A method and system for monitoring the seed guiding performance of a belt-type high-speed seed guiding device, a storage medium, and an optoelectronic monitoring device

Abstract

A method and system for monitoring the guiding performance of a belt-type high-speed seed guiding device, a storage medium and a photoelectric monitoring device, belonging to the field of agricultural machinery. The present invention is proposed to solve the problem that the current pulse waveform has interference and cannot accurately identify the deviation pulse generated when the seed grain passes, and the monitoring system cannot accurately capture the seed grain. The monitoring method is based on bilateral pulse value analysis and energy mask smoothing algorithm to extract the pulse characteristics of belt-type seed guiding grains, and the pulses after smoothing the two channels are subtracted to obtain the grain-specific pulses after the initial highlighting of the seed grain characteristics, which are used to accurately extract the characteristics of the unbalanced pulses caused when the seed grain passes through the monitoring point; according to the energy mask smoothing algorithm, the grain-specific pulses obtained after the above steps are processed are secondary smoothed, and the time interval between two adjacent unbalanced fluctuations is calculated, and the plant spacing X is calculated in real time, thereby realizing the monitoring of the performance of belt-type high-speed seed guiding. The present invention can capture the accuracy of the pulse deviation caused by the passage of seeds, thereby completing the monitoring of parameters such as sowing amount, qualified rate, missed sowing rate and reseeding rate.

Description

Technical Field

[0001] The present invention belongs to the field of agricultural machinery, and relates to an optoelectronic monitoring technology for the seed metering performance of a belt-type seed guiding device. Specifically, it relates to a method and system for monitoring the seed guiding performance of a belt-type high-speed seed guiding device, a storage medium, and an optoelectronic monitoring device. Background Art

[0002] The belt-type seed guiding device can ensure the uniform distribution of seeds in the sowing area. When the device is working, the seeds rotate with the seed conveying belt to the seed dropping port and are dropped into the seed furrow, ensuring the smoothness of the seeds falling into the seed furrow and effectively avoiding the collision or bouncing of the seeds with the inner wall of the seed guiding pipe and the soil of the seedbed. It can adapt to high-speed sowing operations. This belt-type seed guiding device can avoid the interference of the seed position caused by the vibration of the seeder or other reasons, thus ensuring the accuracy and effect of sowing.

[0003] The seed guiding process is the most important link in the entire sowing process, and the seed guiding process can determine the landing posture and uniformity of the seeds. [4][3] Belt-type seed guiding is a method that uses a seed guiding belt to fully constrain the seeds and keep them in an orderly state. [4][9] This method can be achieved through full-degree-of-freedom constraints or full constraints with only local degrees of freedom. Belt-type seed guiding can ensure the uniform distribution of seeds during sowing and effectively avoid the collision or bouncing of the seeds with the inner wall of the seed guiding pipe and the soil of the seedbed. Due to its unique seed receiving mechanism, it can also avoid the re-sowing of two seeds in the same seed cavity. This active seed guiding method can avoid the interference of the seed position caused by the vibration of the seeder or other reasons, thus ensuring the accuracy and effect of sowing. However, due to the need to constrain the seeds, the belt-type seed guiding pipe will have a constraint mechanism in the seed dropping area, and the traditional monitoring method will be interfered by the constraint mechanism, and there is no relevant supporting monitoring system in China. Therefore, it is an urgent technical problem to study a monitoring system for the seed guiding performance suitable for the belt-type high-speed seed guiding device.

[0004] From the perspective of the monitoring mechanism of traditional seed guiding devices, it is mainly divided into visual type

[10]

[12] , capacitive type

[13]

[15] , piezoelectric type

[16]

[18] and optoelectronic type

[19]

[21] There are 4 types. Visual monitoring uses a visual camera to collect the image information of the falling seeds, and through the computer's processing, screening, and obtaining the position information of adjacent seeds. Leemans et al.

[22] The seed metering monitoring system based on computer vision was studied. By using a vision camera to identify the falling seeds, the performance state of the seed guide of the seed metering device was monitored, and it performed well in field operations. However, when using a belt for high-speed sowing, the interval between the supporting plates is small, the seeds fall quickly, and the visual monitoring and recognition operations are relatively slow, making it impossible to process so many features in a short time. Capacitive monitoring mainly relies on the change in the dielectric constant generated by the seeds passing through the capacitor plates, and the change in capacitance is transmitted to the single-chip microcomputer in the form of high and low level signals through a conversion circuit for acquisition. Chen Jianguo, etc.

[23] A high-precision wheat seed number monitoring system was designed based on a capacitive sensor, realizing the accurate monitoring of the sowing amount of a wheat precision seeder. Zhou Liming, etc.

[24]

[26] A seed metering performance monitoring system for a corn planter was designed based on capacitive signals. By obtaining the constraint conditions of the length of the capacitor sensor plate, the reliability of the seed metering performance monitoring of the corn planter was improved. Since the seed cavity of the belt-type seed guiding device is small, the required size of the capacitor plate is small, and the initial capacitance and the change in capacitance are very small, which are difficult to detect and install. Piezoelectric monitoring is to generate a change in the piezoelectric value by the falling seeds hitting the piezoelectric sensor, and it is converted into a voltage pulse signal through a signal processing circuit. Huang Dongyan, etc.

[27] A seed metering monitoring system based on polyvinylidene fluoride piezoelectric film was designed, which converts the physical quantity of single-grain dropping of the seed metering device into a pulse voltage signal, and can monitor the seed metering state in real time and automatically in a harsh environment. Since the belt-type seed guiding device is a fully constrained seed guiding device and the seeds are transported to the seed dropping place through the seed conveying belt, it is impossible to use the seeds contacting the piezoelectric element to monitor the seed position. Photoelectric monitoring uses a light-emitting tube to collect the change in the pulse signal generated by the seeds blocking the light source. Ji Chao, etc.

[28] A seed metering quality monitoring system for a corn no-till precision seeder was developed. By designing and optimizing the photoelectric monitoring probe, the real-time monitoring of the operation quality of corn no-till precision seeding was realized. Cay, etc.

[29] An indoor photoelectric wheat planter seed metering monitoring system was studied, and the monitoring system was composed of photodiodes. By comparing the monitoring results of 10 kinds of seeds with different physical properties, the advancement of the system was verified. Since both the seeds on the seed conveying side and the seed belt supporting plates of the belt-type seed guiding device pass through the monitoring point, the traditional metering pulse or voltage change method of photoelectric monitoring cannot be used. However, due to the characteristics of fast monitoring speed and strong sensitivity of the photoelectric sensor, it can be selected as the basic sensing element for the research of the monitoring method.

[0005] In summary, it can be seen that although certain progress has been made in the research on seed guiding devices at home and abroad, the seed metering performance monitoring systems for seed guiding devices mainly focus on unconstrained hollow seed guiding devices and underconstrained seed guiding devices. Due to the characteristics of the belt-type seed guiding device that require restricting the position and state of seeds, there will be a seed restraint mechanism during its seed dropping process. This leads to the situation that seed metering monitoring methods such as photoelectric, capacitive, and vision-based methods will be blocked and interfered by the restraint mechanism, and it is impossible to accurately identify the difference between seeds and the restraint mechanism. Therefore, it is extremely crucial to develop an optoelectronic monitoring system and its monitoring method specifically applicable to belt-type seed guiding devices. In the prior art, few methods specifically for optoelectronic monitoring of belt-type seed guiding devices have been proposed. Summary of the Invention

[0006] The technical problem to be solved by the present invention is:

[0007] Aiming at the problems that there are interferences in the current pulse waveform and it is impossible to accurately identify the deviation pulses generated when seeds pass through, and the monitoring system cannot accurately capture seeds, a seed guiding performance monitoring method and system, a storage medium, and an optoelectronic monitoring device for a belt-type high-speed seed guiding device are provided.

[0008] The technical solution adopted by the present invention to solve the above technical problems is:

[0009] A seed guiding performance monitoring method for a belt-type high-speed seed guiding device, the monitoring method is based on bilateral pulse value analysis and energy mask smoothing algorithm to realize the extraction of the pulse characteristics of belt-type seed guiding seeds, and the process is as follows:

[0010] First, smooth the pulses on both sides of the seed guiding belt containing "burrs" (the pulse values of the two-channel original channels include the seed delivery side and the rotary side) respectively to reduce the noise and random fluctuations of the original pulses (such as Figure 8 ), so that the data is smoother and more stable;

[0011] Then, subtract the pulses smoothed by the two channels to obtain the grain characteristic pulses after initially highlighting the seed characteristics, which are used to accurately extract the characteristics of the imbalance pulses caused when the seeds pass through the monitoring point (such as Figure 9 );

[0012] Then, according to the energy mask smoothing algorithm, perform secondary smoothing on the grain characteristic pulses obtained after the above steps to obtain pulses that only highlight the seed characteristics (such as Figure 10 ), in the pulses, the seed characteristic pulses are imbalance fluctuations, and the others are balance fluctuations;

[0013] When another adjacent imbalance fluctuation appears, calculate the time interval between the two adjacent imbalance fluctuations, that is, the adjacent seed dropping time interval T F (such as Formula 1-5, Figure 5 ),

[0014] After determining the balance fluctuation, continuously collect the pulse value data on the rotary side, and calculate the time interval T between two adjacent light-shielding times of the supporting piece on the rotary side. P Calculate T. P After calculating T, F calculate the linear velocity V of the seed guiding belt, and calculate the plant spacing X in real time through T, P V, so as to realize the performance monitoring of the high-speed belt-type seed guiding. This part of the content corresponds to Figure 5 judging whether the pulse signals are equal, and the unbalance wave processing and balance wave processing in

[0015] Furthermore, the specific implementation process of the monitoring method is as follows:

[0016] Smooth the pulse values of the two original channels to obtain new pulses. The calculation formula for smoothing the pulse values of the two channels is:

[0017]

[0018] In the formula:

[0019] B j ——The pulse value on the seed output side at the jth sampling point, unit: V;

[0020] C j ——The pulse value (right side) on the rotary side at the jth sampling point, unit: V;

[0021] i, j——Sampling point index

[0022] B smooth,i ——The pulse value after smoothing on the seed output side at the ith sampling point, unit: V;

[0023] C smooth,i ——The pulse value after smoothing on the rotary side at the ith sampling point, unit: V;

[0024] N——Smoothing window size

[0025] The new pulse is a smoothed pulse based on the original pulse, which is realized by averaging all the values within a window N of data points. The "glitches" interference is filtered out from the smoothed pulses.

[0026] Subtract the pulses after smoothing the two channels to obtain the pulse difference after highlighting the features. The calculation process is shown in formula (11):

[0027] D smooth,i =B smooth,i -C smooth,i (8)

[0028] In the formula, D smooth,i ——The difference after smoothing the bilateral pulses at the ith sampling point, unit: V

[0029] An energy mask smoothing algorithm is given. By setting an energy threshold, all interfering waveforms are resolved. Only when the signal energy exceeds this threshold will the signal be smoothed; the part of the signal with large intensity and containing important information is highlighted; the grain pulse signal is made to conform to this characteristic. Among them, the short-time energy refers to the sum of the signal energy within a short-time window, which is an index to measure the intensity of the signal within this time window. The formula for calculating the short-time energy is:

[0030]

[0031] In the formula, E i —— The short-time energy value of the i-th sampling point;

[0032] The energy threshold is a preset boundary used to judge whether the short-time energy of the signal is large enough, that is, a meaningful signal rather than noise; the setting of the energy threshold is used to determine the retention tendency of the data. The formula for calculating the energy threshold is

[0033] θ = mean(E) × K (10)

[0034] In the formula: θ —— Energy threshold, K —— Preset coefficient of the energy threshold;

[0035] The mask signal is a binary signal representing the part of the original signal where the short-time energy exceeds the energy threshold; if the short-time energy of the signal exceeds the threshold, the mask signal is 1 at the corresponding time point, otherwise it is 0. Therefore, the mask signal can be used to "filter" the original signal and only retain the high-energy part. The formula for calculating the mask signal is

[0036]

[0037] In the formula: M i —— The mask signal of the i-th sampling point;

[0038] D masked,i =D smooth,i ×M i (12)

[0039] In the formula: D masked,i —— The difference value after applying the mask to the i-th sampling point, in volts;

[0040]

[0041] In the formula: D’ sm ooth,i —— The value after re-smoothening the mask data of the i-th sampling point, in volts;

[0042] N' —— The window size of the secondary smoothing process.

[0043] Furthermore, the value of the smoothing process window size N is taken as 50.

[0044] Furthermore, the method includes the step of synchronously collecting pulse data of photoelectric emitting tubes on both sides before extracting the pulse characteristics of the belt-type seed guide grains, and judging whether the pulse signals are equal based on the above steps. If so, balanced fluctuation processing is performed, the pulse data on the right side is collected, the data is processed by an algorithm, the linear speed V of the seed guide belt is calculated, and the plant spacing X is calculated in real time (Formulas 1 and 2). Otherwise, unbalanced fluctuation processing is performed, the linear speed V of the seed guide belt is calculated, and finally the plant spacing X is calculated in real time.

[0045] Furthermore, by monitoring the time interval of the seed belt support, the calculation formula of the seed belt linear speed V can be deduced as follows:

[0046]

[0047] Where S is the distance between two adjacent brackets, in m

[0048] T P ——The time interval between two consecutive shading of the rotating side support, in seconds

[0049] Therefore, the actual plant spacing value can be calculated without monitoring the speed of the belt-type seed guide device drive motor. The calculation formula for the spacing between two adjacent seeds, that is, the plant spacing X, is:

[0050]

[0051] Where T F ——The time interval between two pulse deviations, in s

[0052] The monitoring system automatically stores each plant spacing value. Suppose the theoretical plant spacing of the target crop is X T , the sowing spacing is P, if X∈[0,0.5X T ], it is counted as a replay. If the number of replays is P mult , then the replay rate I mult The calculation formula is

[0053]

[0054] If X∈[1.5X T ,+∞], it is counted as missed broadcast. If the number of missed broadcasts is P miss , then the missed broadcast rate I miss The calculation formula is

[0055]

[0056] If X∈[0.5X T ,1.5X T ], then it is considered qualified. If the number of qualified times is P q , then the qualified rate Iq The calculation formula is

[0057]

[0058] To determine whether the plant spacing X meets the threshold is based on the technical conditions of JB / T 10293—2013 Single-grain (precision) seeder.

[0059] Further, before the step of synchronously collecting the pulse data of the two-sided photoelectric pair-emitting tubes, steps of initializing the single-chip microcomputer and enabling the external interrupt are executed. As Figure 5 shown.

[0060] A monitoring system for the seeding performance of a belt-type high-speed seed guiding device, the system having program modules corresponding to the steps of the above technical solution, and executing the steps in the above method for monitoring the seeding performance of the belt-type high-speed seed guiding device when running.

[0061] A computer-readable storage medium, the computer-readable storage medium storing a computer program, the computer program being configured to implement the steps of the method for monitoring the seeding performance of the belt-type high-speed seed guiding device when called by a processor.

[0062] An optoelectronic monitoring device, characterized in that: the optoelectronic monitoring device includes at least one processor (main controller), and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for monitoring the seeding performance of the belt-type high-speed seed guiding device.

[0063] Further, the processor is a main controller, and further includes a main controller circuit connected to the main controller and a main controller peripheral circuit. The main controller selects an STM32F103RBT6 type single-chip microcomputer. The main controller peripheral circuit includes a photoelectric emission circuit, a seed-transporting-side photoelectric receiving circuit a, and a rotary-side photoelectric receiving circuit b. The photoelectric emission circuit includes two groups of photoelectric emission tube groups, one group is the photoelectric emission tube group on the seed-transporting side, and the other group is the photoelectric emission tube group on the rotary side. Each group of photoelectric emission tube groups includes three photodiodes packaged side by side; the seed-transporting-side photoelectric receiving circuit a includes three photodiodes packaged side by side; the rotary-side photoelectric receiving circuit b includes three photodiodes packaged side by side; to adapt to the seed cavity width of the belt-type high-speed seed guiding device and achieve blind-zone-free monitoring, as Figure 4 shown.

[0064] A belt-type high-speed seed guiding device, the belt-type high-speed seed guiding device including the above optoelectronic monitoring device.

[0065] The present invention has the following beneficial technical effects:

[0066] In view of the phenomenon that both the seed belt supporting piece and the seeds pass through the monitoring point during the seed guiding process of the belt-type high-speed seed guiding device, it is difficult to install the above-mentioned sensing structure on the belt-type seed guiding device and distinguish the change of pulse characteristics, resulting in the difficulty of monitoring the seed guiding performance. The present invention proposes a monitoring system for a belt-type high-speed seed guiding device based on an infrared sensor and a bilateral pulse value analysis and energy masking smoothing algorithm (BPV-EMSA). By comparing the pulse fluctuation time of the photoelectric tubes on the seed output side and the rotary side of the seed guiding belt, the pulse change of the seeds passing through the monitoring point is accurately monitored, so as to monitor the seed guiding performance in the seed guiding belt and the speed of the seed guiding belt. This system performs well in indoor tests and has practical application value. Description of the Drawings

[0067] Figure 1 It is a schematic structural diagram of a belt-type high-speed seed guiding device (in the figure: 1-1 finger wheel, 1-2 seed cleaning claw, 1-3 motor, 1-4 driving pulley, 1-5 seed guiding belt, 1-6 pre-tightening spring, 1-7 driven pulley, 1-8 gearbox, 1-9 seed protecting cover, 1-10 seed guiding belt housing, 1-11 seed dropping plate); Figure 2 It is a schematic monitoring principle diagram of a group of belt-type high-speed seed guiding devices (in the figure: 2-1 photoelectric emitter, 2-2 photoelectric receiver, 2-3 seed belt supporting piece, 2-4 corn seeds); Figure 3 It is a schematic installation diagram of the monitoring system for the belt-type high-speed seed guiding device (in the figure: 3-1 through-hole for opposite shooting on the seed output side, 3-2 through-hole for opposite shooting on the rotary side, 3-3 receiving side of the monitoring module, 3-4 transmitting side of the monitoring module, 3-5 infrared light); Figure 4 It is the main hardware circuit diagram of the monitoring system for the belt-type high-speed seed guiding device; Figure 5 It is the software flow block diagram of the monitoring system for the belt-type high-speed seed guiding device ( Figure 7 It is the result of collecting the pulse data of the two-side photoelectric opposite-shooting tubes); Figure 6 It is a photo of the sampling test of the monitoring system (in the figure: 6-1 represents a virtual oscilloscope, 6-2 represents the software interface for data recording, 6-3 represents a 12V storage battery, 6-4 represents a vSet type air-suction seeder, 6-5 represents the pulse sampling channel on the rotary side, 6-6 represents the pulse sampling channel on the seed output side, 6-7 represents the belt-type seed guiding device); Figure 7 It is a curve graph of the pulse change on both sides of the seed guiding belt; Figure 8 It is a curve graph of the change after the pulse on both sides of the seed guiding belt is smoothed; Figure 9 It is a pulse difference curve graph after taking the difference of the pulses on both sides of the seed guiding belt; Figure 10 It is a pulse difference graph after energy masking smoothing processing; Figure 11 It is a bar graph of the specific parameter indicators for monitoring the seeding rate of each group; Figure 12It is a photo of the performance evaluation test for the monitoring system. Detailed implementation manners

[0068] The following will elaborate on the implementation of the present invention from the following aspects in combination with the attached Figure 1-12 :

[0069] 1 Monitoring principle

[0070] The studied belt-type high-speed seed guiding device

[30] is installed at the upper end of the seed discharging port of the precision metering device passing through the seed plate holes. During sowing, the original seed brush of the metering device is removed. Before the grains leave the seed plate, the main and auxiliary finger wheels take the grains by means of rotational clamping, and then transport and discharge the grains into the seed cavity of the seed guiding belt. The grains are transported to the seed dropping port under the rotation of the seed guiding belt and dropped into the seed furrow.

[0071] Belt-type high-speed seed guiding device

[30] The overall structural schematic diagram is as Figure 1 shown.

[0072] The monitoring principle of the belt-type seed guiding monitoring system mainly utilizes the seed conveying characteristics of the belt-type seed guiding. During its seed metering process, the device is divided into a left seed conveying side and a right rotating side, and the two sides are axisymmetric with respect to the center of the seed belt; in addition to the stable reciprocating movement of the seed guiding belt in the seed cavity on the seed conveying side compared with the rotating side, there are also discharged seeds passing through. Based on this difference, a bilateral pulse comparison monitoring method is proposed. Since only the left side of the belt-type high-speed seed guiding device is the seed conveying side and the right side is the rotating side after seed discharging. When the installation positions of the two groups of photoelectric pairs are fixed at the positions as Figure 2 shown, the seed guiding belt supporting pieces on the left and right sides always block the light beams on both sides at the same time, and there are always no seeds passing through the non-seed dropping area on the right side. Therefore, when the two groups of photoelectric pairs simultaneously generate equal-waveform pulse changes, it means that the seed guiding belt supporting pieces on both sides block the light at the same time, and this is called balanced fluctuation at this time; when the time of the pulse generated by the left photoelectric pair is greater than the time of the pulse fluctuation generated by the right side, it is equivalent to that there are seeds passing through the photoelectric pair in the left seed dropping area at this time, and this is called unbalanced fluctuation at this time; subtracting the right pulse time from the left pulse time is the time for the seeds to pass through the photoelectric pair, and this is called deviation fluctuation at this time. Monitoring the time interval between the occurrences of two deviation fluctuations is the time interval for monitoring adjacent discharged seeds; at the same time, monitoring the pulse changes on the rotating side can obtain the time interval between adjacent seed belt supporting pieces, and then calculate the linear velocity of the seed guiding belt, providing conditions for calculating the actual plant spacing after the seeds are discharged.

[0073] On the premise of being able to accurately capture the deviation pulse, in order to determine the actual plant spacing between two adjacent seeds, it is necessary to clarify the linear velocity of the seed guiding belt. In line with the principle of simplifying the complexity of the monitoring system, it is considered not to install an encoder at the motor. The calculation formula for the linear velocity V of the seed guiding belt can be deduced by monitoring the time interval for the seed belt supporting piece to pass through as

[0074]

[0075] Where S is the distance between two adjacent brackets, in m

[0076] T P ——The time interval between two consecutive shading of the rotating side support, in seconds

[0077] Therefore, the actual plant spacing value can be calculated without monitoring the speed of the belt-type seed guide device drive motor. The calculation formula for the spacing between two adjacent seeds, that is, the plant spacing X, is:

[0078]

[0079] Where T F ——The time interval between two pulse deviations, in s

[0080] The monitoring system automatically stores each plant spacing value. Suppose the theoretical plant spacing of the target crop is X T , the sowing spacing is P, if X∈[0,0.5X T ], it is counted as a replay. If the number of replays is P mult , then the replay rate I mult The calculation formula is

[0081]

[0082] If X∈[1.5X T ,+∞], it is counted as missed broadcast. If the number of missed broadcasts is P miss , then the missed broadcast rate I miss The calculation formula is

[0083]

[0084] If X∈[0.5X T ,1.5X T ], then it is considered qualified. If the number of qualified times is P q , then the qualified rate I q The calculation formula is

[0085]

[0086] 2 System overall design

[0087] 2.1 Monitoring module structure design

[0088] The research conducted an overall design of the external structure of the belt - type seeding monitoring module. Two groups of infrared emitter circuits are encapsulated in the same side housing, and two groups of opposite infrared receiver tubes are encapsulated in the other side housing. The opposed tubes are respectively arranged on the left and right sides of the belt - type seed guiding tube relative to the forward direction of operation. Opposed holes are respectively opened on the seed - conveying side and the rotating side of the high - speed belt - type seed guiding device, exposing the seed - belt support piece and the seed cavity. Then, the housing of the emission side and the receiving side of the designed monitoring module are respectively fixed on the left and right sides of the high - speed belt - type seed guiding device. The opposed probes fully enter the opposed holes and are parallel to the seed - belt support piece. After the module is installed, the seed cavity is re - sealed, so that when the seed - belt support piece passes through the infrared light, the three - way light is blocked simultaneously, generating a change in the photoelectric pulse, accurately identifying the time interval of the same - side support piece passing through the monitoring point and the pulse deviation of different sides. The housing of the seed - guiding belt and the outer housing of the monitoring module are both black, avoiding the interference of natural light. The installation schematic diagram of the belt - type seeding monitoring module is as Figure 3 shown.

[0089] 2.2 Monitoring Circuit Design

[0090] To test the actual application effect, the research designed a monitoring system for the high - speed belt - type seed guiding device. Its hardware circuit composition includes: signal acquisition circuit, main controller circuit, human - machine interaction circuit, power supply circuit, etc. The hardware circuit is mainly composed of key devices such as infrared emitter tubes, infrared receiver tubes, operational amplifiers, core processors, and display screens. The main hardware circuit of the high - speed belt - type seed guiding device monitoring system is as Figure 4 shown.

[0091] The main controller of the high - speed belt - type seed guiding device monitoring system selects the STM32F103RBT6 single - chip microcomputer. This controller has characteristics such as high performance, rich peripheral support, low - power consumption, and a maximum operating frequency of 72 MHz, fully ensuring the system's secondary development ability and the algorithm operation speed requirements. The signal acquisition circuit of the monitoring system uses the LM358 operational amplifier, which is a dual operational amplifier that can effectively increase the amplitude of the photoelectric signal; the circuit also forms a filtering network through capacitors C13 - C16 to remove signal noise and improve signal quality; a feedback network is formed by resistors R10 - R15 to adjust the gain and frequency response of the amplifier. The human - machine interaction module selects the DMT10600T070_01W industrial serial port screen, which has characteristics such as good stability, high cost - performance, data storage function, and strong portability of the operation interface. Considering subsequent actual field operations, the power supply uses 12V power supply. There are also power supply requirements of 5V, 3.3V, and a reference voltage of 1V in the designed circuit, so a power conversion circuit is designed to ensure the power supply requirements of the entire monitoring system hardware circuit.

[0092] 2.3 Software Design

[0093] Based on the hardware circuit design selection and the monitoring principle of the belt-type high-speed seed guiding device, the software system of the monitoring system for the belt-type high-speed seed guiding device was designed. Keil μVision5 was selected as the development environment, and C language was used for programming. The corresponding software flow was designed. The software flow of the monitoring system for the belt-type high-speed seed guiding device is as Figure 5 shown. After the system initializes the parameters, the timer is started. On the one hand, the pulse data of the left and right photoelectric opposed tubes are collected in real time. At the same time, through the timer and the interrupt service program, the system monitors the pulse change and pulse deviation to measure the linear velocity V of the seed guiding belt and the specific moment when the seeds pass, which is used to calculate the falling time interval T F of the seeds and the sowing plant spacing X, and automatically stores the plant spacing value each time. The software program is responsible for evaluating the sowing state according to the collected data, counting the multiple sowing rate, missing sowing rate and qualification rate, and calculating the corresponding sowing performance indicators. During the continuous monitoring process, if the measured plant spacing X is greater than 1.5 times the theoretical plant spacing X T , it is considered a missing sowing; if X is less than 0.5 times X T , it is considered a multiple sowing; those between the two are regarded as qualified sowings. In addition, the software will also provide a visualization interface for real-time data, allowing users to customize operation parameters such as the theoretical plant spacing and operation speed, real-time display parameters such as the sowing qualification rate, multiple sowing rate, missing sowing rate, seed belt speed, sowing amount, and average plant spacing, and provide the function of exporting the sowing plant spacing data, providing equipment support for further optimizing the structure of the belt-type high-speed seed guiding device and improving the operation performance of the high-speed precision seeder in the future.

[0094] 3 Monitoring Algorithm

[0095] 3.1 Sampling Test

[0096] The sampling test of the monitoring system for the belt-type high-speed seed guiding device is to determine the actual sampling situation of the system pulse signal. Therefore, it is necessary to conduct a sampling test on the monitoring system and clarify the actual change of the photoelectric pulse during the system operation for algorithm design.

[0097] The sampling test of the monitoring system for the belt-type high-speed seed guiding device was carried out on November 15, 2023, in the sowing equipment laboratory of the College of Engineering, Heilongjiang Bayi Agricultural University. The belt-type high-speed seed guiding device was connected to the vSet air-suction precision metering seeder produced by Precision Planting Company to enable orderly seed taking, and it was fixed on the JPS-16 computer vision seed metering performance detection test bench. The designed infrared monitoring module was fixed on the upper left and right sides of the belt-type high-speed seed guiding device according to the above installation scheme. The feasibility of the monitoring system was determined by comparing the pulse changes of the left and right two infrared monitoring points when no object passed through the infrared monitoring point, when the seed belt support passed through the infrared monitoring point, and when the seeds passed through the infrared monitoring point. The pulse signal changes of the two channels input to the PC0 and PC1 pins of the single-chip microcomputer on the left and right sides of the seed guiding belt were monitored by using the LOTO-OSC482 virtual oscilloscope produced by Xi'an Letuo Lean Information Technology Co., Ltd. At the same time, the pulse data was saved by using the data recording software equipped with the digital oscilloscope. The sampling test of the monitoring system for the belt-type high-speed seed guiding device is as Figure 6 shown. The partial pulse changes of the two channels measured in the test are as Figure 7 shown.

[0098] As Figure 7 shown, when the pulse curves on both sides in the figure rise or fall at the same time and the amplitudes are similar, it indicates that the support pieces on both sides of the seed guiding belt simultaneously block the infrared beam and no seeds pass through. In this case, the pulse waveforms are synchronous and the amplitude-frequency is stable, which is the balanced pulse. The amplitude of the seed conveying side at the moment of 0.06 s is greater than that of the rotating side, which means that seeds pass through the seed conveying side, resulting in a longer time for the left beam to be blocked. At this time, the pulse fluctuation duration of the seed conveying side of the pulse waveform is greater than that of the rotating side. This waveform shows asymmetric fluctuation, where the amplitude and duration of one side are different from those of the other side, indicating the unbalanced pulse state when the seeds pass through. However, the pulse fluctuation of the first seed guiding belt support piece after the seeds pass through on the seed conveying side is in a strong interference state, and there is a large interference in the whole waveform acquisition process, with many "spikes" appearing, which will affect the judgment of the unbalanced pulse. Therefore, it is necessary to filter and algorithmically process the collected signals to complete the accurate monitoring of the seeds by the system.

[0099] 3.2 Research on the BPV-EMSA algorithm

[0100] Aiming at the problems of interference in the current pulse waveform and the inability to accurately identify the deviation pulses generated when seeds pass through, a monitoring algorithm for the belt-type high-speed seed guiding device based on bilateral pulse value analysis and energy mask smoothing algorithm (BPV-EMSA) is proposed. First, it is necessary to smooth the pulses with "burrs". The main purpose of this step is to reduce the noise and random fluctuations of the original pulses, make the data smoother and more stable, highlight the main trends and patterns of the data, and at the same time expect to suppress transient pulse interference and improve the interpretability and analysis accuracy of the data. The following BPV-EMSA algorithm simulation experiments are all carried out in the Matlab2020b simulation environment. The calculation formula for smoothing the pulse values of the two channels is

[0101]

[0102] In the formula:

[0103] B j —— The pulse value on the seed delivery side at the jth sampling point, unit: V

[0104] C j —— The pulse value on the rotary side at the jth sampling point, unit: V

[0105] i, j —— Sampling point index

[0106] B smooth,i —— The smoothed pulse value on the seed delivery side at the ith sampling point, unit: V

[0107] C smooth,i —— The smoothed pulse value on the rotary side at the ith sampling point, unit: V

[0108] N —— Smoothing window size

[0109] The new pulse is a smoothed pulse based on the original pulse. This is achieved by averaging all the values within a window N of the data points, where N is taken as 50, and the "burr" interference is filtered out from the smoothed pulses. The change curves of the pulses on both sides of the seed guiding belt after smoothing are as Figure 8 shown.

[0110] To accurately extract the characteristics of the unbalanced pulses caused when seeds pass through the monitoring point, the study subtracted the smoothed pulses of the two channels to obtain the pulse difference after highlighting the characteristics. The calculation process is shown in formula (11). The pulse deviation after subtracting the pulses on both sides of the seed guiding belt is as Figure 9 shown.

[0111] D smooth,i =B smooth,i -C smooth,i (8)

[0112] In the formula, D smooth,i—— The difference after bilateral pulse smoothing at the i-th sampling point, unit: V

[0113] It can be seen from Figure 9 that due to the measurement errors of the two channels, the processed pulses still cannot ensure the signal characteristics of the protruding seed pulses. Therefore, an energy mask smoothing algorithm is studied. By setting an energy threshold, all interfering waveforms are solved. Only when the signal energy exceeds this threshold, the signal will be smoothed. This can highlight the signal part with large intensity and important information. However, the seed pulse signal exactly conforms to this characteristic. Among them, the short-time energy refers to the sum of the signal energy within a short-time window, which is an index to measure the intensity of the signal within this time window. The short-time energy calculation formula is

[0114]

[0115] In the formula, E i —— The short-time energy value at the i-th sampling point

[0116] The energy threshold is a preset boundary used to judge whether the short-time energy of the signal is large enough, that is, a meaningful signal rather than noise. The setting of the energy threshold is crucial for subsequent signal processing steps because it determines the retention tendency of the data. The energy threshold calculation formula is

[0117] θ = mean(E) × K (10)

[0118] In the formula: θ —— Energy threshold, K —— Preset coefficient of energy threshold;

[0119] The mask signal is a binary signal representing the part of the short-time energy in the original signal that exceeds the energy threshold. If the short-time energy of the signal exceeds the threshold, the mask signal is 1 at the corresponding time point, otherwise it is 0. Therefore, the mask signal can be used to "filter" the original signal and only retain the high-energy part. The mask signal calculation formula is

[0120]

[0121] In the formula: M i —— The mask signal at the i-th sampling point;

[0122] D masked,i = D smooth,i × M i (12)

[0123] In the formula: D masked,i —— The difference after applying the mask at the i-th sampling point, unit: V

[0124]

[0125] In the formula: D’ smooth,i——The value of the mask data of the i-th sampling point after smoothing, in V

[0126] N'——The window size of the secondary smoothing process.

[0127] In statistics, it is common practice to use multiples of the standard deviation to set the threshold (such as 2 times the standard deviation). This can filter out noise while retaining most of the data signal. If the value is too small, the calculation may increase, and if the value is too large, the waveform will be distorted. Therefore, after weighing the calculation time and waveform fidelity, the preset coefficient K of the energy threshold is 2. From a macro perspective, the value of the smoothing window depends on the number of "burrs" in the waveform. The more "burrs" there are, the larger the smoothing window value should be. Therefore, the choice of window size depends on the characteristics of the data and the required degree of smoothing. Larger windows can provide smoother results, but will reduce the sensitivity and time resolution of the data. Conversely, smaller windows can maintain higher sensitivity, but may not be able to effectively reduce noise. Therefore, through initial trial and error experiments, weighing the calculation time, the window size N' of the secondary smoothing process was set to 25. The pulse difference after energy mask smoothing is as follows Figure 10 shown.

[0128] After processing the pulses on both sides of the seed guide belt based on the BPV-EMSA algorithm, the noise and random fluctuations of the original pulses are reduced, making the data smoother and more stable, highlighting the main trends and patterns of the deviation pulse data, and suppressing transient pulse interference, which can improve the interpretability of the data and the accuracy of the analysis, and can obtain the characteristics of the seed pulse, that is, the unbalanced pulse, which can enable the monitoring system to accurately capture the seeds.

[0129] 4 Experimental results and analysis

[0130] In order to verify the accuracy of the pulse deviation captured by the monitoring system due to the passage of seeds when the belt-type high-speed seed guide device is operating, and thus complete the monitoring of parameters such as seeding amount, qualified rate, missed seeding rate and reseeding rate, the monitoring system accuracy test and monitoring system performance evaluation test were carried out to verify the feasibility and advancement of the monitoring system.

[0131] 4.1 Test materials

[0132] The test material was the "Demeya No. 1" corn seeds bred by KWS Seed Co., Ltd. of Germany. The seeds were screened manually and the purity of the test seeds reached 100%, the moisture content was (12.1±0.11)%, the thousand-grain mass was (295.55±0.48)g, and the average density was 1.228g / cm 3 The seed shape is "horse tooth" shape. 300 corn seeds were selected for size measurement. The average length was 9.33mm, the average width was 7.41mm, and the average thickness was 4.11mm.

[0133] 4.2 Monitoring System Accuracy Test

[0134] The accuracy test of the monitoring system for the belt-type high-speed seed guiding device is to evaluate the monitoring accuracy of the system and its actual application potential. First, it is necessary to verify the monitoring accuracy of the system for the number of discharged seeds, so as to obtain the development value of its subsequent seed guiding performance evaluation system. The performance test of the monitoring system for the belt-type high-speed seed guiding device was carried out on December 4, 2023 in the sowing equipment laboratory of the College of Engineering, Heilongjiang Bayi Agricultural University. A seed receiving box was placed at the end of the belt-type high-speed seed guiding device to facilitate the subsequent test to count the actual number of discharged seeds and recycle them. The test collected the monitoring seeding rates of the system under five groups of different theoretical operating speeds of 12, 13, 14, 15, and 16 km / h. Using 300 corn seeds measured by the material as the test materials, each group of tests was repeated 3 times, and the results were averaged. By comparing the relative error of the monitoring seeding rate compared to the actual seeding rate, the accuracy and reliability of the monitoring system were judged. The results of the seeding rate monitoring accuracy test are shown in Table 1. The specific parameter indicators of each seeding rate monitoring are as Figure 11 shown.

[0135] Table 1 Results of the Seeding Rate Monitoring Accuracy Test

[0136]

[0137] As can be seen from Table 1 and Figure 8 it can be seen that the two control items were tested three times under five groups of operating speeds. The accuracy test of the monitoring system shows that the monitoring accuracy under different operating speeds is higher than 95.9%, the highest accuracy is 97.65%, and the lowest accuracy is 95.99%. And through data analysis, it is found that the monitoring accuracy of the system is not affected by the sowing operation speed, and the relatively high monitoring accuracy can prove that the system can accurately and reliably collect the pulse changes of the seeds passing through the monitoring point.

[0138] 4.3 Monitoring System Performance Evaluation Test

[0139] The performance evaluation test of the monitoring system for the belt-type high-speed seed guiding device is to evaluate the actual operating performance of the system. The research needs to evaluate the error range between the missed seeding rate, the multiple seeding rate, the qualified rate output by the monitoring system and the actual seed metering performance parameters, so as to comprehensively evaluate the seed metering performance evaluation ability of the research system. The performance evaluation test of the monitoring system for the belt-type high-speed seed guiding device was carried out on December 15, 2023 in the sowing equipment laboratory of the College of Engineering, Heilongjiang Bayi Agricultural University. The JPS-16 computer vision seed metering performance detection test bench

[31] It is a high-precision sowing performance test bench developed by Heilongjiang Agricultural Machinery Engineering Science Research Institute. It is an important test equipment for exploring the performance of seeding devices in China. The theoretical plant spacing is set to 25cm. The test collects the qualified rate, missed seeding rate, and reseeding rate of the monitoring system at five different operating speeds of 12, 13, 14, 15, and 16km / h, and compares them with the evaluation indicators output by the JPS-16 computer vision seeding performance test bench. For each group, 300 consecutive plant spacings are taken in the middle, and the test is repeated 3 times at each operating speed, and the results are averaged. The performance evaluation test of the monitoring system is as follows. Figure 12 The test results are shown in Table 2.

[0140] Table 2 Monitoring system performance evaluation test results

[0141]

[0142] As shown in Table 2, the average monitoring error of seeding qualified rate is 2.00%, the average monitoring error of seeding missed rate is 1.45%, and the average monitoring error of seeding reseeding rate is 0.56%; the relative error of seeding qualified rate is not more than 2.23%, the relative error of seeding missed rate is not more than 1.78%, and the relative error of seeding reseeding rate is not more than 1.00%. It can be seen from the test data that with the increase of operation speed and the judgment of the reseeding rate of the monitoring system is not affected by the operation speed, some random interference with large amplitude may occur accidentally during the monitoring process, which will cause the algorithm to still have characteristic pulses outside the seed grain after masking, which will be mistakenly counted as reseeding; in addition, it may happen that the characteristics of the seed grain pulse are not obvious after processing, and it is mistakenly counted as missed seeding. The increase of interference amplitude and waveform burr amplitude may cause statistical errors of missed seeding and reseeding. Further in-depth research on improved algorithms is still needed for this part of the phenomenon in the future.

[0143] 5 Conclusion

[0144] (1) A monitoring method for a belt-type high-speed seed guide device was proposed. The double-side pulse comparison method was studied by using the seed feeding characteristics of the belt-type high-speed seed guide device. The hardware circuit and software flow of the monitoring module of the belt-type high-speed seed guide device were independently designed, and the layout of the monitoring module and the hardware and software design were clarified. The problem that the pulse changes could not be distinguished when the seed guide belt support and the seed grains passed through the monitoring point during the seed guidance process, making it difficult to monitor the seed guidance performance, was solved.

[0145] (2) In view of the current problem of interference in pulse waveform and inability to accurately identify the deviation pulse generated when the seed grain passes through, a belt-type high-speed seed guide device monitoring algorithm based on BPV-EMSA is proposed. The noise and random fluctuations of the original pulse are reduced, making the data smoother and more stable, highlighting the main trends and patterns of the deviation pulse data, and suppressing transient pulse interference, which can improve the interpretability of the data and the accuracy of the analysis.

[0146] (3) The test results show that the monitoring accuracy of the designed belt-type high-speed seed guide device monitoring system is maintained above 95.9% at different operating speeds, with the highest accuracy being 97.65% and the lowest accuracy being 95.99%. By analyzing the data, it is found that the system monitoring accuracy is not affected by the sowing operation speed, proving that the system can accurately collect the pulse changes of the seeds passing through the monitoring points. The average monitoring error of the sowing qualified rate is 2.00%, the average monitoring error of the sowing missed rate is 1.45%, and the average monitoring error of the sowing reseeding rate is 0.56%. The relative error of the sowing qualified rate is no more than 2.23%, the relative error of the sowing missed rate is no more than 1.78%, and the relative error of the sowing reseeding rate is no more than 1.00%. It can be considered for use in actual production operations of the seed drill after further field tests.

[0147] The present invention proposes a belt-type high-speed seed guiding device seed guiding performance monitoring algorithm (method) which is the underlying technical core of the present invention, and various products can be derived based on the algorithm.

[0148] Based on the algorithm (method) proposed in the present invention, a belt-type high-speed seed guide device seed guiding performance monitoring system is developed using a programming language. The system has a program module corresponding to the method steps of the above-mentioned technical solution, and executes the steps in the above-mentioned belt-type high-speed seed guide device seed guiding performance monitoring method during operation.

[0149] The computer program of the developed system (software) is stored on a computer-readable storage medium, and the computer program is configured to implement the steps of the above-mentioned method for monitoring the seed guiding performance of a belt-type high-speed seed guiding device when called by a processor, that is, the present invention is materialized on a carrier to become a computer program product.

[0150] The present invention also provides a photoelectric monitoring device, the photoelectric monitoring device includes at least one processor, and a memory connected to the at least one processor in communication, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned method for monitoring the seed guiding performance of a belt-type high-speed seed guiding device. The photoelectric monitoring device is used as a terminal intelligent product applied by the present invention, and is applied to the seed guiding performance monitoring of the belt-type high-speed seed guiding device.

[0151] The various embodiments of the systems and techniques described herein can be implemented in digital electronic circuitry, integrated circuit systems, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from a storage system, at least one input device, and at least one output device, and transmits the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0152] The computational programs (also referred to as programs, software, software applications, or code) in the present invention include machine instructions for a programmable processor and can implement these computational programs using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. As used herein, the terms “machine-readable medium” and “computer-readable medium” refer to any computer program product, device, and / or apparatus (e.g., magnetic disks, optical disks, memory, programmable logic device PLD) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term “machine-readable signal” refers to any signal for providing machine instructions and / or data to a programmable processor.

[0153] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in this application can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved, all within the scope of protection of the present invention.

[0154] The list of references cited in the present invention is as follows:

[0155] [1] Yang Li, Yan Bingxin, Zhang Dongxing, et al. Research progress on precision planting technology of maize [J]. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(11): 38 - 48.

[0156] Yang Li, Yan Bingxin, Zhang Dongxing, et al. Research progress on precision planting technology of maize [J]. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(11): 38 - 48. (in Chinese)

[0157] [2] Yuan Yanwei, Bai Huijuan, Fang Xianfa, et al. Research progress in corn sowing and measurement and control technology [J]. Transactions of the Chinese Society of Agricultural Machinery, 2018, 49(9): 1-18.

[0158] YUAN YANwei,BAI Huijuan,FANG Xianfa,et al.Research progress on maizeseeding and its measurement and control technology[J].Transactions of theChinese Society for Agricultural Machinery,2018,49(9):1-18.(in Chinese)

[0159] [3] Zheng Juan, Liao Yitao, Liao Qingxi, et al. Research status analysis and trend outlook of seeding technology for planters[J]. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(24): 1-13.

[0160] ZHENG Jun,LIAO Yitao,LIAO Qingxi,et al.Trend analysis and prospects of seed metering technologies[J].Transactions of the CSAE,2022,38(24):1-13.(in Chinese)

[0161] [4] Du Zhaohui, He Xiantao, Yang Li, et al. Research progress in precision variable seeding technology and equipment for corn[J]. Transactions of the Chinese Society of Agricultural Engineering, 2023, 39(9): 1-16.

[0162] DU Zhaohui,HE Xiantao,YANG Li,et al.Research progress on precisionvariable-rate seeding technology and equipment for maize[J].Transactions of the CSAE,2023,39(9):1-16.(in Chinese)

[0163] [5] Liao Yitao, Li Chengliang, Liao Qingxi, et al. Research progress analysis of seeding machine seed guidance technology and devices[J]. Transactions of the Chinese Society of Agricultural Machinery, 2020, 51(12): 1-14.

[0164] LIAO Yitao, LI Chengliang, LIAO Qingxi, et al. Research progress of seed - guiding technology and device of planter[J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(12): 1 - 14. (in Chinese)

[0165] [6] CHEN Xuegeng, ZHONG Luming. Design and test on belt - type seed delivery of air - suction metering device[J]. Transactions of the CSAE, 2012, 28(22): 8 - 15.

[0166] CHEN Xuegeng, ZHONG Luming. Design and test on belt - type seed delivery of air - suction metering device[J]. Transactions of the CSAE, 2012, 28(22): 8 - 15. (in Chinese)

[0167] [7] KANG Jianming, WEN Haojun, WANG Shiguo, et al. Experimental study on impact of belt type conductor delivery on seeding uniformity[J]. Journal of Chinese Agricultural Mechanization, 2015, 36(5): 42 - 45.

[0168] KANG Jianming, WEN Haojun, WANG Shiguo, et al. Experimental study on impact of belt type conductor delivery on seeding uniformity[J]. Journal of Chinese Agricultural Mechanization, 2015, 36(5): 42 - 45. (in Chinese)

[0169] [8] CHEN Chen. Structural design and performance test of belt - type high - speed seed - guiding device for precision planter[D]. Inner Mongolia Agricultural University, 2016.

[0170] CHEN Chen. Structure Design and performance test research of the beltseed guide device of precision planter[D]. Inner Mongolia Agricultural University, 2016. (in Chinese)

[0171] [9] Liu Quanwei. Design and experiment of seed precise delivery mechanism for high - speed planter[D]. China Agriculture University, 2017.

[0172] LIU Quanwei. Design and experiment of seed precise delivery mechanism for high - speed planter[D]. China Agriculture University, 2017. (in Chinese)

[0173]

[10] Ma Huanfei. The software development of visual seed flow detection based on VC++[D]. Shanxi Agricultural University, 2015.

[0174] MA Huanfei. The software development of visual seed flow detection based on VC++[D]. Shanxi Agricultural University, 2015. (in Chinese)

[0175]

[11] NAVID H, EBRAHIMIAN S, GASSEMZADEH H, et al. Laboratory evaluation of seed metering device using image processing method[J]. Australian Journal of Agricultural Engineering, 2011.1 - 4.

[0176]

[12] MANGUSD L, SHARDA A, FLIPPO D, et al. Development of high-speed camera hardware and software package to evaluate real-time electric seed meter accuracy of a variable rate planter. [J]. Computers and Electronics in Agriculture, 2017, 142: 314-325.

[0177]

[13] TAGHINEZHAD J, ALIMARDANI R, JAFARY A. Design a capacitive sensor for rapid monitoring of seed rate of sugarcane planter [J]. Agricultural Engineering International: The CIGR Journal, Agricultural Engineering International: The CIGR Journal, 2013, 15(4): 23-29.

[0178]

[14] Zhu Liang, Wang Guanping, Sun Wei, et al. Development of potato seed-metering state monitoring system based on space capacitance sensors [J]. Transactions of the Chinese Society for Agricultural Engineering, 2021, 37(20): 34-43.

[0179] ZHU Liang, WANG Guanping, SUN Wei, et al. Development of potato seed-metering state monitoring system based on space capacitance sensors [J]. Transactions of the Chinese Society for Agricultural Engineering, 2021, 37(20): 34-43. (in Chinese)

[0180]

[15] Liu Kun. Research on the detection mechanism and method of the metering performance of corn precision seeding device by capacitance method [D]. Heilongjiang Bayi Agricultural University, 2020.

[0181] LIU Kun. Detection mechanism and methodology of seeding performance of corn precision seeding device based on capacitive method[D]. Heilongjiang Bayi Agricultural University, 2020. (in Chinese)

[0182]

[16] Huang Dongyan, Zhu Longtu, Jia Honglei, et al. Remote corn seeding quality monitoring system based on GPS and GPRS[J]. Transactions of the Chinese Society for Agricultural Engineering, 2016, 32(6): 162 - 168.

[0183] HUANG Dongyan, ZHU Longtu, JIA Honglei, et al. Remote monitoring system for corn seeding quality based on GPS and GPRS.[J]. Transactions of the Chinese Society for Agricultural Engineering, 2016, 32(6): 162 - 168. (in Chinese)

[0184]

[17] Wang Jinwu, Zhang Zhao, Wang Fei, et al. Design and experiment of monitoring system for rice hill - direct - seeding based on piezoelectric impact method[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(6): 74 - 84 + 99.

[0185] WANG Jinwu, ZHANG Zhao, WANG Fei, et al. Design and experiment of monitoring system for rice hill - direct - seeding based on piezoelectric impact method[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(6): 74 - 84 + 99. (in Chinese)

[0186]

[18] Zhao Bo, Fan Xueqian, Zhou Liming, et al. Design and experiment of piezoelectric flow sensor for pneumatic seeder[J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(8): 55 - 61.

[0187] Zhao Bo, Fan Xueqian, Zhou Liming, et al. Design and test of piezoelectric flow sensor for pneumatic seeder[J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(8): 55 - 61. (in Chinese)

[0188]

[19] Al-Mallahi A A, Kataoka T. Application of fibre sensor in grain drill to estimate seed flow under field operational conditions[J]. Computers and Electronics in Agriculture, 2016, 121: 412 - 419.

[0189]

[20] Okopnik D L, Falater R. Usage of the DFRobot RB-DFR-49 infrared sensor to detect maize seed passage on a conveyor belt[J]. Computers and Electronics in Agriculture, 2014, 102: 106 - 111.

[0190]

[21] Xie Chunji, Yang Li, Zhang Dongxing, et al. Seeding parameter monitoring method based on laser sensors[J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(3): 140 - 146.

[0191] Xie Chunji, Yang Li, Zhang Dongxing, et al. Seeding parameter monitoring method based on laser sensors[J]. Transactions of the CSAE, 2021, 37(3): 140 - 146. (in Chinese)

[0192]

[22] LEEMANS V, DESTAIN M F. A computer-vision based precision seeddrill guidance assistance[J]. Computers and Electronics in Agriculture, 2007, 59(1-2): 1-12.

[0193]

[23] CHEN Jianguo, LI Yanming, QIN Chenjing, et al. Design and experiment of precision detecting system for wheat-planter seeding quantity[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(1): 66-74.

[0194] CHEN Jianguo, LI Yanming, Qin Chenjing, et al. Design and experiment of precision detecting system for wheat-planter seeding quantity[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(1): 66-74. (in Chinese)

[0195]

[24] ZHOU Liming, ZHANG Xiaochao. Monitor system of precision seeder based on capacitive sensors[J]. Journal of Agricultural Mechanization Research, 2009, 31(11): 37-39.

[0196] ZHOU Liming, ZHANG Xiaochao. Monitor system of precision seeder based on capacitive sensors[J]. Journal of Agricultural Mechanization Research, 2009, 31(11): 37-39.

[0197]

[25] ZHOU Liming, ZHANG Xiaochao, YUAN Yanwei. Design of capacitive metering device for wheat planter[J]. Transactions of the Chinese Society of Agricultural Engineering, 2010, 26(10): 99-103.

[0198] ZHOU Liming, ZHANG Xiaochao, YUAN Yanwei. Design of capacitance seedrate sensor of wheat planter[J]. Transactions of the CSAE, 2010, 26(10): 99 - 103..(in Chinese)

[0199]

[26] Zhou Liming, Wang Shumao, Zhang Xiaochao, et al. Seed monitoring system for corn planter based on capacitance signal[J]. Transactions of the CSAE, 2012, 28(13): 16 - 21.

[0200] ZHOU Liming, WANG Shumao, ZHANG Xiaochao, et al. Seed monitoring system for corn planter based on capacitance signal[J]. Transactions of the CSAE, 2012, 28(13): 16 - 21.(in Chinese)

[0201]

[27] Huang Dongyan, Jia Honglei, Qi Yue, et al. Seeding monitor system for planter based on polyvinylidence fluoride piezoelectric film[J]. Transactions of the CSAE, 2013, 29(23): 15 - 22.

[0202] HUANG Dongyan, JIA Honglei, QI Yue, et al. Seeding monitor system for planter based on polyvinylidence fluoride piezoelectric film[J]. Transactions of the CSAE, 2013, 29(23): 15 - 22.(in Chinese)

[0203]

[28] Ji Chao, Chen Xuegeng, Chen Jincheng, et al. Seed dropping quality monitoring system for no - tillage precision planter of corn[J]. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(8): 1 - 6.

[0204] JI Chao, CHEN Xuegeng, CHEN Jincheng, et al. Monitoring system for working performance of no-tillage corn precision seeder[J]. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(8): 1-6. (in Chinese)

[0205]

[29] CAY A, KOCABIYIK H, KARAASLAN B, et al. Development of an opto - electronic measurement system for planter laboratory tests[J]. Measurement, 2017, 102: 90 - 95.

[0206]

[30] Ma Chengcheng, Yi Shujuan, Tao Guixiang, et al. Mechanism analysis and parameter optimization of corn seeds receiving by rotating clamp of belt - type high - speed seed guiding device[J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(7): 134 - 143.

[0207] MA Chengcheng, YI Shujuan, TAO Guixiang, et al. Mechanism analysis and parameter optimization of corn seeds receiving by rotating clamp of belt - type high - speed seed guiding device[J]. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54(7): 134 - 143. (in Chinese)

[0208]

[31] Wu Zequan, Liu Junjie, Yang Xu. Design of the test bench for JPS - 16 seeder[J]. Journal of Agricultural Mechanization Research, 2011, 33(10): 59 - 62.

[0209] WU Zequan, LIU Junjie, YANG Xu. The Design of JPS-16 Seeder Test-bed[J]. Journal of Agricultural Mechanization Research, 2011, 33(10): 59 - 62. (in Chinese)

Claims

1. A method for monitoring the seed guiding performance of a belt-type high-speed seed guiding device, characterized in that, The monitoring method is based on bilateral pulse value analysis and energy mask smoothing algorithm to extract the pulse characteristics of belt-guided seeds. The process is as follows: First, smooth the pulses on both sides of the seed guide belt with burrs respectively to reduce the noise and random fluctuations of the original pulses and make the data smoother and more stable. The pulses on both sides include the seed delivery side pulse and the rotary side pulse. Then, subtract the pulses smoothed by the two channels to obtain the grain feature pulse after initially highlighting the seed characteristics, which is used to accurately extract the characteristics of the imbalance pulse caused when the seeds pass through the monitoring point. Then, according to the energy mask smoothing algorithm, perform secondary smoothing on the grain feature pulse obtained after the previous steps to obtain a pulse that only highlights the seed characteristics. In the pulse, the seed feature pulse is an imbalance fluctuation, and the others are balance fluctuations. When another adjacent imbalance fluctuation appears, calculate the time interval between two adjacent imbalance fluctuations, that is, the adjacent seed falling time interval T F ; After determining the balance fluctuation, continuously collect the pulse value data on the rotary side, and calculate the time interval T between two adjacent light-shielding times of the supporting piece on the rotary side P 、Calculate T P After that, calculate the wire speed V of the seed guiding belt. Through T F 、T P 、V, calculate the plant spacing X in real time, so as to realize the performance monitoring of the high-speed belt-type seed guiding Judge whether the pulse signals are equal. If so, perform balance fluctuation processing, collect the right-side pulse data, process the data through the algorithm, calculate the linear velocity V of the seed guide belt, and calculate the plant spacing X in real time. Otherwise, perform imbalance fluctuation processing, calculate the linear velocity V of the seed guide belt, and finally calculate the plant spacing X in real time. The calculation formula for the linear velocity V of the seed guide belt can be deduced by monitoring the time interval for the seed belt support piece to pass through as where S is the distance between two adjacent support pieces, with the unit of m. T P —— Time interval between two consecutive light-shielding events of the rotating side support piece, unit: s Therefore, the actual plant spacing value can be calculated without monitoring the speed of the driving motor of the belt-guided seed device. The calculation formula for the spacing between two adjacent seeds, that is, the plant spacing X, is where T F —— time interval between two pulse deviations, unit: s The monitoring system automatically stores the plant spacing value each time; let the theoretical plant spacing of the target crop be X T , the sowing spacing is P, if X∈[0,0.5X T ], it is counted as a replay. If the number of replays is P mult , then the replay rate I mult The calculation formula is If X ∈ [1.5X T , +∞], it is counted as missed broadcast. If the number of missed broadcasts is P miss , then the missed broadcast rate I miss is calculated by the formula If X ∈ [0.5X T , 1.5X T , it is counted as qualified. If the number of qualified times is P q , then the qualification rate I q The calculation formula is 2. The method for monitoring the seed guiding performance of a belt-type high-speed seed guiding device according to claim 1, characterized in that, The specific implementation process of the monitoring method is as follows: Smooth the original pulse values of the two channels to obtain new pulses. The calculation formula for smoothing the pulse values of the two channels is: where: B j —— The pulse value on the seed metering side at the j-th sampling point, unit: V; C j —— The rotational side pulse value at the j-th sampling point, unit: V; i, j are the sampling point indices B smooth,i —— The pulse value after smoothing the seed-planting side of the i-th sampling point, with the unit of V; C smooth,i —— The pulse value after smoothing processing on the rotating side of the i-th sampling point, with the unit of V; N is the smoothing window size The new pulse is a smoothed pulse based on the original pulse, which is achieved by averaging all the values within a window N of data points. The interference of "burrs" is filtered out in the smoothed pulses. Subtract the pulses smoothed by the two channels to obtain the pulse difference after highlighting the characteristics. The calculation process is shown in formula (8): D smooth,i = B smooth,i - C smooth,i (8) where D smooth,i —— the difference after bilateral pulse smoothing at the i-th sampling point, unit: V; Give the energy mask smoothing algorithm. Solve all interference waveforms by setting the energy threshold. Only when the signal energy exceeds this threshold will the signal be smoothed; highlight the signal part with large intensity and containing important information; make the seed pulse signal conform to this feature. Among them, the short-time energy refers to the sum of the signal energy within a short-time window, which is an index to measure the intensity of the signal within this time window. The calculation formula for short-time energy is: where E i —— the short-time energy value of the i-th sampling point; The energy threshold is a preset boundary used to judge whether the short-time energy of the signal is large enough, that is, a meaningful signal rather than noise. The setting of the energy threshold is used to determine the retention tendency of the data. The calculation formula for the energy threshold is θ = mean(E) × K (10) where: θ is the energy threshold, and K is the preset coefficient of the energy threshold. The mask signal is a binary signal representing the part of the original signal where the short-time energy exceeds the energy threshold. If the short-time energy of the signal exceeds the threshold, the mask signal is 1 at the corresponding time point, otherwise it is 0. Therefore, the mask signal can be used to "filter" the original signal and only retain the high-energy part. The calculation formula for the mask signal is Where: M i —— The masking signal at the i-th sampling point; D masked,i = D smooth,i × M i (12) Where: D masked,i —— The difference after applying the mask at the i-th sampling point, in volts; where: D’ smooth,i —— the value after re - smoothing the mask data of the i - th sampling point, unit: V; N' is the window size for secondary smoothing processing.

3. A method for monitoring the seed guiding performance of a belt-type high-speed seed guiding device according to claim 2, characterized in that, The value of the smoothing window size N is 50.

4. A method for monitoring the seed guiding performance of a belt-type high-speed seed guiding device according to claim 1, characterized in that: Before the step of synchronously collecting the pulse data of the photoelectric opposite-irradiation tubes on both sides, the steps of initializing the single-chip microcomputer and enabling the external interrupt are executed.

5. A performance monitoring system for a seed guiding device of a belt-type high-speed seed guiding device, characterized in that: The system has program modules corresponding to the steps of any one of the above claims 1-4, and executes the steps in the above-mentioned method for monitoring the seeding performance of the high-speed belt seeding device when running.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is configured to implement the steps of the method for monitoring the seeding performance of the high-speed belt seeding device according to any one of claims 1-4 when called by a processor.

7. An optoelectronic monitoring device, characterized in that: The photoelectric monitoring device includes at least one processor and a memory communicatively connected to the at least one processor. Among them, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for monitoring the seeding performance of the high-speed belt seeding device according to any one of claims 1-4.

8. The optoelectronic monitoring device according to claim 7, characterized in that, The processor is the main controller, and it also includes a main controller circuit connected to the main controller and a main controller peripheral circuit. The main controller selects the STM32F103RBT6 single-chip microcomputer. The main controller peripheral circuit includes a photoelectric emission circuit, a seed-planting-side photoelectric receiving circuit a, and a rotary-side photoelectric receiving circuit b. The photoelectric emission circuit includes two groups of photoelectric emission tube groups. One group is the photoelectric emission tube group on the seed-planting side, and the other group is the photoelectric emission tube group on the rotary side. Each group of photoelectric emission tube groups includes three photodiodes packaged side by side; the seed-planting-side photoelectric receiving circuit a includes three photodiodes packaged side by side; the rotary-side photoelectric receiving circuit b includes three photodiodes packaged side by side; to adapt to the cavity width of the high-speed belt seeding device and achieve blind-zone-free monitoring.

9. A belt-type high-speed seed guiding device, characterized in that, The high-speed belt seeding device includes the photoelectric monitoring device according to claim 7 or 8.

Images