A soil detection method and system based on soil improvement
Automatic soil sampling over the farmland is solved through the flight sampling device, which solves the problem of dangerous movement of staff during soil sampling in non-grained farmlands and improves the convenience and accuracy of detection.
Patent Information
- Application Number
- CN202411569151.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-05
- Publication Date
- 2025-03-21
- Estimated Expiration
- 2044-11-05
AI Technical Summary
In non-grained farmland, staff need to move to the farmland in person during soil sampling, especially in steep terrain, which is dangerous, which makes soil detection more difficult and troublesome.
The flight sampling device is adopted to obtain the demand detection location, feature identification, field ridge identification path determination, sampling point division and detection sampling path planning, so as to realize the flight sampling device moves over the farmland and automatically performs soil sampling.
It reduces the risk of staff moving on farmland and improves the convenience and accuracy of soil detection, especially in soil detection of steep farmland.
Smart Images
Figure CN119064558B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of soil detection technology, and in particular to a soil detection method and system based on soil improvement. Background Art
[0002] Non-grain farmland refers to the name of farmland that was originally used to grow grain crops but is now used to grow non-grain crops. Due to the long-term conversion of farmland into vegetable fields, orchards, tea gardens, etc., the structure of the soil will change. At this time, the original cultivated layer of the farmland will be destroyed, resulting in severe stripping of the topsoil of some farmland, which cannot meet the growth needs of grain crops. Therefore, it is necessary to improve the soil of non-grain farmland.
[0003] In the related art, in order to improve the soil, it is necessary to detect the specific conditions of the soil, so as to improve it in different ways according to the specific conditions of the soil. When testing the soil, it is necessary to first sample the soil from the farmland. Generally, the soil sampling is carried out by the staff moving to the farmland where the sampling is required, and then determining the soil sampling point from the farmland to carry out the sampling operation.
[0004] In the above-mentioned related technologies, when taking soil samples, workers need to move to the farmland in person. However, for farmland with steeper locations, there is a certain degree of danger when workers move to the farmland, which makes soil sampling more difficult, and thus makes soil testing more troublesome, and there is still room for improvement. Summary of the invention
[0005] In order to facilitate soil detection and processing, the present application provides a soil detection method and system based on soil improvement.
[0006] In the first aspect, the present application provides a soil detection method based on soil improvement, which adopts the following technical solution:
[0007] A soil testing method based on soil improvement, comprising:
[0008] Get the required detection location;
[0009] Control the preset flight sampling device to move to the required detection position and obtain the initial detection image;
[0010] Performing feature recognition in the initial detection image to determine the ridge recognition position, and determining the ridge recognition path according to all the ridge recognition positions;
[0011] Control the flying sampling device to move to any end point of the ridge identification path, and control the flying sampling device to continue to acquire the path identification image to continue to determine the ridge identification position, and update the ridge identification path according to the ridge identification position, until the ridge identification path is completely closed to output the device completion position and the required detection area;
[0012] Determine the sampling points in the required detection area according to the preset sampling division rules, and determine the detection sampling path according to the sampling points and the device completion position;
[0013] The flying sampling device is controlled to move along the detection sampling path to perform soil sampling operations at the sampling points in turn.
[0014] Optionally, the step of determining the ridge identification path according to all ridge identification positions includes:
[0015] Obtaining the distance between each ridge identification position and the boundary line of the initial detection image;
[0016] The ridge identification position where the boundary separation distance is less than the preset reference separation distance is defined as the endpoint identification position;
[0017] Determine the distance between points based on the endpoint identification position and the ridge identification position;
[0018] Constructing an initially empty identification position set for placing the ridge identification position with the endpoint identification position, and placing the corresponding ridge identification position when the distance between the points is less than the preset close distance in the identification position set;
[0019] Continue to determine the distance between the ridge identification positions within the identification position set and the ridge identification positions outside the identification position set, and continue to place the ridge identification positions until the ridge identification positions within the identification position set are not increased;
[0020] Counting the ridge identification positions within the identification position set to determine the number of set positions;
[0021] The number of set positions with the largest value is determined according to a preset sorting rule, and a ridge identification path is constructed according to the ridge identification positions in the identification position set corresponding to the number of set positions with the largest value.
[0022] Optionally, the step of constructing a ridge identification path according to the ridge identification positions in the identification position set corresponding to the number of set positions with the largest value includes:
[0023] Determine the distance between the ridges according to any two ridge identification positions in the identification position set;
[0024] Determine the distance between the ridges with the largest values according to the sorting rules, and construct a position trend line segment according to the two ridge identification positions corresponding to the distance between the ridges with the largest values;
[0025] Projecting each ridge identification position vertically onto the position trend line segment to determine the projection identification position corresponding to the ridge identification position;
[0026] Determine the simulated demand point at any end point of the position trend line segment in the direction of the other end point according to a preset appropriate spacing distance;
[0027] Determine the simulated projection distance according to the projection identification position and the simulated demand point, and determine the simulated projection distance with the smallest value according to the sorting rule, and define the projection identification position corresponding to the simulated projection distance with the smallest value as the demand projection position;
[0028] Re-determine the simulated projection distance according to the required projection position to continue determining the required projection position until the required projection position is the end point of the position trend line segment;
[0029] The ridge identification position corresponding to each required projection position is defined as the required arrival position, and the ridge identification path is determined by connecting the required arrival positions in series.
[0030] Optionally, the step of defining the ridge identification position corresponding to each required projection position as the required arrival position includes:
[0031] The ridge identification position corresponding to each required projection position is defined as the mark arrival position;
[0032] Determine whether a single demand projection location has only a unique identification arrival location;
[0033] If a single demand projection position has only a unique identified arrival position, then the identified arrival position is defined as the demand arrival position;
[0034] If a single demand projection position has not only a unique identification arrival position, then the average interval distance of all demand arrival positions determined before the demand arrival position corresponding to the demand projection position is determined, and the closest demand arrival position before the demand arrival position corresponding to the demand projection position is defined as the previous arrival position;
[0035] Determine the preceding sequence interval distance according to the preceding sequence arrival position and the arrival positions of each mark;
[0036] The difference interval distance is determined by calculation based on the previous interval distance and the mean interval distance, and the difference interval distance with the smallest value is determined based on the sorting rule, and the identified arrival position corresponding to the difference interval distance is defined as the required arrival position.
[0037] Optionally, the sampling partitioning rules include:
[0038] Get the required sampling quantity;
[0039] Randomly generate a number of simulated demand points for demand sampling in the demand detection area, and combine all currently generated simulated demand points to form a sampling simulation combination;
[0040] Construct similar areas with a preset radius based on each simulated demand point as the center point;
[0041] Determine a detection coverage area in the required detection area according to each similar area, and determine a detection coverage area according to the detection coverage area;
[0042] The detection coverage area with the largest value is determined according to the sorting rules, and the sampling simulation combination corresponding to the detection coverage area with the largest value is defined as the sampling use combination, and the simulated demand points within the sampling use combination are determined as sampling points.
[0043] Optionally, after the detection coverage area is determined, the soil detection method based on soil improvement also includes:
[0044] Determine whether there are at least two sampling simulation combinations with the same and largest detection coverage areas;
[0045] If there are not at least two sampling simulation combinations with the same and largest detection coverage area, the sampling simulation combination corresponding to the detection coverage area with the largest value is defined as the sampling use combination;
[0046] If there are at least two sampling simulation combinations with the same and largest detection coverage area, the sampling simulation combination corresponding to the detection coverage area with the largest value is defined as the sampling candidate combination;
[0047] Determine an invalid external area outside the required detection area according to each similar area in the sampling candidate combination, and determine an invalid coverage area according to the invalid external area;
[0048] The invalid coverage area with the smallest value is determined according to the sorting rule, and the sampling candidate combination corresponding to the invalid coverage area with the smallest value is defined as the sampling use combination.
[0049] In the second aspect, the present application provides a soil detection system based on soil improvement, which adopts the following technical solution:
[0050] A soil detection system based on soil improvement, comprising:
[0051] An acquisition module, used to acquire a required detection position;
[0052] A processing module, connected to the acquisition module, for storing and processing information;
[0053] The processing module controls the preset flight sampling device to move to the required detection position and enables the acquisition module to acquire the initial detection image;
[0054] The processing module performs feature recognition in the initial detection image to determine the ridge recognition position, and determines the ridge recognition path according to all the ridge recognition positions;
[0055] The processing module controls the flying sampling device to move to any end point of the ridge identification path, and controls the flying sampling device to continue to acquire the path identification image to continue to determine the ridge identification position, and updates the ridge identification path according to the ridge identification position, until the ridge identification path is completely closed to output the device completion position and the required detection area;
[0056] The processing module determines the sampling points in the required detection area according to the preset sampling division rules, and determines the detection sampling path according to the sampling points and the device completion position;
[0057] The processing module controls the flying sampling device to move along the detection sampling path to perform soil sampling operations at the sampling points in turn.
[0058] In a third aspect, the present application provides a computer storage medium capable of storing corresponding programs, which is convenient for detecting and processing soil, and adopts the following technical solutions:
[0059] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any of the above-mentioned soil detection methods based on soil improvement.
[0060] In summary, the present application includes at least one of the following beneficial technical effects:
[0061] When conducting soil testing and sampling, the flying sampling device can be controlled to move in the farmland to determine the specific area of the farmland, so as to facilitate the subsequent flying sampling device to perform soil sampling operations, so as to facilitate soil testing and processing;
[0062] Analyzing the determined ridge identification position can determine a certain erroneously determined ridge identification position, thereby making the determined ridge identification path more accurate;
[0063] When determining the sampling points, the sampling points can be simulated and distributed in advance to make soil sampling random and reduce the occurrence of inaccurate soil testing due to excessive soil concentration. BRIEF DESCRIPTION OF THE DRAWINGS
[0064] Figure 1 It is a flow chart of soil testing method based on soil improvement.
[0065] Figure 2 This is a schematic diagram of the process of determining the ridge identification path.
[0066] Figure 3 It is a module flow chart of soil testing method based on soil improvement. DETAILED DESCRIPTION
[0067] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-Figure 3 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0068] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.
[0069] The present application embodiment discloses a soil detection method based on soil improvement, referring to Figure 1 , the method flow of the soil detection method based on soil improvement includes the following steps:
[0070] Step S100: Obtain the required detection location.
[0071] The required detection location is the location of the farmland that needs soil testing, which can be manually input by the staff.
[0072] Step S101: controlling a preset flying sampling device to move to a required detection position and acquiring an initial detection image.
[0073] The flying sampling device is a device on a drone that carries a soil sampling instrument. By controlling the flying sampling device to move to the required detection position, a local image of the farmland can be acquired, which is the initial detection image. When the staff cannot directly determine the required detection position, the staff can manually control the flying sampling device to move to reach above the farmland. At this time, the staff can determine the required detection position based on the real-time position of the flying sampling device, and construct a suitable coordinate system based on the required detection position to determine each position point.
[0074] Step S102: performing feature recognition in the initial detection image to determine the ridge recognition position, and determining the ridge recognition path according to all the ridge recognition positions.
[0075] The ridge identification position is the position point of the ridge in the farmland determined by feature recognition, wherein the ridge identification position can be identified by constructing a recognition database by conducting neural network learning on the ridge image in advance, and then inputting the current initial detection image into the recognition database for position recognition; the ridge identification path is the path formed by all ridge identification positions, that is, the local path of the ridge at the farmland position in the image.
[0076] Step S103: Control the flying sampling device to move to any end point of the ridge identification path, and control the flying sampling device to continue to obtain the path identification image to continue to determine the ridge identification position, and update the ridge identification path according to the ridge identification position until the ridge identification path is closed to output the device completion position and the required detection area.
[0077] By controlling the flying sampling device to move along the ridge identification path, the positions of the remaining ridges can be identified and determined, so that the entire ridge of the farmland can be determined. At this time, the area enclosed by the ridge is the area where farmland soil exists in the farmland, which is also the area that needs to be detected; at the same time, when the area is identified and determined, the current position of the flying sampling device is output to facilitate the subsequent control of the flying sampling device, and this position is the completion position of the device.
[0078] Step S104: determining sampling points in the required detection area according to a preset sampling division rule, and determining a detection sampling path according to the sampling points and the device completion position.
[0079] The sampling division rule is a method for determining the sampling points, such as setting a sampling point at a fixed interval, etc. It can also be a method such as step S500-step S504. The specific method can be set by the staff according to the actual situation, which will not be described here. The sampling point is the location point where soil sampling is required; the detection sampling path is the path for the flight sampling device to move from the device completion position to each sampling point in sequence to perform soil sampling operations. In order to reduce the overall moving distance of the flight sampling device, different paths can be generated by randomly sorting the sampling points and combining the device completion position, and then the path with the shortest distance can be selected as the detection sampling path for use.
[0080] Step S105: Control the flying sampling device to move along the detection sampling path to perform soil sampling operations at the sampling points in sequence.
[0081] The flying sampling device is controlled to sample and process the soil to reduce the need for staff to go to the farmland in person. The soil sampling operation is carried out through intelligent equipment to facilitate soil testing and processing.
[0082] The steps of determining the ridge identification path according to all ridge identification positions include:
[0083] Step S200: obtaining the distance between each ridge recognition position and the boundary line of the initial detection image.
[0084] The boundary separation distance is the distance between the pixel point corresponding to each ridge identification position and the closest distance between the four boundary lines of the initial detection image.
[0085] Step S201: defining a ridge identification position whose boundary separation distance is less than a preset reference separation distance as an endpoint identification position.
[0086] The benchmark spacing distance is the maximum boundary spacing distance set by the staff to determine that the ridge identification position is the endpoint of the current local ridge. When the boundary spacing distance is less than the benchmark spacing distance, it means that the ridge identification position is at the edge of the image, that is, the ridge at this position may extend outside the image. At this time, it is defined as the endpoint identification position to distinguish different ridge identification positions for subsequent analysis.
[0087] Step S202: determining the distance between points according to the endpoint identification positions and the ridge identification positions.
[0088] The point spacing distance is the straight-line distance between the defined endpoint identification position and the remaining ridge identification positions.
[0089] Step S203: constructing an initially empty identification position set for placing ridge identification positions with the endpoint identification positions, and placing the corresponding ridge identification positions when the distance between points is less than a preset close distance in the identification position set.
[0090] The identification position set is a set for placing ridge identification positions. The close distance is the maximum point separation distance set by the staff that needs to be met when two ridge identification positions are considered close. That is, when the point separation distance is less than the close distance, it means that the two ridge identification positions are more likely to be on the same ridge. At this time, they are placed in the same identification position set for easy analysis in subsequent steps.
[0091] Step S204: Continue to determine the distance between the ridge identification positions in the identification position set and the ridge identification positions outside the identification position set, and continue to place the ridge identification positions until the number of ridge identification positions in the identification position set does not increase.
[0092] By continuously analyzing the distance between points, the points on the same ridge can be placed in the same identification position set to facilitate subsequent analysis.
[0093] Step S205: Count the ridge identification positions in the identification position set to determine the number of set positions.
[0094] The number of set positions is the total number of ridge identification positions in a single identification position set, and can be determined by counting the ridge identification positions one by one.
[0095] Step S206: determining the number of set positions with the largest value according to a preset sorting rule, and constructing a ridge identification path according to the ridge identification positions in the identification position set corresponding to the number of set positions with the largest value.
[0096] The sorting rule is a method set by the staff to sort the size of values, such as the bubble method. The sorting rule can be used to determine the number of set positions with the largest value, that is, the ridge identification position in the corresponding identification position set is the position on the ridge of the farmland at that time, and the remaining ridge identification positions are obtained by erroneous analysis and need to be eliminated. At this time, the ridge identification path can be constructed normally according to the ridge identification positions in the identification position set, thereby ensuring that the determined ridge identification path is accurate.
[0097] The step of constructing a ridge identification path according to the ridge identification positions in the identification position set corresponding to the number of set positions with the largest value includes:
[0098] Step S300: determining the distance between the ridges according to any two ridge identification positions in the identification position set.
[0099] The distance between ridges is the straight-line distance between any two ridge identification positions in the identification position set.
[0100] Step S301: determining the maximum ridge distance according to the sorting rule, and constructing a position trend line segment according to the two ridge identification positions corresponding to the maximum ridge distance.
[0101] The distance between the ridges with the largest values can be determined by the sorting rules. At this time, a position trend line segment can be constructed according to the corresponding two ridge identification positions. The position trend line segment can reflect the change trend of the ridge and facilitate the determination of the entire ridge identification path. The construction method of the position trend line segment can refer to Figure 2 .
[0102] Step S302: vertically project each ridge identification position onto the position trend line segment to determine the projection identification position corresponding to the ridge identification position.
[0103] The projection identification position is the projection position point obtained by vertically projecting the ridge identification position onto the position trend line segment.
[0104] Step S303: Determine a simulated demand point at any end point of the position trend line segment toward the other end point according to a preset appropriate spacing distance.
[0105] The appropriate interval distance is a fixed distance set in advance by the staff. The simulated demand point is a position point that needs to be determined on the position trend segment when the ridge identification path can be determined better under theoretical circumstances. Taking any end point of the position trend segment as the starting point, a simulated demand point is defined at every appropriate interval until the other end point of the position trend segment.
[0106] Step S304: determining a simulated projection distance according to the projection identification position and the simulated demand point, and determining the simulated projection distance with the smallest value according to the sorting rule, and defining the projection identification position corresponding to the simulated projection distance with the smallest value as the demand projection position.
[0107] The simulated projection distance is the distance between each projection identification position and the simulated demand point. The simulated projection distance with the smallest value can be determined through the sorting rules. That is, the corresponding projection identification position is closest to the required simulated demand point at this time, and it can be defined as the required projection position.
[0108] Step S305: re-determine the simulated projection distance according to the required projection position to continue determining the required projection position until the required projection position is the endpoint of the position trend line segment.
[0109] The subsequent demand projection positions can be determined by the demand projection position close to the starting end point of the position trend line segment, so that the projection identification position that needs to be determined in theoretical circumstances can be determined, which is convenient for subsequent analysis.
[0110] Step S306: defining the ridge identification position corresponding to each required projection position as the required arrival position, and connecting the required arrival positions determined in sequence in series to determine the ridge identification path.
[0111] The required arrival positions are defined to distinguish the ridge identification positions that can theoretically better determine the ridge identification path, so that each required arrival position can be processed in series to obtain a more accurate ridge identification path, thereby controlling the flight sampling device to move better.
[0112] The steps of defining the ridge identification position corresponding to each required projection position as the required arrival position include:
[0113] Step S400: defining the ridge identification position corresponding to each required projection position as the mark arrival position.
[0114] Define the mark arrival position to distinguish different ridge identification positions for subsequent analysis.
[0115] Step S401: Determine whether a single required projection position has only a unique identified arrival position.
[0116] The purpose of the judgment is to find out whether there are at least two ridge identification positions projected on the same position point, so as to facilitate subsequent analysis.
[0117] Step S4011: If a single required projection position has only a unique identified arrival position, the identified arrival position is defined as the required arrival position.
[0118] When a single required projection position has only a unique identified arrival position, it means that there is only a unique identified arrival position that meets the requirements. In this case, it can be defined as the required arrival position.
[0119] Step S4012: If a single demand projection position not only has a unique identified arrival position, then the mean interval distance of all demand arrival positions determined before the demand arrival position corresponding to the demand projection position is determined, and the closest demand arrival position before the demand arrival position corresponding to the demand projection position is defined as the preceding arrival position.
[0120] When a single demand projection position not only has a unique identification arrival position, it means that there are multiple location points that can serve as the demand arrival position, and further analysis is required at this time; the mean interval distance is the average value of the distances between adjacent demand arrival positions among all previously determined demand arrival positions. The preceding arrival position is defined to determine the closest demand arrival position for subsequent analysis.
[0121] Step S402: determining the preceding interval distance according to the preceding arrival position and the arrival positions of each mark.
[0122] The predecessor interval distance is the distance value between the predecessor arrival position and the arrival position of each mark.
[0123] Step S403: Calculate the difference interval distance according to the previous interval distance and the mean interval distance, and determine the difference interval distance with the smallest value according to the sorting rule, and define the marked arrival position corresponding to the difference interval distance as the required arrival position.
[0124] The difference interval distance is the difference between the previous interval distance and the mean interval distance. The difference is an absolute value. The difference interval distance with the smallest value can be determined through the sorting rules. That is, the corresponding identification arrival position meets the distribution requirements of the required arrival position at this time. In this case, it can be defined as the required arrival position.
[0125] The sampling division rules include:
[0126] Step S500: Obtain the required sampling quantity.
[0127] The required sampling quantity is the number of sampling points required entered by the staff.
[0128] Step S501: randomly generating a demand sampling number of simulated demand points in a demand detection area, and combining all currently generated simulated demand points to form a sampling simulation combination.
[0129] The simulated demand points are simulated locations for soil sampling, which are generated using random rules, and the sampling simulation combination is a combination of all simulated demand points.
[0130] Step S502: constructing similar areas with a radius of a preset similar distance based on each simulated demand point as a center point.
[0131] The similarity distance is the maximum distance between two locations set by the staff that is allowed to appear when the soil conditions are considered to be similar. By simulating the demand points to construct similar areas, the distribution of the currently sampled soil can be determined to facilitate subsequent analysis.
[0132] Step S503: determining a detection coverage area in the required detection area according to each similar area, and determining a detection coverage area according to the detection coverage area.
[0133] The detection coverage area is the area of all similar areas within the required detection area, that is, the soil range area that can be covered when soil sampling operations are performed based on simulated demand points in theoretical conditions. The detection coverage area is the area of the detection coverage area.
[0134] Step S504: Determine the detection coverage area with the largest value according to the sorting rule, define the sampling simulation combination corresponding to the detection coverage area with the largest value as the sampling use combination, and determine the simulated demand point in the sampling use combination as the sampling point.
[0135] The sorting rules can be used to determine the detection coverage area with the largest value, which means that the soil detection effect of the farmland can be best at this time. At this time, the corresponding sampling simulation combination is defined as a sampling use combination to distinguish different sampling simulation combinations, so that the simulation demand point corresponding to the sampling use combination can be determined as the sampling point to control the flying sampling device to perform sampling operations.
[0136] After the detection coverage area is determined, soil detection methods based on soil improvement also include:
[0137] Step S600: Determine whether there are at least two sampling simulation combinations with the same and largest detection coverage areas.
[0138] The purpose of the judgment is to find out whether there are multiple sampling simulation combinations that meet the requirements, so as to determine the only sampling combination.
[0139] Step S6001: If there are not at least two sampling simulation combinations with the same and largest detection coverage areas, the sampling simulation combination corresponding to the detection coverage area with the largest value is defined as the sampling use combination.
[0140] When there are not at least two sampling simulation combinations with the same and largest detection coverage area, it means that there is only one sampling simulation combination that meets the requirements. In this case, it can be normally determined as the sampling combination.
[0141] Step S6002: If there are at least two sampling simulation combinations with the same and largest detection coverage areas, the sampling simulation combination corresponding to the detection coverage area with the largest value is defined as the sampling candidate combination.
[0142] When there are at least two sampling simulation combinations with the same and largest detection coverage area, it means that there are multiple sampling simulation combinations that meet the requirements. At this time, they are defined as sampling alternative combinations to distinguish different sampling simulation combinations for subsequent analysis.
[0143] Step S601: determining an invalid external region outside the required detection region according to each similar region in the sampled candidate combination, and determining an invalid coverage area according to the invalid external region.
[0144] The invalid external region is the region outside the required detection region in each similar region, and the invalid coverage area is the area of the invalid external region.
[0145] Step S602: Determine the invalid coverage area with the smallest value according to the sorting rule, and define the sampling candidate combination corresponding to the invalid coverage area with the smallest value as the sampling use combination.
[0146] The invalid coverage area with the smallest value can be determined through the sorting rules, that is, the range of invalid detection when soil sampling is performed under the sampling alternative combination at this time is the smallest, that is, the soil detection effect is the best. At this time, it can be determined as the sampling combination for use.
[0147] Reference Figure 3 Based on the same inventive concept, an embodiment of the present invention provides a soil detection system based on soil improvement, comprising:
[0148] An acquisition module, used to acquire a required detection position;
[0149] A processing module, connected to the acquisition module, for storing and processing information;
[0150] The processing module controls the preset flight sampling device to move to the required detection position and enables the acquisition module to acquire the initial detection image;
[0151] The processing module performs feature recognition in the initial detection image to determine the ridge recognition position, and determines the ridge recognition path according to all the ridge recognition positions;
[0152] The processing module controls the flying sampling device to move to any end point of the ridge identification path, and controls the flying sampling device to continue to acquire the path identification image to continue to determine the ridge identification position, and updates the ridge identification path according to the ridge identification position, until the ridge identification path is completely closed to output the device completion position and the required detection area;
[0153] The processing module determines the sampling points in the required detection area according to the preset sampling division rules, and determines the detection sampling path according to the sampling points and the device completion position;
[0154] The processing module controls the flying sampling device to move along the detection sampling path and perform soil sampling operations at the sampling points in sequence;
[0155] The ridge identification position screening module is used to eliminate some ridge identification positions that are not accurately identified;
[0156] A ridge identification path determination module is used to more accurately determine the ridge identification path according to the ridge identification position;
[0157] A module for determining the required arrival location is used to determine a relatively accurate required arrival location;
[0158] A sampling division rule determination module is used to determine a more appropriate sampling division rule to determine the sampling points;
[0159] The sampling simulation combination screening module is used to screen multiple sampling simulation combinations that meet the requirements.
[0160] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.
[0161] An embodiment of the present invention provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute a soil detection method based on soil improvement.
[0162] Computer storage media include, for example, various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.
Claims
1. A soil detection method based on soil improvement, characterized in that: include: Get the required detection location; Control the preset flight sampling device to move to the required detection position and obtain the initial detection image; Performing feature recognition in the initial detection image to determine the ridge recognition position, and determining the ridge recognition path according to all the ridge recognition positions; Control the flying sampling device to move to any end point of the ridge identification path, and control the flying sampling device to continue to acquire the path identification image to continue to determine the ridge identification position, and update the ridge identification path according to the ridge identification position, until the ridge identification path is completely closed to output the device completion position and the required detection area; Determine the sampling points in the required detection area according to the preset sampling division rules, and determine the detection sampling path according to the sampling points and the device completion position; Controlling the flying sampling device to move along the detection sampling path and sequentially perform soil sampling operations at the sampling points; The steps of determining the ridge identification path according to all ridge identification positions include: Obtaining the distance between each ridge identification position and the boundary line of the initial detection image; The ridge identification position where the boundary separation distance is less than the preset reference separation distance is defined as the endpoint identification position; Determine the distance between points based on the endpoint identification position and the ridge identification position; Constructing an initially empty identification position set for placing the ridge identification position with the endpoint identification position, and placing the corresponding ridge identification position when the distance between the points is less than the preset close distance in the identification position set; Continue to determine the distance between the ridge identification positions within the identification position set and the ridge identification positions outside the identification position set, and continue to place the ridge identification positions until the ridge identification positions within the identification position set are not increased; Counting the ridge identification positions within the identification position set to determine the number of set positions; Determine the number of set positions with the largest value according to a preset sorting rule, and construct a ridge identification path according to the ridge identification positions in the identification position set corresponding to the number of set positions with the largest value; The step of constructing a ridge identification path according to the ridge identification positions in the identification position set corresponding to the number of set positions with the largest value includes: Determine the distance between the ridges according to any two ridge identification positions in the identification position set; Determine the distance between the ridges with the largest values according to the sorting rules, and construct a position trend line segment according to the two ridge identification positions corresponding to the distance between the ridges with the largest values; Projecting each ridge identification position vertically onto the position trend line segment to determine the projection identification position corresponding to the ridge identification position; Determine the simulated demand point at any end point of the position trend line segment in the direction of the other end point according to a preset appropriate spacing distance; Determine the simulated projection distance according to the projection identification position and the simulated demand point, and determine the simulated projection distance with the smallest value according to the sorting rule, and define the projection identification position corresponding to the simulated projection distance with the smallest value as the demand projection position; Re-determine the simulated projection distance according to the required projection position to continue determining the required projection position until the required projection position is the end point of the position trend line segment; The ridge identification position corresponding to each required projection position is defined as the required arrival position, and the ridge identification path is determined by connecting the required arrival positions in series in sequence; The steps of defining the ridge identification position corresponding to each required projection position as the required arrival position include: The ridge identification position corresponding to each required projection position is defined as the mark arrival position; Determine whether a single demand projection location has only a unique identification arrival location; If a single demand projection position has only a unique identified arrival position, then the identified arrival position is defined as the demand arrival position; If a single demand projection position has not only a unique identification arrival position, then the average interval distance of all demand arrival positions determined before the demand arrival position corresponding to the demand projection position is determined, and the closest demand arrival position before the demand arrival position corresponding to the demand projection position is defined as the previous arrival position; Determine the preceding sequence interval distance according to the preceding sequence arrival position and the arrival positions of each mark; The difference interval distance is determined by calculation according to the preceding interval distance and the mean interval distance, and the difference interval distance with the smallest value is determined according to the sorting rule, and the marked arrival position corresponding to the difference interval distance is defined as the required arrival position; The sampling division rules include: Get the required sampling quantity; Randomly generate a number of simulated demand points for demand sampling in the demand detection area, and combine all currently generated simulated demand points to form a sampling simulation combination; Construct similar areas with a preset radius based on each simulated demand point as the center point; Determine a detection coverage area in the required detection area according to each similar area, and determine a detection coverage area according to the detection coverage area; The detection coverage area with the largest value is determined according to the sorting rules, and the sampling simulation combination corresponding to the detection coverage area with the largest value is defined as the sampling use combination, and the simulated demand points within the sampling use combination are determined as sampling points.
2. The soil detection method based on soil improvement according to claim 1, characterized in that: After the detection coverage area is determined, soil detection methods based on soil improvement also include: Determine whether there are at least two sampling simulation combinations with the same and largest detection coverage areas; If there are not at least two sampling simulation combinations with the same and largest detection coverage area, the sampling simulation combination corresponding to the detection coverage area with the largest value is defined as the sampling use combination; If there are at least two sampling simulation combinations with the same and largest detection coverage area, the sampling simulation combination corresponding to the detection coverage area with the largest value is defined as the sampling candidate combination; Determine an invalid external area outside the required detection area according to each similar area in the sampling candidate combination, and determine an invalid coverage area according to the invalid external area; The invalid coverage area with the smallest value is determined according to the sorting rule, and the sampling candidate combination corresponding to the invalid coverage area with the smallest value is defined as the sampling use combination.
3. A soil detection system based on soil improvement, characterized in that: The soil testing method based on soil improvement according to claim 1 or 2 is used to perform: An acquisition module, used to acquire a required detection position; A processing module, connected to the acquisition module, for storing and processing information; The processing module controls the preset flight sampling device to move to the required detection position and enables the acquisition module to acquire the initial detection image; The processing module performs feature recognition in the initial detection image to determine the ridge recognition position, and determines the ridge recognition path according to all the ridge recognition positions; The processing module controls the flying sampling device to move to any end point of the ridge identification path, and controls the flying sampling device to continue to acquire the path identification image to continue to determine the ridge identification position, and updates the ridge identification path according to the ridge identification position, until the ridge identification path is completely closed to output the device completion position and the required detection area; The processing module determines the sampling points in the required detection area according to the preset sampling division rules, and determines the detection sampling path according to the sampling points and the device completion position; The processing module controls the flying sampling device to move along the detection sampling path to perform soil sampling operations at the sampling points in turn.
4. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute any of the methods as claimed in claims 1 to 2.
Citation Information
Patent Citations
Measurement method and system of residual amount of residual plastic film on farmland soil surface layer
CN106644939A
Autonomous path planning method and device for unmanned agricultural machine
CN109240284A