Navigation line generation method and device, storage medium and program product

Abstract

The invention provides a navigation line generation method and device, a storage medium and a program product, and relates to the technical field of agricultural production. According to the method, the target area is segmented from the crop image, the target area refers to the crop area and / or the land area, then the reference line of the target area is extracted, the weight coefficient of the reference line is acquired, and the navigation line of the agricultural equipment can be generated according to the weight coefficient and the reference line, so that the navigation line of the agricultural equipment can be generated in the operation process of the agricultural equipment. The navigation line is dynamically generated by integrating the reference line of the target area, so that the accuracy and the real-time performance of agricultural machinery navigation are improved, the artificial interference is reduced, the adaptive capacity of the system to a complex operation environment is enhanced, the operation cost can be effectively saved, and the operation efficiency can be effectively improved.

Description

Technical Field

[0001] This application relates to the technical field of agricultural production, and more particularly, to a method, device, storage medium, and program product for generating navigation lines. Background Art

[0002] In modern agricultural production, the application of agricultural machinery navigation technology is of great significance for improving operation efficiency, reducing labor costs, and increasing crop yields. Existing agricultural machinery navigation systems usually adopt a combination of the Global Navigation Satellite System and the Inertial Navigation System to achieve high-precision positioning. However, such systems need to determine the navigation lines of the operation area based on the current position of the agricultural machinery before operation, and the agricultural machinery only travels along these preset navigation lines during operation, requiring frequent manual intervention for correction, resulting in low operation efficiency. Summary of the Invention

[0003] The purpose of the embodiments of this application is to provide a method, device, storage medium, and program product for generating navigation lines to improve the problem of frequent manual intervention for correction and low operation efficiency after generating navigation lines in the existing method.

[0004] In a first aspect, the embodiments of this application provide a method for generating a navigation line, the method comprising:

[0005] Obtain a crop image, and segment a target area from the crop image, where the target area refers to a crop area and / or a land area;

[0006] Extract a reference line of the target area, where the reference line characterizes the distribution form of the target area;

[0007] Obtain a weight coefficient of the reference line, where the weight coefficient is used to characterize the influence degree of the reference line on the generation of the navigation line;

[0008] Generate a current first navigation line of the agricultural machinery according to the weight coefficient of the reference line and the reference line.

[0009] In the above implementation process, by segmenting the target area from the crop image, where the target area refers to a crop area and / or a land area, then extracting the reference line of the target area, and obtaining the weight coefficient of the reference line, the navigation line of the agricultural machinery can be generated according to the weight coefficient and the reference line. In this way, during the operation process of the agricultural machinery, the reference line of the target area can be comprehensively used to realize the dynamic generation of the navigation line, improving the accuracy and real-time performance of agricultural machinery navigation, reducing manual interference, enhancing the adaptability of the system to complex operation environments, and effectively saving operation costs and improving operation efficiency.

[0010] Optionally, the obtaining the weight coefficient of the reference line includes:

[0011] Determine the weight coefficient of the reference line according to the information of the reference line, where the information of the reference line includes at least one of confidence, fitting residual, distance from the image center of the crop image, and regional size of the target area. The confidence refers to the credibility of the segmentation result of the target area, and the fitting residual refers to the residual determined when fitting the reference line.

[0012] In the above implementation process, by comprehensively evaluating the weight coefficient of the reference line from multi-dimensional information, the accuracy and reliability of the navigation line generation are improved.

[0013] Optionally, if there are multiple reference lines, before obtaining the weight coefficient of the reference line, it further includes:

[0014] Screen and retain the reference lines that meet the set conditions, where the set conditions include at least one of the following: the confidence is greater than the set confidence, the fitting residual is less than the set residual, the distance is greater than the set distance, and the regional size is less than the set size.

[0015] In the above implementation process, by screening the reference lines, unreliable data can be eliminated, error interference can be reduced, and the accuracy and reliability of the navigation line generation are improved.

[0016] Optionally, the extraction of the reference line of the target area includes:

[0017] Calculate multiple central pixel points of the target area according to the boundary pixel points of the target area;

[0018] Fit the multiple central pixel points to obtain the reference line of the target area.

[0019] In the above implementation process, by calculating the central pixel points from the boundary pixel points and fitting these points to obtain the reference line, the distribution form of the target area can be more accurately reflected, and the accuracy and reliability of the navigation line generation are improved. At the same time, the fitting of the central pixel points reduces the amount of data processing and improves the processing speed and efficiency.

[0020] Optionally, the generation of the current first navigation line of the agricultural machinery equipment according to the weight coefficient of the reference line and the reference line includes:

[0021] Align the reference lines to the same coordinate system, where the origin coordinates of the coordinate system are the optical center point of the image collector for collecting the crop image;

[0022] Use the weight coefficient of the reference line to perform weighted calculation based on the aligned reference lines to generate the current first navigation line of the agricultural machinery equipment.

[0023] In the above implementation process, by aligning the reference line to a unified coordinate system with the optical center of the image collector as the origin and performing weighted calculation based on the weight coefficient to generate the navigation line, the accuracy and stability of the navigation line are effectively improved. This process ensures the fusion of different reference lines under the same benchmark, avoids deviations caused by coordinate differences, makes the generated navigation line closer to the actual crop distribution, and enhances the accuracy and reliability of agricultural machinery operations.

[0024] Optionally, the method further includes:

[0025] Obtain the current positioning result of the agricultural machinery equipment;

[0026] Determine the current final navigation line of the agricultural machinery equipment according to the current positioning result and the current first navigation line.

[0027] In the above implementation process, the current positioning result provides the real-time position information of the agricultural machinery. After being fused with the current first navigation line, it can effectively correct the error of the current first navigation line and ensure the real-time performance and accuracy of the navigation line.

[0028] Optionally, the step of determining the current final navigation line of the agricultural machinery equipment according to the current positioning result and the current first navigation line includes:

[0029] Determine the current second navigation line according to the current positioning result;

[0030] Determine the current final navigation line of the agricultural machinery equipment according to the current first navigation line and the current second navigation line.

[0031] In the above implementation process, by generating the first navigation line based on vision and the second navigation line based on the positioning system respectively, and then determining the final navigation line by integrating the two, the deep fusion and complementarity of multi-source data are realized. The vision navigation line provides the actual distribution information of the crops, while the positioning system navigation line provides a high-precision position reference. The combination of the two not only improves the accuracy and stability of navigation, but also enhances the adaptability of the system to complex farmland environments.

[0032] Optionally, the step of determining the current final navigation line according to the current positioning result and the current first navigation line includes:

[0033] If the current positioning result is unreliable or the deviation between the current positioning result and the final navigation line determined at the previous moment is greater than the set deviation, then determine the current final navigation line as the current first navigation line;

[0034] Wherein, if the current positioning result is reliable and the deviation between the current positioning result and the final navigation line determined at the previous moment is less than or equal to the set deviation, and the confidence level of the current first navigation line is less than the set confidence level, then the current final navigation line is the current second navigation line determined according to the current positioning result.

[0035] In the above implementation process, when the positioning result is unreliable or the deviation from the navigation line at the previous moment is too large, the current first navigation line generated visually is directly adopted, avoiding operation deviation caused by positioning errors; and when the positioning result is reliable and the deviation is within the allowable range, and at the same time the confidence level of the visual navigation line is low, it switches to the current second navigation line based on the positioning result, giving full play to the high-precision advantage of the positioning system. This intelligent switching mechanism effectively copes with various situations in the complex farmland environment, reduces the limitations of a single navigation method, and improves the stability and efficiency of agricultural machinery operations.

[0036] Optionally, if there are multiple reference lines, the information of the reference lines includes confidence levels, and the confidence level of the current first navigation line is obtained by weighted calculation of the confidence levels of the reference lines according to the weight coefficients of the reference lines.

[0037] In the above implementation process, the confidence level of the current first navigation line is determined by weighted calculation of the confidence levels of multiple reference lines, effectively improving the accuracy of the reliability evaluation of the navigation line. This weighting method comprehensively considers the confidence levels and their weights of each reference line, making the confidence level of the navigation line not only reflect the reliability of a single reference line.

[0038] Optionally, the step of segmenting the target area from the crop image includes:

[0039] Determining a segmentation target according to the crop type in the crop image;

[0040] Segmenting the target area where the segmentation target is located from the crop image.

[0041] In the above implementation process, by determining the segmentation target according to the crop type in the crop image and segmenting the target area from the image accordingly, the pertinence and accuracy of image segmentation are significantly improved. This method based on crop type can more accurately identify and extract the areas related to operations, providing a high-quality data basis for subsequent navigation line generation.

[0042] In a second aspect, an embodiment of the present application provides an agricultural machinery device, where the agricultural machinery device includes:

[0043] An image collector installed on the agricultural machinery device for collecting crop images;

[0044] A vision terminal for executing the above navigation line generation method;

[0045] A control system for controlling the driving of agricultural machinery equipment according to the current first navigation line.

[0046] In a third aspect, an embodiment of the present application provides an electronic device, including a processor and a memory. The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the method provided in the first aspect above are run.

[0047] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the method provided in the first aspect above are run.

[0048] In a fifth aspect, an embodiment of the present application provides a computer program product, including computer program instructions. When the computer program instructions are read and run by a processor, the steps in the method provided in the first aspect above are executed.

[0049] Other features and advantages of the present application will be described in the subsequent specification, and, in part, will be obvious from the specification, or will be understood by implementing the embodiments of the present application. The objectives and other advantages of the present application can be achieved and obtained by the structures specifically pointed out in the written specification, claims, and drawings. Description of the Drawings

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0051] Figure 1 A schematic structural diagram of an agricultural machinery equipment provided by an embodiment of the present application;

[0052] Figure 2 A flowchart of a navigation line generation method provided by an embodiment of the present application;

[0053] Figure 3 A schematic diagram of a target area segmented from a crop image provided by an embodiment of the present application;

[0054] Figure 4 A structural block diagram of a navigation line generation device provided by an embodiment of the present application;

[0055] Figure 5 A schematic structural diagram of an electronic device for executing a navigation line generation method provided by an embodiment of the present application. Specific Embodiments

[0056] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application.

[0057] It should be noted that the terms "system" and "network" in the embodiments of the present invention can be used interchangeably. "Multiple" means two or more. In view of this, in the embodiments of the present invention, "multiple" can also be understood as "at least two". "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " generally represents an "or" relationship between the associated objects before and after, unless otherwise specified.

[0058] It should also be noted that all actions of obtaining signals, information, or data in this application are carried out on the premise of complying with the corresponding data protection regulations and policies of the country where it is located and obtaining the authorization given by the owner of the corresponding device.

[0059] The embodiments of the present application provide a navigation line generation method. The method segments a target area from a crop image. The target area refers to a crop area and / or a land area. Then, a reference line of the target area is extracted, and a weight coefficient of the reference line is obtained. According to the weight coefficient and the reference line, a navigation line for agricultural machinery equipment can be generated. In this way, during the operation of the agricultural machinery equipment, the reference line of the target area can be comprehensively used to realize the dynamic generation of the navigation line, improving the accuracy and real-time performance of agricultural machinery navigation, reducing manual interference, enhancing the adaptability of the system to complex working environments, and effectively saving operation costs and improving operation efficiency.

[0060] To facilitate the understanding of the following method embodiments, the agricultural machinery equipment involved in the embodiments of the present application will be introduced first, as Figure 1 As shown, the agricultural machinery equipment 10 includes an image collector 11 installed on the agricultural machinery equipment 10 for collecting crop images. The agricultural machinery equipment 10 further includes a vision terminal 12 connected to the image collector 11 for executing the navigation line generation method of this solution. The agricultural machinery equipment 10 further includes a control system 13 connected to the vision terminal 12 for controlling the agricultural machinery equipment 10 to travel according to the currently generated first navigation line by the vision terminal 12.

[0061] The agricultural machinery equipment 10 refers to the equipment in the agricultural machinery navigation scenario. The agricultural machinery equipment 10 refers to working equipment, such as tractors, harvesters, irrigation machines, etc. The image collector 11 can refer to a camera, such as a monocular camera, or can also refer to a multi-camera, etc., for collecting crop images. For example, during the operation of the agricultural machinery equipment 10, the crop images in front of the agricultural machinery equipment are collected in real time or at regular intervals through the image collector 11.

[0062] In some embodiments, the image collector 11 can be installed on the roof of the tractor or on the front end of the tractor through a customized bracket, with the field of view facing forward, facilitating the capture of crop images in front.

[0063] The vision terminal 12 can perform corresponding processing on the crop images to generate a navigation route for the agricultural machinery device 10. The vision terminal 12 can refer to a terminal with computing capabilities, and it can also include a display unit, through which the navigation route can be displayed, enabling the operator to intuitively view the current navigation route.

[0064] The control system 13 refers to the main control system of the agricultural machinery device 10, such as a processor, etc., which is used to control the driving of the agricultural machinery device 10 according to the current navigation line for operation.

[0065] In some embodiments, the image collector 11 and the vision terminal 12, and the vision terminal 12 and the control system 13 can be connected by a wire harness or wirelessly, without specific special restrictions.

[0066] In some embodiments, the vision terminal 12 and the control system 13 can be integrated on one hardware device in terms of hardware structure, which can be understood as different functional modules of one device. Of course, they can also be separate and independent devices.

[0067] Please refer to the following Figure 2 , Figure 2 which is a flowchart of a navigation line generation method provided by an embodiment of the present application. The method includes the following steps:

[0068] Step S110: Obtain a crop image and segment a target area from the crop image.

[0069] The application scenario of the present application can be an agricultural machinery navigation scenario. When the agricultural machinery device is operating, it needs to drive along the planned navigation line to adapt to the operation process. The method of the present application can also be adapted to other application scenarios, such as a robot patrol scenario, a warehousing and logistics scenario, etc. Only in different scenarios, the segmented target areas are different. For example, in a robot patrol scenario, the robot can patrol along the planned route, and its target area can refer to the area where the patrol target is located. For example, in a factory equipment patrol scenario, the target area can refer to the area where the equipment is located. Another example is in a warehousing and logistics scenario, the target area can refer to the area where the goods are located. For example, in a scenario where the goods are stacked row by row, in this scenario, the warehousing robot can perform goods handling tasks according to the generated navigation line.

[0070] In some embodiments, the method of this solution can be executed by a vision terminal on agricultural machinery equipment or by a server. For example, the agricultural machinery equipment can send the collected crop images to the server. After the server processes the crop images and generates the current first navigation line, it can send the current first navigation line to the agricultural machinery equipment, and the agricultural machinery equipment can then travel according to the current first navigation line.

[0071] For ease of description, an agricultural machinery navigation scenario is used as an example in this application for illustration.

[0072] The crop images can be obtained by an image collector on the agricultural machinery equipment. The execution subject of this method can be the above-mentioned vision terminal. The image collector can transmit the collected crop images to the vision terminal. After receiving the crop images, the vision terminal can first identify the target area in the crop images. The target area can refer to the crop area and / or the land area.

[0073] In some embodiments, the target area can be segmented from the crop images by means of threshold segmentation, edge detection, or a neural network model. The neural network model can be, for example, a convolutional neural network model, a long short-term memory network model, etc. For ease of deployment on the vision terminal, the neural network model can also be some lightweight models, such as the UNet model, the fast convolutional network model, etc.

[0074] In some embodiments, if the crops are relatively short, in this case, the crop images contain the area where the crops are located and the area where the land is located. The area where the crops are located can be called the crop ridges. The crop ridges refer to the planting method in which the crops are arranged in rows according to certain rules in the farmland, which is a common agricultural planting layout. The area between the crop ridges is the land area. In this case, the crop area and the land area can be segmented from the crop images. Of course, in order to reduce the calculation amount, only the crop area or the land area can also be segmented.

[0075] If the crops are relatively tall and dense, such as in a sorghum field or a corn field, etc., it may not be possible to identify the land in the captured crop images at this time because the land may be covered by the crops in the crop images. Therefore, for this scenario, only the crop area can be segmented from the crop images.

[0076] In some embodiments, for ease of rapid processing by the vision terminal, the segmentation target can be determined according to the crop type in the crop images, and then the target area where the segmentation target is located can be segmented from the crop images.

[0077] Among them, the crop type of the crops can be identified from the crop images by means of a neural network model or relevant image recognition algorithms. Or the crop type can also be manually input, so that the vision terminal can quickly obtain the crop type.

[0078] In the above example, if the crop types are some short crops, such as rice, potatoes, leeks, etc., for these crop types, the segmentation targets can be the crops and / or the land. At this time, the crop area and / or the land area can be segmented from the crop image.

[0079] If the crop types are some high-density crops, such as sorghum, corn, etc., and the whole image is full of sorghum or corn, it may not be possible to identify the land. For these crop types, the segmentation targets refer to some characteristics of the crops, such as sorghum spikes or corn spikes, etc. At this time, the crop area can be segmented from the crop image. Of course, the area between two crop areas can also be regarded as other areas here, that is, the crop area and other areas are segmented. The other areas can refer to the land areas blocked by the crops. When determining the reference line later, the reference line can be determined for the crop area and other areas respectively.

[0080] In some embodiments, in order to improve the accuracy of target area segmentation and the accuracy of subsequent navigation line generation, after obtaining the initial crop image collected by the image collector, the initial crop image can also be processed accordingly. For example, the images in a certain area on the left and right sides of the initial crop image can be cropped, and the remaining image can be used as the subsequent crop image for target area segmentation. Or the image in the area close to the center of the initial crop image can also be intercepted as the subsequent crop image.

[0081] Step S120: Extract the reference line of the target area.

[0082] After the target area is segmented, the reference line of the target area can be extracted. The reference line can represent the distribution form of the target area, such as the distribution form of the crop area or the land area along the length direction. These reference lines can be used to assist in the generation of the navigation line. Since the reference line is extracted from the target area, it can represent the actual position and direction of the crop or the land, and reflects the distribution of the crop or the land in the current field of view. Therefore, the reference line can provide an auxiliary reference for the generation of the navigation line, making the generated navigation line closer to the actual crop growth line or the land distribution line, thereby reducing the damage to the crops during the operation process.

[0083] In some embodiments, the reference line can be a straight line, or a curve or a broken line, etc. For the convenience of calculation, a straight line is taken as an example in this solution for illustration.

[0084] It can be understood that if there are multiple target areas segmented above, a reference line will be extracted for each target area, and multiple reference lines will be obtained.

[0085] Step S130: Obtain the weight coefficient of the reference line.

[0086] Among them, the weight coefficient is used to characterize the influence degree of the reference line on the generation of the navigation line, and can also be considered as used to characterize the importance of the reference line in the generation of the navigation line. For example, the larger the weight coefficient, the greater the influence degree; the smaller the weight coefficient, the smaller the influence degree.

[0087] In some embodiments, the weight coefficient may be related to the distance between the reference line and the image center of the crop image or the area size of the target area. For example, the larger the distance, the smaller the weight coefficient; the smaller the distance, the larger the weight coefficient. Or, the larger the area size, the larger the weight coefficient; the smaller the area size, the smaller the weight coefficient, etc.

[0088] In some embodiments, the weight coefficient can also be set manually or evenly distributed. If it is set manually, the extracted reference line can be output and displayed to the user, and the user can configure different weight coefficients for each reference line based on experience. If it is evenly distributed, the weight coefficient can be evenly distributed according to the number of reference lines.

[0089] Step S140: Generate the current first navigation line of the agricultural machinery device according to the weight coefficient of the reference line and the reference line.

[0090] After determining the weight coefficient of each reference line, the navigation line can be generated, which can be called the current first navigation line here. The current first navigation line can be obtained by performing weighted calculation on each reference line according to the weight coefficient. For example, multiple coordinates on each reference line (such as taking coordinate points at a certain interval) can be weighted and averaged to obtain a series of coordinate points, and then these coordinate points can be fitted to obtain the current first navigation line.

[0091] After generating the current first navigation line, the vision terminal can send the current first navigation line to the control system, and the control system can control the agricultural machinery device to travel according to the current first navigation line. It can be understood that the control system can convert each coordinate point of the current first navigation line in the image into a coordinate point in the geographic coordinate system, so as to facilitate the control of the travel of the agricultural machinery device.

[0092] In order to facilitate the real-time adjustment of the travel route of the agricultural machinery device, the image collector can collect images in real time or at regular intervals. For example, the image collector can send several continuously collected images to the vision terminal, and then the vision terminal generates the current first navigation line based on these images obtained at the current moment. The control system controls the agricultural machinery device to travel according to the current first navigation line. At the next moment, the image collector continues to collect new images and send them to the vision terminal, and then regenerates the navigation line. The control system can control the agricultural machinery device to travel according to the new navigation line. In this way, during the travel of the agricultural machinery device, the travel route is controlled by dynamically generating the navigation line to continuously adjust the route, so that the travel route of the agricultural machinery device can adapt to the distribution of crops and improve the operation efficiency.

[0093] In the above implementation process, by segmenting the target area from the crop image, where the target area refers to the crop area and / or the land area, then extracting the reference lines of the target area, and obtaining the weight coefficients of the reference lines, the navigation line of the agricultural machinery equipment can be generated based on the weight coefficients and the reference lines. In this way, during the operation of the agricultural machinery equipment, the reference lines of the target area can be comprehensively used to realize the dynamic generation of the navigation line, improving the accuracy and real-time performance of agricultural machinery navigation, reducing manual interference, enhancing the adaptability of the system to complex operating environments, effectively saving operating costs and improving operating efficiency.

[0094] Based on the above embodiments, in the above method of obtaining the weight coefficients of the reference lines, the weight coefficients of the reference lines can also be determined according to the information of the reference lines, where the information of the reference lines includes at least one of confidence, fitting residual, distance from the center of the crop image, and the area size of the target area. The confidence refers to the credibility of the segmentation result of the target area, and the fitting residual refers to the residual determined when fitting the reference line.

[0095] The methods for determining the weight coefficients based on various information are described below.

[0096] (1) Determine the weight coefficients of the reference lines according to the confidence of the reference lines.

[0097] The higher the confidence, the higher the credibility of the segmentation result of the target area, that is, the more accurate the segmentation of the target area. Then, the higher the confidence of the reference line, the greater the importance for the generation of the navigation line. Therefore, the confidence can be directly used as the weight coefficient of the reference line. For the convenience of subsequent calculation of the navigation line, the confidence of each reference line can be normalized to determine the weight coefficient. The normalization method is as follows:

[0098]

[0099] Among them, q i represents the confidence of the i-th reference line, and q i ' represents the normalized confidence of the i-th reference line, and n represents the total number of reference lines.

[0100] In this way, q i ' can be used as the weight coefficient of the i-th reference line.

[0101] (2) Determine the weight coefficients of the reference lines according to the fitting residual of the reference lines.

[0102] The fitting residual represents the degree of coincidence between the reference line and the actual regional reference line. The larger the fitting residual, the smaller the importance of the reference line for generating the navigation line. The smaller the fitting residual, the greater the importance of the reference line for generating the navigation line. At this time, the reciprocal of the fitting residual can be used as the weight coefficient. For the convenience of subsequent calculation of the navigation line, the fitting residuals of each reference line can be normalized to determine the weight coefficient. The normalization method of its fitting residual is similar to the normalization method of the above confidence level and will not be repeated here. Then, the reciprocal of the normalized fitting residual can be used as the weight coefficient, as is the weight coefficient of the i-th reference line, σ i represents the fitting residual of the i-th reference line after normalization.

[0103] (3) Determine the weight coefficient of the reference line according to the distance between the reference line and the image center of the crop image.

[0104] The image center here can refer to the image center point. When calculating the distance between each reference line and the image center point, the vertical distance from the image center point to each reference line can be calculated.

[0105] Alternatively, a horizontal line can be drawn in the crop image first, and this horizontal line passes through the image center point. Since each reference line is a reference line of the crop area or the land area, its target area is distributed in the vertical direction in the crop image, so the reference line extends in the vertical direction. The horizontal line drawn in this way can intersect with each reference line, and then the distance between the intersection coordinates of each reference line and the horizontal line and the image center point coordinates can be calculated, and this distance can be used as the distance between the reference line and the image center.

[0106] Alternatively, the image center can also refer to the image center line, and this image center line can refer to the image center line in the vertical direction, that is, a vertical line drawn along the image center point in the vertical direction. Then, the distance between this vertical line and each reference line can be calculated. Specifically, in the image coordinates, when the ordinate is the same (any ordinate can be selected), the abscissa of each reference line can be selected, and the length of its abscissa can be used as the distance between each reference line and the image center.

[0107] Theoretically speaking, the farther the reference line is, the smaller the influence degree on the generation of the navigation line. Therefore, a smaller weight coefficient can be configured for the farther reference line. In this case, the reciprocal of the distance can be used as the weight coefficient, and the specific implementation method is similar to the method of determining the weight coefficient by the fitting residual as described above. For example, the distances between each reference line and the image center can be normalized to determine the weight coefficient. The normalization method of its distance is similar to the normalization method of the above confidence level and will not be repeated here. Then, the reciprocal of the normalized distance can be used as the weight coefficient, as is the weight coefficient of the i-th reference line, di It represents the distance between the i-th reference line after normalization and the center of the image.

[0108] (4) Determine the weight coefficient of the reference line according to the area size of the target area.

[0109] The area size of the target area can refer to the area of the target area, or the horizontal size of the target area, such as the width of the target area, or the perimeter of the target area.

[0110] In the captured crop image, for the target area closer to the agricultural machinery equipment, its proportion in the crop image will be larger, and for the farther target area, the proportion will be smaller.

[0111] A larger area size indicates that the reference line of the target area has a greater impact on the generation of the navigation line. Therefore, similar to the way of determining the weight coefficient of the confidence level as described above, the normalized area size can be used as the weight coefficient of the reference line. Its normalization method is similar to that of the confidence level above and will not be repeated here.

[0112] (5) Determine the weight coefficient according to at least two kinds of information among the confidence level, fitting residual, distance, and area size.

[0113] In this implementation, a weight coefficient can be determined respectively first, and then the weight coefficients can be weighted and calculated. For example, for reference line 1, the weight coefficient a1 is determined according to the confidence level, and the weight coefficient a2 is determined according to the fitting residual. Then, a1 and a2 can be weighted and averaged to obtain the final weight coefficient of reference line 1, such as (a1*b + a2*(1 - b)) / 2, where b represents the weight of a1. After normalizing this calculation result, it can be used as the final weight coefficient of reference line 1. The processing of other reference lines is the same, and will not be exemplified one by one here.

[0114] In some embodiments, when determining the weight coefficient according to the confidence level and the fitting residual, the confidence level can also be multiplied by the reciprocal of the fitting residual to obtain the weight coefficient. At this time, the weight coefficient is positively correlated with the confidence level and negatively correlated with the fitting residual. The calculation formula is as follows: q represents the weight coefficient, p represents the confidence level, and σ represents the fitting residual. Then, all weight coefficients can be normalized: Where i represents the i-th reference line, and N represents the number of reference lines. represents the weight coefficient after normalization. represents the weight coefficient before normalization.

[0115] In the above implementation process, by comprehensively evaluating the weight coefficient of the reference line from multi-dimensional information, the accuracy and reliability of the navigation line generation are improved.

[0116] Based on the above embodiments, when splitting the target area, for each recognizable target, a target area is split, and multiple target areas may be obtained in this way, resulting in multiple reference lines being extracted. When there are multiple reference lines, in order to reduce the computational amount and improve the accuracy of the navigation line calculation, the reference lines can also be filtered, such as filtering and retaining the reference lines that meet the set conditions. The set conditions include at least one of the following: the confidence level is greater than the set confidence level, the fitting residual is less than the set residual, the distance is less than the set distance, and the area size is greater than the set size.

[0117] Among them, the set confidence level, the set residual, the set distance, and the set size can be flexibly set according to actual needs.

[0118] In this way, only the reference lines that meet the set conditions need to be retained, and these reference lines can be used as the reference lines for subsequent navigation line calculations. For the reference lines that do not meet the set conditions, they can be deleted.

[0119] In the above implementation process, by filtering the reference lines, unreliable data can be eliminated, error interference can be reduced, and the accuracy and reliability of the navigation line generation can be improved.

[0120] Based on the above embodiments, in the method of extracting the reference line of the target area, multiple central pixel points of the target area can be calculated first according to the boundary pixel points of the target area, and then the multiple central pixel points are fitted to obtain the reference line of the target area.

[0121] As Figure 3 shown, Figure 3 What is shown is the split target area (shaded part). At this time, the target area can refer to the crop area. When extracting the reference line, the boundary pixel points of each target area can be extracted first. For example, for a target area, its boundary can refer to the boundary in the length direction, and then boundary pixel point pairs can be extracted at a certain pixel interval in the length direction. A boundary pixel point pair includes the left and right boundary pixel points of the target area in the horizontal direction. Then the mean values of the respective boundary pixel point pairs can be taken to obtain multiple central pixel points. Then, some fitting algorithms, such as the RANSAC algorithm, can be used to fit the multiple central pixel points to obtain the reference line of the target area. At this time, the reference line can refer to the central reference line of the target area. For all target areas, their reference lines can be obtained in the same way.

[0122] Understandably, multiple central pixel points are calculated here, and the central reference line is fitted. In some other embodiments, the reference line fitted here may not be the central reference line. For example, weighted averaging can also be performed on each pair of boundary pixel points, and the weights here may not be evenly distributed. In this way, the obtained central pixel points may not necessarily be at the center but have a certain offset. At this time, a reference line can also be fitted.

[0123] When fitting here, the above-mentioned fitting residuals can be calculated simultaneously.

[0124] When fitting, a certain number of central pixel points can be randomly selected as the initial samples, and then these initial samples are used to fit a straight line as the candidate reference line. For each calculated central pixel point, calculate its distance to the candidate reference line. These distances are the residuals. Then, count the number of points with residuals less than a certain threshold. These points are considered inliers. Repeat the above steps, each time selecting different sample points and fitting different straight lines, record the number of inliers and the corresponding residuals for each fitting. Among all iterations, select the reference line of the fitting with the largest number of inliers as the final reference line.

[0125] For the finally determined reference line, calculate the residuals of all central pixel points to this reference line, and count the root mean square or other statistics of these residuals, which can be used as the fitting residuals.

[0126] In the above implementation process, calculating the central pixel points through the boundary pixel points and fitting these points to obtain the reference line can more accurately reflect the distribution pattern of the target area, improving the accuracy and reliability of the navigation line generation. At the same time, the fitting of the central pixel points reduces the amount of data processing and improves the processing speed and efficiency.

[0127] Based on the above embodiments, in the method of generating the current first navigation line, for the convenience of calculation, the reference line can be aligned to the same coordinate system first. The origin coordinates of this coordinate system are the optical center point of the image collector used to collect the crop image. Then, using the weight coefficients of the reference line, weighted calculation is performed based on the aligned reference line to generate the current first navigation line of the agricultural machinery equipment.

[0128] After the above reference line is extracted, combined with the height information from the image sensor such as the camera to the ground, using the monocular scale recovery algorithm, the 3D position of the reference line in the camera coordinate system can be obtained. At this time, a 3D reference line can be obtained, and the above coordinate system is the camera coordinate system.

[0129] Aligning the reference line to the same coordinate system here can mean first converting the reference line into a 3D reference line, and then translating each 3D reference line to the optical center point, that is, the camera center point, so that all 3D reference lines pass through the camera center point.

[0130] Then, the weight coefficients of each reference line can be utilized to perform weighted calculation on the reference lines, thereby generating the current first navigation line of the agricultural machinery equipment.

[0131] Specifically, for the aligned reference lines, the corresponding abscissas under the same ordinate can be obtained. The aligned reference lines all pass through the camera center point, and the camera center point is the origin coordinate. These reference lines intersect. Thus, when calculating the navigation line, the abscissas of each reference line under the same ordinate can be weighted and calculated according to the corresponding weight coefficients to obtain a reference coordinate point. Then, the reference coordinate point and the camera center point are fitted into a straight line, and the current first navigation line can be obtained.

[0132] For example, for the aligned reference lines, take the same ordinate Y, calculate the abscissa Xi of each reference line, and then perform weighted average on Xi to obtain the final abscissa X, that is w i represents the weight coefficient of the i-th reference line, and n represents the number of reference lines.

[0133] Then, a straight line is calculated based on the reference coordinate point (X, Y) and the camera center point coordinate (0, 0), and this straight line is the current first navigation line.

[0134] In the above implementation process, by aligning the reference lines to a unified coordinate system with the optical center of the image collector as the origin and performing weighted calculation based on the weight coefficients to generate the navigation line, the accuracy and stability of the navigation line are effectively improved. This process ensures the fusion of different reference lines under the same benchmark, avoids deviations caused by coordinate differences, makes the generated navigation line closer to the actual crop distribution, and enhances the accuracy and reliability of agricultural machinery operations.

[0135] On the basis of the above embodiments, in order to provide more navigation references, the current positioning result of the agricultural machinery equipment can also be obtained, and then the current final navigation line of the agricultural machinery equipment is determined according to the current positioning result and the current first navigation line.

[0136] Among them, the current positioning result can be detected and obtained by a positioning module installed on the agricultural machinery equipment. The positioning module can be, for example, GNSS (Global Navigation Satellite System), INS (Inertial Navigation System), etc.

[0137] The current positioning result includes the current position of the agricultural machinery equipment. Therefore, the current final navigation line can be determined by combining the current positioning result and the current first navigation line. For example, it is judged whether the current position coincides with the current first navigation line or the deviation is within the set range. If so, the current final navigation line can be determined as the current first navigation line. If not, the current first navigation line can be recalculated, such as adjusting the weight coefficients of each reference line. The adjustment strategy can be manual adjustment or reselecting other methods to determine the weight coefficients. Or, it can also be to fit a straight line by combining the current position and the calculated coordinates (X, Y) above as the current final navigation line.

[0138] In the above implementation process, the current positioning result provides the real-time position information of the agricultural machinery. After being fused with the current first navigation line, it can effectively correct the error of the current first navigation line and ensure the real-time performance and accuracy of the navigation line.

[0139] On the basis of the above embodiments, when determining the current final navigation line, the current second navigation line can also be determined according to the current positioning result, and then the current final navigation line of the agricultural machinery equipment can be determined according to the current first navigation line and the current second navigation line.

[0140] The current second navigation line can be determined according to the current positioning result and the heading. The heading can refer to the direction directly in front of the current position of the agricultural machinery equipment or the direction consistent with the current first navigation line.

[0141] The current second navigation line can be generated according to the current position and heading of the agricultural machinery equipment. When determining the current final navigation line, the parameters (such as slope and intercept) of the current first navigation line and the current second navigation line can be weighted and averaged (the weights of the weighted average can be determined according to their confidence levels), and the fused navigation line can be obtained as the current final navigation line.

[0142] In the above implementation process, by separately generating the current first navigation line based on vision and the current second navigation line based on the positioning system, and then comprehensively determining the final navigation line, the deep fusion and complementarity of multi-source data are realized. The vision navigation line provides the actual distribution information of the crops, while the positioning system navigation line provides a high-precision position reference. The combination of the two not only improves the accuracy and stability of the navigation, but also enhances the adaptability of the system to the complex farmland environment.

[0143] Based on the above embodiments, in the method of determining the current final navigation line according to the current positioning result and the current first navigation line, the navigation line can also be selected according to the reliability of the current positioning result. Specifically, if the current positioning result is unreliable or the deviation between the current positioning result and the final navigation line determined at the previous moment is greater than the set deviation, then the current final navigation line is determined to be the current first navigation line. If the current positioning result is reliable, the deviation between the current positioning result and the final navigation line determined at the previous moment is less than or equal to the set deviation, and the confidence level of the current first navigation line is less than the set confidence level, then the current final navigation line is determined to be the current second navigation line determined according to the current positioning result.

[0144] Among them, the current positioning result being unreliable may mean that the current positioning result is a non-fixed solution, and the current positioning result being reliable may mean that the current positioning result is a fixed solution.

[0145] The deviation between the current positioning result and the final navigation line determined at the previous moment refers to the distance between the current position of the agricultural machinery equipment and the final navigation line at the previous moment. For example, the distance in the horizontal direction between the current position and the final navigation line can be calculated here. Specifically, when calculating, a horizontal line can be drawn in the horizontal direction with the current position as the origin. This horizontal line intersects the final navigation line at the previous moment, the intersection point is determined, and the intersection point coordinates are obtained. Then, the distance between the intersection point coordinates and the current position is calculated, and this distance can be used as the distance between the current position and the final navigation line at the previous moment.

[0146] The confidence level of the current first navigation line refers to the degree of credibility of the current first navigation line. When generating the current first navigation line above, the confidence level of the current first navigation line can be output together.

[0147] In some embodiments, the confidence level of the current first navigation line is obtained by weighted calculation of the confidence levels of the reference lines according to the weight coefficients of the reference lines. As in the above example, a confidence level is obtained for each reference line, and the current first navigation line is generated based on these reference lines. The confidence level of the current first navigation line can be obtained by weighted averaging the confidence levels of these reference lines.

[0148] For example, if there are n reference lines, then the n confidence levels are weighted averaged according to the weight coefficients of the reference lines to obtain the confidence level of the current first navigation line.

[0149] Determining the confidence level of the current first navigation line by weighted calculation of the confidence levels of multiple reference lines effectively improves the accuracy of the reliability evaluation of the navigation line. This weighting method comprehensively considers the confidence levels and their weights of each reference line, making the confidence level of the navigation line not only reflect the reliability of a single reference line.

[0150] If the current positioning result is unreliable or the deviation is greater than the set deviation, it indicates that the positioning result may be inaccurate. In this case, the current first navigation line is directly used for navigation. If the current positioning result is reliable and the deviation is small, it means the positioning result is credible. And if the confidence level of the current first navigation line is small at this time, the current second navigation line generated by the current positioning result can be used as the current final navigation line for navigation.

[0151] In addition to the above situations, in other cases, such as when both the positioning result and the current first navigation line are credible, the current first navigation line and the current second navigation line can be directly fused to obtain the current final navigation line.

[0152] In the above implementation process, when the positioning result is unreliable or the deviation from the navigation line at the previous moment is too large, the first navigation line generated by vision is directly adopted, avoiding the operation deviation caused by positioning errors. And when the positioning result is reliable and the deviation is within the allowable range, and at the same time the confidence level of the vision navigation line is low, the second navigation line based on the positioning result is switched to, giving full play to the high-precision advantage of the positioning system. This intelligent switching mechanism effectively copes with various situations in the complex farmland environment, reduces the limitations of a single navigation method, and improves the stability and efficiency of agricultural machinery operations.

[0153] Please refer to Figure 4 , Figure 4 FIG. 200 is a structural block diagram of a navigation line generation device 200 provided by an embodiment of the present application. The device 200 may be a module, a program segment, or code on an electronic device. It should be understood that the device 200 corresponds to the above method embodiment and can execute each step involved in the method embodiment. The specific functions of the device 200 can be seen in the above description. To avoid repetition, the detailed description is appropriately omitted here.

[0154] Optionally, the device 200 includes:

[0155] An area segmentation module 210, configured to obtain a crop image and segment a target area from the crop image, where the target area refers to a crop area and / or a land area;

[0156] A reference line extraction module 220, configured to extract a reference line of the target area, where the reference line characterizes the distribution form of the target area;

[0157] A coefficient acquisition module 230, configured to acquire a weight coefficient of the reference line, where the weight coefficient is used to characterize the influence degree of the reference line on the generation of the navigation line;

[0158] A navigation line generation module 240, configured to generate a current first navigation line of the agricultural machinery according to the weight coefficient of the reference line and the reference line.

[0159] Optionally, the coefficient obtaining module 230 is configured to determine a weight coefficient of the reference line according to information of the reference line, where the information of the reference line includes at least one of confidence, fitting residual, distance from the image center of the crop image, and region size of the target region, the confidence refers to the credibility of the segmentation result of the target region, and the fitting residual refers to the residual determined when fitting the reference line.

[0160] Optionally, if there are multiple reference lines, the reference line extraction module 220 is further configured to screen and retain the reference lines that meet set conditions, where the set conditions include at least one of the following: the confidence is greater than a set confidence, the fitting residual is less than a set residual, the distance is less than a set distance, and the region size is greater than a set size.

[0161] Optionally, the reference line extraction module 220 is configured to calculate a plurality of central pixel points of the target region according to boundary pixel points of the target region; fit the plurality of central pixel points to obtain a reference line of the target region.

[0162] Optionally, the navigation line generation module 240 is configured to align the reference lines to the same coordinate system, where the origin coordinates of the coordinate system are the optical center points of the image collector for collecting the crop image; use the weight coefficients of the reference lines to perform weighted calculation based on the aligned reference lines to generate a current first navigation line of the agricultural machinery device.

[0163] Optionally, the navigation line generation module 240 is further configured to obtain a current positioning result of the agricultural machinery device; determine a current final navigation line of the agricultural machinery device according to the current positioning result and the current first navigation line.

[0164] Optionally, the navigation line generation module 240 is further configured to determine a current second navigation line according to the current positioning result; determine a current final navigation line of the agricultural machinery device according to the current first navigation line and the current second navigation line.

[0165] Optionally, the navigation line generation module 240 is further configured to, if the current positioning result is unreliable or the deviation between the current positioning result and the final navigation line determined at the previous moment is greater than a set deviation, determine the current final navigation line as the current first navigation line;

[0166] where, if the current positioning result is reliable and the deviation between the current positioning result and the final navigation line determined at the previous moment is less than or equal to the set deviation, and the confidence of the current first navigation line is less than the set confidence, the current final navigation line is the current second navigation line determined according to the current positioning result.

[0167] Optionally, if there are multiple reference lines, the information of the reference lines includes confidence levels, and the confidence level of the current first navigation line is obtained by weighted calculation of the confidence levels of the reference lines according to the weight coefficients of the reference lines.

[0168] Optionally, the region segmentation module 210 is configured to determine a segmentation target according to the crop type in the crop image; and segment the target region where the segmentation target is located from the crop image.

[0169] It should be noted that those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0170] Please refer to Figure 5 , Figure 5 FIG. is a schematic structural diagram of an electronic device for executing a navigation line generation method provided by an embodiment of the present application. The electronic device (such as a vision terminal of an agricultural machine device or a server) may include: at least one processor 310, such as a CPU, at least one communication interface 320, at least one memory 330, and at least one communication bus 340. Among them, the communication bus 340 is used to implement connection communication between these components. Among them, the communication interface 320 of the device in the embodiment of the present application is used to perform signaling or data communication with other node devices. The memory 330 may be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. Optionally, the memory 330 may also be at least one storage device located far from the foregoing processor. Computer-readable instructions are stored in the memory 330. When the computer-readable instructions are executed by the processor 310, the electronic device executes the method process shown above.

[0171] It can be understood that Figure 5 the structure shown is only schematic, and the electronic device may further include more or fewer components than those shown in Figure 5 , or have a different configuration from that shown in Figure 5 . Figure 5 Each component shown in

[0172] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it executes the method process executed by the electronic device in the foregoing method embodiment.

[0173] This embodiment discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided in the above method embodiments. For example, it includes:

[0174] Obtain a crop image, and segment a target area from the crop image. The target area refers to a crop area and / or a land area;

[0175] Extract a reference line of the target area, where the reference line characterizes the distribution form of the target area;

[0176] Obtain a weight coefficient of the reference line, where the weight coefficient is used to characterize the influence degree of the reference line on the generation of a navigation line;

[0177] Generate a current first navigation line of the agricultural machinery according to the weight coefficient of the reference line and the reference line.

[0178] In summary, the embodiments of the present application provide a method, device, storage medium and program product for generating a navigation line. By segmenting a target area from a crop image, where the target area refers to a crop area and / or a land area, then extracting a reference line of the target area, and obtaining a weight coefficient of the reference line, a navigation line of the agricultural machinery can be generated according to the weight coefficient and the reference line. In this way, during the operation of the agricultural machinery, the reference line of the target area can be comprehensively used to realize the dynamic generation of the navigation line, improving the accuracy and real-time performance of agricultural machinery navigation, reducing manual interference, enhancing the adaptability of the system to complex operating environments, and effectively saving operating costs and improving operating efficiency.

[0179] In the embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some communication interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0180] In addition, the units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0181] Furthermore, in each embodiment of the present application, each functional module may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.

[0182] In this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0183] The above description is only for the embodiments of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for generating a navigation line, characterized in that, The method includes: Obtaining a crop image and segmenting a target area from the crop image, where the target area refers to a crop area and / or a land area; Extracting a reference line of the target area, where the reference line characterizes the distribution form of the target area; Obtaining a weight coefficient of the reference line, where the weight coefficient is used to characterize the influence degree of the reference line on the generation of a navigation line; Generating a current first navigation line of the agricultural machinery device according to the weight coefficient of the reference line and the reference line.

2. The method according to claim 1, wherein The obtaining of the weight coefficient of the reference line includes: Determining the weight coefficient of the reference line according to the information of the reference line, where the information of the reference line includes at least one of confidence, fitting residual, distance from the image center of the crop image, and area size of the target area, the confidence refers to the credibility of the segmentation result of the target area, and the fitting residual refers to the residual determined when fitting the reference line.

3. The method according to claim 2, wherein If there are multiple reference lines, before obtaining the weight coefficient of the reference line, it further includes: Screening and retaining the reference lines that meet the set conditions, where the set conditions include at least one of the following: the confidence is greater than a set confidence, the fitting residual is less than a set residual, the distance is less than a set distance, and the area size is greater than a set size.

4. The method according to claim 1, wherein The extracting of the reference line of the target area includes: Calculating a plurality of central pixel points of the target area according to the boundary pixel points of the target area; Fitting the plurality of central pixel points to obtain the reference line of the target area.

5. The method according to claim 1, wherein The generating of the current first navigation line of the agricultural machinery device according to the weight coefficient of the reference line and the reference line includes: Aligning the reference line to the same coordinate system, where the origin coordinate of the coordinate system is the optical center point of the image collector for collecting the crop image; Using the weight coefficient of the reference line to perform weighted calculation based on the aligned reference line to generate the current first navigation line of the agricultural machinery device.

6. The method according to claim 1, characterized in that The method further includes: Obtaining the current positioning result of the agricultural machinery device; Determining the current final navigation line of the agricultural machinery device according to the current positioning result and the current first navigation line.

7. The method according to claim 6, wherein The determining of the current final navigation line of the agricultural machinery device according to the current positioning result and the current first navigation line includes: Determining a current second navigation line according to the current positioning result; Determining the current final navigation line of the agricultural machinery device according to the current first navigation line and the current second navigation line.

8. The method according to claim 6, characterized in that, The determining of the current final navigation line according to the current positioning result and the current first navigation line includes: If the current positioning result is unreliable or the deviation between the current positioning result and the final navigation line determined at the previous moment is greater than a set deviation, then determining the current final navigation line as the current first navigation line; Among them, if the current positioning result is reliable and the deviation between the current positioning result and the final navigation line determined at the previous moment is less than or equal to the set deviation, and the confidence level of the current first navigation line is less than the set confidence level, then the current final navigation line is the current second navigation line determined according to the current positioning result.

9. The method according to claim 8, wherein If there are multiple reference lines, the information of the reference lines includes confidence levels, and the confidence level of the current first navigation line is obtained by weighted calculation of the confidence levels of the reference lines according to the weight coefficients of the reference lines.

10. The method according to any one of claims 1-9, characterized in that The segmenting the target area from the crop image includes: Determining a segmentation target according to the crop type in the crop image; Segmenting the target area where the segmentation target is located from the crop image.

11. An agricultural machinery device, characterized in that, The agricultural machinery equipment includes: An image collector installed on the agricultural machinery equipment for collecting crop images; A vision terminal for executing the navigation line generation method according to any one of claims 1-10; A control system for controlling the agricultural machinery equipment to travel according to the current first navigation line.

12. An electronic device, characterized in that, It includes a processor and a memory, and the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method according to any one of claims 1-10 is run.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the method according to any one of claims 1-10 is run.

14. A computer program product, characterized in that, It includes computer program instructions, and when the computer program instructions are read and run by the processor, the method according to any one of claims 1-10 is executed.

Images