perform any labeling;

[0017] Step23. For the gray - scale image with a flowering - stage area: Use the image edge of this gray - scale image as a feature edge, combine the feature edge with a two - dimensional coordinate system, confirm the two - dimensional coordinates associated with different edge points within the feature edge, and then confirm the center point of this gray - scale image by finding the mean coordinate of all two - dimensional coordinates of several groups of edge points;

[0018] Based on the confirmed center point and the labeled flowering - stage area, confirm the point with the shortest straight - line distance and the farthest point from the center point on the edge contour of the flowering - stage area. Denote the point with the shortest straight - line distance as the inner - circle point and the point with the farthest straight - line distance as the outer - circle point;

[0019] Construct a set of circles with the connection line between the inner - circle point and the center point as the radius and the center point as the center. Denote this set of circles as the inner - circle. Then construct a set of circles with the connection line between the outer - circle point and the center point as the radius and the center point as the center. Denote this set of circles as the outer - circle. Denote the area between the inner - circle and the outer - circle as the non - sprinkling area and label it within the associated gray - scale image;

[0020] Step3. Re - analyze the gray - scale image for which the sprinkling range has been confirmed, confirm the water flow spraying speed of the corresponding sprinkler nozzle, and ensure that the spraying range of the sprayed water does not cover the flowering - stage area through the gradually confirmed water flow spraying speed. Based on the confirmation result, generate the execution interval of the corresponding sprinkling area. The specific sub - steps are as follows:

[0021] Step31. According to the preset water flow spraying - speed interval of each sprinkler nozzle, during the process of gradually changing the water flow spraying - speed interval from the minimum value to the maximum value, the sprayed water effectively covers the entire sprinkling area, and the end - point values of the water flow spraying - speed interval are all preset values;

[0022] Step32. Based on the non - sprinkling area confirmed within the corresponding gray - scale image, confirm the inner - circle radius and the outer - circle radius of the non - sprinkling area, lock the feature range belonging to this non - sprinkling area, and then based on the center point confirmed within this gray - scale image, determine the edge point with the farthest straight - line distance from this center point on the feature edge of the gray - scale image. Denote the distance between this edge point and the center point as L;

[0023] Step 33. Uniformly map the numerical range of X1-L to the water flow spraying speed interval, where X1 is a preset value. Confirm the mapping characteristics, set the water flow spraying speed interval as [Vmin, Vmax], use (Vmax - Vmin) ÷ (L - X1) = Tz to confirm the mapping characteristic value Tz, and calibrate the characteristic range as [R1, R2]. Use (R1 - X1) × Tz + Vmin = V R1 and (R2 - X1) × Tz + Vmin = V R2 Lock the spraying speed range associated with the characteristic range [V R1 ,V R2 ;

[0024] Step 34. Confirm the spraying area associated with this grayscale image. Exclude the spraying speed range [V R1 ,V R2 from the water flow spraying speed interval of this spraying area, and use the water flow spraying speed interval after the exclusion process as the execution interval, and control the spray nozzles of the corresponding spraying area to perform relevant executions according to the water flow spraying speed within the execution interval.

[0025] Preferably, the intelligent greenhouse control system includes:

[0026] A grayscale image processing terminal, which acquires the area image associated with each spraying area in the intelligent greenhouse based on a machine vision device, and performs grayscale processing on the acquired area image to obtain a grayscale image;

[0027] A contour feature processing terminal, which confirms several contour areas associated within the grayscale image based on the grayscale value features of different points within the grayscale image;

[0028] A flowering period area calibration terminal, which, based on the different contour areas confirmed within the grayscale image corresponding to the spraying area, identifies whether the corresponding contour area belongs to the flowering period area according to the contour features associated with the different contour areas and performs calibration, and confirms the spraying range for the grayscale image with a flowering period area;

[0029] An execution interval determination terminal, which re-analyzes the grayscale image after the spraying range is confirmed, confirms the water flow spraying speed of the corresponding spray nozzle, and makes the spraying range of the sprayed water not cover the flowering period area through the gradually confirmed water flow spraying speed. Based on the confirmation result, an execution interval for the corresponding spraying area is generated.

[0030] The present invention provides an intelligent greenhouse control system and method. Compared with the prior art, it has the following beneficial effects:

[0031] The present invention acquires the image of the spraying area through a high-definition probe, accurately identifies the contour area by using grayscale processing and the Sobel algorithm, and then accurately calibrates the flowering period area in combination with the grayscale interval of the flower color characteristics;

[0032] On this basis, the non-sprayable areas are reasonably planned, and the spraying speed of the spray nozzles is accurately regulated, effectively avoiding damage to the flowering crops caused by the sprayed water, and greatly improving the growth quality and yield of the crops;

[0033] Compared with the traditional greenhouse spraying, the incidence of problems such as damaged flowers and pollination failure caused by improper spraying is significantly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic flow chart of the method of the present invention;

[0035] Figure 2 It is a schematic diagram for determining the non-sprayable area of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0037] The First Embodiment

[0038] Please refer to Figure 1 , the present application provides an intelligent greenhouse control method, including the following steps:

[0039] Step1. Obtain the area images associated with each spraying area in the intelligent greenhouse, perform grayscale processing on the obtained area images to obtain grayscale images, and then based on the grayscale value characteristics of different points in the grayscale images, confirm several contour areas associated in the grayscale images. Specifically, a corresponding water delivery cross beam is arranged at the upper position inside the intelligent greenhouse, and several spray nozzles are arranged below the corresponding cross beam. A high-definition probe is synchronously arranged on one side of the spray nozzle, and the detection range of the high-definition probe is the same as the spraying range of the spray nozzle. The main function of the high-definition probe is to identify whether the relevant crops planted in the corresponding area belong to the seedling stage or the flowering stage. The spraying methods associated with the seedling stage and the flowering stage are different, and the corresponding spray nozzles can effectively adjust the spraying range below, changing from a small range to a large range. When the spraying range changes, only the spraying speed needs to be changed. The intensity of the spraying speed can change the distance of the corresponding dust, so as to effectively change the corresponding spraying range;

[0040] Among them, the specific method for performing grayscale processing on the area image is:

[0041] Based on the different area images obtained for each spraying area, confirm the RGB values associated with different pixel points in the specified area image: Using HD i= 0.299R + 0.587G + 0.114B to confirm the gray value HD associated with different pixel points i , where 0.299, 0.587, and 0.114 are all weight factors preset for the corresponding pixel features. Here, i represents different pixel points associated within the corresponding regional image, and different gray values HD are confirmed based on different pixel points i , and perform grayscale processing on this regional image to generate a grayscale image belonging to this regional image;

[0042] The specific method for confirming the contour area of the grayscale image is as follows:

[0043] Use the Sobel algorithm to confirm the vertical gradient and the vertical gradient associated with different pixel points in the grayscale image, and confirm the comprehensive gradient ZH associated with the corresponding pixel points based on the vertical gradient and the vertical gradient i , where X i is the vertical gradient, and its Y i is the vertical gradient. Specifically, the Sobel algorithm takes the corresponding pixel point as the midpoint, combines the specific pixel values of the surrounding eight groups of pixel points, and assigns different weight factors in gradient confirmation. The value range of the weight factor is: [-2, 2], and includes the value 0. Then, sum up the pixel values with the assigned weight factors to confirm the corresponding vertical gradient and vertical gradient. Since the method of the Sobel algorithm for confirming pixel point gradients is relatively common in the prior art, it will not be elaborated here;

[0044] Compare the confirmed comprehensive gradient ZH i with the preset value Y1. Pixel points that satisfy: ZH i ≥ Y1 are marked as gradient pixel points. Otherwise, no marking is performed. Y1 is a preset determination threshold, generally taking the value 5, which is determined by the operator in advance according to experience;

[0045] Mark the continuous gradient pixel points as the feature contour (that is, adjacent gradient pixel points are all the corresponding feature contours). Mark the area included in the complete and closed feature contour as the contour area in the grayscale image. Otherwise, no specific marking of the contour area is performed (that is, in the case of non-closed, no marking is required. Normally, the non-closed case belongs to the empty area. There are also corresponding feature contours around the air area. However, due to the specific angle during the acquisition of the corresponding image, it will limit the inability to obtain the entire empty area, resulting in the contour around the corresponding contour area unable to form a closed loop. Normally, for the corresponding green plants or flower areas, there will be corresponding contours);

[0046] Step 2. Based on the different contour regions identified in the grayscale image corresponding to the spraying area, identify whether the corresponding contour region belongs to the flowering period region and mark it according to the contour features associated with the different contour regions, and confirm the spraying range of the grayscale image with the flowering period region. The specific sub-steps for confirmation are as follows:

[0047] Step 21. For several groups of contour regions identified in a single group of grayscale images, confirm the grayscale values associated with the pixel points inside the corresponding contour regions, and confirm the minimum grayscale value and the maximum grayscale value therefrom, generate the grayscale features of the corresponding contour regions, and sequentially confirm the grayscale features associated with the different contour regions;

[0048] Step 22. Compare the grayscale features associated with the different contour regions with a preset numerical range. The end point values of the numerical range are both preset values, and their specific values are determined by the operator according to experience. The numerical range is the grayscale feature associated with the corresponding flower color feature, and the specific confirmation of the numerical range can be based on the grayscale representation related to the corresponding flower:

[0049] If the grayscale feature belongs to the numerical range, mark this contour region as the flowering period region;

[0050] If the grayscale feature does not belong to the numerical range, identify the intersection range between the grayscale feature and the corresponding numerical range. If the proportion of the intersection range in the entire grayscale feature range exceeds 90% (that is: intersection range ÷ total grayscale feature range > 90%), mark this contour region as the flowering period region; otherwise (that is, the corresponding proportion does not exceed 90%), do not perform any marking;

[0051] Step 23. Combine Figure 2 , for the grayscale image with the flowering period region: take the image edge of this grayscale image as a feature edge, and combine the feature edge with the two-dimensional coordinate system (it can be randomly combined as long as the internal edge points of the feature edge have coordinate representations in the two-dimensional coordinate system, that is, the corresponding edge points have corresponding two-dimensional coordinates), confirm the two-dimensional coordinates associated with the different edge points inside the feature edge, and then confirm the center point of this grayscale image by confirming the mean coordinates of all the two-dimensional coordinates of several groups of edge points;

[0052] Based on the confirmed center point and the marked flowering period region, confirm the point with the shortest straight-line distance and the point with the longest straight-line distance from the center point on the edge contour of the flowering period region. Denote the point with the shortest straight-line distance as the inner circle point and the point with the longest straight-line distance as the outer circle point;

[0053] Construct a set of circles with the line connecting the inner marked points and the center point as the radius and the center point as the center. Denote these circles as the inner circle. Then, construct another set of circles with the line connecting the outer marked points and the center point as the radius and the center point as the center. Denote these circles as the outer circle. Denote the area between the inner circle and the outer circle as the non-sprayable area and mark it within the associated grayscale image;

[0054] Specifically, according to the confirmed flowering period area in the corresponding grayscale image, in order to effectively protect the flowers from damage, during spraying, it is necessary to control that the area associated with the flowers is not sprayed with sprinkler water. Then, during the spraying control process, confirm the specific spraying speed of the corresponding spray nozzle in the spray area associated with the corresponding grayscale image to determine the specific dust emission, effectively avoid such flowering period areas, and effectively confirm the spraying logic associated with the specified grayscale image;

[0055] Step3. Re-analyze the grayscale image for which the spraying range has been confirmed, confirm the water flow speed of the corresponding spray nozzle, and make the spraying range of the sprinkler water not cover the flowering period area through the gradually confirmed water flow speed. Based on the confirmation result, generate the execution interval for the corresponding spray area. The specific sub-steps for generating the execution interval are as follows:

[0056] Step31. According to the preset water flow speed interval of each spray nozzle, during the process of the water flow speed interval gradually changing from the minimum value to the maximum value, the generated sprinkler water effectively sprays and covers the entire spray area. The end values of the water flow speed interval are both preset values, which are determined by the operator in advance based on experience;

[0057] Step32. Based on the non-sprayable area confirmed in the corresponding grayscale image, confirm the inner circle radius and outer circle radius of the non-sprayable area, lock the characteristic range belonging to this non-sprayable area, and then based on the center point confirmed in this grayscale image, determine the edge point on the characteristic edge of the grayscale image that is the farthest from this center point in a straight line distance. Denote the distance between this edge point and the center point as L;

[0058] Step33. Map the numerical range of X1 - L and the water flow speed interval together. X1 is a preset value, generally taking 5 cm. Confirm the mapping characteristics, denote the water flow speed interval as [Vmin, Vmax], use (Vmax - Vmin) ÷ (L - X1) = Tz to confirm the mapping characteristic value Tz, and denote the characteristic range as [R1, R2]. Use (R1 - X1) × Tz + Vmin = V R1 and (R2 - X1) × Tz + Vmin = V R2 Lock the spray speed range [V R1 , V R2 associated with the characteristic range;

[0059] Step 34. Confirm the spray area associated with this grayscale image, and exclude the spray speed range [V R1 , V R2 from the water flow spray speed range of this spray area. Then, use the water flow spray speed range after the exclusion process as the execution range, and control the spray nozzles of the corresponding spray area to perform relevant operations according to the water flow spray speed within the execution range, so as to ensure that the sprayed water generated by the corresponding spray nozzles does not cover the flowering period area. For example, assume that the confirmed spray speed range [V R1 , V R2 is [2, 4], and the associated water flow spray speed range is [1, 10]. Then, after the specific exclusion of the corresponding spray speed range, the corrected execution range is [1, 2), (4, 10]. In the subsequent spray execution process, the specific state of such spray speed execution can effectively ensure that the corresponding water flow will not be sprayed on the corresponding flowering period area.

[0060] Second Embodiment

[0061] The intelligent greenhouse control system includes:

[0062] A grayscale image processing terminal, which acquires the area image associated with each spray area in the intelligent greenhouse based on a machine vision device, and performs grayscale processing on the acquired area image to obtain a grayscale image;

[0063] A contour feature processing terminal, which confirms several contour areas associated in the grayscale image based on the grayscale value features of different points in the grayscale image;

[0064] A flowering period area calibration terminal, which, based on the different contour areas confirmed in the grayscale image corresponding to the spray area, identifies whether the corresponding contour area belongs to the flowering period area and calibrates it according to the contour features associated with the different contour areas, and confirms the spray range of the grayscale image with a flowering period area;

[0065] An execution range determination terminal, which re-analyzes the grayscale image for which the spray range has been confirmed, confirms the water flow spray speed of the corresponding spray nozzle, and ensures that the spray range of the sprayed water does not cover the flowering period area through the gradually confirmed water flow spray speed. Based on the confirmation result, an execution range corresponding to the spray area is generated.

[0066] Some of the data in the above formula are numerically calculated after removing their dimensions, and the content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

[0067] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The intelligent greenhouse control method is characterized in that, It includes the following steps: Step 1: Obtain the area images associated with each sprinkler area in the intelligent greenhouse, perform grayscale processing on the obtained area images to obtain grayscale images, and then based on the grayscale value characteristics of different points in the grayscale images, confirm several contour areas associated in the grayscale images; Step 2: Based on the different contour areas confirmed in the grayscale images corresponding to the sprinkler areas, according to the contour characteristics associated with the different contour areas, identify whether the corresponding contour areas belong to the flowering period areas and mark them, and confirm the sprinkling range of the grayscale images with flowering period areas; Step 3: Re-analyze the grayscale images for which the sprinkling range has been confirmed, confirm the water spray speed of the corresponding sprinkler nozzles, and make the sprinkling range of the sprinkling water not cover the flowering period areas through the gradually confirmed water spray speeds. Based on the confirmation results, generate the execution intervals for the corresponding sprinkler areas.

2. The intelligent greenhouse control method according to claim 1, wherein In Step 1, the specific method for performing grayscale processing on the obtained area images is: Based on the different area images obtained for each sprinkler area, the RGB values associated with different pixel points in the specified area image are confirmed: Using HD i = 0.299R + 0.587G + 0.114B to confirm the grayscale value HD i , where 0.299, 0.587, and 0.114 are all weight factors preset for the corresponding pixel features, where i represents different pixel points associated within the corresponding area image. Based on the different grayscale values confirmed for different pixel points HD i , this area image is grayscale processed to generate a grayscale image belonging to this area image.

3. The intelligent greenhouse control method according to claim 2, wherein In Step 1, the specific method for confirming the contour areas in the grayscale images is: The Sobel algorithm is used to confirm the vertical gradient and the vertical gradient associated with different pixel points in the grayscale image, and the comprehensive gradient ZH associated with the corresponding pixel points is confirmed based on the vertical gradient and the vertical gradient i , where X i is the vertical gradient, and its Y i is the vertical gradient; Compare the confirmed comprehensive gradient ZH i with the preset value Y1. If it satisfies: ZH i ≥Y1, mark the pixel points as gradient pixel points; otherwise, do not perform any marking. Here, Y1 is the preset determination threshold. Calibrate the characteristic contours of the continuously appearing gradient pixel points, and mark the area included in the complete and closed characteristic contour as the contour area in the grayscale image. Otherwise, no specific calibration of the contour area is performed.

4. The intelligent greenhouse control method according to claim 1, wherein In Step 2, the specific method for identifying whether the corresponding contour areas belong to the flowering period areas is: Step 21: For several groups of contour areas confirmed in a single group of grayscale images, confirm the grayscale values associated with the pixel points inside the corresponding contour areas, and then confirm the minimum grayscale value and the maximum grayscale value from them, generate the grayscale characteristics of the corresponding contour areas, and sequentially confirm the grayscale characteristics associated with different contour areas; Step 22: Compare the grayscale characteristics associated with different contour areas with a preset numerical interval, and the end point values of the numerical interval are both preset values: If the grayscale characteristic belongs to the numerical interval, mark this contour area as the flowering period area; If the grayscale characteristic does not belong to the numerical interval, identify the intersection range between the grayscale characteristic and the corresponding numerical interval. If the proportion of the intersection range in the entire grayscale characteristic range exceeds 90%, mark this contour area as the flowering period area. Otherwise, no marking is performed.

5. The intelligent greenhouse control method according to claim 4, wherein, In Step 2, the specific method for confirming the sprinkling range of the grayscale images is: Step 23: For the grayscale images with flowering period areas: Use the image edge of this grayscale image as a characteristic edge, combine the characteristic edge with the two-dimensional coordinate system, confirm the two-dimensional coordinates associated with different edge points inside the characteristic edge, and then confirm the center point of this grayscale image by confirming the average coordinates of all the two-dimensional coordinates of several groups of edge points; Based on the confirmed center point and the marked flowering period areas, confirm the point with the shortest straight-line distance and the farthest point from the center point on the edge contour of the flowering period area. Mark the point with the shortest straight-line distance as the inner circle point and the point with the farthest straight-line distance as the outer circle point; Construct a set of circles with the line connecting the inner dot and the center point as the radius and the center point as the center of the circle. Denote this set of circles as the inner circle. Then, construct another set of circles with the line connecting the outer dot and the center point as the radius and the center point as the center of the circle. Denote this set of circles as the outer circle. Denote the area included between the inner circle and the outer circle as the non-sprayable area, and mark it within the associated grayscale image.

6. The intelligent greenhouse control method according to claim 1, wherein, In the said Step 3, the specific sub-steps for generating the execution interval are as follows: Step 31: According to the preset water flow spray speed interval of each spray nozzle, during the process of gradually changing from the minimum value to the maximum value of the water flow spray speed interval, the generated spray water effectively sprays and covers the entire spray area, and the end point values of the water flow spray speed interval are all preset values; Step 32: Based on the non-sprayable area confirmed within the corresponding grayscale image, confirm the inner circle radius and the outer circle radius of the non-sprayable area, lock the characteristic range belonging to this non-sprayable area, and then based on the center point confirmed within this grayscale image, determine the edge point on the characteristic edge of the grayscale image that is the farthest from this center point in a straight line distance. Denote the distance between this edge point and the center point as L; Step 33. Uniformly map the numerical range of X1-L, where X1 is a preset value, confirm the mapping characteristics, and set the water flow spray speed range as [Vmin, Vmax]. Use (Vmax - Vmin) ÷ (L - X1) = Tz to confirm the mapping characteristic value Tz, and calibrate the characteristic range as [R1, R2]. Use (R1 - X1) × Tz + Vmin = V R1 and (R2 - X1) × Tz + Vmin = V R2 Lock the spray speed range [V R1 , V R2 associated with the characteristic range; Step34. Confirm the spray area associated with this grayscale image, and exclude the spray speed range [V R1 , V R2 from the water flow spray speed range of this spray area. Take the water flow spray speed range after the exclusion process as the execution range, and control the spray nozzles of the corresponding spray area to perform relevant operations according to the water flow spray speed within the execution range.

7. Intelligent greenhouse control system, which operates according to the intelligent greenhouse control method described in any one of claims 1-6, characterized in that, Including: The grayscale image processing terminal acquires the area image associated with each spray area in the intelligent greenhouse based on the machine vision device, and performs grayscale processing on the acquired area image to obtain a grayscale image; The contour feature processing terminal confirms several contour areas associated within the grayscale image based on the grayscale value features of different points within the grayscale image; The flowering period area calibration terminal, based on the different contour areas confirmed within the grayscale image corresponding to the spray area, identifies whether the corresponding contour area belongs to the flowering period area and calibrates it according to the contour features associated with the different contour areas, and confirms the spray range of the grayscale image with the flowering period area; The execution interval determination terminal re-analyzes the grayscale image for which the spray range has been confirmed, confirms the water flow spray speed of the corresponding spray nozzle, and makes the spray range of the spray water not cover the flowering period area through the gradually confirmed water flow spray speed. Based on the confirmation result, generate the execution interval for the corresponding spray area.

Images