Color correction method, system and device based on cooperative flight of double unmanned aerial vehicles, medium and product

Abstract

The invention discloses a color correction method, system and device based on cooperative flight of double unmanned aerial vehicles, a medium and a product, and relates to the technical field of agricultural detection. According to the method, a double-unmanned-aerial-vehicle cooperative flight mode is adopted, one unmanned aerial vehicle (namely the first unmanned aerial vehicle) flies at a high height for aerial photographing, the other unmanned aerial vehicle (namely the second unmanned aerial vehicle) carries a color card to synchronously fly at a low height, multiple farmland images containing the color card are obtained, a color correction matrix is constructed based on the RGB value of the color card in the farmland images and the actual RGB value of the color card, and the color correction matrix is obtained. Color correction under a dynamic illumination condition is realized, the problem of color distortion in a low-altitude remote sensing imaging process of the unmanned aerial vehicle is overcome, the reusability of cross-sensor data is improved, accurate acquisition of crop colors by the unmanned aerial vehicle in a field environment is realized, and reliable data support is provided for accurate crop management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural detection, in particular to a color correction method, system, device, medium and product based on cooperative flight of double unmanned aerial vehicles. BACKGROUND

[0002] In modern agricultural production and breeding research, the accurate acquisition of plant organ (such as leaf, flower, fruit, etc.) color is increasingly important, which not only can represent the state of plant suffering from diseases, water deficit and nutrient stress, but also is often used as a key indicator for variety selection to screen new varieties meeting market demand. In traditional agricultural production, farmers and breeders rely on visual observation to evaluate plant color. However, this subjective determination method is not only inefficient in large-area farmland or large quantities of sample scenarios, but also is easily affected by the experience level and visual physiological differences of observers, resulting in significant individual bias in color perception.

[0003] In recent years, unmanned aerial vehicle low-altitude remote sensing technology with high-resolution RGB camera has been widely applied in precision breeding and crop monitoring due to its high spatial resolution, flexible operation mode and wide coverage. Compared with ground measurement methods with limited coverage and satellite remote sensing with insufficient spatio-temporal resolution, the unmanned aerial vehicle platform can quickly acquire large-scale farmland images, providing an efficient technical approach for crop color parameter analysis. In addition, compared with traditional manual sampling or high-cost manned aerial measurement, unmanned aerial vehicle remote sensing significantly reduces operation cost and is more suitable for normal agricultural applications. However, the unmanned aerial vehicle images acquired in the field environment are easily affected by the coupling of multiple factors such as sensor characteristics, flight height, lighting conditions, imaging angle and camera parameters, resulting in color distortion within a single flight and between different flight times. Related researchers have revealed the problem of significant color deviation caused by environmental factors through multiple aerial photography and orthographic reconstruction experiments in multiple fields and multiple times. Without effective color correction means, the crop color data extracted by unmanned aerial vehicles will have consistency and accuracy defects, which not only affects the reuse, sharing and quantitative analysis of data, but also seriously restricts its reliability in crop management decision-making. SUMMARY

[0004] The purpose of the present application is to provide a color correction method, system, device, medium and product based on cooperative flight of double unmanned aerial vehicles to overcome the problem of color distortion in unmanned aerial vehicle low-altitude remote sensing imaging process and realize accurate acquisition of unmanned aerial vehicle crop color in field environment.

[0005] To achieve the above purpose, the present application provides the following solutions.

[0006] In a first aspect, the present application provides a color correction method based on cooperative flight of double unmanned aerial vehicles, comprising:

[0007] The first unmanned aerial vehicle and the second unmanned aerial vehicle are controlled to fly synchronously to collect a plurality of farmland images containing color cards; the first unmanned aerial vehicle is provided with an RGB sensor, and the second unmanned aerial vehicle is provided with a color card; during flight, the first unmanned aerial vehicle is located above the second unmanned aerial vehicle;

[0008] A color card region in each farmland image is identified;

[0009] A color correction matrix of each farmland image is constructed according to RGB values of each color block in the color card region in each farmland image and standard RGB values of each color block in a standard reference color card;

[0010] Each farmland image is color corrected based on the color correction matrix of the farmland image to obtain a plurality of corrected farmland images.

[0011] In a second aspect, the present application provides a color correction system based on cooperative flight of double unmanned aerial vehicles, comprising: a first unmanned aerial vehicle, a second unmanned aerial vehicle and a general control module;

[0012] The first unmanned aerial vehicle is provided with an RGB sensor, and the second unmanned aerial vehicle is provided with a color card;

[0013] The general control module is connected with control ends of the first unmanned aerial vehicle and the second unmanned aerial vehicle respectively, and the general control module is also connected with the RGB sensor; the general control module is used to execute the color correction method based on cooperative flight of double unmanned aerial vehicles according to any one of claims 1-6.

[0014] In a third aspect, the present application provides a computer device, comprising: a memory, a processor and a computer program stored in the memory and executable on the processor; the processor executes the computer program to realize the color correction method based on cooperative flight of double unmanned aerial vehicles.

[0015] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program; when the computer program is executed by a processor, the color correction method based on cooperative flight of double unmanned aerial vehicles is realized.

[0016] In a fifth aspect, the present application provides a computer program product, comprising a computer program; when the computer program is executed by a processor, the color correction method based on cooperative flight of double unmanned aerial vehicles is realized.

[0017] According to the specific embodiments provided by the present application, the present application has the following technical effects.

[0018] This application provides a color correction method, system, device, medium, and product based on dual-UAV cooperative flight. The application first controls a first and a second UAV to fly synchronously, acquiring multiple farmland images containing color charts. Then, it identifies the color chart regions in each farmland image. Based on the RGB values ​​of each color block in the color chart region of each farmland image and the standard RGB values ​​of each color block in the standard reference color chart, a color correction matrix is ​​constructed for each farmland image. Color correction is then performed on each farmland image based on its color correction matrix, resulting in multiple corrected farmland images. This application employs a dual-UAV cooperative flight method, where one UAV (the first UAV) flies at a higher altitude for aerial photography, while the other UAV (the second UAV) carries a color chart and flies synchronously at a lower altitude, obtaining multiple farmland images containing color charts. Based on the RGB values ​​of the color charts in the farmland images and the actual RGB values ​​of the color charts, a color correction matrix is ​​constructed to achieve color correction under dynamic lighting conditions. This overcomes the color distortion problem during low-altitude remote sensing imaging by UAVs, improves the reusability of cross-sensor data, and enables accurate acquisition of crop colors by UAVs in field environments, providing reliable data support for precision crop management. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a color correction method based on cooperative flight of two unmanned aerial vehicles (UAVs) according to an embodiment of this application.

[0021] Figure 2 This is a schematic diagram of two unmanned aerial vehicles (UAVs) cooperating in flight, provided as an embodiment of this application.

[0022] Figure 3 This is a schematic diagram of waypoint layout provided in one embodiment of this application.

[0023] Figure 4 This is a schematic diagram of the farmland image processing flow provided in an embodiment of this application.

[0024] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application.

[0025] Explanation of reference numerals in the attached diagram: 1. First UAV; 2. Second UAV; 3. RGB sensor; 4. Double-sided foam adhesive; 5. Fixing color chart. Detailed Implementation

[0026] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0027] The above-mentioned purposes, features and advantages of the present application will be more apparent and understandable. The present application will be further described in detail below with reference to the drawings and specific embodiments.

[0028] For the color deviation of the image obtained by the unmanned aerial vehicle remote sensing caused by the change of light or camera parameter, the related research has tried to use the image color feature migration method. Such method usually selects the reference image as the color reference by manual or automatic selection, extracts the color feature by means of machine learning or deep learning algorithm and determines the conversion rule, and then maps the color style of other images to the reference image. Although such method can improve the color consistency of the image in a single flight, its effect is highly dependent on the representativeness of the reference image, and the effect is not good when the environmental light changes greatly. In addition, the existing research lacks quantitative evaluation of the accuracy of the image color, and it is difficult to meet the demand of color correction in agricultural application. In order to ensure the accuracy of the image color, using standard color card for color correction is a common method in photography, printing and dental medicine, which realizes accurate correction by comparing the color difference between the original color card and the standard color card. In agricultural research, the color card is usually fixed beside the crops for shooting, and then the color correction is realized by post-processing algorithm. However, this method faces practical challenges in unmanned aerial vehicle remote sensing: the single color card placed at a fixed position cannot cover all the flight areas, and cannot effectively correct the color error caused by the change of dynamic light and camera parameter in the flight process.

[0029] The present application proposes a color correction method, system, device, medium and product based on the cooperative flight of double unmanned aerial vehicles, aiming at solving the problem of accurate acquisition of crop color in field environment. The method realizes color correction under dynamic light condition and improves the reusability of cross-sensor data by innovative design of double-machine cooperative flight, which provides reliable data support for accurate crop management.

[0030] In one exemplary embodiment, a color correction method based on the cooperative flight of double unmanned aerial vehicles is provided, as shown in Figure 1 The method includes the following steps 101-104.

[0031] Step 101, controlling the first unmanned aerial vehicle and the second unmanned aerial vehicle to fly synchronously to collect multiple farmland images containing color cards; the first unmanned aerial vehicle is equipped with an RGB sensor, and the second unmanned aerial vehicle is equipped with a color card; during flight, the first unmanned aerial vehicle is located above the second unmanned aerial vehicle.

[0032] Step 102, identifying the color card region in each farmland image.

[0033] Step 103, constructing a color correction matrix for each farmland image according to the RGB values of each color block in the color card region in each farmland image and the standard RGB values of each color block in the standard reference color card.

[0034] Step 104, performing color correction on each farmland image based on the color correction matrix of each farmland image to obtain multiple corrected farmland images.

[0035] Implementing the above steps 101-104 can achieve color correction under dynamic lighting conditions and overcome the problem of color distortion in the process of unmanned aerial vehicle low-altitude remote sensing imaging.

[0036] In another exemplary embodiment, a dual-unmanned aerial vehicle cooperative flight strategy is designed, as shown in Figure 2 The flight heights of the first unmanned aerial vehicle 1 and the second unmanned aerial vehicle 2 are set to h1 (15 m) and h2 (6.5 m), respectively. The first unmanned aerial vehicle 1 is equipped with an RGB sensor 3 to perform farmland orthographic image acquisition tasks, and the second unmanned aerial vehicle 2 has a color card 5 fixed on the top by double-sided foam tape 4. The color card 5 is a standard color card (ColorChecker Classic, X-Rite, Michigan, USA), and the installation stability of the color card 5 is verified by pre-experiments to ensure the horizontal posture of the color card during the entire flight. Through the dual-unmanned aerial vehicle cooperative flight strategy, the color card 5 carried by the second unmanned aerial vehicle 2 is always located in the center region of the field of view of the RGB sensor 3 of the first unmanned aerial vehicle 1.

[0037] In another exemplary embodiment, the dual-unmanned aerial vehicle cooperative flight task is implemented by calling the DJI cloud API (v1.4.0, Shenzhen, China) through an R language script (v4.1.2) to generate a flight path KML file, which ensures that the KML files of the first unmanned aerial vehicle 1 and the second unmanned aerial vehicle 2 only differ in flight height, and the flight waypoints and flight speed are consistent. To solve the time difference problem of the dual-unmanned aerial vehicles from the takeoff point to the first flight waypoint, a synchronization calibration point A0 is added 1 m in front of the first waypoint A1, Figure 3 A1-A7 in are the seven waypoints set by the unmanned aerial vehicle when performing an exemplary aerial photography task, wherein A1 is the first waypoint. When any unmanned aerial vehicle arrives at the synchronization calibration point first, it automatically hovers and waits until the dual-unmanned aerial vehicles are synchronized, and then the subsequent task can be continued. The waypoint diagram is shown in Figure 3 .

[0038] In another exemplary embodiment, a single flight mission of a UAV can obtain hundreds of images of farmland containing color cards. In order to achieve consistency correction of image colors, it is necessary to accurately identify the location of the color card in each image. Artificial processing is time-consuming and laborious. Therefore, the present scheme constructs an automatic process for color card recognition and color correction. First, an image containing a complete and unobstructed color card region is selected from the images taken by the first UAV as a target image, and the color card region in the target image is cropped out as a color card template image. The remaining images are used as images to be processed.

[0039] The implementation principle of the above SIFT algorithm is as follows: first, a multi-scale Gaussian pyramid is constructed for the image, and a Gaussian difference (DoG) image is generated by difference between adjacent scale images to extract local extreme points in the scale space; then, each candidate key point is precisely positioned, low-contrast points and edge response points are removed, and a 128-dimensional rotation-invariant feature descriptor is generated according to the gradient direction of the neighborhood.

[0040] In another exemplary embodiment, the above step 102 can be replaced by steps 201-206 as follows.

[0041] Step 201: Select an image containing a complete and unobstructed color card region from a plurality of farmland images as a target image, and select the farmland images other than the target image as images to be processed.

[0042] Step 202, cropping a color card region from the target image as a color card template image.

[0043] Step 203, using a SIFT algorithm to respectively extract feature points of the color card template image and the image to be processed, to obtain feature information of the color card template image and feature information of the image to be processed; the feature information includes feature points and descriptors of the feature points.

[0044] Step 204, using a brute force matcher to match the feature information of the color card template image and the feature information of the image to be processed, to obtain a matching pair.

[0045] Step 205, determining a homographic transformation matrix between the color card template image and the image to be processed based on the matching pair.

[0046] Step 206, projecting the color card template image to the image to be processed based on the homographic transformation matrix, to obtain a color card region in the image to be processed.

[0047] Wherein, the above-mentioned step of using a brute force matcher to match the feature information of the color card template image and the feature information of the image to be processed to obtain a matching pair specifically includes:

[0048] Respectively calculating the Euclidean distance between the descriptor of feature point i of the color card template image and the descriptor of each feature point of the image to be processed, to obtain the feature point in the image to be processed that has the closest Euclidean distance to the descriptor of feature point i, and to form a matching pair with feature point i; i = 1, 2, …, I, I represents the number of feature points of the color card template image;

[0049] Using a ratio test method to filter the I matching pairs, to obtain a plurality of filtered matching pairs.

[0050] In another exemplary embodiment, a pixel region is extracted near the center position of each color block (about 80% of its area), and the average RGB value is calculated to reduce edge interference. The RGB value of the extracted image color block is fitted with the true RGB value of the standard reference color card to construct a color correction matrix (CCM), and the farmland image is color corrected accordingly.

[0051] In another exemplary embodiment, the above-mentioned step 103 can be replaced by steps 301-305 as follows.

[0052] Step 301, color card division is performed on the color card region in the farmland image, and the average value of the RGB values of the pixel points in the center region of each color block is calculated as the RGB value of each color block.

[0053] Step 302, based on the RGB value of each color block, determine the color blocks of black and white as two matching color blocks.

[0054] Step 303, compare the RGB values of the two matching color blocks with the standard RGB values of each color block in the standard reference color card, and determine the correspondence between each color block in the color card region of the farmland image and each color block in the standard reference color card.

[0055] Step 304, according to the correspondence between each color block in the color card region of the farmland image and each color block in the standard reference color card, determine the standard RGB value corresponding to the RGB value of each color block in the color card region of the farmland image.

[0056] Step 305, taking the minimization of the difference between the RGB value of each color block in the color card region of the farmland image and its corresponding standard RGB value as the target, construct the color correction matrix of the farmland image.

[0057] In the embodiments of the present application, the linear mapping method based on the color correction matrix is used for image correction. The standard color card contains 24 precisely designed color blocks, and the RGB values of the color blocks under the CIE (International Commission on Illumination) D65 standard light source present a strictly regulated three-dimensional color space distribution. The color is used as the standard RGB value of the color block. The CCM method establishes the color mapping relationship between the two matrices by minimizing the difference between the RGB value of the color block obtained by the RGB sensor (i.e. the image matrix A) and the standard RGB value (i.e. the standard matrix C). This method corrects the color difference of other regions of the image through interpolation operation while correcting the standard color blocks in the correction color card.

[0058] The mathematical principle is described as follows: the actual RGB value of the color block is constructed as a 24x3 standard matrix C, and the RGB value of the color block obtained by the RGB sensor is constructed as a 24x3 image matrix A. In order to compensate for the non-linear color deviation of the imaging device (i.e. the RGB sensor), the image matrix A is expanded to a 24x4 matrix by introducing a bias term. By solving equations (1)-(2), the color correction matrix M is obtained:

[0059] C=A x M (1)

[0060] M=(A T x A) -1 x A T x C (2)

[0061] The original farmland image I (w x h x 3) is expanded to a (w x h) x 3 two-dimensional matrix I ' O , and after adding a unit bias term, a (w x h) x 4 matrix I 'O ' The corrected matrix dimension of I is (w x h) x 4, and the corrected farmland image I is reshaped to w x h x 3 dimensions. ' C C

[0062] I′ C =I″ O ×M (3)

[0063] In another exemplary embodiment, the obtained aerial images of the unmanned aerial vehicle need to be spliced into an orthographic image large image for subsequent image processing and analysis. If the original images containing color cards are directly spliced, the image splicing algorithm will identify the color cards as image features, resulting in many color cards in the spliced orthographic image. Therefore, as shown in FIG. 3, a mask needs to be generated according to the identified color card area after color correction. Considering that the size of the unmanned aerial vehicle (including the arm and the propeller) is larger than that of the color card body, the length and width of the mask are respectively extended by 30% and 50%, so as to completely shield the unmanned aerial vehicle in the image. The splicing of the image after mask processing can obtain a large field orthographic image without color card interference. Figure 4

[0064] In another exemplary embodiment, after the step 104, the method further comprises the following steps 105-106.

[0065] Step 105: Masking the color card area in each corrected farmland image with a size of 1.3L x 1.5D, respectively, to obtain a plurality of farmland images after mask processing; wherein L is the length of the color card area in the corrected farmland image, and D is the width of the color card area in the corrected farmland image.

[0066] Step 106: Splicing the plurality of farmland images after mask processing to obtain a large field orthographic image.

[0067] According to the specific embodiments provided by the present application, the present application has the following technical effects.

[0068] Compared with the image style transfer method used in the current unmanned aerial vehicle remote sensing, the embodiments of the present application use a standard color card as a reference for color correction, providing a standard value for the color of the farmland image, and improving the color consistency and accuracy between the farmland images. The style transfer in the existing remote sensing can only improve the color consistency between the images, but cannot ensure the accuracy of the color.

[0069] ​​​Compared with the traditional color card correction method, the application embodiment innovatively proposes a scheme of using two unmanned aerial vehicles to fly cooperatively and synchronously to ensure that each image obtained in the unmanned aerial vehicle remote sensing contains a standard color card, and improve the accuracy of color correction.

[0070] Based on the same inventive concept, the application embodiment also provides a color correction system based on cooperative flight of dual unmanned aerial vehicles for implementing the color correction method based on cooperative flight of dual unmanned aerial vehicles. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, and therefore the specific limitations in one or more color correction system embodiments based on cooperative flight of dual unmanned aerial vehicles provided below can refer to the limitations of the color correction method based on cooperative flight of dual unmanned aerial vehicles described above, and will not be described here again.

[0071] In an exemplary embodiment, a color correction system based on cooperative flight of dual unmanned aerial vehicles is provided, including a first unmanned aerial vehicle, a second unmanned aerial vehicle, and a total control module.

[0072] The first unmanned aerial vehicle is provided with an RGB sensor, and the second unmanned aerial vehicle is provided with a color card.

[0073] The total control module is connected to the control end of the first unmanned aerial vehicle and the second unmanned aerial vehicle respectively, and is also connected to the RGB sensor. The total control module is used to execute the color correction method based on cooperative flight of dual unmanned aerial vehicles described above.

[0074] In an exemplary embodiment, a computer device can be a server or a terminal, and its internal structure diagram can be as shown in Figure 5 The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement a color correction method based on cooperative flight of dual unmanned aerial vehicles.

[0075] Those skilled in the art can understand, Figure 5The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps in the above method embodiments.

[0076] In an exemplary embodiment, a computer readable storage medium is provided, storing a computer program, which is executed by a processor to implement the steps in the above method embodiments.

[0077] In an exemplary embodiment, a computer program product is provided, including a computer program, which is executed by a processor to implement the steps in the above method embodiments.

[0078] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties.

[0079] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0080] The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a blockchain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0081] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described, but as long as the combinations of the technical features do not exist, they should be considered as the scope of the present application.

[0082] The principles and implementation modes of the present application are described by applying specific examples herein. The above description of the embodiments is only used to help understand the method and its core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range can be changed. In conclusion, the content of the present application should not be understood as a limitation.

Claims

1. A color correction method based on cooperative flight of two unmanned aerial vehicles, characterized in that, The method comprises the following steps: controlling the first unmanned aerial vehicle and the second unmanned aerial vehicle to fly synchronously to collect a plurality of farmland images containing color cards; the first unmanned aerial vehicle is provided with an RGB sensor, and the second unmanned aerial vehicle is provided with a color card; during flight, the first unmanned aerial vehicle is located above the second unmanned aerial vehicle; identifying a color card region in each farmland image; constructing a color correction matrix of each farmland image according to the RGB values of each color block in the color card region of each farmland image and the standard RGB values of each color block in a standard reference color card; performing color correction on each farmland image based on the color correction matrix of each farmland image to obtain a plurality of corrected farmland images.

2. The color correction method based on cooperative flight of dual unmanned aerial vehicles according to claim 1, characterized in that, The identification of the color card region in each farmland image comprises the following steps: selecting one image containing a complete and unobstructed color card region from the plurality of farmland images as a target image, and taking the farmland images other than the target image as to-be-processed images; cropping the color card region from the target image as a color card template image; performing feature point extraction on the color card template image and the to-be-processed images respectively by using a SIFT algorithm to obtain feature information of the color card template image and feature information of the to-be-processed images; the feature information comprises feature points and descriptors of the feature points; matching the feature information of the color card template image and the feature information of the to-be-processed images by using a brute force matcher to obtain a matching pair; determining a homographic transformation matrix between the color card template image and the to-be-processed images based on the matching pair; projecting the color card template image to the to-be-processed images based on the homographic transformation matrix to obtain the color card region in the to-be-processed images. 3.The color correction method based on the cooperative flight of dual UAVs according to claim 1, characterized in that, The matching of the feature information of the color card template image and the feature information of the to-be-processed images by using the brute force matcher to obtain a matching pair comprises the following steps: calculating the Euclidean distance between the descriptor of feature point i of the color card template image and the descriptors of each feature point of the to-be-processed images to obtain the feature point in the to-be-processed images closest to the descriptor of feature point i, and forming a matching pair with feature point i; i = 1, 2, …, I, I represents the number of feature points of the color card template image; filtering the I matching pairs by using a ratio test method to obtain a plurality of filtered matching pairs.

4. The color correction method based on cooperative flight of dual UAVs according to claim 1, characterized in that, The construction of the color correction matrix of each farmland image according to the RGB values of each color block in the color card region of each farmland image and the standard RGB values of each color block in a standard reference color card comprises the following steps: dividing the color card region in the farmland image into color blocks, and calculating the average value of the RGB values of the pixel points in the central region of each color block as the RGB value of each color block; determining the color blocks of black and white as two matching color blocks based on the RGB values of each color block; comparing the RGB values of the two matching color blocks with the standard RGB values of each color block in a standard reference color card to determine the correspondence between each color block in the color card region of the farmland image and each color block in the standard reference color card. According to the correspondence between each color block of the color card region in the farmland image and each color block in the standard reference color card, the standard RGB value corresponding to the RGB value of each color block of the color card region in the farmland image is determined; A color correction matrix of the farmland image is constructed by minimizing the difference between the RGB value of each color block of the color card region in the farmland image and the corresponding standard RGB value.

5. The color correction method based on cooperative flight of dual UAVs according to claim 1, characterized in that, Each farmland image is color corrected based on the color correction matrix of each farmland image, and a plurality of corrected farmland images are obtained, and then the method further comprises: The color card region in each corrected farmland image is masked with a size of 1.3Lx1.5D, and a plurality of mask-processed farmland images are obtained; wherein L is the length of the color card region in the corrected farmland image, and D is the width of the color card region in the corrected farmland image; The plurality of mask-processed farmland images are spliced to obtain a large field orthographic image.

6. The color correction method based on cooperative flight of dual UAVs according to claim 1, characterized in that, The first unmanned aerial vehicle and the second unmanned aerial vehicle are controlled to fly synchronously to collect a plurality of farmland images containing color cards, specifically comprising: A synchronous calibration point is added before the first waypoint; The first unmanned aerial vehicle and the second unmanned aerial vehicle are made to fly synchronously with the same flight speed by using the hovering waiting mode, taking the synchronous calibration point as the starting point, and using the RGB sensor carried on the first unmanned aerial vehicle to collect a plurality of farmland images containing color cards.

7. A color correction system based on cooperative flight of two unmanned aerial vehicles, characterized in that, It comprises: A first unmanned aerial vehicle, a second unmanned aerial vehicle and a total control module; The first unmanned aerial vehicle is provided with an RGB sensor, and the second unmanned aerial vehicle is provided with a color card; The total control module is connected with the control end of the first unmanned aerial vehicle and the second unmanned aerial vehicle respectively, and is also connected with the RGB sensor, and is used to execute the color correction method based on the cooperative flight of the double unmanned aerial vehicles according to any one of claims 1-6.

8. A computer device comprising: A memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the color correction method based on the cooperative flight of the double unmanned aerial vehicles according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the color correction method based on the cooperative flight of the double unmanned aerial vehicles according to any one of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the color correction method based on the cooperative flight of the double unmanned aerial vehicles according to any one of claims 1-6.

Images