CN118334650A - Modeling method, system and equipment for grain moisture detection - Google Patents
Modeling method, system and equipment for grain moisture detection Download PDFInfo
- Publication number
- CN118334650A CN118334650A CN202410764061.9A CN202410764061A CN118334650A CN 118334650 A CN118334650 A CN 118334650A CN 202410764061 A CN202410764061 A CN 202410764061A CN 118334650 A CN118334650 A CN 118334650A
- Authority
- CN
- China
- Prior art keywords
- grain
- detection
- samples
- sample
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 288
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000012549 training Methods 0.000 claims abstract description 133
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 115
- 238000012937 correction Methods 0.000 claims abstract description 113
- 230000003595 spectral effect Effects 0.000 claims abstract description 89
- 238000001035 drying Methods 0.000 claims abstract description 55
- 238000013507 mapping Methods 0.000 claims abstract description 22
- 230000007613 environmental effect Effects 0.000 claims description 27
- 238000012795 verification Methods 0.000 claims description 17
- 238000005259 measurement Methods 0.000 claims description 13
- 238000001228 spectrum Methods 0.000 claims description 13
- 238000003384 imaging method Methods 0.000 claims description 11
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 11
- 238000005070 sampling Methods 0.000 claims description 10
- 230000001419 dependent effect Effects 0.000 claims description 8
- 238000002329 infrared spectrum Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 239000000463 material Substances 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 238000010191 image analysis Methods 0.000 claims description 3
- 238000002360 preparation method Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 description 14
- 238000010276 construction Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 6
- 230000007547 defect Effects 0.000 description 5
- 230000009897 systematic effect Effects 0.000 description 5
- 238000003860 storage Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000011897 real-time detection Methods 0.000 description 3
- 241000209149 Zea Species 0.000 description 2
- 235000005824 Zea mays ssp. parviglumis Nutrition 0.000 description 2
- 235000002017 Zea mays subsp mays Nutrition 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000000701 chemical imaging Methods 0.000 description 2
- 235000005822 corn Nutrition 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005842 biochemical reaction Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 238000007602 hot air drying Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004060 metabolic process Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000000985 reflectance spectrum Methods 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Landscapes
- Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)
Abstract
The invention discloses a modeling method, a modeling system and modeling equipment for grain moisture detection. The modeling method comprises the following steps: preparing a sample; calculating a correction model: aiming at M grain samples, respectively acquiring the water content of each sample based on a drying detection method and a capacitive rapid detection method, and generating error coordinate mapping points so as to obtain a correction model; automatically training a detection model: for N grain samples, detecting the water content of the samples by a capacitive rapid detection method, extracting spectral image data of the samples to obtain spectral image texture features, inputting the water content of the samples and the spectral image texture features into an initial multiple regression model for training and learning to obtain a grain water content detection model; and (5) verifying and optimizing the detection model. Through the scheme, the model is automatically optimized through the correction link, the system error of the training data is eliminated, the training data acquisition efficiency and the training data acquisition precision are improved, and the method can be widely applied to rapid and accurate modeling of various grain moisture.
Description
Technical Field
The invention relates to the field of rapid analysis and detection, in particular to a modeling method, a modeling system and electronic equipment for grain moisture detection.
Background
The moisture content of grain is a key factor affecting the quality of grain during storage, transportation and processing. The grain harvesting season, the processing mode process and the grain storage environment are affected, and the moisture content in the grain is generally changed to a certain extent. Too high water content can promote the vigorous metabolism of grains, so that the grains are rotted, mildewed, vermin or other biochemical reactions occur. Detecting the moisture content of grain is an important task in this field. In the prior art, the conventional method for detecting the moisture content of grains mainly comprises the following steps: a grain moisture detection method based on constant temperature drying technology, a grain moisture detection method based on capacitance/resistance measurement, and a grain moisture detection method based on big data detection model. The three methods have the following characteristics: the grain moisture detection method of the constant temperature drying technology is high in detection accuracy, but usually needs more perfect detection and experimental environment, is long in detection period, low in sample detection efficiency, cannot meet the production requirements of real-time detection in the scenes of production lines, warehouses, transportation and the like, and is extremely large in workload especially in big data acquisition work; the capacitance/resistance measurement technology has higher detection efficiency and quick output result, but can realize real-time detection requirement, but is greatly influenced by environment, instruments and operation, so that detection errors are large, and the state characteristics (such as the characteristics of volume, density, shape, internal arrangement condition and the like) of a detected target have obvious influence on the detection result, which are important factors causing result errors, and the adverse factors are difficult to overcome in actual detection operation; the grain moisture detection method based on the big data detection model can be convenient, quick and lossless, but the accuracy and repeatability of the grain moisture detection method are highly dependent on whether the detection model is complete or not, and the generation of the detection model and the accuracy of the grain moisture detection method are highly dependent on accurate basic data, such as water content data detection result data; when the calibrated detection result data is inaccurate (for example, the detection result with a large error is introduced by adopting a capacitance/resistance measurement method in the prior art), or the calibrated detection result is insufficient in quantity to cover all cases (the basic data is too little), the detection model is deviated, so that the detection result is inaccurate finally. Accordingly, there is a need for research and improvement in methods, systems and apparatus for grain moisture detection.
The information disclosed in the background section of the invention is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a modeling method for grain moisture detection, which comprises the following steps:
s100: sample preparation: s random grain samples are obtained, and the S random grain samples are divided into M grain samples for calculating a correction model, N grain samples for training a detection model and L grain samples for verifying a detection result in a random extraction mode; wherein S, M, N, L are natural numbers, M < N, (M+N+L) < S;
s110: calculating a correction model: for the M grain samples, respectively acquiring the water content of the samples based on a drying detection method and a capacitive rapid detection method, generating error coordinate mapping points, and fitting by using a least square method to obtain a correction model;
S120: training a detection model: detecting the water content of the samples by a capacitive rapid detection method aiming at the N grain samples, extracting spectral image data of the samples to obtain spectral image texture features, inputting the water content of the samples and the spectral image texture features into an initial multiple regression model for training and learning to obtain a grain water content detection model; the initial multiple regression model incorporates a correction model for error correction operation of training data;
S130: verification and output detection model: and aiming at the L grain samples, acquiring the water content of the samples based on a drying detection method, outputting the predicted water content of the samples based on a detection model, and verifying the detection model based on an estimated standard error.
Further, in the step S110, the calculation correction model specifically includes: taking the M grain samples, dividing any grain Sample in the M grain samples into p subsamples { Sample 1,Sample2,……Samplep }, wherein the grain water content of each subsample in the p subsamples is the same, and p is a natural number; taking one Sample 1 in the p subsamples, and detecting the water content of the subsamples by a drying detection method to obtain subsamples, drying and measuring the water contentTaking the rest p-1 parts in the p parts of subsamplesThe average moisture content of the subsamples is obtained by detecting the average moisture content value of the subsamples by a capacitance type rapid detection methodRecording and drying to measure water contentCorresponding capacitance measurement of average water contentError value of (2)=Obtaining a sample error coordinate mapping point corresponding to the sample,) Performing the steps for each grain sample in the M grain samples to obtain M sample error coordinate mapping points, and fitting the M sample error coordinate mapping points by using a least square method to obtain a correction modelWherein, the method comprises the steps of, wherein,Is thatIs used for correcting the error of the (a),To fit a curve.
Further, in the step S120, the training detection model specifically includes: taking the N grain samples, and equally dividing any grain Sample in the N grain samples into 2 sub-samples { Sample a1,Samplea2 }, wherein the water content of the sub-samples Sample a1 is the same as that of the sub-Sample a2; sampling a1, and detecting the water content by a capacitive rapid detection method; Taking a sub-Sample a2, extracting spectral image data of the sub-Sample a2, preprocessing the spectral image data, and extracting a spectral image texture feature C= { C 1,c2,……,cn }, wherein C i (i is more than or equal to 1 and less than or equal to n) is a spectral feature value with a sequence number i; will be%C) adding training set T as training data; executing the steps on each grain sample in the N grain samples to obtain N groups of training data so as to add a training set T; inputting the training set T into an initial multiple regression model for training and learning to obtain a detection model for outputting the water content of grains; the initial multiple regression model incorporates a correction modelAnd the error correction operation is used for carrying out error correction operation on the training data.
Further, in the step S130, the verification and output detection model is specifically: taking the L grain samples, and equally dividing each grain Sample in the L grain samples into 2 sub-samples { Sample b1,Samplb2 }, wherein the water content of the sub-samples Sample b1 is the same as that of the sub-Sample b2; sampling Sample b1, and detecting the water content by a drying detection method to obtain a sub Sample, drying and measuring the water content; Taking a subsamples Sample b2, extracting spectral image data of the subsamples Sample b2 to obtain spectral image texture features C ', and inputting the spectral image texture features C' into a detection model to obtain a model output resultValidating the estimated standard errorWhere k is the number of elements of the spectral image texture feature C'; if the standard error is within the threshold range, outputting a trained detection model for detecting the moisture content of the grains; otherwise, correcting the learning sample optimization model.
Further, a spectral image of the grain sample is obtained through an imaging system, and further, the characteristic data of the spectral image of the grain sample are extracted, wherein the characteristic data of the spectral image are grain images and near infrared spectrums.
Further, in the step S110, different environmental parameters are recorded to generate an environmental parameter set e= { E 1,e2,……,ek }, and the average water content is measured by the environmental parameters and the capacitanceIs an independent variable and an error valueTraining multiple regression correction models for dependent variablesReplacing the correction model in step S110Wherein, the method comprises the steps of, wherein,To correct the model predicted correction errors.
Further, in the step S120, the current environmental parameters are recorded to generate a current environmental parameter set E', which will be ((-)E'), C) adding training data into a training set T; the initial multiple regression model incorporates a multiple regression correction modelAnd the error correction operation is used for carrying out error correction operation on the training data.
Further, the environmental parameters include at least temperature and humidity.
Further, the preprocessing of the spectral image data includes: and filtering, denoising and normalizing the spectrum image data.
The invention also provides a modeling system for grain moisture detection, comprising:
The sample acquisition module is used for acquiring S pieces of random grain samples, and dividing the S pieces of random grain samples into M pieces of grain samples for calculating a correction model, N pieces of grain samples for training a detection model and L pieces of grain samples for verifying a detection result according to a random extraction mode; wherein M < N, (m+n+l) < S;
The correction model acquisition module is used for calculating a correction model: for the M grain samples, respectively acquiring the water content of the samples based on a drying detection method and a capacitive rapid detection method, generating error coordinate mapping points, and fitting by using a least square method to obtain a correction model;
The detection model acquisition module is used for training a detection model: detecting the water content of the samples by a capacitive rapid detection method aiming at the N grain samples, extracting spectral image data of the samples to obtain spectral image texture features, inputting the water content of the samples and the spectral image texture features into an initial multiple regression model for training and learning to obtain a grain water content detection model; the initial multiple regression model incorporates a correction model for error correction operation of training data;
The verification and output module is used for verifying and outputting a detection model: and aiming at the L grain samples, acquiring the water content of the samples based on a drying detection method, outputting the predicted water content of the samples based on a detection model, and verifying the detection model based on an estimated standard error.
Further, in the correction model obtaining module, the calculation correction model specifically includes: taking the M grain samples, and dividing any grain Sample in the M grain samples into p subsamples { Sample 1,Sample2,……Samplep }, wherein the grain water content of each subsamples in the p subsamples is the same; taking one Sample 1 in the p subsamples, and detecting the water content of the subsamples by a drying detection method to obtain subsamples, drying and measuring the water contentTaking the rest p-1 parts in the p parts of subsamplesThe average moisture content of the subsamples is obtained by detecting the average moisture content value of the subsamples by a capacitance type rapid detection methodRecording and drying to measure water contentCorresponding capacitance measurement of average water contentError value of (2)=Obtaining a sample error coordinate mapping point corresponding to the sample,) Performing the steps for each grain sample in the M grain samples to obtain M sample error coordinate mapping points, and fitting the M sample error coordinate mapping points by using a least square method to obtain a correction modelWherein, the method comprises the steps of, wherein,Is thatIs used for correcting the error of the (a),Fitting a curve;
In the detection model acquisition module, the training detection model specifically includes: taking the N grain samples, and equally dividing any grain Sample in the N grain samples into 2 sub-samples { Sample a1,Samplea2 }, wherein the water content of the sub-samples Sample a1 is the same as that of the sub-Sample a2; sampling a1, and detecting the water content by a capacitive rapid detection method ; Taking a sub-Sample a2, extracting spectral image data of the sub-Sample a2, preprocessing the spectral image data, and extracting a spectral image texture feature C= { C 1,c2,……,cn }, wherein C i (i is more than or equal to 1 and less than or equal to n) is a spectral feature value with a sequence number i; will be%C) adding training set T as training data; executing the steps on each grain sample in the N grain samples to obtain N groups of training data so as to add a training set T; inputting the training set T into an initial multiple regression model for training and learning to obtain a detection model for outputting the water content of grains; the initial multiple regression model incorporates a correction modelThe error correction operation is used for carrying out error correction operation on the training data;
In the verification and output module, the verification and output detection model specifically includes: taking the L grain samples, and equally dividing each grain Sample in the L grain samples into 2 sub-samples { Sample b1,Sampleb2 }, wherein the water content of the sub-samples Sample b1 is the same as that of the sub-Sample b2; sampling Sample b1, and detecting the water content by a drying detection method to obtain a sub Sample, drying and measuring the water content ; Taking a subsamples Sample b2, extracting spectral image data of the subsamples Sample b2 to obtain spectral image texture features C ', and inputting the spectral image texture features C' into a detection model to obtain a model output resultValidating the estimated standard errorWhere k is the number of elements of the spectral image texture feature C'; if the standard error is within the threshold range, outputting a trained detection model for detecting the moisture content of the grains; otherwise, correcting the learning sample optimization model.
Further, a spectral image of the grain sample is obtained through an imaging system, and further, the characteristic data of the spectral image of the grain sample are extracted, wherein the characteristic data of the spectral image are grain images and near infrared spectrums.
Further, the correction model obtaining module is further configured to record different environmental parameters to generate an environmental parameter set e= { E 1,e2,……,ek }, measure the average water content by using the environmental parameters and the capacitanceIs an independent variable and an error valueTraining multiple regression correction models for dependent variablesReplacing the correction model in step S110Wherein, the method comprises the steps of, wherein,Correcting errors predicted for the correction model;
the detection model acquisition module is also used for recording the current environment parameters to generate a current environment parameter set E', and determining the current environment parameters (the current environment parameters are (-)) E'), C) adding training data into a training set T; the initial multiple regression model incorporates a multiple regression correction modelThe error correction operation is used for carrying out error correction operation on the training data;
further, the environmental parameter includes at least temperature, humidity.
Further, the preprocessing of the spectral image data includes: and filtering, denoising and normalizing the spectrum image data.
The invention also provides detection model construction equipment applied to a modeling method for grain moisture detection, which comprises the following steps: the device comprises a main control module, a capacitive moisture calibration module, a visible light-near infrared light source module, an imaging module and a sample transmission module.
The main control module comprises a main control board and is used for storing basic data, analyzing, calculating, training and optimizing a model; the main control module comprises an image analysis and processing module, and is used for processing, collecting and analyzing spectral images of grain samples aiming at video streams and extracting corresponding spectral image characteristics;
The capacitive moisture calibration module comprises a cylindrical capacitive sensor and a resonance detection circuit, and a funnel device is arranged below the cylindrical capacitive sensor to facilitate material removal; the capacitive moisture calibration module is used for detecting the moisture content of different grain samples through different dielectric constants of grains in the sensor and inputting the detection result to the main control module;
the visible-near infrared light source and the imaging module are used for acquiring image information and near infrared spectrum information of the grain sample;
the sample transmission module is used for placing and conveying samples.
The invention also provides correction model construction equipment applied to the modeling method for grain moisture detection, which comprises the following steps: the system comprises a main control module, a capacitive moisture calibration module and a drying moisture calibration module;
The main control module comprises a main control board and is used for storing basic data, analyzing, calculating, training and optimizing a model;
The capacitive moisture calibration module comprises a cylindrical capacitive sensor and a resonance detection circuit, and a funnel device is arranged below the cylindrical capacitive sensor to facilitate material removal; the capacitive moisture calibration module is used for detecting the moisture content of different grain samples through different dielectric constants of grains in the sensor and inputting the detection result to the main control module;
The drying type moisture calibration module is used for detecting the moisture content of different grain samples and inputting the detection result to the main control module.
The invention also provides a modeling system for grain moisture detection, which consists of detection model construction equipment applied to a modeling method for grain moisture detection and correction model construction equipment applied to the modeling method for grain moisture detection.
The invention provides a modeling method, a modeling system and modeling equipment for grain moisture detection, and the modeling method, the modeling system and modeling equipment can at least achieve the following effects: (1) The spectrum detection method is applied to the detection of the grain moisture content, so that the grain moisture content can be rapidly and nondestructively predicted based on a spectrum image and a constructed detection model, and the detection model is trained by adopting big data obtained by capacitance measurement, so that the model precision requirement is met; (2) When the training data is constructed, the capacitance measurement result is used as basic training data, and meanwhile, the training data is corrected based on the correction model, so that the system error of the measurement data is eliminated, the training data acquisition efficiency and the training data acquisition precision are improved, and the defect that the training data is acquired based on the prior art is avoided.
Drawings
The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular descriptions of exemplary embodiments of the invention as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts throughout the exemplary embodiments of the invention.
FIG. 1 shows a flow chart of the steps of a grain sample moisture detection modeling method according to one embodiment of the invention.
FIG. 2 illustrates a block diagram of a modeling system for grain moisture detection in accordance with one embodiment of the present invention.
Fig. 3 shows a schematic diagram of a multispectral curve of grain samples of different moisture content according to one embodiment of the invention.
Reference numerals illustrate:
201. A sample acquisition module; 202. a correction model acquisition module; 203. a detection model acquisition module; 204. and a verification and output module.
Detailed Description
Preferred embodiments of the present invention will be described in more detail below. While the preferred embodiments of the present invention are described below, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein.
The invention provides a modeling method, a modeling system and modeling equipment for grain moisture detection, which can meet the requirement of rapid nondestructive detection of grain moisture content.
Example 1
In a first aspect, an embodiment of the present invention provides a modeling method for grain moisture detection, including:
s100: sample preparation: s random grain samples are obtained, and the S random grain samples are divided into M grain samples for calculating a correction model, N grain samples for training a detection model and L grain samples for verifying a detection result in a random extraction mode; wherein S, M, N, L are natural numbers, M < N, (M+N+L) < S;
In the prior art, the grain moisture detection method based on the constant temperature drying technology has high detection accuracy, but has the defects of overlong detection period, complex detection and experimental environment, difficult implementation in application scenes of real-time detection, inconvenient acquisition of a large amount of detection data based on the drying technology, and rapid increase of workload caused by the application of the grain moisture detection method to training scenes of a large data model requiring a large amount of basic detection data. In contrast to the capacitive detection technology, the capacitive detection technology has the advantages of higher detection efficiency, quick output result, low detection accuracy and instability. Furthermore, the grain moisture detection method based on the big data detection model can be convenient and quick and lossless, but the detection accuracy is highly dependent on whether the detection model is complete or not.
The invention considers the characteristics of various technologies in the prior art, and is characterized in that the invention uses the characteristics of high efficiency, simple operation and quick output result of the capacitance detection technology, uses the detection result based on the capacitance detection technology as big data training data for model training, and simultaneously, uses the characteristics of high accuracy of the drying detection technology to set a correction model to correct the detection result of the capacitance detection technology, thus overcoming the defect of low detection efficiency of the drying detection technology, overcoming the defect of inaccurate capacitance detection technology, improving the accuracy and generation efficiency of the training data, and being beneficial to efficiently generating a relatively accurate grain moisture content detection model. Therefore, when the test samples are set, M grain samples with smaller quantity are selected from the total grain samples and used for calculating the correction model, and after the correction model is formed, N (N > M) grain samples with larger quantity are selected from the total grain samples and used for quickly generating a large quantity of training data to train the detection model. In addition, in order to ensure the detection accuracy of the detection model, L grain samples with smaller quantity are selected from the total grain samples to verify whether the detection model meets the actual production requirement. If the detection model meets the requirement, the scheme of the invention outputs the detection model for detecting the water content of the grains; otherwise, the training data, the training parameters and the training algorithm are optimized to obtain an optimized detection model. In one example, the method of optimizing the training data is to set a more accurate correction model, provide environmental parameters as training data, and so on; the optimization method of the training parameters and training algorithm is to increase environmental parameters, adjust weights, loss functions, etc., and model parameter optimization methods in the prior art can be utilized, which are not limited herein.
S110: calculating a correction model: for the M grain samples, respectively acquiring the water content of the samples based on a drying detection method and a capacitive rapid detection method, generating error coordinate mapping points, and fitting by using a least square method to obtain a correction model;
in one example, in the step S110, the calculated correction model is specifically: taking the M grain samples, dividing any grain Sample in the M grain samples into p subsamples { Sample 1,Sample2,……Samplep }, wherein the grain water content of each subsample in the p subsamples is the same, and p is a natural number; taking one Sample 1 in the p subsamples, and detecting the water content of the subsamples by a drying detection method to obtain subsamples, drying and measuring the water content Taking the rest p-1 parts in the p parts of subsamplesThe average moisture content of the subsamples is obtained by detecting the average moisture content value of the subsamples by a capacitance type rapid detection methodRecording and drying to measure water contentCorresponding capacitance measurement of average water contentError value of (2)=Obtaining a sample error coordinate mapping point corresponding to the sample,) Performing the steps for each grain sample in the M grain samples to obtain M sample error coordinate mapping points, and fitting the M sample error coordinate mapping points by using a least square method to obtain a correction modelWherein, the method comprises the steps of, wherein,Is thatIs used for correcting the error of the (a),Fitting a curve;
in the above method, an example may be that each grain sample is divided into 3 sub-samples, and since the water content of each sample is constant, the water content of the sub-samples does not change (3 sub-samples are identical in water content index, but are not required to be identical in quality index). Wherein 1 part is used for detecting the water content based on a drying detection method, 2 parts is used for detecting the water content based on a capacitance type rapid detection method, and the average value of the water contents detected for a plurality of times is calculated as a final detection result. The method for taking the result mean value improves the detection accuracy and the stability of the detection result of the capacitance detection method. The detection result of the drying detection method is taken as a reference to calculate the error value of the capacitance measurement method, and meanwhile, the detection results of different drying detection methods all have corresponding error values, the error represents the system error of the detection system, the correction is carried out based on the error value, the system error of the capacitance detection method can be eliminated, and specifically, the system error can be caused by one or a combination of the following aspects: (1) Detecting a state of the apparatus, for example, a state of aging of the apparatus, which is a state capable of causing the inside of the subject to assume a specific arrangement with a high probability, resulting in a systematic deviation of the detection result; (2) The influence of equipment, ambient temperature and ambient humidity on the detected target can cause systematic deviation of detection results; (3) Different habits of inspectors and different equipment operation methods lead to systematic deviation of detection results; (4) The detection flows with different characteristics lead to systematic deviation of detection results. The system error can lead the measured water content value to be higher, and can lead the measured water content value to be lower, so that the error can not be avoided in the measuring system in theory, and the detection result of the drying detection method is introduced as a reference, thereby being capable of improving and even avoiding the error to a great extent.
S120: training a detection model: detecting the water content of the samples by a capacitive rapid detection method aiming at the N grain samples, extracting spectral image data of the samples to obtain spectral image texture features, inputting the water content of the samples and the spectral image texture features into an initial multiple regression model for training and learning to obtain a grain water content detection model; the initial multiple regression model incorporates a correction model for error correction operation of training data;
In the step S120, the training detection model specifically includes: taking the N grain samples, and equally dividing any grain Sample in the N grain samples into 2 sub-samples { Sample a1,Samplea2 }, wherein the water content of the sub-samples Sample a1 is the same as that of the sub-Sample a2; sampling a1, and detecting the water content by a capacitive rapid detection method ; Taking a sub-Sample a2, extracting spectral image data of the sub-Sample a2, preprocessing the spectral image data, and extracting a spectral image texture feature C= { C 1,c2,……,cn }, wherein C i (i is more than or equal to 1 and less than or equal to n) is a spectral feature value with a sequence number i; will be%C) adding training set T as training data; executing the steps on each grain sample in the N grain samples to obtain N groups of training data so as to add a training set T; inputting the training set T into an initial multiple regression model for training and learning to obtain a detection model for outputting the water content of grains; the initial multiple regression model incorporates a correction modelThe error correction operation is used for carrying out error correction operation on the training data;
Preprocessing spectral image data includes: the spectral image data is subjected to filtering, denoising and normalization treatment, and the aim of preprocessing is to realize data alignment and simplify the acquisition mode of the texture characteristics of the spectral image. In one example, extracting spectral image data includes: dividing the multispectral image by a typical discriminant analysis and binarization method to obtain a detection target area; graying treatment is carried out on the detection target area, and the spectral reflectivity of the detection target area is calculated, namely spectral image data; in one example, the spectral reflectance of the detection target region is calculated by the following formula:
；
Wherein S k is the spectral reflectance of the grain sample in the kth wave band, k is a positive integer less than the number of the appointed wave bands, the number of the appointed wave bands can take a value of 19, I (i, j) is the gray value of the (i, j) th pixel after the multi-spectrum image is grayed, and m, n are the number of rows and the number of columns of the image respectively.
Multispectral images at 19 discrete wavelengths at 405, 435, 450, 470, 505, 525, 570, 590, 630, 645, 660, 700, 780, 850, 870, 890, 910, 940, and 970 nm can be acquired by a multispectral system. Wavelength information at 19 wavelengths of a specific multispectral imaging system. In order to ensure that light is uniformly scattered when photographing, the LED lamps are uniformly distributed inside the integrating sphere. The corn sample is photographed on a stage within the integrating sphere and each LED lamp produces a multispectral image with an image resolution of 2056 x 2056. The system will obtain 19 photos at different wavelengths superimposed into one multispectral image containing 19 wavelength information. The average reflectance spectra of the corn kernels at different hot air drying periods are shown in fig. 3.
In the method, the training set T contains the water content measured by a capacitance type rapid detection method for N grain samples and the corresponding spectral image texture characteristics as training data; in a more preferable example, the training set T further includes sample image data corresponding to the measured water content as training data. Carrying out multiple regression model training on the multidimensional training data to obtain a final detection model, wherein the input data of the detection model is an extracted spectrum image texture feature set, and outputting a water content value predicted by the detection model; based on corrective models prior to model trainingAnd carrying out error correction operation on the water content in the input training data, so as to eliminate the systematic error of the capacitive rapid detection method in the step and improve the accuracy of the detection model.
S130: verification and output detection model: and aiming at the L grain samples, acquiring the water content of the samples based on a drying detection method, outputting the predicted water content of the samples based on a detection model, and verifying the detection model based on an estimated standard error.
In the step S130, the verification and output detection model specifically includes: taking the L grain samples, and equally dividing each grain Sample in the L grain samples into 2 sub-samples { Sample b1,Sampleb2 }, wherein the water content of the sub-samples Sample b1 is the same as that of the sub-Sample b2; sampling Sample b1, and detecting the water content by a drying detection method to obtain a sub Sample, drying and measuring the water content; Taking a subsamples Sample b2, extracting spectral image data of the subsamples Sample b2 to obtain spectral image texture features C ', and inputting the spectral image texture features C' into a detection model to obtain a model output resultValidating the estimated standard errorWhere k is the number of elements of the spectral image texture feature C'; if the standard error is within the threshold range, outputting a trained detection model for detecting the moisture content of the grains; otherwise, correcting the learning sample optimization model.
In the method, the training data is subjected to the correction operation through the correction model, so that the detection result of the accurate drying detection method is directly used as the expected result when the model is verified, and the detection result of the conventional capacitance detection method is not used as the expected result, and therefore, the verification method is related to the accuracy of the correction model and the detection model, and the model verification efficiency is improved. The estimation standard error expresses the deviation degree of the model predicted value and the expected value, and can be used for evaluating the accuracy of the model to a certain extent.
Because the training data is generated based on capacitance detection, the main factors causing errors can be considered, and when the main factors are considered in the constructed model to perform simulation or training, the output of the final model can avoid the influence of the main factors to a large extent, so that the model accuracy is improved.
In a preferred example, the main factors causing the error are ambient temperature, ambient humidity.
In a preferred example, in the correction model construction process, the training detection model process, and the verification detection model process, the main factors causing errors are detected, and the values of the main factors causing errors in the three processes are kept the same.
In a preferred example, in the step S110, the different environmental parameters are recorded to generate an environmental parameter set e= { E 1,e2,……,ek }, and the average water content is measured with the environmental parameters and the capacitanceIs an independent variable and an error valueTraining multiple regression correction models for dependent variablesReplacement of the correction model in the original step S110Wherein, the method comprises the steps of, wherein,Correcting errors predicted for the correction model; in step S120, the current environmental parameters are recorded to generate a current environmental parameter set E', which is to be used (aE'), C) adding training data into a training set T; the initial multiple regression model incorporates a multiple regression correction modelAnd the error correction operation is used for carrying out error correction operation on the training data. In the example, the environment parameters are used as training parameters in the process of constructing the correction model and training the detection model, so that the model accuracy is further improved.
Example 2
In a second aspect, embodiments of the present invention provide a modeling system for grain moisture detection, comprising:
The sample acquisition module is used for acquiring S pieces of random grain samples, and dividing the S pieces of random grain samples into M pieces of grain samples for calculating a correction model, N pieces of grain samples for training a detection model and L pieces of grain samples for verifying a detection result according to a random extraction mode; wherein S, M, N, L are natural numbers, M < N, (M+N+L) < S;
The correction model acquisition module is used for calculating a correction model: for the M grain samples, respectively acquiring the water content of the samples based on a drying detection method and a capacitive rapid detection method, generating error coordinate mapping points, and fitting by using a least square method to obtain a correction model;
The detection model acquisition module is used for training a detection model: detecting the water content of the samples by a capacitive rapid detection method aiming at the N grain samples, extracting spectral image data of the samples to obtain spectral image texture features, inputting the water content of the samples and the spectral image texture features into an initial multiple regression model for training and learning to obtain a grain water content detection model; the initial multiple regression model incorporates a correction model for error correction operation of training data;
The verification and output module is used for verifying and outputting a detection model: and aiming at the L grain samples, acquiring the water content of the samples based on a drying detection method, outputting the predicted water content of the samples based on a detection model, and verifying the detection model based on an estimated standard error.
Further, in the correction model obtaining module, the calculation correction model specifically includes: taking the M grain samples, and dividing any grain Sample in the M grain samples into p subsamples { Sample 1,Sample2,……Samplep }, wherein the grain water content of each subsamples in the p subsamples is the same; taking one Sample 1 in the p subsamples, and detecting the water content of the subsamples by a drying detection method to obtain subsamples, drying and measuring the water contentTaking the rest p-1 parts in the p parts of subsamplesThe average moisture content of the subsamples is obtained by detecting the average moisture content value of the subsamples by a capacitance type rapid detection methodRecording and drying to measure water contentCorresponding capacitance measurement of average water contentError value of (2)=Obtaining a sample error coordinate mapping point corresponding to the sample,) Performing the steps for each grain sample in the M grain samples to obtain M sample error coordinate mapping points, and fitting the M sample error coordinate mapping points by using a least square method to obtain a correction modelWherein, the method comprises the steps of, wherein,Is thatIs used for correcting the error of the (a),Fitting a curve;
In the detection model acquisition module, the training detection model specifically includes: taking the N grain samples, and equally dividing any grain Sample in the N grain samples into 2 sub-samples { Sample a1,Samplea2 }, wherein the water content of the sub-samples Sample a1 is the same as that of the sub-Sample a2; sampling a1, and detecting the water content by a capacitive rapid detection method ; Taking a sub-Sample a2, extracting spectral image data of the sub-Sample a2, preprocessing the spectral image data, and extracting a spectral image texture feature C= { C 1,c2,……,cn }, wherein C i (i is more than or equal to 1 and less than or equal to n) is a spectral feature value with a sequence number i; will be%C) adding training set T as training data; executing the steps on each grain sample in the N grain samples to obtain N groups of training data so as to add a training set T; inputting the training set T into an initial multiple regression model for training and learning to obtain a detection model for outputting the water content of grains; the initial multiple regression model incorporates a correction modelThe error correction operation is used for carrying out error correction operation on the training data;
In the verification and output module, the verification and output detection model specifically includes: taking the L grain samples, and equally dividing each grain Sample in the L grain samples into 2 sub-samples { Sample b1,Sampleb2 }, wherein the water content of the sub-samples Sample b1 is the same as that of the sub-Sample b2; sampling Sample b1, and detecting the water content by a drying detection method to obtain a sub Sample, drying and measuring the water content ; Taking a subsamples Sample b2, extracting spectral image data of the subsamples Sample b2 to obtain spectral image texture features C ', and inputting the spectral image texture features C' into a detection model to obtain a model output resultValidating the estimated standard errorWhere k is the number of elements of the spectral image texture feature C'; if the standard error is within the threshold range, outputting a trained detection model for detecting the moisture content of the grains; otherwise, correcting the learning sample optimization model.
Further, a spectral image of the grain sample is obtained through an imaging system, and further, the characteristic data of the spectral image of the grain sample are extracted, wherein the characteristic data of the spectral image are grain images and near infrared spectrums.
Further, the correction model obtaining module is further configured to record different environmental parameters to generate an environmental parameter set e= { E 1,e2,……,ek }, measure the average water content by using the environmental parameters and the capacitanceIs an independent variable and an error valueTraining multiple regression correction models for dependent variablesReplacing the correction model in step S110Wherein, the method comprises the steps of, wherein,Correcting errors predicted for the correction model;
the detection model acquisition module is also used for recording the current environment parameters to generate a current environment parameter set E', and determining the current environment parameters (the current environment parameters are (-)) E'), C) adding training data into a training set T; the initial multiple regression model incorporates a multiple regression correction modelThe error correction operation is used for carrying out error correction operation on the training data;
further, the environmental parameter includes at least temperature, humidity.
Further, the preprocessing of the spectral image data includes: and filtering, denoising and normalizing the spectrum image data.
Example 3
In a third aspect, an embodiment of the present invention provides a detection model construction apparatus applied to a modeling method for grain moisture detection, including:
The device comprises a main control module, a capacitive moisture calibration module, a visible light-near infrared light source module, an imaging module and a sample transmission module.
The main control module comprises a main control board and is used for storing basic data, analyzing, calculating, training and optimizing a model; the main control module comprises an image analysis and processing module, and is used for processing, collecting and analyzing spectral images of grain samples aiming at video streams and extracting corresponding spectral image characteristics; the main control model is based on a data output detection model of the capacitive moisture calibration module, the visible light-near infrared light source module and the imaging module;
The capacitive moisture calibration module comprises a cylindrical capacitive sensor and a resonance detection circuit, and a funnel device is arranged below the cylindrical capacitive sensor to facilitate material removal; the capacitive moisture calibration module is used for detecting the moisture content of different grain samples through different dielectric constants of grains in the sensor and inputting the detection result to the main control module;
the visible-near infrared light source and the imaging module are used for acquiring image information and near infrared spectrum information of the grain sample;
the sample transmission module is used for placing and conveying samples.
Example 4
In a fourth aspect, an embodiment of the present invention provides a corrective model construction apparatus applied to a modeling method for grain moisture detection, including: the system comprises a main control module, a capacitive moisture calibration module and a drying moisture calibration module;
The main control module comprises a main control board and is used for storing basic data, analyzing, calculating, training and optimizing a model; the main control model is based on a data output correction model of the capacitive moisture calibration module and the drying moisture calibration module;
The capacitive moisture calibration module comprises a cylindrical capacitive sensor and a resonance detection circuit, and a funnel device is arranged below the cylindrical capacitive sensor to facilitate material removal; the capacitive moisture calibration module is used for detecting the moisture content of different grain samples through different dielectric constants of grains in the sensor and inputting the detection result to the main control module;
The drying type moisture calibration module is used for detecting the moisture content of different grain samples and inputting the detection result to the main control module.
Example 5
In a fifth aspect, embodiments of the present invention provide a modeling system for grain moisture detection, which is composed of a detection model construction apparatus applied to a modeling method for grain moisture detection and a correction model construction apparatus applied to a modeling method for grain moisture detection.
The invention provides a modeling method, a modeling system and modeling equipment for grain moisture detection, and the modeling method, the modeling system and modeling equipment can at least achieve the following effects: (1) The spectrum detection method is applied to the detection of the grain moisture content, so that the grain moisture content can be rapidly predicted based on a spectrum image and a constructed detection model, and the detection model is trained by adopting big data obtained by capacitance measurement, so that the model precision requirement is met; (2) When the training data is constructed, the capacitance measurement result is used as basic training data, and meanwhile, the training data is corrected based on the correction model, so that the system error of the measurement data is eliminated, the training data acquisition efficiency and the training data acquisition precision are improved, and the defect that the training data is acquired based on the prior art is avoided.
Example 6
The invention provides an electronic device comprising a memory, a storage, a control unit and a control unit, wherein executable instructions are stored in the memory; and the processor runs executable instructions in the memory to realize the grain moisture detection modeling method for the capacitance and multispectral imaging.
An electronic device according to an embodiment of the invention includes a memory and a processor.
The memory is for storing non-transitory computer readable instructions. In particular, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like.
The processor may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities, and may control other components in the electronic device to perform the desired functions. In one embodiment of the invention, the processor is configured to execute the computer readable instructions stored in the memory.
It should be understood by those skilled in the art that, in order to solve the technical problem of how to obtain a good user experience effect, the present embodiment may also include well-known structures such as a communication bus, an interface, and the like, and these well-known structures are also included in the protection scope of the present invention.
The detailed description of the present embodiment may refer to the corresponding description in the foregoing embodiments, and will not be repeated herein.
Claims (10)
1. A modeling method for grain moisture detection, comprising:
S100: sample preparation: s random grain samples are obtained, the S random grain samples are divided into M grain samples for calculating a correction model, N grain samples for training a detection model and L grain samples for verifying a detection result in a random extraction mode, wherein S, M, N, L are natural numbers, M < N, (M+N+L) < S;
s110: calculating a correction model: for the M grain samples, respectively acquiring the water content of the samples based on a drying detection method and a capacitive rapid detection method, generating error coordinate mapping points, and fitting by using a least square method to obtain a correction model;
S120: training a detection model: detecting the water content of the samples by a capacitive rapid detection method aiming at the N grain samples, extracting spectral image data of the samples to obtain spectral image texture features, inputting the water content of the samples and the spectral image texture features into an initial multiple regression model for training and learning to obtain a grain water content detection model; the initial multiple regression model incorporates a correction model for error correction operation of training data;
S130: verification and output detection model: and aiming at the L grain samples, acquiring the water content of the samples based on a drying detection method, outputting the predicted water content of the samples based on a detection model, and verifying the detection model based on an estimated standard error.
2. The modeling method for grain moisture detection of claim 1, wherein:
In step S110, the calculated correction model specifically includes: taking the M grain samples, dividing any grain Sample in the M grain samples into p subsamples { Sample 1,Sample2,……Samplep }, wherein the grain water content of each subsample in the p subsamples is the same, and p is a natural number; taking one Sample1 in the p subsamples, and detecting the water content of the subsamples by a drying detection method to obtain subsamples, drying and measuring the water content Taking the rest p-1 parts in the p parts of subsamplesThe average moisture content of the subsamples is obtained by detecting the average moisture content value of the subsamples by a capacitance type rapid detection methodRecording and drying to measure water contentCorresponding capacitance measurement of average water contentError value of (2)=Obtaining a sample error coordinate mapping point corresponding to the sample,) Performing the steps for each grain sample in the M grain samples to obtain M sample error coordinate mapping points, and fitting the M sample error coordinate mapping points by using a least square method to obtain a correction modelWherein, the method comprises the steps of, wherein,Is thatIs used for correcting the error of the (a),Fitting a curve;
In step S120, the training detection model specifically includes: taking the N grain samples, and equally dividing any grain Sample in the N grain samples into 2 sub-samples { Sample a1,Samplea2 }, wherein the water content of the sub-samples Sample a1 is the same as that of the sub-Sample a2; sampling a1, and detecting the water content by a capacitive rapid detection method ; Taking a sub-Sample a2, extracting spectral image data of the sub-Sample a2, preprocessing the spectral image data, and extracting spectral image texture features C= { C1, C2, … …, cn }, wherein ci is a spectral feature value with a sequence number i, and i is more than or equal to 1 and less than or equal to n; will be%C) adding training set T as training data; executing the steps on each grain sample in the N grain samples to obtain N groups of training data so as to add a training set T; inputting the training set T into an initial multiple regression model for training and learning to obtain a detection model for outputting the water content of grains; the initial multiple regression model incorporates a correction modelThe error correction operation is used for carrying out error correction operation on the training data;
In step S130, the verification and output detection model specifically includes: taking the L grain samples, and equally dividing each grain Sample in the L grain samples into 2 sub-samples { Sample b1,Sampleb2 }, wherein the water content of the sub-samples Sample b1 is the same as that of the sub-Sample b2; sampling Sample b1, and detecting the water content by a drying detection method to obtain a sub Sample, drying and measuring the water content ; Taking a subsamples Sample b2, extracting spectral image data of the subsamples Sample b2 to obtain spectral image texture features C ', and inputting the spectral image texture features C' into a detection model to obtain a model output resultValidating the estimated standard errorWhere k is the number of elements of the spectral image texture feature C'; if the standard error is within the threshold range, outputting a trained detection model for detecting the moisture content of the grains; otherwise, correcting the learning sample optimization model.
3. A modeling method for grain moisture detection according to claim 2, comprising: and acquiring a spectrum image of the grain sample through an imaging system, and further extracting characteristic data of the spectrum image of the grain sample, wherein the characteristic data of the spectrum image are a grain image and a near infrared spectrum.
4. A modeling method for grain moisture detection according to claim 2, comprising:
In step S110, different environmental parameters are recorded to generate environmental parameter sets E= { E1, E2, … …, ek }, and the average water content is measured by the environmental parameters and the capacitance Is an independent variable and an error valueTraining multiple regression correction models for dependent variablesReplacing the correction model in step S110Wherein, the method comprises the steps of, wherein,Correcting errors predicted for the correction model;
in step S120, the current environmental parameters are recorded to generate a current environmental parameter set E '= { E 1',e2',……,ek' }, to be (" E'), C) adding training data into a training set T; the initial multiple regression model incorporates a multiple regression correction modelAnd the error correction operation is used for carrying out error correction operation on the training data.
5. The modeling method for grain moisture detection of claim 4, wherein: the environmental parameters include at least temperature and humidity.
6. A modeling method for grain moisture detection as defined in claim 2 wherein the preprocessing of the spectral image data comprises: and filtering, denoising and normalizing the spectrum image data.
7. A modeling system for grain moisture detection, comprising:
The sample acquisition module is used for acquiring S pieces of random grain samples, and dividing the S pieces of random grain samples into M pieces of grain samples for calculating a correction model, N pieces of grain samples for training a detection model and L pieces of grain samples for verifying a detection result according to a random extraction mode; wherein M < N, (m+n+l) < S;
The correction model acquisition module is used for calculating a correction model: for the M grain samples, respectively acquiring the water content of the samples based on a drying detection method and a capacitive rapid detection method, generating error coordinate mapping points, and fitting by using a least square method to obtain a correction model;
The detection model acquisition module is used for training a detection model: detecting the water content of the samples by a capacitive rapid detection method aiming at the N grain samples, extracting spectral image data of the samples to obtain spectral image texture features, inputting the water content of the samples and the spectral image texture features into an initial multiple regression model for training and learning to obtain a grain water content detection model; the initial multiple regression model incorporates a correction model for error correction operation of training data;
The verification and output module is used for verifying and outputting a detection model: and aiming at the L grain samples, acquiring the water content of the samples based on a drying detection method, outputting the predicted water content of the samples based on a detection model, and verifying the detection model based on an estimated standard error.
8. A grain moisture detection apparatus applied to the method of any one of claims 1 to 6, comprising: the device comprises a main control module, a capacitive moisture calibration module, a visible light-near infrared light source module, an imaging module and a sample transmission module;
The main control module comprises a main control board and is used for storing basic data, analyzing and calculating, training and optimizing a model; the main control module comprises an image analysis and processing module, and is used for processing, collecting and analyzing spectral images of grain samples aiming at video streams and extracting corresponding spectral image characteristics;
The capacitive moisture calibration module comprises a cylindrical capacitive sensor and a resonance detection circuit, and a funnel device is arranged below the cylindrical capacitive sensor to facilitate material removal; the capacitive moisture calibration module is used for detecting the moisture content of different grain samples through different dielectric constants of grains in the sensor and inputting the detection result to the main control module;
the visible light-near infrared light source and the imaging module are used for acquiring image information and near infrared spectrum information of the grain sample;
the sample transmission module is used for placing and conveying samples.
9. A grain moisture detection apparatus for data interaction with the grain moisture detection apparatus of claim 8, applied to the method of any one of claims 1 to 6, comprising a main control module, a capacitive moisture calibration module, and a dry moisture calibration module;
The main control module comprises a main control board and is used for storing basic data, analyzing and calculating, training and optimizing a model;
The capacitive moisture calibration module comprises a cylindrical capacitive sensor and a resonance detection circuit, and a funnel device is arranged below the cylindrical capacitive sensor to facilitate material removal; the capacitive moisture calibration module is used for detecting the moisture content of different grain samples through different dielectric constants of grains in the sensor and inputting the detection result to the main control module;
The drying type moisture calibration module is used for detecting the moisture content of different grain samples and inputting the detection result to the main control module.
10. A modeling system for grain moisture detection, characterized by: the modeling system is composed of the grain moisture detection apparatus of claim 8 and the grain moisture detection apparatus of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410764061.9A CN118334650B (en) | 2024-06-14 | 2024-06-14 | Modeling method, system and equipment for grain moisture detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410764061.9A CN118334650B (en) | 2024-06-14 | 2024-06-14 | Modeling method, system and equipment for grain moisture detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118334650A true CN118334650A (en) | 2024-07-12 |
| CN118334650B CN118334650B (en) | 2024-08-16 |
Family
ID=91776356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410764061.9A Active CN118334650B (en) | 2024-06-14 | 2024-06-14 | Modeling method, system and equipment for grain moisture detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118334650B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113049530A (en) * | 2021-03-17 | 2021-06-29 | 北京工商大学 | Single-seed corn seed moisture content detection method based on near-infrared hyperspectrum |
| CN113504251A (en) * | 2021-08-13 | 2021-10-15 | 河南工业大学 | Grain moisture rapid detection method and system based on radio frequency signals |
| CN115937610A (en) * | 2023-01-05 | 2023-04-07 | 鞍钢集团矿业有限公司 | Iron ore concentrate water content online detection method based on image characteristics |
| US20230292647A1 (en) * | 2020-06-16 | 2023-09-21 | Dark Horse Technologies Ltd | System and Method for Crop Monitoring |
| US20230316555A1 (en) * | 2020-08-14 | 2023-10-05 | Agriculture Victoria Services Pty Ltd | System and Method for Image-Based Remote Sensing of Crop Plants |
| WO2023222456A1 (en) * | 2022-05-19 | 2023-11-23 | Yara International Asa | Method and system for grain moisture measurement |
| CN117593742A (en) * | 2023-11-30 | 2024-02-23 | 航天信息股份有限公司 | Method, device, medium and equipment for detecting imperfect rate of grains |
| CN118053517A (en) * | 2024-02-05 | 2024-05-17 | 长春理工大学 | Grain moisture model building method based on optimized support vector regression |
-
2024
- 2024-06-14 CN CN202410764061.9A patent/CN118334650B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230292647A1 (en) * | 2020-06-16 | 2023-09-21 | Dark Horse Technologies Ltd | System and Method for Crop Monitoring |
| US20230316555A1 (en) * | 2020-08-14 | 2023-10-05 | Agriculture Victoria Services Pty Ltd | System and Method for Image-Based Remote Sensing of Crop Plants |
| CN113049530A (en) * | 2021-03-17 | 2021-06-29 | 北京工商大学 | Single-seed corn seed moisture content detection method based on near-infrared hyperspectrum |
| CN113504251A (en) * | 2021-08-13 | 2021-10-15 | 河南工业大学 | Grain moisture rapid detection method and system based on radio frequency signals |
| WO2023222456A1 (en) * | 2022-05-19 | 2023-11-23 | Yara International Asa | Method and system for grain moisture measurement |
| CN115937610A (en) * | 2023-01-05 | 2023-04-07 | 鞍钢集团矿业有限公司 | Iron ore concentrate water content online detection method based on image characteristics |
| CN117593742A (en) * | 2023-11-30 | 2024-02-23 | 航天信息股份有限公司 | Method, device, medium and equipment for detecting imperfect rate of grains |
| CN118053517A (en) * | 2024-02-05 | 2024-05-17 | 长春理工大学 | Grain moisture model building method based on optimized support vector regression |
Non-Patent Citations (2)
| Title |
|---|
| JING ZHANG ET AL.: "Raipid determination of protein, starch and moisture content in wheat flour by near-infrared hyperspectral imaging", 《JOURNAL OF FOOD COMPOSITION AND ANALYSIS》, 6 January 2023 (2023-01-06), pages 1 - 11 * |
| 王乐军等: "植物性食品原料水分含量检测研究进展", 《食品与机械》, 31 August 2023 (2023-08-31), pages 218 - 226 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118334650B (en) | 2024-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Guo et al. | Nondestructive measurement of soluble solids content of kiwifruits using near-infrared hyperspectral imaging | |
| CN110687072B (en) | Calibration set and verification set selection and modeling method based on spectral similarity | |
| Dong et al. | Nondestructive determination of soluble solids content of ‘Fuji’apples produced in different areas and bagged with different materials during ripening | |
| CN105842190B (en) | A kind of method for transferring near infrared model returned based on spectrum | |
| Li et al. | Wavelength selection method for near-infrared spectroscopy based on standard-sample calibration transfer of mango and apple | |
| CN114676636A (en) | Grassland area soil moisture rapid inversion method integrating vegetation and habitat characteristics | |
| Duan et al. | Sensitive variables extraction, non-destructive detection and visualization of total viable count (TVC) and pH in vacuum packaged lamb using hyperspectral imaging | |
| CN116026795A (en) | Rice grain quality character nondestructive prediction method based on reflection and transmission spectrum | |
| CN115809427A (en) | Mixed gas identification method based on mixed strategy optimization BP neural network | |
| CN110609011A (en) | Near-infrared hyperspectral detection method and system for starch content of single-kernel corn seeds | |
| CN117517245B (en) | Method, system, apparatus and medium for evaluating tea flavor, aroma and overall sensory | |
| CN114527082A (en) | Sweet corn seed germination prediction method and device | |
| Sun et al. | Nondestructive identification of barley seeds varieties using hyperspectral data from two sides of barley seeds | |
| Liu et al. | Detection of Apple Taste Information Using Model Based on Hyperspectral Imaging and Electronic Tongue Data. | |
| CN118334650B (en) | Modeling method, system and equipment for grain moisture detection | |
| CN112945901A (en) | Method for detecting quality of ensiled soybeans based on near infrared spectrum | |
| CN116662751A (en) | Tobacco leaf moisture content detection method for removing abnormal samples based on principal component analysis and lever value method | |
| CN110632024B (en) | Quantitative analysis method, device and equipment based on infrared spectrum and storage medium | |
| CN115541531A (en) | Method for predicting protein content in feed based on two-dimensional correlation spectrum | |
| CN115049902B (en) | Visual prediction method, system, equipment and storage medium for water content of citrus blade | |
| CN116310788A (en) | Plant leaf nitrogen content hyperspectral modeling method based on map fusion | |
| CN115327023A (en) | Plant leaf nitrogen content detection method and system | |
| CN114813593A (en) | Method for detecting total acid content of fermented grains based on hyperspectral imaging technology | |
| CN114942233A (en) | Near infrared spectrum characteristic wavelength selection method, device, equipment and storage medium | |
| CN114544530A (en) | Tobacco shred component detection method and device and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |