A non-destructive testing device and method for crop seeds

Abstract

The application discloses a kind of nondestructive testing device and method for crop seed, it is related to crop detection technical field, the device discloses complex nondestructive testing determination module, according to production batch selection module, complex nondestructive testing procedure execution module, the device of the present application first determines whether the subsequent complex nondestructive testing project needs to be executed to the seed sample of this batch, avoid high-cost detection resource investment to inferior batch and cause invalid detection, improve resource utilization, by basic nondestructive testing similarity algorithm, quickly select high consistency historical production batch as reference, intelligently generate priority adaptive complex nondestructive testing procedure, preferably for the complex quality risk (such as molecular structure anomaly, cell activity is insufficient) that seed sample can exist match detection item order, realize the targeting and high efficiency of complex detection, while reducing the cost of detection, improve the comprehensiveness and accuracy of seed quality evaluation.

Description

Technical Field

[0001] This invention relates to the field of crop testing technology, and more specifically, to a non-destructive testing device and method for crop seeds. Background Technology

[0002] To address the core problems of rigid testing procedures, inefficient data utilization, and one-sided quality assessment in existing technologies, this invention provides a non-destructive testing device and method for crop seeds. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a non-destructive testing device and method for crop seeds.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A non-destructive testing device for crop seeds includes a complex non-destructive testing judgment module, a production batch selection module, and a complex non-destructive testing procedure execution module.

[0006] The complex non-destructive testing decision module randomly selects multiple seed samples after each batch of crop seeds has been produced, generates a basic non-destructive testing dataset for each seed sample, and determines whether to perform complex non-destructive testing steps on the batch of crop seeds.

[0007] Based on the production batch selection module, after determining that a complex non-destructive testing step will be performed on the seed sample of the production batch, the production batch for non-destructive testing is selected from all the production batches corresponding to the previous crop seeds.

[0008] The complex nondestructive testing procedure execution module determines the complex nondestructive testing procedure for the seed sample in the current production batch based on the selected nondestructive testing basis production batch, and performs various complex nondestructive tests on the seed sample in the current production batch according to the complex nondestructive testing procedure.

[0009] Furthermore, to determine whether to perform complex non-destructive testing (NDT) steps on the seed samples of the production batch: import the basic NDT dataset of each seed sample into the basic NDT model, the basic NDT model derives the basic NDT analysis values ​​of various sub-samples, sum and average the basic NDT analysis values ​​of all seed samples to calculate the comprehensive basic NDT analysis value, set the comprehensive basic NDT threshold, and when the comprehensive basic NDT analysis value is greater than or equal to the comprehensive basic NDT threshold, it is determined whether to perform complex NDT steps on the seed samples of the production batch.

[0010] Furthermore, a basic non-destructive testing dataset for each seed sample is generated: various basic non-destructive tests are performed on each seed sample to generate a basic non-destructive testing dataset for each seed sample.

[0011] Furthermore, the steps for generating a basic nondestructive testing dataset for a seed sample are as follows: select a seed sample, obtain the detection data of the seed sample for each basic nondestructive test, and combine the detection data of each basic nondestructive test into a dataset to form a basic nondestructive testing dataset.

[0012] Furthermore, select the production batch for non-destructive testing from all previous production batches of crop seeds: obtain the non-destructive testing criteria value corresponding to each previous production batch in the system, set the non-destructive testing criteria threshold, and mark the production batch as the non-destructive testing criteria production batch when the non-destructive testing criteria value is greater than or equal to the non-destructive testing criteria threshold.

[0013] Furthermore, the steps for obtaining the non-destructive testing baseline value corresponding to the production batch are as follows: Select a production batch, label all seed samples in the production batch as control seed samples, determine the basic non-destructive testing baseline value for each control seed sample, sum and average the basic non-destructive testing baseline values ​​for each control seed sample, and calculate the non-destructive testing baseline value corresponding to the production batch.

[0014] Furthermore, the steps for determining the basic NDT control value of the reference seed sample are as follows: Identify all seed samples in the current production batch; select one control seed sample; determine the basic NDT similarity between this control seed sample and each seed sample; sum and average all basic NDT similarities to calculate the average basic NDT similarity, and label it as... The basic nondestructive testing similarity is calculated by pairwise absolute difference calculation of all basic nondestructive testing similarities, and the average basic nondestructive testing similarity is calculated by summing and averaging all basic nondestructive testing similarities. ,pass Calculate the baseline non-destructive testing control value for this control seed sample. .

[0015] Furthermore, the steps for determining the basic nondestructive testing similarity between the control seed sample and a seed sample are as follows: obtain the basic nondestructive testing dataset of the control seed sample, select a seed sample, obtain the basic nondestructive testing dataset of the seed sample, vectorize the detection data of each basic nondestructive test in the two basic nondestructive testing datasets, and calculate the basic nondestructive testing similarity using the similarity calculation formula.

[0016] Furthermore, the steps for determining the complex non-destructive testing procedures for seed samples in the current production batch are as follows: obtain the number of complex non-destructive tests performed for each non-destructive test based on the production batch, and then determine the complex non-destructive testing procedure value for each complex non-destructive test. Sort the complex non-destructive tests in descending order of the complex non-destructive testing procedure values, and generate complex non-destructive testing procedures based on the sorting order.

[0017] The steps for obtaining the number of complex non-destructive testing (NDT) operations corresponding to a production batch are as follows: Select an NDT production batch and select a complex NDT. If the seed sample in the NDT production batch has performed the complex NDT, the number of complex NDT operations is increased by one. If the seed sample in the NDT production batch has not performed the complex NDT, the number of complex NDT operations is 0.

[0018] The steps for determining the complex test procedure value of a complex non-destructive test are as follows: Select a complex non-destructive test, sum the number of times each non-destructive test is performed for the corresponding complex non-destructive test based on the production batch, and calculate the complex test procedure value of the complex non-destructive test.

[0019] Furthermore, a non-destructive testing method for crop seeds includes the following steps:

[0020] Step 1: After each batch of crop seeds is produced, determine whether to perform complex non-destructive testing on that batch of crop seeds.

[0021] Step 2: After determining whether to perform complex non-destructive testing on the seed samples of this production batch, select the production batch for which non-destructive testing is to be performed;

[0022] Step 3: Determine the complex non-destructive testing procedures for seed samples in the current production batch;

[0023] Step 4: Perform various complex non-destructive tests on the seed samples in the current production batch according to the complex non-destructive testing procedures.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] The apparatus and method of this invention, through the cooperation of various modules, first performs basic non-destructive testing on seed samples in a batch. Based on the test data of the basic non-destructive testing, it determines whether subsequent complex non-destructive testing items need to be performed on the seed samples of that batch. This avoids the waste of high-cost testing resources on inferior batches, thus improving resource utilization. Through the basic non-destructive testing similarity algorithm, it quickly selects historical production batches with high consistency as references and intelligently generates complex non-destructive testing procedures with appropriate priorities. It prioritizes the matching of test item order to complex quality risks that may exist in seed samples (such as abnormal molecular structure or insufficient cell activity), achieving targeted and efficient complex testing. This reduces testing costs while improving the comprehensiveness and accuracy of seed quality assessment. Attached Figure Description

[0026] Figure 1 This is a mind diagram illustrating the principle of the device of the present invention;

[0027] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0028] Example 1: As Figure 1 A non-destructive testing device for crop seeds (non-destructive testing refers to the detection and evaluation of seed quality, vigor, composition, internal structure and other characteristics without damaging the integrity and physiological activity of the seeds), including a complex non-destructive testing judgment module, a production batch selection module, and a complex non-destructive testing procedure execution module.

[0029] The complex non-destructive testing (NDT) decision module, after each batch of crop seeds has been produced (defined as a collection of seeds of the same variety, origin, generation, production season, and processing batch), randomly selects multiple seed samples from the batch based on a pre-set sample extraction ratio (the pre-set sample extraction ratio is the result of a comprehensive consideration of multiple factors, including statistical principles, national standards, seed characteristics, testing requirements, and quality risk control). Basic NDT is performed on each seed sample, generating a basic NDT dataset for each seed sample, and determining whether to perform complex NDT steps on the batch of crop seeds.

[0030] Determine whether to perform complex non-destructive testing (NDT) on the seed samples of the production batch: Import the basic NDT dataset of each seed sample into the basic NDT model. The basic NDT model derives the basic NDT analysis values ​​of various sub-samples. Sum the basic NDT analysis values ​​of all seed samples and calculate the average value to obtain the comprehensive basic NDT value. Set the comprehensive basic NDT threshold (pre-set). If the comprehensive basic NDT value is greater than or equal to the comprehensive basic NDT threshold, determine whether to perform complex NDT on the seed samples of the production batch (otherwise, there is no need to perform complex NDT on the seed samples of the production batch, because the seed samples of the production batch cannot meet the basic quality requirements, and there is no need to analyze their complex quality data).

[0031] The steps for building a basic nondestructive testing (NDT) model are as follows: First, a deep learning model is built. Multiple basic NDT datasets are collected and used as the foundation for training the deep learning model. During this process, each basic NDT dataset is assigned a basic NDT analysis value, ranging from 1 to 100. The magnitude of the basic NDT analysis value has a clear meaning; a larger value indicates a higher basic quality requirement for the seed sample under various basic NDT tests. Next, the multiple basic NDT datasets are divided into training, validation, and test sets according to a specific ratio of 60%:20%:20%. The deep learning model is first repeatedly trained using the training set. During training, the validation set is used to validate the model's performance during the training phase. Based on the validation results, the model parameters are adjusted promptly to optimize the model's structure, making it more accurate and stable. Finally, the basic NDT model is successfully built.

[0032] Perform basic non-destructive testing (NDT) on each seed sample to generate a basic NDT dataset for each seed sample. This involves performing various basic NDTs on each seed sample (including but not limited to: machine vision inspection, near-infrared spectroscopy moisture detection, and vibration spectroscopy inspection). Machine vision inspection analyzes the appearance features of the seed sample using two-dimensional images. Near-infrared spectroscopy moisture detection quantifies the moisture content and uniformity of the seed sample. Vibration spectroscopy analyzes the saturation and internal structural integrity using the seed sample's natural vibration frequencies. All basic NDTs share the characteristics of being simple and fast, allowing for rapid detection of relevant features associated with the seed sample. This process further generates a basic NDT dataset for each seed sample.

[0033] The steps for generating a basic non-destructive testing dataset for a seed sample are as follows: Select a seed sample and obtain the detection data for each basic non-destructive test (each basic non-destructive test corresponds to the corresponding detection data, such as the machine vision inspection project corresponding to the detection data of roundness, aspect ratio, and breakage rate, the near-infrared spectroscopy moisture detection project corresponding to the detection data of moisture content and moisture standard deviation, and the vibration spectroscopy detection project corresponding to the detection data of fundamental frequency and vibration decay rate). Combine the detection data of each basic non-destructive test into a dataset to form a basic non-destructive testing dataset.

[0034] Based on the production batch selection module, after determining that a complex non-destructive testing step will be performed on the seed sample of the production batch, the production batch for non-destructive testing is selected from all the production batches corresponding to the previous crop seeds.

[0035] Select the production batch for non-destructive testing from all previous production batches of crop seeds: obtain the non-destructive testing criteria value corresponding to each previous production batch in the system, set the non-destructive testing criteria threshold (pre-set), and mark the production batch as the non-destructive testing criteria production batch when the non-destructive testing criteria value is greater than or equal to the non-destructive testing criteria threshold (otherwise, do not mark it).

[0036] The steps for obtaining the non-destructive testing baseline value corresponding to the production batch are as follows: Select a production batch, label all seed samples in the production batch as control seed samples, determine the basic non-destructive testing baseline value for each control seed sample, sum and average the basic non-destructive testing baseline values ​​for each control seed sample, and calculate the non-destructive testing baseline value corresponding to the production batch.

[0037] The steps for determining the basic NDT control value of the reference seed sample are as follows: Identify all seed samples in the current production batch; select one control seed sample; determine the basic NDT similarity between this control seed sample and each seed sample; sum and average all basic NDT similarities to calculate the average basic NDT similarity, and label it as follows: The basic nondestructive testing similarity is calculated by pairwise absolute difference calculation of all basic nondestructive testing similarities, and the average basic nondestructive testing similarity is calculated by summing and averaging all basic nondestructive testing similarities. ,pass Calculate the baseline non-destructive testing control value for this control seed sample. .

[0038] The steps for determining the basic nondestructive testing similarity between a control seed sample and a seed sample are as follows: obtain the basic nondestructive testing dataset of the control seed sample, select a seed sample, obtain the basic nondestructive testing dataset of the seed sample, vectorize the detection data of each basic nondestructive test in the two basic nondestructive testing datasets, and calculate the basic nondestructive testing similarity using the similarity calculation formula.

[0039] The detection data of each basic nondestructive test in the two basic nondestructive testing datasets are vectorized, and the similarity of the basic nondestructive tests is calculated using the similarity calculation formula: The detection data of each basic nondestructive test in the control seed sample are vectorized and transformed into a feature vector A=(a1,a2,...,a... n The detection data of various basic non-destructive tests in the seed sample are vectorized and transformed into a feature vector B=(b1,b2,...,b...). n Using the similarity formula Calculate the basic nondestructive testing similarity.

[0040] The complex non-destructive testing (NDT) procedure execution module determines the complex NDT procedure for seed samples in the current production batch based on the selected NDT production batch. (The complex NDT procedure includes various complex NDT tests arranged in a specific order, including but not limited to: Fourier transform infrared spectroscopy (FTIR) and fluorescence imaging. FTIR provides molecular structure information of chemical components without damaging seed integrity, while fluorescence imaging assesses cell-level activity by specifically labeling the physiological state of living seed cells with fluorescent dyes, all without compromising seed integrity.) According to the complex nondestructive testing procedure, various complex nondestructive tests are performed on the seed samples in the current production batch. (For example, if the first item in the complex nondestructive testing procedure is Fourier transform infrared spectroscopy, then Fourier transform infrared spectroscopy is performed on all seed samples first. After all seed samples in the current production batch have completed Fourier transform infrared spectroscopy, if a seed sample does not meet the complex quality standard of Fourier transform infrared spectroscopy, then there is no need to perform subsequent complex nondestructive tests on the seed samples in the current production batch. If it meets the complex quality standard of Fourier transform infrared spectroscopy, then the next complex nondestructive test in the sequence is performed on the seed samples in the current production batch.)

[0041] The steps for determining the complex nondestructive testing procedures for seed samples in the current production batch are as follows: obtain the number of complex nondestructive tests performed for each complex nondestructive test corresponding to the production batch, and then determine the complex nondestructive testing procedure value for each complex nondestructive test. Sort the complex nondestructive tests in descending order of the complex nondestructive testing procedure values, and generate complex nondestructive testing procedures based on the sorting order.

[0042] The steps for obtaining the number of complex NDT executions for a production batch based on an NDT are as follows: Select a production batch based on an NDT and select a complex NDT. If the seed sample in the production batch has undergone the complex NDT, increment the number of complex NDT executions by one. If the seed sample in the production batch has not undergone the complex NDT (because the seed sample in the production batch did not meet the corresponding complex quality standard in the previous complex NDT project, the complex NDT was not performed), the number of complex NDT executions is 0.

[0043] The steps for determining the complex test procedure value of a complex non-destructive test are as follows: Select a complex non-destructive test, sum the number of times each non-destructive test is performed for the corresponding complex non-destructive test based on the production batch, and calculate the complex test procedure value of the complex non-destructive test.

[0044] The aforementioned device, through the cooperation of its various modules, first performs basic non-destructive testing on seed samples in a batch. Based on the data from these basic non-destructive tests, it determines whether subsequent complex non-destructive testing items need to be performed on the seed samples in that batch. This avoids the waste of high-cost testing resources on inferior batches, thus improving resource utilization. By using a basic non-destructive testing similarity algorithm, it quickly selects historical production batches with high consistency as references and intelligently generates priority-adapted complex non-destructive testing procedures. It prioritizes the matching of testing item order to potential complex quality risks in seed samples (such as abnormal molecular structure or insufficient cell activity), achieving targeted and efficient complex testing. This reduces testing costs while improving the comprehensiveness and accuracy of seed quality assessment.

[0045] Example 2: Figure 2 A non-destructive testing method for crop seeds, the steps of which are as follows:

[0046] Step 1: After each batch of crop seeds is produced, determine whether to perform complex non-destructive testing on that batch of crop seeds.

[0047] Step 2: After determining whether to perform complex non-destructive testing on the seed samples of this production batch, select the production batch for which non-destructive testing is to be performed;

[0048] Step 3: Determine the complex non-destructive testing procedures for seed samples in the current production batch;

[0049] Step 4: Perform various complex non-destructive tests on the seed samples in the current production batch according to the complex non-destructive testing procedures.

[0050] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0051] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0052] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0053] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0054] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0055] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A non-destructive testing apparatus for crop seeds, characterized by, The method comprises the following steps: a complex non-destructive testing determination module, after the completion of production of each production batch of crop seeds, randomly selects a plurality of seed samples, generates a basic non-destructive testing data set for each seed sample, and determines whether to perform a complex non-destructive testing step on the production batch of crop seeds; a production batch selection module, after determining to perform a complex non-destructive testing step on the seed sample of the production batch, selects a non-destructive testing basis production batch from all the production batches corresponding to the previous crop seeds; the non-destructive testing basis production batch is selected from all the production batches corresponding to the previous crop seeds: the non-destructive testing basis value corresponding to each production batch of the system is obtained, a non-destructive testing basis threshold value is set, when the non-destructive testing basis value is greater than or equal to the non-destructive testing basis threshold value, the production batch is marked as a non-destructive testing basis production batch; the non-destructive testing basis value corresponding to the production batch is obtained as follows: a production batch is selected, all seed samples in the production batch are marked as control seed samples, the basic non-destructive testing control value of each control seed sample is determined, the basic non-destructive testing control value of each control seed sample is calculated by summing and averaging, and the non-destructive testing basis value corresponding to the production batch is calculated; The determination step of the basic nondestructive detection control value of the control seed sample is as follows: all seed samples in the current production batch are determined, a control seed sample is selected, the basic nondestructive detection similarity between the control seed sample and each seed sample is determined, all the basic nondestructive detection similarities are calculated by summation and averaging, the average basic nondestructive detection similarity is calculated, and is marked as All the basic nondestructive detection similarities are calculated by two-by-two absolute difference value, the basic nondestructive detection gap degree is calculated, all the basic nondestructive detection gap degrees are calculated by summation and averaging, the average basic nondestructive detection gap degree is calculated, and is marked as The basic nondestructive detection control value of the control seed sample is calculated ; a complex non-destructive testing procedure execution module, according to the selected non-destructive testing basis production batch, determines the complex non-destructive testing procedure of the seed sample in the current production batch, and performs each complex non-destructive testing on the seed sample in the current production batch according to the complex non-destructive testing procedure; the determination step of the complex non-destructive testing procedure of the seed sample in the current production batch is as follows: the complex detection execution times of each complex non-destructive testing corresponding to each non-destructive testing basis production batch are obtained, and then the complex detection procedure values of each complex non-destructive testing are determined, the complex detection procedure values are sorted in descending order, and the complex non-destructive testing procedure is generated based on the sorting order; the obtaining step of the complex detection execution times of one non-destructive testing basis production batch corresponding to one complex non-destructive testing is as follows: one non-destructive testing basis production batch is selected, and one complex non-destructive testing is selected, when the seed sample in the non-destructive testing basis production batch has performed the complex non-destructive testing, the complex detection execution times is increased by one, and when the seed sample in the non-destructive testing basis production batch has not performed the complex non-destructive testing, the complex detection execution times is 0; the determination step of the complex detection procedure value of one complex non-destructive testing is as follows: one complex non-destructive testing is selected, the complex detection execution times of each non-destructive testing basis production batch corresponding to the complex non-destructive testing is calculated by summing, and the complex detection procedure value of the complex non-destructive testing is calculated.

2. The non-destructive testing device for crop seeds according to claim 1, characterized in that determining whether to perform a complex non-destructive testing step on the seed sample of the production batch: the basic non-destructive testing data set of each seed sample is respectively introduced into the basic non-destructive testing model, the basic non-destructive testing model outputs the basic non-destructive analysis value of each seed sample, the basic non-destructive analysis values of all seed samples are calculated by summing and averaging, the comprehensive basic non-destructive analysis value is calculated, the comprehensive basic non-destructive analysis threshold value is set, when the comprehensive basic non-destructive analysis value is greater than or equal to the comprehensive basic non-destructive analysis threshold value, it is determined to perform a complex non-destructive testing step on the seed sample of the production batch.

3. The non-destructive testing device for crop seeds according to claim 1, characterized in that Generating the basic non-destructive testing data set of each seed sample: performing various basic non-destructive tests on each seed sample to further generate the basic non-destructive testing data set of each seed sample.

4. The non-destructive testing device for crop seeds according to claim 3, characterized in that The generation steps of the basic non-destructive testing data set of a seed sample are as follows: selecting a seed sample, obtaining the detection data of the seed sample for various basic non-destructive tests, and combining the detection data of various basic non-destructive tests into a basic non-destructive testing data set in the form of a data set.

5. The non-destructive testing device for crop seeds according to claim 1, characterized in that, The determination steps of the basic non-destructive testing similarity between the control seed sample and a seed sample are as follows: obtaining the basic non-destructive testing data set of the control seed sample, selecting a seed sample, obtaining the basic non-destructive testing data set of the seed sample, performing vectorization processing on the detection data of various basic non-destructive tests in the two basic non-destructive testing data sets, and calculating the basic non-destructive testing similarity by using a similarity calculation formula.

6. A method for non-destructive testing of crop seeds, applied to a device for non-destructive testing of crop seeds according to any one of claims 1 to 5, characterized in that, The steps are as follows: Step one: after the production of each production batch of crop seeds is completed, it is determined whether to perform the complex non-destructive testing step on the production batch of crop seeds; Step two: after it is determined to perform the complex non-destructive testing step on the production batch of seed samples, the non-destructive testing basis production batch is selected; Step three: determining the complex non-destructive testing procedure of the seed sample in the current production batch; Step four: performing various complex non-destructive tests on the seed sample in the current production batch according to the complex non-destructive testing procedure.

Images