Seed metering quality detection method, device, equipment, medium and product

Abstract

The invention relates to a seed metering quality detection method, device and equipment, a medium and a product. The method comprises the following steps: acquiring a seed image sequence, wherein seed images in the seed image sequence are acquired in a seed metering process; position information sets corresponding to multiple seeds are determined from the seed image sequence, the multiple seeds are different seeds, and the position information set corresponding to each seed comprises position information of the seeds in at least three seed images; for each seed, determining landing position information and landing time information of the seed according to the position information set corresponding to the seed and the acquisition time information corresponding to the seed image from which the position information in the position information set comes; and according to the landing position information and the landing time information of each seed, determining the seeding spacing between the adjacent seeds, and determining a seeding quality detection result according to the seeding spacing between the adjacent seeds. By adopting the method, the seeding quality detection precision can be improved.

Description

Technical Field

[0001] This application relates to the field of seed metering detection technology, and in particular to a seed metering quality detection method, apparatus, equipment, medium and product. Background Technology

[0002] With the development of intelligent agricultural machinery technology, computer vision inspection technology, due to its characteristics of non-contact measurement, strong real-time performance, and comprehensive information collection, has been gradually applied to the field of agricultural equipment performance testing. Among them, the seed quality testing of air suction seed metering devices, as a key link to ensure accurate sowing, has put forward high requirements for the accuracy of the testing technology.

[0003] In traditional technology, the detection of seed metering quality of air-suction seed meterers mainly adopts a conveyor belt detection device. The core of this device is to guide the seeds discharged by the seed meterer onto a conveyor belt that has been pre-coated with hydraulic oil. The viscosity of the hydraulic oil is used to fix the spacing of the seeds during the metering process. Then, a camera scans the seed images on the surface of the conveyor belt to extract information such as seed position and spacing to complete the quality assessment.

[0004] However, traditional detection methods cannot accurately detect the seed metering quality of the seed metering device while reducing seed loss. Summary of the Invention

[0005] Therefore, it is necessary to provide a seeding quality detection method, apparatus, equipment, medium, and product that can improve the accuracy of seeding quality detection in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a method for detecting seeding quality, including:

[0007] Obtain the seed image sequence, where the seed images are collected during the seeding process;

[0008] Determine the location information sets corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the location information set corresponding to each seed includes the location information of the seed in at least three seed images.

[0009] For each seed, the landing location and landing time information of the seed are determined based on the location information set corresponding to the seed and the acquisition time information of the seed image from which the location information in the location information set comes.

[0010] Based on the landing location and landing time information of each seed, the spacing between adjacent seeds is determined, and the seeding quality test results are determined based on the spacing between adjacent seeds.

[0011] In one embodiment, determining a set of location information corresponding to multiple seeds from a seed image sequence includes: for each seed image, performing seed recognition on the seed image to obtain a seed recognition result corresponding to the seed image, the seed recognition result containing location information corresponding to the seeds in the seed image; for adjacent first and second seed images in the seed image sequence, determining multiple seed combinations, each seed combination containing one seed from the first seed image and one seed from the second seed image; for each seed combination, determining the pixel distance between two seeds in the seed combination based on the location information; if the pixel distance is less than a distance threshold, determining the two seeds in the seed combination as the same seed; and obtaining a set of location information corresponding to the same seed based on the location information corresponding to each seed.

[0012] In one embodiment, seed recognition is performed on a seed image to obtain a seed recognition result corresponding to the seed image, including: converting the seed image into a binary image and preprocessing the binary image; identifying connected regions in the preprocessed binary image and calculating the area of ​​each connected region; determining connected regions with an area greater than an area threshold as seeds and obtaining the location information of the connected regions as the location information of the seeds; and obtaining the seed recognition result corresponding to the seed image based on the location information of the seeds.

[0013] In one embodiment, the seed's landing location information and landing time information are determined based on the seed's location information set and the acquisition time information corresponding to the seed image from which the location information in the location information set originates. This includes: fitting the seed's motion trajectory based on the seed's location information set and the acquisition time information corresponding to the seed image from which the location information in the location information set originates, to obtain the seed's initial location information, first axial velocity, second axial initial velocity, and second axial acceleration; acquiring the ground image location, and calculating the seed's landing location information and landing time information based on the ground image location, initial location information, first axial velocity, second axial initial velocity, and second axial acceleration.

[0014] In one embodiment, the location information includes lateral position; determining the planting spacing between adjacent seeds based on the landing position information and landing time information of each seed includes: sorting the lateral positions of each seed according to the landing time information of each seed, and using the difference in lateral positions between adjacent seeds as the initial spacing between adjacent seeds; determining the time difference between adjacent seeds based on the landing time information of each seed, and determining the compensation spacing between adjacent seeds based on the vehicle speed and the time difference; and determining the planting spacing between adjacent seeds based on the initial spacing and the compensation spacing.

[0015] In one embodiment, determining the seeding quality test result based on the seeding spacing between adjacent seeds includes: counting the total number of spacings associated with the seeding spacing; filtering the seeding spacing between adjacent seeds according to the spacing range, and counting the number of target spacings that meet the spacing range; calculating the seeding qualification rate based on the total number of spacings and the number of target spacings, and determining the seeding quality test result based on the seeding qualification rate.

[0016] Secondly, this application also provides a seeding quality detection device, comprising:

[0017] The acquisition module is used to acquire the seed image sequence, in which the seed images are collected during the seeding process;

[0018] The matching module is used to determine the location information sets corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the location information set corresponding to each seed includes the location information of the seed in at least three seed images.

[0019] The calculation module is used to determine the landing location and landing time of each seed based on the location information set corresponding to the seed and the acquisition time information of the seed image from which the location information in the location information set comes.

[0020] The detection module is used to determine the spacing between adjacent seeds based on the landing location and landing time information of each seed, and to determine the seeding quality detection result based on the spacing between adjacent seeds.

[0021] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0022] Obtain the seed image sequence, where the seed images are collected during the seeding process;

[0023] Determine the location information sets corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the location information set corresponding to each seed includes the location information of the seed in at least three seed images.

[0024] For each seed, the landing location and landing time information of the seed are determined based on the location information set corresponding to the seed and the acquisition time information of the seed image from which the location information in the location information set comes.

[0025] Based on the landing location and landing time information of each seed, the spacing between adjacent seeds is determined, and the seeding quality test results are determined based on the spacing between adjacent seeds.

[0026] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0027] Obtain the seed image sequence, where the seed images are collected during the seeding process;

[0028] Determine the location information sets corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the location information set corresponding to each seed includes the location information of the seed in at least three seed images.

[0029] For each seed, the landing location and landing time information of the seed are determined based on the location information set corresponding to the seed and the acquisition time information of the seed image from which the location information in the location information set comes.

[0030] Based on the landing location and landing time information of each seed, the spacing between adjacent seeds is determined, and the seeding quality test results are determined based on the spacing between adjacent seeds.

[0031] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0032] Obtain the seed image sequence, where the seed images are collected during the seeding process;

[0033] Determine the location information sets corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the location information set corresponding to each seed includes the location information of the seed in at least three seed images.

[0034] For each seed, the landing location and landing time information of the seed are determined based on the location information set corresponding to the seed and the acquisition time information of the seed image from which the location information in the location information set comes.

[0035] Based on the landing location and landing time information of each seed, the spacing between adjacent seeds is determined, and the seeding quality test results are determined based on the spacing between adjacent seeds.

[0036] The aforementioned seeding quality detection method, apparatus, equipment, medium, and product acquire seed movement information by collecting seed image sequences during the seeding process, thus reducing seed loss. Simultaneously, by analyzing the position information and acquisition time of each seed in multiple image frames, its landing position and time are calculated, thereby accurately determining the seeding spacing between adjacent seeds and improving the detection accuracy of seeding quality. Ultimately, this achieves accurate detection of seeding quality while reducing seed loss. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is an application environment diagram of the seeding quality detection method in one embodiment;

[0039] Figure 2 This is a flowchart illustrating a seeding quality detection method in one embodiment;

[0040] Figure 3 This is a schematic diagram of the structure of the seed metering detection device in one embodiment;

[0041] Figure 4 This is a flowchart illustrating the seeding process in one embodiment;

[0042] Figure 5 This is a flowchart illustrating the seeding quality detection method in another embodiment;

[0043] Figure 6 This is a structural block diagram of a seeding quality detection device in one embodiment;

[0044] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0046] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0047] The seeding quality detection method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, the seed metering test equipment communicates with a computer via a network. The seed metering test equipment sends the acquired seed image sequence to the computer. The computer acquires the seed image sequence, where the seed images were collected during the seed metering process. From the seed image sequence, it determines multiple sets of location information corresponding to different seeds. Each seed's location information set includes its position information in at least three seed images. For each seed, based on its corresponding location information set and the acquisition time information of the seed image from which the location information in the location information set originates, it determines the seed's landing location information and landing time information. Based on the landing location information and landing time information of each seed, it determines the seeding spacing between adjacent seeds. Based on the seeding spacing between adjacent seeds, it determines the seed metering quality detection result. The computer sends the seed metering quality detection result to the seed metering test equipment so that the seed metering test equipment can display the seed metering quality detection result. The seed metering device can be used to test the seed metering quality. It includes a camera sensor to capture images of the seed falling process, and the camera sensor can be positioned within the device directly opposite the seed falling plane. The computer device can be a terminal or a server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, and tablets. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. It is worth noting that the computer device can be located inside the seed metering device or independently of it; this application does not impose any limitations on this.

[0048] In one exemplary embodiment, such as Figure 2 As shown, a method for detecting seeding quality is provided, which is applied to... Figure 1 Taking the terminal in the example, the explanation includes:

[0049] Step 201: Obtain the seed image sequence. The seed images in the seed image sequence are collected during the seeding process.

[0050] The seed image sequence can be a collection of multiple seed images arranged in chronological order of acquisition time; the seed image can be an image of a seed falling in the air during the seeding process.

[0051] Optionally, after the seed metering device starts seeding, images of the seed falling process can be collected in real time by the seed metering device detection equipment. Each frame of the collected image contains at least one seed in the air, and all the collected images are stored sequentially according to the generation timestamp to form the seed image sequence.

[0052] It is worth noting that seed metering detection equipment, such as Figure 3 As shown, the seeding process is as follows: Figure 4 As shown, the seed metering detection device includes a camera sensor. The camera sensor can be calibrated before the seed metering device starts planting. Specifically, a calibration board with a black and white checkerboard pattern is placed at a distance D1 from the camera sensor plane. Each square on the calibration board has a side length l. The vertex positions of the black and white squares are found by observing the strong gradient changes of the black and white squares in the image. The gradient change matrix M describing the gradient change at pixel (x, y) in the image is:

[0053]

[0054] Among them, I x and I y Both are image gradients. For corner points, both eigenvalues ​​of M are large, indicating the simultaneous existence of horizontal and vertical gradients. After finding the corner points, the distance between two corner points on the same edge of the same cell can be calculated as n pixels, yielding the length d of the seed drop plane corresponding to each pixel on the camera sensor. Assuming the distance from the seed metering plane to the camera sensor plane is D2, the actual length d corresponding to one pixel in the image during detection is:

[0055]

[0056] For example, to determine the position of the virtual ground in the image, a white background can be used. The row number of the lowest point of the seed metering device is R1. If the position at a depth of h (e.g., 2cm) below the seed metering device is defined as the ground, then the corresponding row number of the virtual ground in the image is:

[0057]

[0058] Step 202: Determine the location information sets corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the location information set corresponding to each seed includes the location information of the seed in at least three seed images.

[0059] The location information set can be a data set established for each seed, containing its coordinate information in a single frame or multiple frames of images.

[0060] Optionally, each frame of the seed image in the seed image sequence can be preprocessed to identify all potential seed regions and calculate their positions through connected component analysis; then, the position points belonging to the same seed in different frames can be matched to construct a set of position information corresponding to each seed.

[0061] Step 203: For each seed, determine the seed's landing location information and landing time information based on the seed's corresponding location information set and the acquisition time information of the seed image from which the location information in the location information set originates.

[0062] Among them, the acquisition time information can be the timestamp when the seed image corresponding to the location information was acquired; the landing location information can be the pixel coordinates of the seed at the moment of contact with the ground; and the landing time information can be the time point at which the seed contacts the ground.

[0063] Alternatively, for each seed, its trajectory already exists. That is, the set of location information corresponding to the seed, and the time t when the seed image corresponding to the location information was acquired. i The seed's motion in the air should satisfy uniform acceleration, with the acceleration direction along the y-axis. Therefore, a quadratic regression method can be used to fit this parabola to determine the pixel position coordinates and time point at the instant the seed contacts the ground.

[0064] In one embodiment, determining the seed's landing location and landing time information based on the seed's location information set and the acquisition time information corresponding to the seed image from which the location information in the location information set originates includes: fitting the seed's motion trajectory based on the seed's location information set and the acquisition time information corresponding to the seed image from which the location information in the location information set originates, to obtain the seed's initial location information, first axial velocity, second axial initial velocity, and second axial acceleration; acquiring the ground image location, and calculating the seed's landing location and landing time information based on the ground image location, initial location information, first axial velocity, second axial initial velocity, and second axial acceleration.

[0065] The initial position information can be the horizontal and vertical pixel coordinates of the seed corresponding to the zero point of time in the fitted parabola; the first axial velocity can be the horizontal uniform motion velocity of the seed; the second axial initial velocity can be the initial motion velocity of the seed in the vertical falling direction; the second axial acceleration can be the uniform acceleration of the seed in the vertical falling direction; and the ground image position can be the virtual reference pixel coordinates of the image corresponding to the ground, determined based on the camera calibration results and the height of the seed metering device above the ground.

[0066] Optionally, for a single seed, all horizontal and vertical pixel coordinates and the acquisition time of the corresponding seed image are extracted from its location information set. Based on a model where the first axis is uniform linear motion and the second axis is uniformly accelerated linear motion, these discrete spatiotemporal points are fitted using the least squares method to obtain the initial position information, the velocity of the first axis, the initial velocity of the second axis, and the acceleration of the second axis. Subsequently, a predetermined ground image position is obtained and substituted into the uniformly accelerated motion equation of the second axis to solve for candidate solutions for the landing time. Extraneous roots outside the detection time range are filtered out to obtain the landing time information. Finally, the landing time information is substituted into the uniform motion equation of the first axis to calculate the horizontal pixel coordinates, thus obtaining the landing position information.

[0067] It is worth noting that the falling trajectory of each seed satisfies the following:

[0068] x i =x0+v x t i

[0069] y i =y0+v y0 t i +1 / 2at i 2

[0070] Where, x i Let be the lateral position of the i-th seed; the acceleration is only on the y-axis and is constant. The seed's initial velocity along the x-axis and y-axis v can be solved using a least-squares closed-form solution (linear regression). y and acceleration a on the y-axis y The specific solution method includes: solving for lateral motion, that is, first finding the intermediate variables, the expression of which is:

[0071]

[0072] Then the velocity of the seed on the x-axis (v) can be calculated. x ) and the lateral position (x0) at t=0, v x The expressions for x0 of the two are as follows:

[0073]

[0074] For solving the longitudinal motion problem, the unknowns can be expressed as a parameter vector, where the unknowns include the initial velocity (v) of the seed in the vertical downward direction. y0 The initial position (y0) of the seed in the vertical falling direction and the uniform acceleration (a) of the seed in the vertical falling direction; the expression for the parameter vector can be:

[0075]

[0076] For example, it is also possible to determine the seed trajectory based on (x1, y1)...(x i ,y i ...(x) n ,y n The design matrix for the time ti in which these frames occur:

[0077]

[0078] The least squares solution to the above content is:

[0079]

[0080] At this point, the velocity v along the x-axis was calculated. x The initial velocity v along the y-axis y acceleration a on the y-axis y .

[0081] Step 204: Determine the seed spacing between adjacent seeds based on the landing location and landing time information of each seed, and determine the seeding quality test result based on the seed spacing between adjacent seeds.

[0082] Among them, adjacent seeds can be different seeds that land consecutively after being sorted according to their landing time; the seed spacing can be the straight-line distance formed by adjacent seeds at the moment of landing; and the seeding quality detection result can be the detection result of whether the seeding quality of the seed metering device meets the requirements.

[0083] Optionally, all seeds are first sorted in ascending order according to their landing time information to determine adjacent seed pairs that land consecutively; for each pair of adjacent seeds, the seeding spacing between their landing positions is calculated, and then the seeding quality test result is determined based on the seeding spacing between each pair of adjacent seeds.

[0084] It is worth noting that the seed metering quality of the seed metering device mainly depends on the uniformity of the seeds discharged. This uniformity is reflected in the spacing between the seeds. Therefore, by analyzing the seed spacing between each pair of adjacent seeds, it can be determined whether the seed metering device is uniform, thus enabling the detection of the seed metering quality.

[0085] In one embodiment, the location information includes lateral position; determining the planting spacing between adjacent seeds based on the landing position information and landing time information of each seed includes: sorting the lateral positions of each seed according to the landing time information of each seed, and using the difference in lateral positions between adjacent seeds as the initial spacing between adjacent seeds; determining the time difference between adjacent seeds based on the landing time information of each seed, and determining the compensation spacing between adjacent seeds based on the vehicle speed and the time difference; and determining the planting spacing between adjacent seeds based on the initial spacing and the compensation spacing.

[0086] The lateral position can be the horizontal coordinate of the seed at the moment it lands; the initial spacing can be the absolute value of the difference in lateral position between adjacent seeds at the moment they land; and the compensation spacing can be the distance the vehicle travels within the time difference between the landing of adjacent seeds.

[0087] Optionally, first extract the landing time information and corresponding lateral position of all seeds, sort the seeds from early to late according to the landing time to determine consecutive adjacent seed pairs; calculate the difference in lateral position of each pair of adjacent seeds and take the absolute value to obtain the initial spacing of the pair of seeds; simultaneously calculate the difference in landing time information of the pair of seeds, and combine it with the pre-set uniform forward speed of the vehicle to calculate the compensation spacing; finally, according to the actual length corresponding to a pixel on the image determined by camera calibration, convert the initial spacing into the actual spacing, and then add it to the compensation spacing to obtain the actual planting spacing of the pair of adjacent seeds.

[0088] It is worth noting that after solving for the velocity vx along the x-axis, the initial velocity vy along the y-axis, and the acceleration ay along the y-axis, the time and position of the seed reaching the ground can be calculated. The method for calculating the time and position of the seed reaching the ground is as follows:

[0089]

[0090] Where x0 and y0 are the coordinates of the seed at t=0; R is the row number of the virtual ground in the image; by solving this quadratic equation, the time t when the i-th seed falls to the ground can be obtained. i , and its x on the ground i Position, calculate t i The specific method is as follows:

[0091]

[0092] Among them, removing those not in (t0,t) N After identifying extraneous roots within the range, the calculated t can be used as a basis. i Calculate the x of the seed falling on the ground i Location: Finally, the distribution in the seed space can be determined:

[0093]

[0094] Among them, v m The vehicle's forward speed is set in the program; Δx i Δt represents the difference in lateral position between adjacent seeds. i The time difference between adjacent seeds; Δs i d is the spacing between adjacent seeds; d is the actual length corresponding to one pixel on the image.

[0095] In one embodiment, determining the seeding quality test result based on the seeding spacing between adjacent seeds includes: counting the total number of spacings associated with the seeding spacing; filtering the seeding spacing between adjacent seeds according to the spacing range, and counting the number of target spacings that meet the spacing range; calculating the seeding qualification rate based on the total number of spacings and the number of target spacings, and determining the seeding quality test result based on the seeding qualification rate.

[0096] Among them, the total number of spacings can be the total number of planting spacings corresponding to all consecutive adjacent seed pairs; the spacing range can be a pre-set interval threshold for determining whether the planting spacing is qualified; the target spacing number can be the number of qualified spacings within the spacing range; and the planting qualification rate can be the ratio of the target spacing number to the total number of spacings.

[0097] Optionally, first summarize the planting spacing of all adjacent seed pairs determined by the landing time, and count the total number of spacings; then call the preset spacing range (e.g., the qualified spacing range for corn planting is 18-22cm), compare the matching of each planting spacing with the range, and count the number of target spacings that meet the planting spacing range; determine the planting qualification rate by the ratio of the number of target spacings to the total number of spacings, and finally combine the preset grade standards (e.g., a qualification rate of ≥95% is "high quality" and 85%-94% is "good") to output the planting quality test results including the qualification rate and quality grade.

[0098] In the aforementioned seeding quality detection method, seed movement information is obtained by acquiring seed image sequences during the seeding process, reducing seed loss. Simultaneously, by analyzing the position information and acquisition time of each seed in multiple image frames, its landing position and time are calculated, thereby accurately determining the seeding spacing between adjacent seeds and improving the detection accuracy of seeding quality. Ultimately, this method achieves accurate seeding quality detection while reducing seed loss.

[0099] In an exemplary embodiment, determining a set of location information corresponding to multiple seeds from a seed image sequence includes: for each seed image, performing seed recognition on the seed image to obtain a seed recognition result corresponding to the seed image, the seed recognition result containing location information corresponding to the seeds in the seed image; for adjacent first and second seed images in the seed image sequence, determining multiple seed combinations, each seed combination containing one seed from the first seed image and one seed from the second seed image; for each seed combination, determining the pixel distance between two seeds in the seed combination based on the location information; if the pixel distance is less than a distance threshold, determining the two seeds in the seed combination as the same seed; and obtaining a set of location information corresponding to the same seed based on the location information corresponding to each seed.

[0100] The first seed image and the second seed image can be two adjacent frames in the seed image sequence ordered by acquisition time; the seed combination can be a pairing set composed of a single seed in the first seed image and a single seed in the second seed image; the pixel distance can be the straight-line distance between the position information (pixel coordinates) of two seeds in the same seed combination; the distance threshold can be a preset pixel distance threshold used to determine whether adjacent frame seeds are the same seed.

[0101] Optionally, seed identification is first performed on each seed image in the seed image sequence to obtain the seed identification result and the position information of each seed in each frame image; then, the seed image sequence is traversed in the order of acquisition time, and adjacent first and second seed images are selected to construct all possible seed combinations of each seed in the first seed image and each seed in the second seed image; for each seed combination, the pixel distance corresponding to the position information of the two seeds in the combination is calculated based on the distance formula between two points; the calculated pixel distance is compared with a preset distance threshold, and if the pixel distance is less than the distance threshold, the two seeds in the seed combination are determined to be the same seed; finally, all position information of the same seed identified in each adjacent frame image is summarized and integrated to form the position information set corresponding to the seed, and the construction of all seed position information sets is completed in sequence.

[0102] In one embodiment, obtaining the seed recognition result corresponding to the seed image by performing seed recognition on the seed image includes: converting the seed image into a binary image and preprocessing the binary image; identifying connected regions in the preprocessed binary image and calculating the area of ​​each connected region; determining connected regions with an area greater than an area threshold as seeds and obtaining the location information of the connected regions as the location information of the seeds; and obtaining the seed recognition result corresponding to the seed image based on the location information of the seeds.

[0103] The area threshold can be a pre-set critical value based on the actual size of the seed and the camera calibration results, used to distinguish between the seed and interference items.

[0104] Optionally, the seed image can be converted into a binary image first, and then morphological preprocessing (erosion followed by dilation) can be performed on the binary image to optimize the image quality. Subsequently, all connected regions in the preprocessed binary image are identified, the pixel area of ​​each connected region is calculated and compared with a preset area threshold, and connected regions with an area greater than the area threshold are identified as seeds. At the same time, the position information of the connected region is extracted as the position information of the corresponding seed. Finally, the position information of all seeds in a single frame image is summarized to obtain the seed recognition result corresponding to the seed image.

[0105] It is worth noting that the overall flowchart of the measurement process is as follows: Figure 5As shown, to ensure processing speed, a dual-threaded programming design can be used to decouple data acquisition and data processing operations. In thread 1, a queue data structure is used to store image information. In thread 2, a pre-set threshold is used to convert the black-and-white image into a binary image; pixels larger than the threshold are set to 1, and pixels smaller than the threshold are set to 0. Then, an erosion-dilation algorithm is used to remove noise from the image. Next, all isolated islands with an area greater than a certain threshold are found in the image, typically set to 20-50. Finally, the coordinates (x, y, y) of the lowest point of each isolated island (i.e., connected region) are found. m,i ,y m,i ), where m represents the m-th image and i represents the i-th island. After the set detection time, the camera stops acquiring data and waits for all image data in the thread 2 queue to finish processing. After processing, the position information of all seeds has been recorded. Next, it is necessary to match the falling trajectory of each seed in all images, that is, to determine the same seed in different images and then accurately fit the falling trajectory of the seed. In this step, for each previously recorded coordinate (x... m,i ,y m,i ), find the record in the (m+1)th image that is in (x m,i ,y m,i Seek seed positions that are less than D pixels away (typically 40 pixels) and identify these two seeds as the same seed. After the traversal is complete, the trajectory of each seed can be found, thus determining the set of positional information corresponding to multiple seeds.

[0106] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0107] Based on the same inventive concept, this application also provides a seeding quality detection device for implementing the seeding quality detection method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more seeding quality detection device embodiments provided below can be found in the limitations of the seeding quality detection method described above, and will not be repeated here.

[0108] In one exemplary embodiment, such as Figure 6 As shown, a seeding quality detection device is provided, comprising: an acquisition module 61, a matching module 62, a calculation module 63, and a detection module 64, wherein:

[0109] The acquisition module 61 is used to acquire a seed image sequence, wherein the seed images in the seed image sequence are collected during the seeding process;

[0110] Matching module 62 is used to determine the location information sets corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the location information set corresponding to each seed includes the location information of the seed in at least three seed images.

[0111] The calculation module 63 is used to determine the landing location information and landing time information of each seed based on the location information set corresponding to the seed and the acquisition time information of the seed image from which the location information in the location information set comes.

[0112] The detection module 64 is used to determine the seed spacing between adjacent seeds based on the landing location and landing time information of each seed, and to determine the seeding quality detection result based on the seed spacing between adjacent seeds.

[0113] In one embodiment, the matching module 62 is further configured to:

[0114] For each seed image, seed recognition is performed to obtain the seed recognition result corresponding to the seed image. The seed recognition result contains the location information of the seeds in the seed image.

[0115] For adjacent first and second seed images in a seed image sequence, multiple seed combinations are determined, each seed combination containing one seed from the first seed image and one seed from the second seed image;

[0116] For each seed combination, determine the pixel distance between two seeds in the seed combination based on the location information;

[0117] If the pixel distance is less than the distance threshold, the two seeds in the seed combination are identified as the same seed;

[0118] Based on the location information corresponding to the same seed, obtain the set of location information corresponding to the seed.

[0119] In one embodiment, the matching module 62 is further configured to:

[0120] The seed image is converted into a binary image, and the binary image is preprocessed.

[0121] Identify connected regions in the preprocessed binary image and calculate the area of ​​each connected region;

[0122] Connected regions with an area greater than the area threshold are identified as seeds, and the location information of the connected regions is obtained as the location information of the seeds.

[0123] The seed recognition result corresponding to the seed image is obtained based on the seed's location information.

[0124] In one embodiment, the calculation module 63 is further configured to:

[0125] The motion trajectory of the seed is fitted by the set of location information corresponding to the seed and the acquisition time information of the seed image from which the location information in the set of location information comes, and the initial position information, first axial velocity, second axial initial velocity and second axial acceleration of the seed are obtained.

[0126] Obtain the ground image position, and calculate the seed's landing position and landing time based on the ground image position, initial position information, first axial velocity, second axial initial velocity, and second axial acceleration.

[0127] In one embodiment, the detection module 64 is further configured to:

[0128] The horizontal positions of each seed are sorted according to the landing time information of each seed, and the difference in the horizontal position between adjacent seeds is used as the initial spacing between adjacent seeds.

[0129] The time difference between adjacent seeds is determined based on the landing time information of each seed, and the compensation spacing between adjacent seeds is determined based on the vehicle speed and the time difference.

[0130] The spacing between adjacent seeds is determined based on the initial spacing and the compensation spacing.

[0131] In one embodiment, the detection module 64 is further configured to:

[0132] Count the total number of spacings associated with the seeding spacing;

[0133] The spacing between adjacent seeds is filtered according to the spacing range, and the number of seeding spacings that meet the target spacing range is counted.

[0134] The seeding qualification rate is calculated based on the total number of spacings and the target number of spacings, and the seeding quality inspection results are determined based on the seeding qualification rate.

[0135] Each module in the aforementioned seed quality detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0136] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a seed quality detection method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0137] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0138] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0139] Obtain a seed image sequence, wherein the seed images in the seed image sequence are collected during the seeding process;

[0140] From the seed image sequence, determine a set of location information corresponding to multiple seeds respectively. The multiple seeds are different seeds, and the set of location information corresponding to each seed includes the location information of the seed in at least three seed images.

[0141] For each seed, the landing location information and landing time information of the seed are determined based on the location information set corresponding to the seed and the acquisition time information corresponding to the seed image from which the location information in the location information set originates.

[0142] Based on the landing location information and landing time information of each seed, the seeding spacing between adjacent seeds is determined, and the seeding quality detection result is determined based on the seeding spacing between adjacent seeds.

[0143] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0144] For each seed image, seed recognition is performed to obtain the corresponding seed recognition result, which includes the position information of the seeds in the seed image. For adjacent first and second seed images in the seed image sequence, multiple seed combinations are determined, each seed combination containing one seed from the first seed image and one seed from the second seed image. For each seed combination, the pixel distance between the two seeds in the seed combination is determined based on the position information. If the pixel distance is less than a distance threshold, the two seeds in the seed combination are identified as the same seed. Based on the position information corresponding to the same seed, a set of position information corresponding to the seed is obtained.

[0145] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0146] The seed image is converted into a binary image and preprocessed. Connected regions in the preprocessed binary image are identified and the area of ​​each connected region is calculated. Connected regions with an area greater than the area threshold are identified as seeds, and the location information of the connected regions is obtained as the location information of the seeds. The seed identification result corresponding to the seed image is obtained based on the location information of the seeds.

[0147] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0148] The seed's trajectory is fitted by the set of location information corresponding to the seed and the acquisition time information of the seed image from which the location information in the set of location information is obtained, and the initial position information, first axial velocity, second axial initial velocity and second axial acceleration of the seed are obtained; the ground image position is obtained, and the landing position information and landing time information of the seed are calculated based on the ground image position, initial position information, first axial velocity, second axial initial velocity and second axial acceleration.

[0149] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0150] The seeds are sorted horizontally according to their landing time information. The difference in horizontal position between adjacent seeds is used as the initial spacing between them. The time difference between adjacent seeds is determined based on the landing time information of each seed, and the compensation spacing between adjacent seeds is determined based on the vehicle speed and the time difference. The seeding spacing between adjacent seeds is determined based on the initial spacing and the compensation spacing.

[0151] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0152] The total number of spacings associated with the seed spacing is counted; the seed spacing between adjacent seeds is screened according to the spacing range, and the number of seed spacings that meet the target spacing range is counted; the seed spacing qualification rate is calculated based on the total number of spacings and the number of target spacings, and the seed spacing quality test results are determined based on the seed spacing qualification rate.

[0153] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0154] Obtain the seed image sequence, where the seed images are collected during the seeding process;

[0155] Determine the location information sets corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the location information set corresponding to each seed includes the location information of the seed in at least three seed images.

[0156] For each seed, the landing location and landing time information of the seed are determined based on the location information set corresponding to the seed and the acquisition time information of the seed image from which the location information in the location information set comes.

[0157] Based on the landing location and landing time information of each seed, the spacing between adjacent seeds is determined, and the seeding quality test results are determined based on the spacing between adjacent seeds.

[0158] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0159] For each seed image, seed recognition is performed to obtain the corresponding seed recognition result, which includes the position information of the seeds in the seed image. For adjacent first and second seed images in the seed image sequence, multiple seed combinations are determined, each seed combination containing one seed from the first seed image and one seed from the second seed image. For each seed combination, the pixel distance between the two seeds in the seed combination is determined based on the position information. If the pixel distance is less than a distance threshold, the two seeds in the seed combination are identified as the same seed. Based on the position information corresponding to the same seed, a set of position information corresponding to the seed is obtained.

[0160] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0161] The seed image is converted into a binary image and preprocessed. Connected regions in the preprocessed binary image are identified and the area of ​​each connected region is calculated. Connected regions with an area greater than the area threshold are identified as seeds, and the location information of the connected regions is obtained as the location information of the seeds. The seed identification result corresponding to the seed image is obtained based on the location information of the seeds.

[0162] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0163] The seed's trajectory is fitted by the set of location information corresponding to the seed and the acquisition time information of the seed image from which the location information in the set of location information is obtained, and the initial position information, first axial velocity, second axial initial velocity and second axial acceleration of the seed are obtained; the ground image position is obtained, and the landing position information and landing time information of the seed are calculated based on the ground image position, initial position information, first axial velocity, second axial initial velocity and second axial acceleration.

[0164] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0165] The seeds are sorted horizontally according to their landing time information. The difference in horizontal position between adjacent seeds is used as the initial spacing between them. The time difference between adjacent seeds is determined based on the landing time information of each seed, and the compensation spacing between adjacent seeds is determined based on the vehicle speed and the time difference. The seeding spacing between adjacent seeds is determined based on the initial spacing and the compensation spacing.

[0166] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0167] The total number of spacings associated with the seed spacing is counted; the seed spacing between adjacent seeds is screened according to the spacing range, and the number of seed spacings that meet the target spacing range is counted; the seed spacing qualification rate is calculated based on the total number of spacings and the number of target spacings, and the seed spacing quality test results are determined based on the seed spacing qualification rate.

[0168] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0169] Obtain the seed image sequence, where the seed images are collected during the seeding process;

[0170] Determine the location information sets corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the location information set corresponding to each seed includes the location information of the seed in at least three seed images.

[0171] For each seed, the landing location and landing time information of the seed are determined based on the location information set corresponding to the seed and the acquisition time information of the seed image from which the location information in the location information set comes.

[0172] Based on the landing location and landing time information of each seed, the spacing between adjacent seeds is determined, and the seeding quality test results are determined based on the spacing between adjacent seeds.

[0173] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0174] For each seed image, seed recognition is performed to obtain the corresponding seed recognition result, which includes the position information of the seeds in the seed image. For adjacent first and second seed images in the seed image sequence, multiple seed combinations are determined, each seed combination containing one seed from the first seed image and one seed from the second seed image. For each seed combination, the pixel distance between the two seeds in the seed combination is determined based on the position information. If the pixel distance is less than a distance threshold, the two seeds in the seed combination are identified as the same seed. Based on the position information corresponding to the same seed, a set of position information corresponding to the seed is obtained.

[0175] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0176] The seed image is converted into a binary image and preprocessed. Connected regions in the preprocessed binary image are identified and the area of ​​each connected region is calculated. Connected regions with an area greater than the area threshold are identified as seeds, and the location information of the connected regions is obtained as the location information of the seeds. The seed identification result corresponding to the seed image is obtained based on the location information of the seeds.

[0177] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0178] The seed's trajectory is fitted by the set of location information corresponding to the seed and the acquisition time information of the seed image from which the location information in the set of location information is obtained, and the initial position information, first axial velocity, second axial initial velocity and second axial acceleration of the seed are obtained; the ground image position is obtained, and the landing position information and landing time information of the seed are calculated based on the ground image position, initial position information, first axial velocity, second axial initial velocity and second axial acceleration.

[0179] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0180] The seeds are sorted horizontally according to their landing time information. The difference in horizontal position between adjacent seeds is used as the initial spacing between them. The time difference between adjacent seeds is determined based on the landing time information of each seed, and the compensation spacing between adjacent seeds is determined based on the vehicle speed and the time difference. The seeding spacing between adjacent seeds is determined based on the initial spacing and the compensation spacing.

[0181] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0182] The total number of spacings associated with the seed spacing is counted; the seed spacing between adjacent seeds is screened according to the spacing range, and the number of seed spacings that meet the target spacing range is counted; the seed spacing qualification rate is calculated based on the total number of spacings and the number of target spacings, and the seed spacing quality test results are determined based on the seed spacing qualification rate.

[0183] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0184] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0185] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0186] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for detecting seeding quality, characterized in that, The method includes: Obtain a seed image sequence, wherein the seed images in the seed image sequence are collected during the seeding process; From the seed image sequence, determine a set of location information corresponding to multiple seeds respectively. The multiple seeds are different seeds, and the set of location information corresponding to each seed includes the location information of the seed in at least three seed images. For each seed, the landing location information and landing time information of the seed are determined based on the location information set corresponding to the seed and the acquisition time information corresponding to the seed image from which the location information in the location information set originates. Based on the landing location information and landing time information of each seed, the seeding spacing between adjacent seeds is determined, and the seeding quality detection result is determined based on the seeding spacing between adjacent seeds.

2. The method according to claim 1, characterized in that, Determine a set of location information corresponding to multiple seeds from the seed image sequence, including: For each seed image, seed recognition is performed on the seed image to obtain the seed recognition result corresponding to the seed image. The seed recognition result includes the position information of the seeds in the seed image respectively. For adjacent first and second seed images in the seed image sequence, multiple seed combinations are determined, each seed combination containing one seed from the first seed image and one seed from the second seed image; For each seed combination, the pixel distance between two seeds in the seed combination is determined based on the location information; If the pixel distance is less than a distance threshold, the two seeds in the seed combination are identified as the same seed; Based on the location information corresponding to the same seed, obtain the set of location information corresponding to the seed.

3. The method according to claim 2, characterized in that, The step of performing seed recognition on the seed image to obtain the seed recognition result corresponding to the seed image includes: The seed image is converted into a binary image, and the binary image is preprocessed. Identify connected regions in the preprocessed binary image and calculate the area of ​​each connected region; The connected regions with an area greater than the area threshold are identified as seeds, and the location information of the connected regions is obtained as the location information of the seeds. The seed identification result corresponding to the seed image is obtained based on the location information of the seed.

4. The method according to any one of claims 1-3, characterized in that, The step of determining the landing location information and landing time information of the seed based on the location information set corresponding to the seed and the acquisition time information corresponding to the seed image from which the location information in the location information set originates, includes: The motion trajectory of the seed is fitted based on the set of location information corresponding to the seed and the acquisition time information of the seed image from which the location information in the set of location information originates, to obtain the initial position information, first axial velocity, second axial initial velocity and second axial acceleration of the seed. Obtain the ground image position, and calculate the seed's landing position and landing time information based on the ground image position, the initial position information, the first axial velocity, the second axial initial velocity, and the second axial acceleration.

5. The method according to any one of claims 1 to 3, characterized in that, The location information includes lateral position; determining the planting spacing between adjacent seeds based on the landing location information and landing time information of each seed includes: The horizontal positions of each seed are sorted according to the landing time information of each seed, and the difference in the horizontal positions between adjacent seeds is used as the initial spacing between adjacent seeds. The time difference between adjacent seeds is determined based on the landing time information of each seed, and the compensation distance between adjacent seeds is determined based on the vehicle speed and the time difference. The planting spacing between adjacent seeds is determined based on the initial spacing and the compensation spacing.

6. The method according to any one of claims 1 to 3, characterized in that, The determination of seeding quality detection results based on the seeding spacing between adjacent seeds includes: Count the total number of spacings associated with the seeding spacing; The spacing between adjacent seeds is filtered according to the spacing range, and the number of seeding spacings that meet the target spacing range is counted. The seeding qualification rate is calculated based on the total number of spacings and the target number of spacings, and the seeding quality inspection result is determined based on the seeding qualification rate.

7. A seeding quality detection device, characterized in that, The device includes: An acquisition module is used to acquire a seed image sequence, wherein the seed images in the seed image sequence are collected during the seeding process; The matching module is used to determine a set of location information corresponding to multiple seeds from the seed image sequence. The multiple seeds are different seeds, and the set of location information corresponding to each seed includes the location information of the seed in at least three seed images. The calculation module is used to determine the landing location information and landing time information of each seed based on the location information set corresponding to the seed and the acquisition time information corresponding to the seed image from which the location information in the location information set comes; The detection module is used to determine the seed spacing between adjacent seeds based on the landing location information and landing time information of each seed, and to determine the seeding quality detection result based on the seed spacing between adjacent seeds.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images