Soil nutrient content measuring method, device and system

Abstract

The invention provides a soil nutrient content measuring method, device and system, and belongs to the technical field of soil detection.The method comprises the steps that a region of interest of soil at a target sampling point is determined, and the region of interest is divided into a plurality of grid regions; receiving an RGB image and a depth image of soil at the target sampling point; calculating the score of each grid region according to the RGB image and the depth image; taking the grid region with the highest score as an optimal insertion point of the sensor; after the coordinates of the optimal insertion point are output, sensing data obtained at the optimal insertion point are received; and obtaining the soil nutrient content of the target sampling point according to the sensing data. The optimal insertion point of the sensor is determined based on the score of the grid region, the soil nutrient content of the target sampling point is determined based on the sensing data of the optimal insertion point, it is ensured that the collected sensing data can comprehensively reflect the real soil characteristics of the target sampling point, and the accuracy and reliability of a measurement result are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of soil detection, and particularly relates to a soil nutrient content measurement method, device and system. BACKGROUND

[0002] Soil nutrient content measurement can accurately understand the soil condition and guide agricultural workers to formulate appropriate fertilization strategies, which is an important measure to improve crop yield. Soil nutrient content measurement is usually used to measure soil alkali nitrogen, available phosphorus, available potassium and other nutrient information.

[0003] In the prior art, sensor detection technology is used to measure soil nutrient content. First, the sampling points to be detected need to be determined, and then sensors are deployed for detection. However, in the prior art, there is a lack of effective planning method in the sensor positioning and deployment link, and it is difficult to determine the best insertion position of the sensor according to the actual conditions such as the complex terrain of the field soil, the vegetation distribution, the stone interference and the like, which leads to the fact that the collected data cannot fully reflect the real characteristics of the soil, and the accuracy and reliability of the measurement results are reduced. SUMMARY

[0004] The present application provides a soil nutrient content measurement method, device and system, which aims to solve the problem that the prior art is difficult to determine the best insertion position of the sensor according to the actual conditions such as the complex terrain of the field soil, the vegetation distribution, the stone interference and the like.

[0005] In a first aspect, the present application provides a soil nutrient content measurement method, comprising: determining a region of interest of the soil at a target sampling point, and dividing the region of interest into a plurality of grid regions; receiving an RGB image and a depth image of the soil at the target sampling point; calculating a score of each grid region according to the RGB image and the depth image; taking the grid region with the highest score as the best insertion point of the sensor; after outputting the coordinates of the best insertion point, receiving sensor data obtained at the best insertion point; obtaining the soil nutrient content of the target sampling point according to the sensor data.

[0006] As an embodiment, the score of the grid region is calculated, specifically comprising: calculating a plurality of score indicators of the grid region according to the RGB image and the depth image, the plurality of score indicators including soil surface flatness, hard object area, vegetation coverage area, soil humidity uniformity, soil humidity degree and soil texture uniformity; weighting and summing the plurality of score indicators to obtain the score of the grid region.

[0007] As an embodiment, the soil surface flatness of the grid region is calculated, specifically including: obtaining a point cloud subset of the grid region according to the depth image; calculating the angle between the normal vector of each point in the point cloud subset and the theoretical vertical direction; calculating the variance of the angles of all points in the point cloud subset; normalizing the variance to obtain the soil surface flatness of the grid region.

[0008] As an embodiment, the hard object area or vegetation coverage area of the grid region is calculated, specifically including: segmenting the RGB image of the grid region to obtain soil regions, hard object regions and vegetation regions in the grid region; calculating the area of the hard object region or the area of the vegetation region; normalizing the area of the hard object region or the area of the vegetation region as the hard object area or the vegetation coverage area of the grid region.

[0009] As an embodiment, the soil moisture degree of the grid region is calculated, specifically including: converting the RGB image into an HSV image; calculating the ratio of the S channel and the V channel of each pixel in the grid region based on the HSV image; calculating the average value of the ratio of all pixels in the grid region, and normalizing the average value as the soil moisture degree of the grid region.

[0010] As an embodiment, the soil moisture uniformity of the grid region is calculated, specifically including: taking the grid region as a first grid region, calculating the similarity between the soil moisture degree of the first grid region and the soil moisture degree of a second grid region around the first grid region; normalizing the similarity of the first grid region as the soil moisture uniformity of the first grid region.

[0011] As an embodiment, the soil texture uniformity of the grid region is calculated, specifically including: calculating the gray level co-occurrence matrix of the soil surface layer of the grid region according to the RGB image; normalizing the texture features of the gray level co-occurrence matrix as the soil texture uniformity of the grid region.

[0012] As an embodiment, the sensing data includes soil moisture, soil temperature, soil conductivity, soil pH value, and potential of soil oxidation-reduction reaction system. The neural network model is trained based on a data set obtained from the sensing data and the soil nutrient content data. The neural network model is trained based on a data set obtained from the sensing data and the soil nutrient content data.

[0013] In a second aspect, the present application further provides a soil nutrient content measuring device, comprising a division module, an image receiving module, a score calculation module, an optimal insertion point determination module, a sensing data receiving module, and a nutrient content obtaining module. The division module is configured to determine a region of interest of the soil at the target sampling point and divide the region of interest into a plurality of grid regions. The image receiving module is configured to receive an RGB image and a depth image of the soil at the target sampling point. The score calculation module is configured to calculate a score of each grid region based on the RGB image and the depth image. The optimal insertion point determination module is configured to determine the grid region with the highest score as the optimal insertion point of the sensor. The sensing data receiving module is configured to receive sensing data obtained at the optimal insertion point after outputting the coordinates of the optimal insertion point. The nutrient content obtaining module is configured to obtain the soil nutrient content at the target sampling point based on the sensing data.

[0014] As an embodiment, the score calculation module comprises an index calculation module and a weighted summation module. The index calculation module is configured to calculate a plurality of score indexes of the grid region based on the RGB image and the depth image, the plurality of score indexes comprising soil surface flatness, hard object area, vegetation coverage area, soil humidity uniformity, soil humidity degree, and soil texture uniformity. The weighted summation module is configured to perform weighted summation on the plurality of score indexes to obtain the score of the grid region.

[0015] In a third aspect, the present application further provides a soil nutrient content measuring system, comprising a vehicle frame and a control terminal, an image acquisition component, a sensor driving component, and a sensor assembly mounted on the vehicle frame. The image acquisition component is arranged on the bottom surface of the main body of the vehicle frame, and the field of view of the image acquisition component covers the soil surface of the target sampling point. The sensor driving component drives the sensor assembly to move relative to the soil surface, so that the plurality of sensors in the sensor assembly detect the soil of the target sampling point. The control terminal is signal connected with the image acquisition component and the sensor assembly, respectively, and is configured to execute the soil nutrient content measuring method described above.

[0016] In a fourth aspect, the present application provides an electronic device, which comprises a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor implements any of the soil nutrient content measurement methods described above when executing the computer program.

[0017] In a fifth aspect, the present application provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement any of the soil nutrient content measurement methods described above.

[0018] In a sixth aspect, the present application provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement any of the soil nutrient content measurement methods described above. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings described below are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0020] Figure 1 is one of the flowcharts of the soil nutrient content measurement method provided by the present application; Figure 2 is another flowchart of the soil nutrient content measurement method provided by the present application; Figure 3 is one of the flowcharts of the grid area calculation provided by the present application; Figure 4 is one of the structural diagrams of the soil nutrient content measurement device provided by the present application; Figure 5 is one of the structural diagrams of the soil nutrient content measurement system provided by the present application; Figure 6 is a bottom view of the soil nutrient content measurement system shown in Figure 5 Figure 7 is a front view of the soil nutrient content measurement system shown in Figure 5 Figure 8 is one of the operation flowcharts of the soil nutrient content measurement system provided by the present application; Figure 9 is one of the grid division diagrams provided by the present application; Figure 10 is one of the structural diagrams of the electronic device provided by the present application. ​​DETAILED DESCRIPTION

[0021] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions will be described clearly and completely in the present application with reference to the drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all of them. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0022] It should be noted that in the description of the present application, the terms "comprising", "containing" or any other variants thereof are intended to cover the non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or equipment including the element. The terms "upper", "lower" and the like indicate the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. Unless otherwise specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be connected inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0023] The terms "first", "second" and the like in the present application are used to distinguish similar objects, and are not used to describe a particular order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second" and the like are generally a class, and do not limit the number of objects, for example, the first object can be one or more. In addition, "and / or" means at least one of the connected objects, and the character " / ", generally means that the front and rear associated objects are in a "or" relationship.

[0024] The following will be described in conjunction with Figures 1 to 10 The soil nutrient content measurement method, device and system provided by the present application are described.

[0025] It should be noted that the soil nutrient content measurement system provided by the embodiment of the present application includes a mechanical structure for controlling the movement of the sensor and a soil nutrient content measurement device, so that the soil nutrient content measurement system is used to realize the preparation before measurement, the collection of sensor data and the processing of sensor data, and finally the soil nutrient content is obtained.

[0026] The soil nutrient content measurement method provided by the embodiment of the present application is realized based on the soil nutrient content measurement device. The soil nutrient content measurement method can automatically identify the best insertion point of the sensor on the soil at the target sampling point, so that the sensor component collects the sensor data at the best insertion point, which fully reflects the real characteristics of the soil, and improves the accuracy and reliability of the soil nutrient content measurement result.

[0027] Figure 1 is one of the flowcharts of the soil nutrient content measurement method provided by the present application. Figure 2 is another flowchart of the soil nutrient content measurement method provided by the present application.

[0028] As shown in Figure 1 , the soil nutrient content measurement method provided by the present application comprises: S110: determining a region of interest (ROI) of the soil at the target sampling point, and dividing the region of interest into a plurality of grid regions.

[0029] S120: receiving an RGB image and a depth image of the soil at the target sampling point.

[0030] S130: calculating the score of the soil in each grid region according to the RGB image and the depth image.

[0031] S140: taking the grid region with the highest score as the best insertion point of the sensor.

[0032] If there are multiple grid regions with the highest score, the grid region closest to the current position of the sensor component is selected as the best insertion point.

[0033] S150: after outputting the coordinates of the best insertion point, receiving the sensor data obtained at the best insertion point.

[0034] S160: obtaining the soil nutrient content of the target sampling point according to the sensor data.

[0035] It should be noted that the soil nutrient content measurement system is installed with an image acquisition component (such as a depth real sense camera) and a sensor component. The image acquisition component is used to collect the RGB image and the depth image of the soil, and the sensor component is used to obtain the sensor data.

[0036] In combination Figure 2In step S110, before each soil nutrient content measurement operation and after the soil nutrient content measurement system reaches the target sampling point, the soil nutrient content measurement device automatically obtains a region of interest (ROI) within the soil surface area displayed in the soil surface image based on the field of view of the image acquisition component and the soil surface image captured by the image acquisition component. The ROI is located within the soil surface area and covers the soil surface area as much as possible without being obscured by the mechanical structure of the soil nutrient content measurement system. It is understood that after the ROI is automatically obtained, the position and size of the ROI can be manually adjusted if fine-tuning is required.

[0037] After determining the ROI (region of interest), the ROI is divided into a grid based on the measurement area size of the sensor assembly on the soil nutrient content measurement system. All grid regions are equal in length and width. Each grid region represents a sensor insertion point, and the coordinates of the center point of the grid region are the coordinates of the center point where the sensor assembly is inserted into the soil. It should be noted that the number of grid regions is dynamically adjusted based on the size of the soil surface area.

[0038] After obtaining the RGB and depth images from the image acquisition component, the soil nutrient content measurement device then scores each grid area and selects the grid area with the highest score as the optimal sensor insertion point. After the soil nutrient content measurement system obtains sensor data from the sensor assembly, the soil nutrient content measurement device calculates the soil nutrient content at the target sampling point based on the sensor data.

[0039] In the embodiment of the present application, the soil at the target sampling point is divided into a grid and each grid area is scored based on the soil image to determine the optimal insertion point of the sensor. The soil nutrient content at the target sampling point is calculated based on the sensor data of the optimal insertion point, avoiding interference from factors such as complex soil terrain, vegetation distribution, and stones, ensuring that the collected sensor data can fully reflect the actual soil characteristics of the target sampling point, and improving the accuracy and reliability of the measurement results.

[0040] In step S130 , the scoring of the grid area is achieved by combining the results of image processing and depth information analysis with a multi-index fusion strategy.

[0041] In one possible embodiment, Figure 3 As shown, in step S130, calculating the score of the grid area specifically includes: S310: Calculate multiple scoring indicators of the grid area based on the RGB image and the depth image. The multiple scoring indicators include flatness of soil surface, ), stone area (Stone area, ), vegetation coverage area (Vegetation coverage area, ), soil moisture level (Soil moisture level, ), soil moisture uniformity (Soil moisture uniformity, ), soil texture uniformity (Soil texture uniformity, ).

[0042] The soil surface flatness is used to avoid poor sensor contact caused by the inclination of the sensor assembly. The stone area is used to reduce the risk of damage to the sensor by hard objects. The vegetation coverage area is used to reduce the risk of damage to the sensor by crops. The soil moisture level is used to ensure that the soil moisture of the measurement area is greater than a threshold value. The soil moisture uniformity is used to ensure that the measurement area is representative. The soil texture uniformity is used to reflect the consistency of the soil structure.

[0043] S320: Weighted sum of the plurality of scoring indicators to obtain the score of the grid area.

[0044] The weight of each scoring indicator is dynamically adjusted according to different regions, different soil types, different crops on the soil, different sensor types, etc. The adjustment of the weight can be set according to human experience or recalibrated through experiments.

[0045] The embodiments of the present application analyze the RGB image and the depth image of the soil, obtain various characteristics of the soil, and score the grid area through a plurality of indicators that jointly affect the measurement results, to comprehensively evaluate the measurement characteristics of the grid area and accurately obtain the best insertion point, thereby improving the representativeness and authenticity of the soil during measurement and reducing the wear of the sensor components.

[0046] In one possible embodiment, the soil surface flatness of the grid area is calculated, specifically including: P1: Obtain a point cloud subset of the grid area according to the depth image.

[0047] It should be noted that the image acquisition component is installed directly opposite the ground, and the coordinate system of the image acquisition component is composed of three direction coordinate axes X, Y and Z, wherein the Z axis is perpendicular to the soil surface. The corresponding point cloud data is obtained according to the depth image collected by the image acquisition component, and the point cloud data is divided according to the grid area, and each grid area corresponds to a point cloud subset. The coordinates of each point in the point cloud subset are (x, y, z), wherein x and y represent the horizontal and vertical coordinates of the pixel point in the image, respectively, and z is the depth value of the pixel point.

[0048] P2: Calculate the angle between the normal vector of each point in the point cloud subset and the theoretical vertical direction.

[0049] The soil surface flatness is mainly evaluated according to the normal vector of the point in the point cloud subset. The normal vector is a vector perpendicular to the local surface, and the direction of the normal vector of the point on the flat area tends to be consistent (close to the Z-axis direction), while the direction of the normal vector of the point on the concave-convex area is discrete. Therefore, the soil surface flatness can be evaluated based on the angle between the normal vector of each point in the point cloud subset and the ideal vertical direction (i.e. the Z-axis, the unit vector is [0, 0, 1]).

[0050] wherein, assuming that the normal vector of the point is n i = [n x , n y , n z ], the angle between the normal vector of the point and the Z-axis is : .

[0051] P3: According to the variance of the angle of all points in the point cloud subset.

[0052] The variance of the angle reflects the dispersion degree of the normal vector of the point, thereby reflecting the flatness of the soil surface. The smaller the variance, the flatter the soil surface; the larger the variance, the more uneven the soil surface.

[0053] The variance of the angle of all points in the point cloud subset is : (1) wherein, is the average value of the angle of all points in the point cloud subset, and N is the number of points in the point cloud subset.

[0054] P4: Normalize the variance to obtain the soil surface flatness of the grid area .

[0055] wherein, the normalization is performed by using the following formula: (2) wherein, is the maximum allowed error of the test calibration.

[0056] It should be noted that in actual work, there may be outliers of points leading to a large angle, based on such consideration, the outlier points with an angle greater than a threshold (such as 60°) can be removed to avoid the interference of the edge of hard objects such as stones.

[0057] In addition, if the grid area is on a slope (such as the overall normal vector is inclined), the weight of the soil surface flatness can be reduced in step S320, and the weight of other scoring indicators can be increased according to the actual situation.

[0058] The embodiment of the application obtains a point cloud subset of the grid region through a depth image, and reflects the flatness of the soil surface according to the variance of the included angle between the normal vector of the point and the Z axis, thereby improving the accuracy of the flatness of the soil surface through fine-grained calculation, and evaluating the influence of the sensor assembly inserted into the soil in a tilted manner on the removal of poor effects through the flatness of the soil surface.

[0059] In a possible embodiment, the hard object area or the vegetation coverage area of the grid region is calculated, specifically including: Q1: segmenting the RGB image of the grid region to obtain a soil region (without hard objects such as stones, interference such as vegetation), a hard object region and a vegetation region in the grid region, and using 0 to represent soil pixels, 1 to represent hard object pixels, and 2 to represent vegetation pixels in the image pixels.

[0060] In a possible embodiment, a network model such as U-Net or Mask R-CNN is used for semantic segmentation to realize the segmentation of the RGB image, that is, the RGB image is input into a pre-trained semantic segmentation model to obtain the pixel coordinates of the soil region without interference, the hard object region and the vegetation region in the image.

[0061] Q2: calculating the area of the hard object region or the area of the vegetation region: (3) wherein, the number of pixel points of the hard object in the grid region, the number of all pixel points in the grid region.

[0062] (4) wherein, the number of pixel points of the vegetation in the grid region.

[0063] Q3: normalizing the area of the hard object region or the area of the vegetation region as the hard object area or the vegetation coverage area of the grid region.

[0064] It can be understood that according to actual needs, when the hard object area is very small (for example, it does not affect the measurement of the sensor), the small hard object can be classified as soil, and then the formula is used to calculate S a .

[0065] The embodiment of the application divides the grid area into different types of areas according to the surface cover, and reflects the measurement influence of hard objects and vegetation on the grid area by calculating the area of the hard objects and the vegetation, so as to evaluate the loss risk of the sensor caused by hard objects such as stones and crops.

[0066] In a possible embodiment, the soil moisture degree of the grid area is calculated, specifically including: R1: converting the RGB image into an HSV image.

[0067] It can be understood that the RGB image can be pre-processed before conversion, for example, Gaussian blur denoising, to improve the accuracy of image processing.

[0068] R2: calculating the ratio of the values of the S channel and the V channel of each pixel in the grid area based on the HSV image.

[0069] Soil with high moisture usually appears dark and color-saturated in the RGB image, and the dark color usually appears as low brightness of the V channel in the HSV color space, and the color saturation usually appears as high S channel value in the HSV color space, so the S / V ratio is used to measure the soil moisture degree. For each pixel, the calculation formula is: (5) wherein S is the value of the S channel after the RGB image is converted into the HSV color space, and V is the value of the V channel after the RGB image is converted into the HSV color space.

[0070] R3: calculating the average value of the ratio of all pixels in the grid area, and taking the normalized result of the average value as the soil moisture degree of the grid area .

[0071] The larger the value is, the higher the soil moisture is; The smaller the value is, the drier the soil is.

[0072] The embodiment of the application accurately reflects the soil moisture degree by converting the RGB image into the HSV color space and capturing the S channel and V channel values related to the soil moisture.

[0073] In a possible embodiment, the soil moisture uniformity of the grid area is calculated, specifically including: S1: taking the grid area as a first grid area, calculating the similarity (for example, cosine similarity) between the soil moisture degree of the first grid area and the soil moisture degree of a second grid area around the first grid area.

[0074] S2: normalizing the similarity of the first grid area as the soil moisture uniformity of the first grid area .

[0075] The higher the similarity of the grid region, the better the soil structure consistency of the first grid region, the better the soil moisture uniformity, and the more representative the grid region in the ROI region.

[0076] The embodiment of the application evaluates the soil structure consistency by the similarity between the soil moisture degree of the grid region and the soil moisture degree of the surrounding grid region, as an index for selecting a representative grid region in the ROI region.

[0077] In a possible embodiment, the soil texture uniformity of the grid region is calculated, specifically including: U1: calculating the gray-level co-occurrence matrix (GLCM) of the soil surface layer of the grid region according to the RGB image.

[0078] U2: normalizing the texture features of the gray-level co-occurrence matrix as the soil texture uniformity of the grid region .

[0079] The texture features of the gray-level co-occurrence matrix are: (6) wherein, , and respectively represent the probability of the occurrence of the gray level, the gray level and the gray level and

[0080] The texture features of the gray-level co-occurrence matrix reflect the uniformity of the gray distribution. The higher the texture features, the more consistent the texture, indicating that the soil structure consistency is better.

[0081] In combination with the soil texture uniformity , the region with a smaller contrast value and better soil structure consistency can be preferentially selected, so as to avoid the deviation of the sensor measurement data caused by the too large difference in soil structure, thereby ensuring that the sensor can select a more representative grid region.

[0082] The embodiment of the application reflects the soil structure consistency from another angle according to the texture features of the gray-level co-occurrence matrix, and comprehensively reflects the soil structure consistency with the soil moisture uniformity, thereby improving the accuracy of the evaluation of the soil structure consistency.

[0083] In a possible embodiment, the sensor data includes soil moisture, soil temperature, soil conductivity, soil PH value, and the potential of the soil oxidation-reduction reaction system.

[0084] On the basis of the above, when obtaining the soil nutrient content of the target sampling point according to the sensing data, the sensing data is normalized and input into the neural network model to obtain the soil nutrient content output by the neural network model (for example, a convolutional receptive field (RF) neural network model), and the soil nutrient content includes soil alkali nitrogen, available phosphorus, available potassium, and the like.

[0085] The neural network model is obtained by training a data set based on sensing data (soil moisture, soil temperature, soil conductivity, soil pH, and potential of a soil oxidation-reduction reaction system) and soil nutrient content data (soil alkali nitrogen, available phosphorus, available potassium, and the like).

[0086] During training, the sensing data is normalized, and the Z-score standardization method is specifically used to normalize the sensing data, so that the data presents a normal distribution with a mean of 0 and a standard deviation of 1. After normalization, a large amount of data is concentrated around the mean, so that the machine learning algorithm can stably process the outliers in the data, accelerate the convergence speed, and improve the training efficiency.

[0087] The Z-score standardization formula is (7) wherein, is the mean of the same type of sensing data, is the standard deviation of the same type of sensing data.

[0088] In actual work, the obtained sensing data is input into the neural network model to calculate the soil nutrient content such as soil alkali nitrogen, available phosphorus, and available potassium, and the latitude and longitude information and the soil nutrient content information of the target sampling point are transmitted to the storage unit for storage.

[0089] The embodiment of the application constructs a neural network model, calculates the soil nutrient content according to the sensing data, and improves the accuracy of the soil nutrient content measurement.

[0090] Based on the above, the application further provides a soil nutrient content measurement device. The soil nutrient content measurement device and the soil nutrient content measurement method described above can be mutually corresponding and referred to.

[0091] As an example, as shown in Figure 4 The soil nutrient content measurement device includes a division module 410, an image receiving module 420, a score calculation module 430, an optimal insertion point determination module 440, a sensing data receiving module 450, and a nutrient content obtaining module 460.

[0092] The dividing module 410 is configured to determine a region of interest of the soil at the target sampling point, and divide the region of interest into a plurality of grid regions.

[0093] The image receiving module 420 is configured to receive an RGB image and a depth image of the soil at the target sampling point.

[0094] The score calculating module 430 is configured to calculate a score of each grid region according to the RGB image and the depth image.

[0095] The optimal insertion point determining module 440 is configured to determine the grid region with the highest score as the optimal insertion point of the sensor.

[0096] The sensor data receiving module 450 is configured to receive sensor data obtained at the optimal insertion point after outputting the coordinates of the optimal insertion point.

[0097] The nutrient content obtaining module 460 is configured to obtain the soil nutrient content of the target sampling point according to the sensor data.

[0098] The embodiments of the present application divide the soil at the target sampling point into grid regions, and score each grid region based on the soil image, to determine the optimal insertion point of the sensor, and obtain the soil nutrient content of the target sampling point based on the sensor data of the optimal insertion point, so as to avoid the interference of factors such as complex terrain, vegetation distribution, and stones of the soil, ensure that the collected sensor data can fully reflect the real soil characteristics of the target sampling point, and improve the accuracy and reliability of the measurement result.

[0099] In a possible embodiment, the score calculating module 430 includes an index calculating module 4301 and a weighted sum module 4302.

[0100] The index calculating module 4301 is configured to calculate a plurality of score indexes of the grid region according to the RGB image and the depth image, and the plurality of score indexes include soil surface flatness, hard object area, vegetation coverage area, soil humidity uniformity, soil humidity degree, and soil texture uniformity.

[0101] The weighted sum module 4302 is configured to perform weighted sum on the plurality of score indexes to obtain the score of the grid region.

[0102] The embodiments of the present application analyze the RGB image and the depth image of the soil, obtain various characteristics of the soil, score the grid region by combining the plurality of indexes affecting the measurement result, comprehensively evaluate the measurement characteristics of the grid region, accurately obtain the optimal insertion point, improve the representativeness and authenticity of the soil during measurement, and reduce the loss of the sensor component.

[0103] In a possible embodiment, the index calculating module 4301 is specifically configured to calculate the soil surface flatness of the grid region, and specifically includes: obtaining a point cloud subset of the grid region according to the depth image; calculating an angle between a normal vector of each point in the point cloud subset and a theoretical vertical direction; obtaining a variance of the angle of all points in the point cloud subset; normalizing the variance to obtain a soil surface flatness of the grid region.

[0104] The embodiment of the present application obtains a point cloud subset of the grid region according to the depth image, and reflects the flatness of the soil surface according to the variance of the angle between the normal vector of the point and the Z-axis, improves the accuracy of the soil surface flatness through fine-grained calculation, and evaluates the influence of the sensor assembly inserted into the soil in a tilted manner through the soil surface flatness.

[0105] In a possible embodiment, the index calculation module 4301 is specifically configured to calculate a hard object area or a vegetation coverage area of the grid region, and specifically includes: segmenting an RGB image of the grid region to obtain a soil region, a hard object region and a vegetation region in the grid region; calculating an area of the hard object region or an area of the vegetation region; normalizing the area of the hard object region or the area of the vegetation region as the hard object area or the vegetation coverage area of the grid region.

[0106] The embodiment of the present application divides the grid region into different types of regions according to the surface coverings, reflects the influence of the hard object and the vegetation on the measurement of the grid region by calculating the areas of the hard object and the vegetation, and evaluates the loss risk of the sensor caused by the hard object such as stones and crops.

[0107] In a possible embodiment, the index calculation module 4301 is specifically configured to calculate a soil moisture degree of the grid region, and specifically includes: converting the RGB image into an HSV image; calculating a ratio of the values of the S channel and the V channel of each pixel in the grid region based on the HSV image; calculating an average value of the ratio of all pixels in the grid region, and normalizing the average value as the soil moisture degree of the grid region.

[0108] The embodiment of the present application accurately reflects the soil moisture degree by converting the RGB image into the HSV color space and capturing the S channel and the V channel values related to the soil moisture.

[0109] In a possible embodiment, the index calculation module 4301 is specifically configured to calculate a soil moisture uniformity of the grid region, and specifically includes: The similarity between the soil moisture degree of the grid region and the soil moisture degree of the second grid region around the first grid region is calculated as the first grid region is taken as a first grid region; The similarity of the first grid region is normalized as the soil moisture uniformity of the first grid region.

[0110] The embodiment of the present application evaluates the consistency of the soil structure through the similarity between the soil moisture degree of the grid region and the soil moisture degree of the surrounding grid region, as an index for selecting a representative grid region in the ROI region.

[0111] In a possible embodiment, the index calculation module 4301 is specifically configured to calculate the soil texture uniformity of the grid region, and specifically includes: The gray level co-occurrence matrix of the soil surface layer of the grid region is calculated according to the RGB image; The texture features of the gray level co-occurrence matrix are normalized as the soil texture uniformity of the grid region.

[0112] The embodiment of the present application reflects the consistency of the soil structure from another angle according to the texture features of the gray level co-occurrence matrix, and comprehensively reflects the consistency of the soil structure with the soil moisture uniformity, thereby improving the accuracy of the evaluation of the consistency of the soil structure.

[0113] In a possible embodiment, when the soil nutrient content of the target sampling point is obtained according to the sensing data, the sensing data is input into the neural network model after being normalized, and the soil nutrient content output by the neural network model is obtained, including soil alkali nitrogen, available phosphorus, and available potassium.

[0114] The embodiment of the present application constructs a neural network model, and calculates the soil nutrient content according to the sensing data, thereby improving the accuracy of the measurement of the soil nutrient content.

[0115] Based on the above, the present application further provides a soil nutrient content measurement system, which is used to perform the whole process of the soil nutrient content.

[0116] As shown in Figures 5 to 7 The soil nutrient content measurement system includes a vehicle frame 2, and a control terminal 1, an image acquisition component 16, a sensor driving component, and a sensor assembly installed on the vehicle frame 2.

[0117] The vehicle frame 2 comprises a main body 3 and four wheels 10 arranged below the main body 3, which are used to support and move the main body 3. A DC motor 9 is arranged on each wheel 10 to drive the wheel 10 and provide driving force for the movement of the vehicle frame 2. Four first servo motors 8 are arranged on the main body 3, each of which is connected to a support frame of one of the wheels 10 to accurately control the steering angle of the soil nutrient content measurement system. A damping spring 7 is arranged between the four first servo motors 8 and the main body 3 to reduce damage to electrical components caused by vibration of the vehicle body and improve the stability of the soil nutrient content measurement system in field operation. A battery 23 is arranged on the main body 3, and an angle iron is arranged around the battery 23 to block the battery 23 and prevent the battery from overturning during movement of the main body 3. A control terminal 1 is fixed to the main body 3. The control terminal 1 is connected to the image acquisition component 15 and the sensor assembly, and is used to execute the soil nutrient content measurement method described above and control the execution components for driving the movement of the vehicle frame and the sensor driving component.

[0118] As shown in Figure 6 , the image acquisition component 16 is arranged on the bottom surface of the main body 3 (preferably at the geometric center of the bottom surface), and the field of view of the image acquisition component covers the soil surface of the target sampling point. The coordinate system of the image acquisition component consists of three directional coordinate axes X, Y and Z, wherein the Z axis is perpendicular to the soil surface.

[0119] In one possible embodiment, the image acquisition component 16 is a depth real-sensing camera, which can simultaneously acquire RGB images and depth images.

[0120] The sensor driving component drives the sensor assembly to move relative to the soil surface, so that the multiple sensors in the sensor assembly detect the soil of the target sampling point.

[0121] Specifically, the sensor driving component can drive the sensor assembly to realize three degrees of freedom of movement. The sensor driving component comprises a second servo motor 15, a coupling 24, a first lead screw 12, a lead screw sliding table 13, a linear slide rail 17, a third servo motor 19, a coupling 18, a stepping motor 20 and an electric push rod 21.

[0122] The first lead screw 12 and the linear slide rail 17 are arranged in parallel on the bottom surface of the main body 3 and are located on the inner side of the wheel 10. One end of the first lead screw 12 is rotatably connected to the main body through a fixing piece 11, and the other end is fixed to the coupling 24. The coupling 24 is connected to the output end of the second servo motor 15, and the second servo motor 15 drives the first lead screw 12 to rotate through the coupling 24.

[0123] The screw rod sliding table 13 is arranged between the first screw rod 12 and the linear sliding rail 17, and a second screw rod perpendicular to the linear sliding rail 17 and an inner thread matched with the second screw rod are arranged in the screw rod sliding table 13. A screw rod sliding block 14 is fixed to the first end of the screw rod sliding table 13, and an inner thread matched with the first screw rod 12 and a mechanism (such as a bearing) rotationally connected with the second screw rod are arranged in the screw rod sliding block 14. The third servo motor 19 is slidably arranged on the linear sliding rail 17. The output end of the third servo motor 19 is fixedly connected with the second screw rod through the shaft coupling 18. Thus, the operation of the second servo motor 15 can drive the first screw rod 12 to rotate, so as to drive the screw rod sliding table 13 to drive the third servo motor 19 and the shaft coupling 18 to move along the first screw rod 12 and the linear sliding rail 17. The operation of the third servo motor 19 can drive the second screw rod to rotate, so as to drive the screw rod sliding table 13 to move between the first screw rod 12 and the linear sliding rail 17.

[0124] The screw rod sliding table 13 is arranged between the first screw rod 12 and the linear sliding rail 17, and a second screw rod perpendicular to the linear sliding rail 17 and an inner thread matched with the second screw rod are arranged in the screw rod sliding table 13. A screw rod sliding block 14 is fixed to the first end of the screw rod sliding table 13, and an inner thread matched with the first screw rod 12 and a mechanism (such as a bearing) rotationally connected with the second screw rod are arranged in the screw rod sliding block 14. The third servo motor 19 is slidably arranged on the linear sliding rail 17. The output end of the third servo motor 19 is fixedly connected with the second screw rod through the shaft coupling 18. Thus, the operation of the second servo motor 15 can drive the first screw rod 12 to rotate, so as to drive the screw rod sliding table 13 to drive the third servo motor 19 and the shaft coupling 18 to move along the first screw rod 12 and the linear sliding rail 17. The operation of the third servo motor 19 can drive the second screw rod to rotate, so as to drive the screw rod sliding table 13 to move between the first screw rod 12 and the linear sliding rail 17.

[0125] The sensor assembly includes a fixed block 5 fixed to the lower end (i.e. the output end) of the electric push rod 21, and a soil oxidation-reduction potential sensor 4, a soil multi-parameter sensor 6 and a contact sensor 22 fixed to the fixed block. The lower ends of the soil oxidation-reduction potential sensor 4 and the soil multi-parameter sensor 6 are lower than the lower end of the fixed block 5 and are at the same height. The contact sensor 22 is fixed to the bottom surface of the fixed block 5, and the lower end of the contact sensor 22 is higher than the lower ends of the soil oxidation-reduction potential sensor 4 and the soil multi-parameter sensor 6.

[0126] Through the superposition of the three-direction movements, the fixed block 5 can drive the soil oxidation-reduction potential sensor 4, the soil multi-parameter sensor 6 and the contact sensor 22 to move arbitrarily in the space below the main body.

[0127] Based on the above, as shown in Figure 8 When the soil nutrient content measuring system is used to measure the soil nutrient content, the following steps are included: S810: The vehicle frame is moved to directly above the target sampling point according to the latitude and longitude information of the target sampling point.

[0128] In a possible embodiment, a global positioning system (GPS) is arranged on the vehicle frame, the latitude and longitude information of the target sampling point is input into the control terminal, and the control terminal controls the direct current motor 9 and the first servo motor 8 to move the vehicle frame to directly above the target sampling point.

[0129] S820: Above the target sampling point, the control terminal determines a region of interest (ROI) according to the field of view of the image acquisition component and the soil surface image collected by the image acquisition component, and divides the region of interest (ROI) into a plurality of grid regions according to the size of the measurement region of the sensor assembly. Figure 9 An example of grid division is shown.

[0130] S830: The image acquisition component 16 collects the RGB image and depth image of the soil within its field of view and transmits them to the control terminal.

[0131] S840: The control terminal determines one of the grid regions as the optimal insertion point of the sensor assembly according to the RGB image and the depth image, and takes the center point position of the optimal insertion point as the target position.

[0132] Specifically, after obtaining the image pixel coordinates and depth value of the optimal insertion point, in order to obtain the coordinate mapping relationship of the RGB image and the depth image in the original data, the image is registered according to the camera intrinsic parameter matrix, distortion coefficient and other information of the camera calibration parameters, the conversion from pixel coordinates to three-dimensional world coordinates of the grid region corresponding to the optimal insertion point is realized, and the optimal insertion point position information and depth information in the three-dimensional world coordinates are obtained. The center point position of the optimal insertion point in the three-dimensional world coordinates is taken as the target position.

[0133] S850: The control terminal controls the second servo motor 15 and the third servo motor 19 to move the output end position of the electric push rod 21 to the target position.

[0134] S860: The control terminal controls the stepping motor 20 to drive the electric push rod 21 to move downward and drive the sensor assembly to insert into the soil.

[0135] Specifically, the control terminal determines the distance of the downward extension of the electric push rod 21 according to the depth information of the optimal insertion point d d-d 1 + △ d ), wherein d is the distance from the camera to the soil surface, d 1 is the distance from the camera to the electric push rod, and △ d is the depth information of the insertion of the probes at the lower end of the soil oxidation-reduction potential sensor 4 and the soil multi-parameter sensor 6 into the soil. After the probes of the soil oxidation-reduction potential sensor 4 and the soil multi-parameter sensor 6 reach the soil surface, they need to be further inserted by a certain distance to ensure that the sensor probes are covered by the soil, and the insertion depth ensures accurate measurement and reflects the soil parameter information. Generally △d is the length of the probe of the sensor. The electric push rod 21 extends downward by d-d 1 + △ d ​After the soil redox potential sensor 4 and the soil multi-parameter sensor 6 reach the optimal measurement depth, the electric push rod 21 sends a signal to the control terminal 1, indicating that the probes of the soil redox potential sensor 4 and the soil multi-parameter sensor 6 reach the optimal measurement depth, at which time the electric push rod 21 pauses to extend downward. At the same time, in the process of extending downward, the contact sensor 22 gradually approaches and inserts into the soil, and the contact sensor 22 is subjected to the resistance of the soil, and when the force reaches 1 N, the contact sensor 22 sends a signal to the control terminal. When the control terminal receives the signals of the electric push rod and the contact sensor, it indicates that the probes of the sensors reach the optimal measurement depth, and the electric push rod pauses to extend downward, at which time the soil redox potential sensor 4 and the soil multi-parameter sensor 6 can normally carry out measurement work.

[0136] The depth information obtained by the depth real-sense camera and the force information of the contact sensor are combined with each other to provide double position information for the extension movement of the electric push rod, improve the safety of the sensor, and avoid the risk that the sensor cannot be inserted due to hard objects such as stones in the surface layer of the soil and the risk of damage to the sensor.

[0137] S870: After the sensor assembly transmits the obtained sensing data to the control terminal, the control terminal receives the sensing data, and obtains the soil nutrient content of the target sampling point according to the sensing data.

[0138] The soil nutrient content measurement system provided in the application integrates the image acquisition component, the sensor assembly and the control terminal on the vehicle frame, realizes the field movement of the device, the accurate positioning and measurement of the insertion point and the rapid data processing by using the same device, completes the whole process of soil nutrient content measurement, realizes the accurate control of the sensor and the rapid, high-precision and low-cost measurement of the soil nutrient content, and improves the efficiency of soil nutrient content measurement.

[0139] Figure 10 is a structural schematic diagram of an electronic device provided by the application, as Figure 10As shown, the electronic device can include a processor 1010, a communications interface 1020, a memory 1030, and a communications bus 1040, wherein the processor 1010, the communications interface 1020, and the memory 1030 complete mutual communication through the communications bus 1040. The processor 1010 can invoke a logical instruction in the memory 1030 to execute a soil nutrient content measurement method, which includes determining a region of interest of soil at a target sampling point and dividing the region of interest into a plurality of grid regions; receiving an RGB image and a depth image of the soil at the target sampling point; calculating a score of each grid region according to the RGB image and the depth image; taking the grid region with the highest score as the best insertion point of a sensor; after outputting the coordinates of the best insertion point, receiving sensor data obtained at the best insertion point; and obtaining the soil nutrient content of the target sampling point according to the sensor data.

[0140] In addition, the logical instruction in the memory 1030 described above can be implemented in the form of a software functional unit and sold or used as an independent product, and can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0141] On the other hand, the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions, when the program instructions are executed by a computer, the computer can execute the soil nutrient content measurement method provided by the above-mentioned embodiments, which includes: determining a region of interest of soil at a target sampling point and dividing the region of interest into a plurality of grid regions; receiving an RGB image and a depth image of the soil at the target sampling point; calculating a score of each grid region according to the RGB image and the depth image; taking the grid region with the highest score as the best insertion point of a sensor; after outputting the coordinates of the best insertion point, receiving sensor data obtained at the best insertion point; and obtaining the soil nutrient content of the target sampling point according to the sensor data.

[0142] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the soil nutrient content measurement method provided by any of the above embodiments, and the method comprises: determining a region of interest of soil at a target sampling point, and dividing the region of interest into a plurality of grid regions; receiving an RGB image and a depth image of the soil at the target sampling point; calculating a score of each grid region according to the RGB image and the depth image; taking the grid region with the highest score as the best insertion point of the sensor; after outputting the coordinates of the best insertion point, receiving sensor data obtained at the best insertion point; and obtaining the soil nutrient content of the target sampling point according to the sensor data.

[0143] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e., they can be located in one place, or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0144] From the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in terms of the contribution to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer readable storage medium, such as a read-only memory (ROM) / random access memory (RAM), a magnetic disk, an optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0145] From the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in terms of the contribution to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer readable storage medium, such as a read-only memory (ROM) / random access memory (RAM), a magnetic disk, an optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0146] Finally, it should be noted that the above examples are only used to illustrate the technical solutions of the present application, and are not intended to limit the same; although the present application has been described in detail with reference to the foregoing examples, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for measuring soil nutrient content, characterized in that: include: Determine an area of ​​interest of the soil at a target sampling point, and divide the area of ​​interest into a plurality of grid areas; Receive RGB images and depth images of the soil at the target sampling point; Calculating a score for each grid area based on the RGB image and the depth image; The grid area with the highest score is used as the optimal insertion point for the sensor; After outputting the coordinates of the optimal insertion point, receiving sensor data obtained at the optimal insertion point; The soil nutrient content of the target sampling point is obtained based on the sensor data.

2. The soil nutrient content measurement method according to claim 1, characterized in that: Calculating the score of the grid area specifically includes: Calculating a plurality of scoring indicators for the grid area based on the RGB image and the depth image, the plurality of scoring indicators including soil surface flatness, hard object area, vegetation cover area, soil moisture uniformity, soil moisture degree, and soil texture uniformity; The multiple scoring indicators are weighted and summed to obtain a score for the grid area.

3. The soil nutrient content measuring method according to claim 2, characterized in that: Calculating the soil surface flatness of the grid area specifically includes: Obtaining a point cloud subset of the grid area based on the depth image; Calculate the angle between the normal vector of each point in the point cloud subset and the theoretical vertical direction; Based on the variance of the angles of all points in the point cloud subset; The variance is normalized to obtain the soil surface flatness of the grid area.

4. The method for measuring soil nutrient content according to claim 2, wherein: Calculating the hard object area or the vegetation cover area in the grid area specifically includes: Segmenting the RGB image of the grid area to obtain a soil area, a hard object area, and a vegetation area within the grid area; calculating the area of ​​the hard object area or the area of ​​the vegetation area; The area of ​​the hard object region or the area of ​​the vegetation region is normalized to serve as the hard object area or the vegetation coverage area of ​​the grid region.

5. The method for measuring soil nutrient content according to claim 2, wherein: Calculating the soil moisture level in the grid area includes: Convert the RGB image into an HSV image; Calculate the ratio of the S channel and the V channel values ​​of each pixel in the grid area based on the HSV image; The average value of the ratios of all pixels in the grid area is calculated, and the result of normalizing the average value is used as the soil moisture level of the grid area.

6. The method for measuring soil nutrient content according to claim 2, wherein: Calculating the soil moisture uniformity of the grid area specifically includes: Taking the grid area as a first grid area, calculating the similarity between the soil moisture level of the first grid area and the soil moisture level of a second grid area surrounding the first grid area; The similarity of the first grid area is normalized to obtain the soil moisture uniformity of the first grid area.

7. The method for measuring soil nutrient content according to claim 1, wherein: The sensor data includes soil moisture, soil temperature, soil conductivity, soil pH value, and the potential of the soil redox reaction system; Normalizing the sensor data and inputting it into a neural network model to obtain soil nutrient content output by the neural network model, wherein the soil nutrient content includes soil alkaline-hydrolyzable nitrogen, available phosphorus, and available potassium; The neural network model is obtained by training a data set based on sensor data and soil nutrient content data.

8. A soil nutrient content measuring device, characterized in that: It includes a division module, an image receiving module, a scoring calculation module, an optimal insertion point determination module, a sensor data receiving module and a nutrient content acquisition module; The division module is used to determine the region of interest of the soil at the target sampling point and divide the region of interest into a plurality of grid areas; The image receiving module is used to receive the RGB image and depth image of the soil at the target sampling point; The score calculation module is used to calculate the score of each grid area based on the RGB image and the depth image; The optimal insertion point determination module is used to select the grid area with the highest score as the optimal insertion point of the sensor; The sensor data receiving module is used to receive sensor data obtained at the optimal insertion point after outputting the coordinates of the optimal insertion point; The nutrient content acquisition module is used to obtain the soil nutrient content of the target sampling point based on the sensor data.

9. The soil nutrient content measuring device according to claim 8, characterized in that: The scoring calculation module includes an index calculation module and a weighted summation module; The index calculation module is used to calculate multiple scoring indicators of the grid area based on the RGB image and the depth image, and the multiple scoring indicators include soil surface flatness, hard object area, vegetation cover area, soil moisture uniformity, soil moisture degree and soil texture uniformity; The weighted summation module is used to perform weighted summation on the multiple scoring indicators to obtain the score of the grid area.

10. A soil nutrient content measurement system, characterized in that: It includes a vehicle frame and a control terminal, an image acquisition component, a sensor driving component and a sensor assembly installed on the vehicle frame; The image acquisition component is arranged on the bottom surface of the main body of the frame, and the field of view of the image acquisition component covers the soil surface of the target sampling point; The sensor driving component drives the sensor assembly to move relative to the soil surface, so that the multiple sensors in the sensor assembly detect the soil at the target sampling point; The control terminal is signal-connected to the image acquisition component and the sensor assembly respectively, and the control terminal is used to execute the soil nutrient content measurement method according to any one of claims 1 to 7.

Images