A digital breeding method for vegetables

Abstract

The present application belongs to the technical field of digital breeding of vegetables, and discloses a digital breeding method of vegetables. The method comprises the following steps: collecting images of vegetable seeds, identifying excellent shapes of the vegetable seeds, weighing the vegetable seeds, and screening excellent vegetable seeds according to the identification results and the weighing structure; monitoring the pH value, nitrogen, phosphorus, potassium, conductivity, humidity and temperature of the soil for breeding the vegetable seeds; monitoring the air temperature, light, wind direction and environmental data for breeding the vegetable seeds; quantitatively watering the vegetable seeds for breeding; and monitoring the excellent germination value for breeding the vegetable seeds. The present application obtains objective self parameters and soil variable parameters of the vegetable at the current time by the quantitative watering method for breeding the vegetable seeds, and performs watering according to the water demand predicted by the pre-trained RBF neural network algorithm, so that the vegetable can be accurately and quantitatively watered, thereby achieving the purposes of saving water resources and increasing the yield of the vegetable.

Description

Technical Field

[0001] This invention belongs to the field of digital vegetable breeding technology, and in particular relates to a digital vegetable breeding method. Background Technology

[0002] Vegetable breeding is the technology of selecting and propagating superior plant varieties or improving the genetic characteristics of vegetables to cultivate high-yield and high-quality varieties; it is also known as vegetable variety improvement. Vegetable breeding is a technology for improving the genetic patterns of economic traits in vegetables and is an important component of crop breeding. When determining breeding objectives for vegetables, in addition to considering the general requirements of crop breeding, the following characteristics are often emphasized: ① The variety of vegetables far exceeds that of food crops and other economic crops; people's consumption needs are also more complex and diverse. This necessitates increasing the variety of vegetables in breeding. At the same time, most vegetable products have a high water content and are not resistant to storage and transportation, requiring breeding to provide varieties with different maturity periods (early, mid, and late maturing), resistance to storage and transportation, and suitability for various processing methods, in order to provide a reasonable variety combination for vegetable production and ensure a balanced year-round supply and export trade needs. ② Vegetables, as a by-product of food, are a major source of various vitamins, amino acids, minerals, carbohydrates, and other nutrients for humans; therefore, quality breeding is particularly important for vegetable crops. ③ Vegetables are mostly cultivated in specialized fields, with concentrated planting in the same area year after year, which easily leads to the accumulation of pathogens. Long-term, high-volume application of pesticides has screened pathogens and increased their resistance. Therefore, disease-resistant breeding is not only important for reducing losses and improving quality, but also for reducing pesticide use, mitigating environmental pollution, and lowering production costs. However, most existing digital vegetable breeding methods rely on farmers' experience to determine the amount of water needed, lacking scientific judgment and methods. Excessive or insufficient watering, or improper watering methods, not only lead to low water resource utilization but also reduce vegetable yield. Furthermore, the inability to monitor and control the growth environment data in real time during the breeding process results in suboptimal breeding conditions. Visual observation of vegetable germination during breeding lacks specific quantitative standards, leading to inaccurate results and significant subjective influence.

[0003] Based on the above analysis, the problems and shortcomings of the existing technology are as follows:

[0004] (1) Most of the existing digital breeding methods for vegetables rely on the experience of farmers to judge the amount of water to be poured. There is a lack of scientific watering judgment and watering methods. Too much or too little watering and unreasonable watering methods will not only reduce the utilization rate of water resources, but also reduce the yield of vegetables.

[0005] (2) The inability to control the growth environment data in real time during the vegetable breeding process results in the vegetable breeding environment not being in the best condition. The germination of vegetables is observed by the naked eye during the vegetable breeding process, but there are no specific quantitative standards, so the results are not accurate and are greatly affected by subjective factors. Summary of the Invention

[0006] To address the problems existing in the prior art, this invention provides a digital breeding method for vegetables.

[0007] This invention is implemented as follows: A digital breeding method for vegetables includes:

[0008] Step 1: Collect images of vegetable seeds, analyze the images using image processing techniques, extract the morphological, color, and texture features of the seeds, identify the superior appearance of the vegetable seeds based on the extracted features, weigh the vegetable seeds, and select superior vegetable seeds based on the identification results and weighing results.

[0009] Step two: Use a sensor network to monitor the pH, nitrogen, phosphorus, potassium, electrical conductivity, humidity, and temperature of the soil used for vegetable seed breeding; and use environmental monitoring equipment to monitor the air temperature, light intensity, and wind direction data for vegetable seed breeding.

[0010] Step 3: Quantitatively water the vegetable seeds and monitor the germination rate of the vegetables.

[0011] Furthermore, the method for quantitative watering in vegetable seed breeding is as follows:

[0012] (1) Configure the ultrasonic detector parameters and obtain the ultrasonic frequency data emitted by the vegetables at the current moment through the ultrasonic detector parameters; perform noise reduction processing on the ultrasonic frequency data; determine whether the vegetables meet the watering conditions at the current moment based on the different ultrasonic frequency data emitted by the vegetables at different water content periods; if the watering conditions are met, obtain the objective self-parameters of the vegetables and the variable parameters of the soil at the current moment.

[0013] (2) Based on the objective self-parameters of the vegetables at the current moment and the variable parameters of the soil, the water requirement is predicted using a pre-trained RBF neural network algorithm; the vegetables are watered according to the predicted water requirement.

[0014] Furthermore, the objective intrinsic parameters of the vegetable at the current moment include at least one of the vegetable type and soil type; the variable parameters of the soil include at least one of the soil moisture content and soil temperature.

[0015] Furthermore, the determination of whether the vegetables meet the watering conditions at the current moment is achieved in the following way:

[0016] The first frequency difference is determined based on the average value of the ultrasonic frequency data of the vegetables at the current moment and the ultrasonic frequency data of the vegetables when they are dehydrated, which were collected in advance.

[0017] The second frequency difference is determined based on the average value of the ultrasonic frequency data of the vegetables at the current moment and the ultrasonic frequency data of the vegetables when they are not short of water.

[0018] If the first frequency difference is less than or equal to the second frequency difference, then the watering conditions are met.

[0019] Conversely, if the conditions for watering are not met, then it is determined that the conditions for watering are not met.

[0020] Furthermore, the training method for the RBF neural network includes:

[0021] The sample input sequence and the associated sample output sequence are divided into two parts: one part is used as the training data sequence and the other part is used as the test data sequence.

[0022] The RBF neural network algorithm is trained using the training data sequence to obtain the RBF neural network algorithm.

[0023] The trained RBF neural network algorithm was tested using the test data sequence.

[0024] Determine whether the accuracy of the test results is below 90%;

[0025] If so, the learning rate of the RBF neural network algorithm is modified, and the RBF neural network algorithm is retrained using the sample input sequence until the accuracy of the test is not less than 90%.

[0026] Furthermore, the method also includes:

[0027] After obtaining the predicted water requirement, the vegetables are watered precisely and quantitatively according to the predicted water requirement.

[0028] Furthermore, acquiring the ultrasonic frequency data emitted by the vegetables at the current moment includes:

[0029] Real-time acquisition of ultrasonic frequency data emitted by vegetables at the current moment;

[0030] Alternatively, it can receive a watering command, start watering in response to the command, and pause watering when the watering duration reaches the preset duration, while acquiring the ultrasonic frequency data emitted by the vegetables at the current moment.

[0031] Furthermore, the method for monitoring the germination quality of vegetable breeding is as follows:

[0032] 1) Construct a vegetable breeding database and store the collected breeding data into the vegetable breeding database; collect growth environment data of multiple vegetable breeds that have undergone the same pretreatment during the breeding process; regulate the growth environment data according to the pre-constructed vegetable breeding guidance parameter table;

[0033] 2) Collect growth images of multiple vegetable varieties during the breeding process; obtain growth status data of different vegetable varieties based on the collected growth images; evaluate the germination status of vegetables based on the growth status data and the data in the pre-constructed vegetable breeding guidance parameter table at intervals of breeding days, and calculate the germination excellence value of different vegetable varieties.

[0034] Furthermore, the method also includes the following steps:

[0035] Based on the calculated germination quality values ​​of vegetables, calculate the overall germination quality values ​​of different vegetable breeds;

[0036] The vegetable varieties with the highest overall germination quality were selected for planting.

[0037] Furthermore, the vegetables in each vegetable breeding group are placed along the same straight line, and growth images of the vegetables are collected perpendicular to the growth direction of the vegetable buds.

[0038] Methods for regulating growth environment data include:

[0039] Get the current number of breeding days;

[0040] Based on the pre-constructed vegetable breeding guidance parameter table, obtain the control range of growth environment data corresponding to the current breeding days;

[0041] Determine whether the collected current growth environment data is within the control range of the current growth environment data corresponding to the current breeding days. If so, there is no need to adjust the current growth environment data. Otherwise, adjust the current growth environment data so that the current growth environment data conforms to the control range of the current breeding days.

[0042] Methods for regulating growth environment data also include the following steps:

[0043] Obtain the data type of the growth environment that needs to be regulated and the values ​​that need to be regulated;

[0044] Send a control command to the growth environment data control device corresponding to the data type of growth environment that needs to be controlled. The control command carries the value that needs to be controlled.

[0045] The method for obtaining growth status data of different vegetable varieties includes the following sub-steps:

[0046] Based on a pre-trained vegetable sprout recognition model, vegetable sprout feature images are extracted from vegetable growth images;

[0047] Vegetable sprout growth status data are obtained from vegetable sprout feature images;

[0048] Based on a pre-trained vegetable sprout pathology recognition model, obtain the pathology data of vegetable sprouts in the feature images of vegetable sprouts;

[0049] The data on the growth status of vegetable sprouts include: the number of vegetable sprouts, the diameter of the vegetable sprout stem, the length of the vegetable sprout stem, the number of vegetable sprout leaves, and the area of ​​the vegetable sprout leaves.

[0050] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:

[0051] This invention utilizes a quantitative watering method for vegetable seed breeding to obtain the objective intrinsic parameters of vegetables and variable soil parameters at the current moment. Watering is then performed based on the water requirement predicted by a pre-trained RBF neural network algorithm, enabling precise and quantitative watering of vegetables, thereby saving water resources and increasing vegetable yield. Simultaneously, a method for monitoring the germination quality of vegetable seedlings allows for real-time adjustment of the growth environment data during the breeding process, ensuring the optimal breeding environment. Furthermore, by collecting germination images of vegetables, the germination status is accurately analyzed, improving the accuracy of the analysis results. Attached Figure Description

[0052] Figure 1 This is a flowchart of the digital breeding method for vegetables provided in an embodiment of the present invention;

[0053] Figure 2 This is a flowchart of a quantitative watering method for vegetable seed breeding provided in an embodiment of the present invention;

[0054] Figure 3 This is a flowchart of a method for monitoring the germination quality of vegetable breeding, provided in an embodiment of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0056] like Figure 1 As shown, the present invention provides a digital breeding method for vegetables, comprising the following steps:

[0057] S101: Collect vegetable seed images, analyze the vegetable seed images using image processing technology, extract the morphological features, color features, and texture features of the seeds, identify the superior appearance of the vegetable seeds based on the extracted features, weigh the vegetable seeds, and select superior vegetable seeds based on the identification results and weighing results.

[0058] S102 uses a sensor network to monitor the pH, nitrogen, phosphorus, potassium, electrical conductivity, humidity, and temperature of the soil used for vegetable seed breeding; and uses environmental monitoring equipment to monitor the air temperature, light intensity, and wind direction data for vegetable seed breeding.

[0059] S103 involves quantitative watering for vegetable seed breeding and monitoring the germination quality of vegetable seeds.

[0060] like Figure 2 As shown, the quantitative watering method for vegetable seed breeding provided by this invention is as follows:

[0061] S201, Configure the ultrasonic detector parameters, obtain the ultrasonic frequency data emitted by the vegetables at the current moment through the ultrasonic detector parameters; perform noise reduction processing on the ultrasonic frequency data; determine whether the vegetables meet the watering conditions at the current moment based on the different ultrasonic frequency data emitted by the vegetables at different water content periods; if the watering conditions are met, obtain the objective self-parameters of the vegetables and the variable parameters of the soil at the current moment.

[0062] S202, based on the objective parameters of the vegetables at the current moment and the variable parameters of the soil, a pre-trained RBF neural network algorithm is used to predict the water requirement; the vegetables are watered according to the predicted water requirement.

[0063] The objective parameters of the vegetables at the current moment provided by the present invention include at least one of the following: vegetable type and soil type; the variable parameters of the soil include at least one of the following: soil moisture content and soil temperature.

[0064] The present invention provides a method for determining whether vegetables meet the watering requirements at the current moment:

[0065] The first frequency difference is determined based on the average value of the ultrasonic frequency data of the vegetables at the current moment and the ultrasonic frequency data of the vegetables when they are dehydrated, which were collected in advance.

[0066] The second frequency difference is determined based on the average value of the ultrasonic frequency data of the vegetables at the current moment and the ultrasonic frequency data of the vegetables when they are not short of water.

[0067] If the first frequency difference is less than or equal to the second frequency difference, then the watering conditions are met.

[0068] Conversely, if the conditions for watering are not met, then it is determined that the conditions for watering are not met.

[0069] The training method for the RBF neural network provided by this invention includes:

[0070] The sample input sequence and the associated sample output sequence are divided into two parts: one part is used as the training data sequence and the other part is used as the test data sequence.

[0071] The RBF neural network algorithm is trained using the training data sequence to obtain the RBF neural network algorithm.

[0072] The trained RBF neural network algorithm was tested using the test data sequence.

[0073] Determine whether the accuracy of the test results is below 90%;

[0074] If so, the learning rate of the RBF neural network algorithm is modified, and the RBF neural network algorithm is retrained using the sample input sequence until the accuracy of the test is not less than 90%.

[0075] The method provided by this invention also includes:

[0076] After obtaining the predicted water requirement, the vegetables are watered precisely and quantitatively according to the predicted water requirement.

[0077] The present invention provides data for obtaining the ultrasonic frequency emitted by vegetables at the current moment, including:

[0078] Real-time acquisition of ultrasonic frequency data emitted by vegetables at the current moment;

[0079] Alternatively, it can receive a watering command, start watering in response to the command, and pause watering when the watering duration reaches the preset duration, while acquiring the ultrasonic frequency data emitted by the vegetables at the current moment.

[0080] like Figure 3 As shown, the method for monitoring the germination quality of vegetable breeding provided by this invention is as follows:

[0081] S301, Construct a vegetable breeding database and store the collected breeding data into the vegetable breeding database; collect growth environment data of multiple vegetable breeds that have undergone the same pretreatment during the breeding process; regulate the growth environment data according to the pre-constructed vegetable breeding guidance parameter table;

[0082] S302: Collect growth images of multiple vegetable varieties during the breeding process; obtain growth status data of different vegetable varieties based on the collected growth images; evaluate the germination status of vegetables based on the growth status data and the data in the pre-constructed vegetable breeding guidance parameter table at intervals of breeding days, and calculate the germination excellence value of different vegetable varieties.

[0083] The method provided by this invention further includes the following steps:

[0084] Based on the calculated germination quality values ​​of vegetables, calculate the overall germination quality values ​​of different vegetable breeds;

[0085] The vegetable varieties with the highest overall germination quality were selected for planting.

[0086] In this invention, the vegetables in each vegetable breeding group are placed along the same straight line, and growth images of the vegetables are collected perpendicular to the growth direction of the vegetable buds.

[0087] Methods for regulating growth environment data include:

[0088] Get the current number of breeding days;

[0089] Based on the pre-constructed vegetable breeding guidance parameter table, obtain the control range of growth environment data corresponding to the current breeding days;

[0090] Determine whether the collected current growth environment data is within the control range of the current growth environment data corresponding to the current breeding days. If so, there is no need to adjust the current growth environment data. Otherwise, adjust the current growth environment data so that the current growth environment data conforms to the control range of the current breeding days.

[0091] Methods for regulating growth environment data also include the following steps:

[0092] Obtain the data type of the growth environment that needs to be regulated and the values ​​that need to be regulated;

[0093] Send a control command to the growth environment data control device corresponding to the data type of growth environment that needs to be controlled. The control command carries the value that needs to be controlled.

[0094] The method for obtaining growth status data of different vegetable varieties includes the following sub-steps:

[0095] Based on a pre-trained vegetable sprout recognition model, vegetable sprout feature images are extracted from vegetable growth images;

[0096] Vegetable sprout growth status data are obtained from vegetable sprout feature images;

[0097] Based on a pre-trained vegetable sprout pathology recognition model, obtain the pathology data of vegetable sprouts in the feature images of vegetable sprouts;

[0098] The data on the growth status of vegetable sprouts include: the number of vegetable sprouts, the diameter of the vegetable sprout stem, the length of the vegetable sprout stem, the number of vegetable sprout leaves, and the area of ​​the vegetable sprout leaves.

[0099] This invention applies a quantitative watering method for vegetable seed breeding to obtain the objective intrinsic parameters of the vegetables and the variable parameters of the soil at the current moment. Watering is then performed based on the water requirement predicted by a pre-trained RBF neural network algorithm, enabling precise and quantitative watering of vegetables, thereby saving water resources and increasing vegetable yield. Simultaneously, a method for monitoring the germination quality of vegetable seedlings allows for real-time adjustment of the growth environment data during the breeding process, ensuring the optimal breeding environment. Furthermore, by collecting germination images of the vegetables, the germination status is accurately analyzed, improving the accuracy of the analysis results.

[0100] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0101] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A digital breeding method for vegetables, characterized in that, The digital breeding method for vegetables includes the following steps: Step 1: Collect images of vegetable seeds, analyze the images using image processing technology, extract the morphological, color, and texture features of the seeds, identify the superior appearance of the vegetable seeds based on the extracted features, weigh the vegetable seeds, and select superior vegetable seeds based on the identification results and weighing results. Step two: Use a sensor network to monitor the pH, nitrogen, phosphorus, potassium, electrical conductivity, humidity, and temperature of the soil used for vegetable seed breeding; and use environmental monitoring equipment to monitor the air temperature, light intensity, and wind direction data for vegetable seed breeding. Step 3: Quantitatively water the vegetable seeds and monitor the germination rate of the vegetables. The method for quantitative watering in vegetable seed breeding is as follows: (1) Configure the ultrasonic detector parameters and obtain the ultrasonic frequency data emitted by the vegetables at the current moment through the ultrasonic detector parameters; perform noise reduction processing on the ultrasonic frequency data; determine whether the vegetables meet the watering conditions at the current moment based on the different ultrasonic frequency data emitted by the vegetables at different water content periods; if the watering conditions are met, obtain the objective self-parameters of the vegetables and the variable parameters of the soil at the current moment. (2) Based on the objective self-parameters of the vegetables at the current moment and the variable parameters of the soil, the water requirement is predicted using a pre-trained RBF neural network algorithm; the vegetables are watered according to the predicted water requirement.

2. The vegetable digital breeding method as described in claim 1, characterized in that, The objective parameters of the vegetable at the current moment include at least one of the vegetable type and soil type; the variable parameters of the soil include at least one of the soil moisture content and soil temperature.

3. The vegetable digital breeding method as described in claim 1, characterized in that, The determination of whether the vegetables meet the watering requirements at the current moment is achieved in the following way: The first frequency difference is determined based on the average value of the ultrasonic frequency data of the vegetables at the current moment and the ultrasonic frequency data of the vegetables when they are dehydrated, which were collected in advance. The second frequency difference is determined based on the average value of the ultrasonic frequency data of the vegetables at the current moment and the ultrasonic frequency data of the vegetables when they are not short of water. If the first frequency difference is less than or equal to the second frequency difference, then the watering conditions are met. Conversely, if the conditions for watering are not met, then it is determined that the conditions for watering are not met.

4. The digital breeding method for vegetables as described in claim 1, characterized in that, The training method for the RBF neural network includes: The sample input sequence and the associated sample output sequence are divided into two parts: one part is used as the training data sequence and the other part is used as the test data sequence. The RBF neural network algorithm is trained using the training data sequence to obtain the RBF neural network algorithm; The trained RBF neural network algorithm was tested using the test data sequence. Determine whether the accuracy of the test results is below 90%; If so, the learning rate of the RBF neural network algorithm is modified, and the RBF neural network algorithm is retrained using the sample input sequence until the accuracy of the test is not less than 90%.

5. The digital breeding method for vegetables as described in claim 1, characterized in that, The method further includes: After obtaining the predicted water requirement, the vegetables are watered precisely and quantitatively according to the predicted water requirement.

6. The digital breeding method for vegetables as described in claim 1, characterized in that, The acquisition of ultrasonic frequency data emitted by the vegetables at the current moment includes: Real-time acquisition of ultrasonic frequency data emitted by vegetables at the current moment; Alternatively, it can receive a watering command, start watering in response to the command, and pause watering when the watering duration reaches the preset duration, while acquiring the ultrasonic frequency data emitted by the vegetables at the current moment.

7. The vegetable digital breeding method as described in claim 1, characterized in that, The method for monitoring the germination quality of vegetable breeding is as follows: 1) Construct a vegetable breeding database and store the collected breeding data into the vegetable breeding database; collect growth environment data of multiple vegetable breeds that have undergone the same pretreatment during the breeding process; regulate the growth environment data according to the pre-constructed vegetable breeding guidance parameter table; 2) Collect growth images of multiple vegetable varieties during the breeding process; obtain growth status data of different vegetable varieties based on the collected growth images; evaluate the germination status of vegetables based on the growth status data and the data in the pre-constructed vegetable breeding guidance parameter table at intervals of breeding days, and calculate the germination excellence value of different vegetable varieties.

8. The vegetable digital breeding method as described in claim 7, characterized in that, The method further includes the following steps: Based on the calculated germination quality values ​​of vegetables, calculate the overall germination quality values ​​of different vegetable breeds; The vegetable varieties with the highest overall germination quality were selected for planting.

9. The digital breeding method for vegetables as described in claim 7, characterized in that, In each vegetable breeding group, the vegetables were placed in the same straight line, and growth images of the vegetables were collected perpendicular to the growth direction of the vegetable buds. Methods for regulating growth environment data include: Get the current number of breeding days; Based on the pre-constructed vegetable breeding guidance parameter table, obtain the control range of growth environment data corresponding to the current breeding days; Determine whether the collected current growth environment data is within the control range of the current growth environment data corresponding to the current breeding days. If so, there is no need to adjust the current growth environment data. Otherwise, adjust the current growth environment data so that the current growth environment data conforms to the control range of the current breeding days. Methods for regulating growth environment data also include the following steps: Obtain the data type of the growth environment that needs to be regulated and the values ​​that need to be regulated; Send a control command to the growth environment data control device corresponding to the data type of growth environment that needs to be controlled, and the control command carries the value that needs to be controlled. The method for obtaining growth status data of different vegetable varieties includes the following sub-steps: Based on a pre-trained vegetable sprout recognition model, vegetable sprout feature images are extracted from vegetable growth images; Vegetable sprout growth status data are obtained from vegetable sprout feature images; Based on a pre-trained vegetable sprout pathology recognition model, obtain the pathology data of vegetable sprouts in the feature images of vegetable sprouts; The data on the growth status of vegetable sprouts include: the number of vegetable sprouts, the diameter of the vegetable sprout stem, the length of the vegetable sprout stem, the number of vegetable sprout leaves, and the area of ​​the vegetable sprout leaves.

Images