Method for calculating covering state of greenhouse top covering object based on unmanned aerial vehicle image recognition

Abstract

The invention discloses a method for calculating the covering state of a greenhouse top covering object based on unmanned aerial vehicle image recognition. The method comprises the following specific calculation steps: S1, carrying out structural analysis on a greenhouse house structure; s2, the unmanned aerial vehicle shoots the state picture data of the greenhouse sunshade net cloth, further conversion calculation is carried out to obtain the real covering state of the rolling covering object, and further conversion is carried out according to the covering object state to obtain the rolling opening degree and the covering area of the covering object; and S3, calculating the operation efficiency phi (t) of the sun-shading net of the sunlight greenhouse. According to the invention, the unmanned aerial vehicle platform is utilized to realize greenhouse covering rolling state monitoring and effect evaluation, and the method has important practical significance for realizing effective control and improving the greenhouse covering rolling operation efficiency. The research takes the sunshade net cloth rolling operation of the sunlight greenhouse as an example, and realizes the method process of identifying, calculating and evaluating the rolling state of the covering by the unmanned aerial vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of greenhouse top covering monitoring, and in particular to a method for calculating the covering state of a greenhouse top covering based on unmanned aerial vehicle image recognition. Background Art

[0002] As a conventional structural component of the global greenhouse system, covering materials play an important role in isolating the external environment to a certain extent and forming a relatively independent and controllable greenhouse internal environment. Greenhouse coverings enable agricultural activities to continue and allow the internal environment to be manipulated, and have become a traditional and important component of agricultural facilities. The installation location and form of covering materials usually vary significantly due to their functional properties and the type of greenhouse they belong to. Due to their structural characteristics, large multi-span greenhouse structures often require the construction of external support structures to achieve the rolling operation of coverings. For single-structure greenhouses such as solar greenhouses, the rolling operation of coverings is mostly achieved by relying on their basic structural surfaces. There are even assembled solar greenhouses that are completely composed of metal frames and coverings. Compared with other types of greenhouses, solar greenhouses also rely more on the rolling operation of coverings to regulate the internal environment of the greenhouse.

[0003] The three types of coverings on the front roof of a solar greenhouse (insulation blankets, light-transmitting films, and shade mesh) are particularly representative in greenhouse production. As a routine and necessary task in greenhouse production, rolling up coverings is essential for improving the greenhouse environment and increasing greenhouse production efficiency. In summer greenhouse production, it is usually necessary to cover the top of the greenhouse with a colored net at noon to reduce indoor solar radiation and slow down the rise in air temperature. Among them, the summer heat load of a solar greenhouse is relatively high, and covering with a colored shade net is a key production measure for summer production in a solar greenhouse. The rolling colored net is installed on the front roof on the south side of the solar greenhouse, and the mechanism of rolling / unrolling the covering is similar in various greenhouses. It is difficult to effectively monitor and evaluate the current rolling status of the solar greenhouse covering, and it is impossible to effectively adjust the internal temperature of the greenhouse according to the covering situation. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems in the related art to a certain extent. To this end, one purpose of the present invention is to propose a method for calculating the covering state of the greenhouse roof cover based on drone image recognition, which solves the problem that the current rolling state of the solar greenhouse cover is difficult to effectively monitor and evaluate.

[0005] According to the present invention, a method for calculating the covering state of a greenhouse roof based on drone image recognition is characterized in that the specific calculation steps are as follows:

[0006] S1: Structural analysis of the greenhouse structure, statistics of the size and angle of the greenhouse structure, and statistics of the roll-up range of the greenhouse roof covering;

[0007] S2: The drone captures the greenhouse sunshade cloth status image data, and further converts and calculates the real covering status of the rolled covering, and further converts the roll opening and covering area of ​​the covering according to the covering status;

[0008] S3: On the roof above and below the sunshade cloth, there are fixed-size upper and lower ventilation membranes. The transmittance of the sunshade cloth is input into the calculation system to calculate the operating efficiency of the rolled sunshade cloth in the solar greenhouse.

[0009]

[0010] Among them, Q s (t) is the total amount of solar radiation received by the ground planting area of ​​the solar greenhouse covered with shade nets, J·s -1 ;Q o (t) is the total solar radiation at the same time without roller blinds, J·s -1 ; I(t) is the real-time illumination radiation, W·m -2 ; γ is the transmittance of the transparent film, that is, the transmittance of the upper ventilation membrane and the lower ventilation membrane; τ is the shading rate of the shading net, which is considered to be a constant value of the overall shading net on the front roof of the solar greenhouse.

[0011] In some embodiments of the present invention, the size and angle statistics of the greenhouse house structure and the rolling range statistics of the greenhouse top cover are performed, and the specific parameter numbers are as follows: greenhouse length S m, the width of the entire south roof L m and the greenhouse height H m, o and R are set to be the arc center and radius of the greenhouse arched front roof, θ1 and l1 are the central angle and arc length of the insulation rolling range, θ2, θ3 and θ4 are the central angles of the rolling range of the upper vent, the sunshade net and the lower vent, l2, l3 and l4 are the arc lengths corresponding to θ2, θ3 and θ4;

[0012] The position of the covering can be determined by the opening K x The standard calculation formula is determined as:

[0013]

[0014] Among them, K x is the opening of different coverings, x = 1, 2, 3, 4 represent the insulation blanket, upper vent, sunshade net and lower vent respectively; θ x and Δθ x Indicates the change in the center angle of the front roof and the center angle of the corresponding covering, rad; l x and Δlx represents the corresponding arc length, m; A x and ΔA x Represents the corresponding coverage area, m 2 .

[0015] In other embodiments of the present invention, the roll opening and the covering area of ​​the covering are further converted according to the state of the covering, and the arc equation of the corresponding light-receiving surface is constructed according to the simplified scheme of the front roof:

[0016] x 2 +(z+RH)) 2 =R 2

[0017]

[0018]

[0019] x and z are the width and height of the light-receiving arc surface in spatial coordinates.

[0020] In some other embodiments of the present invention, the starting point of the covering is point b (x b ,z b ), the cover is placed at point c (x c ,z c ), the arc distance between the two points is:

[0021]

[0022] In other embodiments of the present invention, the end point of the shading net is f(x f , z f ), real-time location (x f (t), z f (t)), and obtain the actual opening degree K of the sunshade net s (t):

[0023]

[0024] In other embodiments of the present invention, when sunlight passes through the shading net, the diffuse sky radiation is greatly absorbed and reflected to form diffuse sky transmission, a part of the direct radiation is also absorbed and reflected to form a diffuse beam, and another part of the direct radiation is reflected by the tiny mirror surface of the shading net. Only a small part of the direct radiation passes through the mesh of the shading net and the transparent material to reach the ground in the greenhouse, and the real-time shadow area A md1 To a large extent, it represents the actual operating effect of the shadow net, which can be calculated using the following equation:

[0025] A md1 (t) = L md1 (t)×Smd (t)=(L m1 -H m ×tanα(t))×(S m -H m ×tanσ(t)

[0026] The real-time shadow area A can be obtained by the above formula d (t):

[0027]

[0028] In the present invention, the unmanned aerial vehicle platform is used to monitor the rolling state of greenhouse coverings and evaluate their effects, which has important practical significance for achieving effective control and improving the efficiency of greenhouse covering rolling operations. The study will take the rolling operation of sunshade mesh in a solar greenhouse as an example, and implement the method process of unmanned aerial vehicle identification, calculation and evaluation of the rolling state of coverings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0030] Figure 1 This is a schematic diagram of the calculation principle of the covering status of the greenhouse top cover based on drone image recognition proposed by the present invention.

[0031] Figure 2 This is a schematic diagram of the covering effect on the top cover of a greenhouse based on drone image recognition proposed by the present invention.

[0032] Figure 3 This is a schematic diagram of the rolling state of a greenhouse top cover based on drone image recognition proposed by the present invention.

[0033] Figure 4 This is a schematic diagram of the actual coverage area of ​​the greenhouse top covering based on drone image recognition proposed in the present invention.

[0034] Figure 5 A schematic diagram of the rolling range of a greenhouse top cover based on drone image recognition proposed in the present invention. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0036] Using drones to collect and identify the rolling status of greenhouse top covers is applicable to most greenhouses and can also be applied to large-scale cluster greenhouse production, as follows Figure 1 The figure shows the workflow of using drones to identify the covering status of a solar greenhouse. The operator collects aerial images of the greenhouse through a portable drone system (drone and ground station), and the image data is transmitted to the monitoring center through a base station or a dedicated network. During the image analysis process, a single greenhouse target object is extracted and the covering on the front roof of the solar greenhouse is identified. Finally, the rolling status of the covering and its impact on the internal environment are evaluated.

[0037] exist Figure 1 Where o and R (m) are the arc center and radius of the arched front roof of the solar greenhouse; θ1 (rad) and l1 (m) are the central angle and arc length of the insulation roll-up range; similarly, θ2, θ3 and θ4 are the central angles of the roll-up range of the upper vent, sunshade net and lower vent; l2, l3 and l4 are the corresponding arc lengths (positions in Figure 1 center upper right corner).

[0038] The present invention proposes a method for identifying and analyzing a single solar greenhouse and its covering. The related theories and algorithms involved can also be extended to the production of cluster solar greenhouses and other types of greenhouses. The three coverings on the front roof of the solar greenhouse are made of different materials and colors according to their different environmental control functions, and the actual rolling range and application season scenarios are also different. On this basis, the drone can be used to evaluate the real-time rolling status of the three coverings (sunshade nets and upper and lower ventilation membranes, the upper and lower ventilation membranes are fixed in size and cannot be rolled up), and some parameters are determined during the flight process for further theoretical analysis.

[0039] Experimental, solar greenhouse construction parameters: length S (unit: m), width L (unit: m) of the entire south roof and height H (unit: m). Most greenhouses currently use standardized construction techniques. The construction size parameters of clustered solar greenhouses are usually consistent and are usually easy to measure and obtain.

[0040] Experimental, rolling state of the top cover of the solar greenhouse: real-time width (position information) and deformation state information of the cover.

[0041] The arched front roof of the experimental solar greenhouse is simplified to an arc to facilitate our research in the theoretical analysis process, so that the center o of the arc on the shed surface and the corresponding arc radius R can be determined. The position of the covering can be determined by the opening K x The standard calculation formula is determined as:

[0042]

[0043] Where K xis the opening of different coverings, x = 1, 2, 3, 4, representing the thermal blanket, upper vent, sunshade net and lower vent; θ x and Δθ x Indicates the change in the center angle of the front roof and the center angle of the corresponding covering, rad; l x and Δl x represents the corresponding arc length, m; A x and ΔA x Represents the corresponding coverage area, m 2 .

[0044] Among them, the upper and lower vents of the solar greenhouse are usually kept fully open during the summer production period. Therefore, the present invention will focus on the rolling operation of the sunshade net, specifically including the specific working principle and calculation processing of analyzing the image information obtained by the drone, aiming to assist production personnel to use the drone to quickly evaluate and adjust the rolling state of the sunshade net, so as to improve the operating efficiency of the sunshade net in the summer solar greenhouse.

[0045] Working principle of greenhouse shading net: determination of solar altitude angle and solar azimuth angle

[0046] The sunshade net is mainly used to reduce the indoor solar radiation level and slow down the temperature rise at noon. It is necessary to first calculate the real-time solar illumination outside the greenhouse. The solar altitude angle α refers to the angle between the incident direction of sunlight and the horizontal plane, which is obtained by equation (2).

[0047] sinα=cosβcosδcosω+sinβsinδ (2)

[0048] The solar azimuth angle σ refers to the angle between the projection of sunlight on the ground and the local meridian, which can be approximated as the angle between the horizontal projection of sunlight and the south direction. Among them, south-west is a positive value and south-east is a negative value. It is obtained by equations (3) and (4).

[0049]

[0050] Where β represents the local north latitude; δ is the solar declination angle; ω represents the time angle. δ is the angle between the Earth's equatorial plane and the line connecting the Sun and the center of the Earth, obtained by equation (5).

[0051]

[0052] Where N is the day of the year (January 1 corresponds to 1, January 2 corresponds to 2, and so on. December 31 corresponds to 365 in a normal year and 366 in a leap year. The solar altitude, solar azimuth, and solar radiation irradiance at the upper atmospheric interface at a given time are calculated.

[0053] According to equations (2), (3) and (4), ω is a key parameter related to the values ​​of α and σ, which also determines the time when the colored sunshade net is rolled up and down, which is obtained by equation (6)

[0054]

[0055] Where Tm is the local true solar time; Tp is the local mean solar time, t is Beijing time; γ is the longitude of the local location. The time difference e refers to the difference between the true solar time and the mean solar time, which is usually determined by the empirical equation:

[0056] e≈9.87sin2B-7.53cosB-1.5sinB (7)

[0057] Where B is a constant related to N, and its value is 1.978×(N-81).

[0058] like Figure 2 , the external solar radiation collection of greenhouse buildings mainly consists of three parts: direct radiation, diffuse radiation, and diffuse solar radiation reflected from the external ground. In actual production, the commonly used photoelectric total solar radiation sensors mainly include direct radiation and diffuse radiation. Due to its relatively small proportion and inconvenient measurement, the diffuse reflection of solar radiation from the external ground is ignored. Usually, when external solar radiation passes through transparent materials (glass and transparent films) on the surface of the greenhouse, it is weakened by absorption, reflection and refraction. The basic working principle of the horizontal shading net with external structure commonly used in various greenhouse productions is explained below:

[0059] like Figure 2 , when sunlight passes through the shading net, the diffuse sky radiation is greatly absorbed and reflected, forming a diffuse sky transmission. A part of the direct radiation is also absorbed and reflected to form a diffuse beam, and another part of the direct radiation is reflected by the tiny mirror of the shading net. Only a small part of the direct radiation passes through the mesh of the shading net and the transparent material to reach the ground inside the greenhouse. Finally, under the influence of the shading net, a shaded area is formed on the ground. Therefore, the real-time shadow area A md1 To a large extent, it represents the actual operating effect of the shadow net, which can be calculated using the following equation:

[0060] A md1 (t) = L md1 (t)×S md (t)=(L m1 -H m ×tanα(t))×(S m -H m ×tanσ(t)) (8)

[0061] from Figure 2It can be seen from equation (8) that rolling up the shade net on a plane makes it easier to use drones to calculate and evaluate the actual operating effect of the shade net, such as a horizontal shade net. Based on the above basic working principle analysis, a calculation method for evaluating the rolling state of a more complex solar greenhouse shade net is proposed.

[0062] like Figure 3 The arched structure of the front roof of the solar greenhouse makes it impossible to directly obtain the size parameters of the rolled covering material through the drone. The image data of the solar greenhouse sunshade cloth status captured by the drone needs to be further converted and calculated to obtain its true covering status. The following will take the current common solar greenhouse external sunshade covering method as an example to explain the detailed method.

[0063] The roll opening and blanket area of ​​the covering can be further converted according to the state of the covering. First, the arc equation of the corresponding light-receiving surface needs to be constructed according to the simplified scheme of the front roof:

[0064] x 2 +(z+(RH)) 2 =R 2 (9)

[0065]

[0066]

[0067] like Figure 3 As shown, point b (x b ,z b ) and point c (x c ,z c ) correspond to the starting and ending positions of the cover on the south roof. Using the two points and the center of the circle to form an isosceles triangle, we get the corresponding trigonometric function:

[0068]

[0069] Length S (unit: m), width L (unit: m) and height H (unit: m) of the entire south roof.

[0070] Then, point b (x b ,z b ) and point c (x c ,z c )The arc distance between two points is:

[0071]

[0072] L P1 is the horizontal width of the upper vent, L P is the horizontal width of the cover.

[0073] According to equation (13), the parameter A obtained from the drone image is p It can be converted into the actual coverage area A, and then the true opening value is calculated by equation (1). When the rolling cover is significantly deformed, the above calculation method can be further developed through finite element analysis theory to obtain the relevant formula for calculating the actual coverage area, such as Figure 3 shown.

[0074] The calculation formula involved is as follows: Similarly, the parameter A obtained through drone images pc (Photo coverage) can be converted into the actual coverage area A through equation (14): c ;

[0075]

[0076] Among them, X(y) is the actual curve fitting equation of the deformed cover of the solar greenhouse; X'(y) is the curve fitting equation of the deformed cover photographed by the drone. X(y) is an equation that passes through the actual covered area of ​​each greenhouse (not appearing under the curve, i.e. Figure 4 The intersection of the middle covered area and the uncovered area is a straight line. Figure 5 is a straight line), and X'(y) is the area photographed by the helicopter. In fact, the intersection of the covered area and the uncovered area is not completely a straight line, that is, Figure 4 X(y) and X'(y) are calculation formulas given based on actual conditions.

[0077] Evaluating the rolling status of summer production mulch in solar greenhouses

[0078] like Figure 5 , for the summer production of daylight, the rolling state of the sunshade net and ventilation film (upper and lower vents) is the key factor affecting its indoor light temperature environment. Among them, the opening of the upper vent and the lower vent (the vent position is the ventilation film) is usually a constant value, which can be calculated by the method proposed in Section 4.3.1. The transparent film is fixed, which also corresponds to the rolling range of the sunshade net, and the coordinates of the two points are considered to be known values.

[0079] like Figure 5 As shown, point f(x f , z f ) is the end point of the shading net, and the real-time position (x f (t), z f (t)) can be calculated by methods (9)(10)(11)(12)(13). In addition, the actual opening degree K of the sunshade net can be obtained. s (t).

[0080]

[0081] Due to the shading of the east and west walls of the solar greenhouse, the effect of the change of the solar azimuth angle on the shadow area inside the solar greenhouse is ignored. According to equation (8), the real-time shadow area A can be obtained: d (t).

[0082]

[0083] Where Q s (t) is the total amount of solar radiation received by the ground planting area of ​​the solar greenhouse covered with shade nets, J·s -1 ;Q o (t) is the total solar radiation at the same time without roller blinds, J·s -1 ; I(t) is the real-time illumination radiation, W·m -2 ; γ is the transmittance of the transparent film; τ is the shading rate of the shading net, which is considered to be a constant value of the overall shading net on the front roof of the solar greenhouse, ignoring the influence of the stacking phenomenon of shading in some areas.

[0084] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for calculating the covering state of a greenhouse roof based on drone image recognition, characterized in that: The specific calculation steps are as follows: S1: Structural analysis of the greenhouse structure, statistics of the size and angle of the greenhouse structure, and statistics of the roll-up range of the greenhouse roof covering; S2: The drone captures the greenhouse sunshade cloth status image data, and further converts and calculates the real covering status of the rolled covering, and further converts the roll opening and covering area of ​​the covering according to the covering status; S3: On the roof above and below the sunshade cloth, there are fixed-size upper and lower ventilation membranes. The transmittance of the sunshade cloth is input into the calculation system to calculate the operating efficiency of the rolled sunshade cloth in the solar greenhouse. Among them, Q s (t) is the total amount of solar radiation received by the ground planting area of ​​the solar greenhouse covered with shade nets, J·s -1 ;Q o (t) is the total solar radiation at the same time without roller blinds, J·s -1 ; I(t) is the real-time illumination radiation, W·m -2 ; γ is the transmittance of the transparent film, that is, the transmittance of the upper ventilation membrane and the lower ventilation membrane; τ is the shading rate of the shading net, which is considered to be a constant value of the overall shading net on the front roof of the solar greenhouse.

2. The method for calculating the covering state of a greenhouse top cover based on drone image recognition according to claim 1 is characterized in that: The size and angle statistics of the greenhouse structure and the rolling range of the greenhouse roof covering are counted. The specific parameter numbers are as follows: greenhouse length Sm, the width of the entire south roof Lm and the greenhouse height Hm. Set o and R to be the arc center and radius of the greenhouse arched front roof, θ1 and l1 to be the central angle and arc length of the insulation rolling range, θ2, θ3 and θ4 to be the central angle of the rolling range of the upper vent, sunshade net and lower vent, l2, l3 and l4 to be the arc lengths corresponding to θ2, θ3 and θ4; The position of the covering can be determined by the opening K x The standard calculation formula is determined as: Among them, K x is the opening of different coverings, x = 1, 2, 3, 4 represent the insulation blanket, upper vent, sunshade net and lower vent respectively; θ x and Δθ x Indicates the change in the center angle of the front roof and the center angle of the corresponding covering, rad; l x and Δl x represents the corresponding arc length, m; A x and ΔA x Represents the corresponding coverage area, m 2 .

3. The method for calculating the covering state of the greenhouse top cover based on drone image recognition according to claim 2 is characterized in that: The roll opening and covering area of ​​the covering are further converted according to the state of the covering, and the arc equation of the corresponding light-receiving surface is constructed according to the simplified scheme of the front roof: x 2 +(z+(R-H)) 2 =R 2 x and z are the width and height of the light-receiving arc surface in spatial coordinates.

4. The method for calculating the covering state of the greenhouse top cover based on drone image recognition according to claim 3 is characterized in that: The starting point of the covering is point b (x b ,z b ), the cover is placed at point c (x c ,z c ), the arc distance between the two points is:

5. The method for calculating the covering state of a greenhouse top cover based on drone image recognition according to claim 2 is characterized in that: The end point of the shading net is f(x f , z f ), real-time location (x f (t), z f (t)), and obtain the actual opening degree K of the sunshade net s (t):

6. The method for calculating the covering state of the greenhouse top cover based on drone image recognition according to claim 5 is characterized in that: When sunlight passes through the shading net, the diffuse sky radiation is greatly absorbed and reflected, forming a diffuse sky transmission. A part of the direct radiation is also absorbed and reflected to form a diffuse beam, and another part of the direct radiation is reflected by the tiny mirror of the shading net. Only a small part of the direct radiation passes through the mesh of the shading net and the transparent material to reach the ground in the greenhouse. The real-time shadow area A md1 To a large extent, it represents the actual operating effect of the shadow net, which can be calculated using the following equation: A md1 (t)=L md1 (t)×S md (t)=(L m1 -H m ×tanα(t))×(S m -H m ×tanσ(t)) The real-time shadow area A can be obtained by the above formula d (t):

Images