Cereal fatty acid value detection method and device based on multi-dimensional spectrum

Abstract

The invention provides a cereal fatty acid value detection method and device based on a multi-dimensional spectrum, and relates to the technical field of grain and oil detection.The method comprises the steps that a plurality of light beams penetrate through to-be-detected cereals, and the multi-dimensional spectrum corresponding to the to-be-detected cereals is obtained; inputting the multi-dimensional spectrum corresponding to the to-be-detected grain into a constructed grain fatty acid detection model corresponding to the color of the to-be-detected grain to obtain a fatty acid value of the to-be-detected grain; the cereal fatty acid detection model is constructed through an actual fatty acid value of a cereal sample and a multi-dimensional spectrum corresponding to the cereal sample. According to the method, pretreatment and professional operation are not needed, the detection time is 1-2 minutes, interference of grain colors is avoided, rapid nondestructive detection of the grain fatty acid is achieved, and the detection efficiency and accuracy of the grain fatty acid value are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of grain and oil detection, and particularly relates to a method and device for detecting the fatty acid value of grains based on multi-dimensional spectra. Background Art

[0002] The fatty acid value of grains (such as corn and paddy rice, etc.) can be used to judge the degree of deterioration of grain quality, and is used as an appropriate storage index for paddy rice and corn in the national standard grain storage determination rules.

[0003] Common detection means for the fatty acid value of grains are: manual titration or automatic instrument titration, and automatic instrument titration is further divided into photometric titration and potentiometric titration. However, the detection efficiency of both manual titration and automatic instrument titration is not high, the operation methods of operators have a great influence on the detection results, and the detection results are prone to large deviations. Summary of the Invention

[0004] The present invention provides a method and device for detecting the fatty acid value of grains based on multi-dimensional spectra, so as to solve the defects of low efficiency and low accuracy in the detection of the fatty acid value of grains in the prior art, and achieve the improvement of the efficiency and accuracy of the detection of the fatty acid value of grains.

[0005] The present invention provides a method for detecting the fatty acid value of grains based on multi-dimensional spectra, including: Passing multiple light beams through the grains to be detected to obtain the multi-dimensional spectra corresponding to the grains to be detected; Inputting the multi-dimensional spectra corresponding to the grains to be detected into a constructed grain fatty acid detection model corresponding to the color of the grains to be detected to obtain the fatty acid value of the grains to be detected; the grain fatty acid detection model is constructed by the actual fatty acid value of grain samples and the multi-dimensional spectra corresponding to the grain samples.

[0006] In some embodiments, the method further includes: Using a light wavelength of 680nm - 780nm to pass through the grains to be detected to obtain the spectral data of the grains to be detected; Inputting the spectral data into a constructed grain color determination model to obtain the color of the grains to be detected; wherein, the grain color determination model is constructed by the spectral data of grain samples of different colors at a light wavelength of 680nm - 780nm.

[0007] In some embodiments, the method further includes: Obtaining the multi-dimensional spectra corresponding to grain samples of different colors; Construct a grain fatty acid detection model corresponding to each grain color based on the proportions of the grain samples of different colors, the multi-dimensional spectra corresponding to the grain samples of different colors, and the actual fatty acid values of the grain samples of different colors.

[0008] In some embodiments, the method further includes: When the proportions of the grain samples of different colors are determined, divide the multi-dimensional spectra corresponding to the grain samples of each color into multiple segments; Establish a multiple regression model by respectively combining the actual fatty acid values of the grain samples of different colors with all segments, any segment, and segment combinations of the multi-dimensional spectra corresponding to the grain samples of different colors; Use the multiple regression model with the optimal model performance as the grain fatty acid detection model.

[0009] In some embodiments, the obtaining of the multi-dimensional spectra corresponding to the grain samples of different colors includes: Divide the grain samples of each color into multiple sub-grain samples; Pass the multiple light beams through the multiple sub-grain samples respectively to obtain the multi-dimensional spectra corresponding to each sub-grain sample; Combine the multi-dimensional spectra corresponding to the multiple sub-grain samples respectively to obtain the multi-dimensional spectra corresponding to the grain samples of each color.

[0010] In some embodiments, the method further includes: Spectrally split the light waveband from visible light to near-infrared light in the range of 680 nm - 1080 nm to generate the multiple light beams.

[0011] In some embodiments, when the grain is paddy rice, the paddy rice spectral combination corresponding to the multiple regression model with the optimal model performance is: 725 - 840 nm, 832 - 946 nm, 960 - 1050 nm.

[0012] In some embodiments, when the grain is corn, the corn spectral combination corresponding to the multiple regression model with the optimal model performance is: 692 - 780 nm, 898 - 1050 nm.

[0013] The present invention also provides a device for detecting the fatty acid value of grains based on multi-dimensional spectra, including: A first acquisition module for passing multiple light beams through the grain to be detected to obtain the multi-dimensional spectra corresponding to the grain to be detected; The first input module is configured to input the multi-dimensional spectrum corresponding to the grain to be detected into a constructed grain fatty acid detection model corresponding to the color of the grain to be detected, so as to obtain the fatty acid value of the grain to be detected; the grain fatty acid detection model is constructed by the actual fatty acid value of a grain sample and the multi-dimensional spectrum corresponding to the grain sample.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method for detecting the fatty acid value of grains based on multi-dimensional spectrum as described in any one of the above is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for detecting the fatty acid value of grains based on multi-dimensional spectrum as described in any one of the above is implemented.

[0016] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, the method for detecting the fatty acid value of grains based on multi-dimensional spectrum as described in any one of the above is implemented.

[0017] The method and device for detecting the fatty acid value of grains based on multi-dimensional spectrum provided by the present invention construct a grain fatty acid detection model through the actual fatty acid value of a grain sample and the multi-dimensional spectrum corresponding to the grain sample, input the multi-dimensional spectrum corresponding to the grain to be detected into the constructed grain fatty acid detection model corresponding to the color of the grain to be detected, and obtain the fatty acid value of the grain to be detected. Without pretreatment and the operation of professional personnel, the detection time is 1-2 minutes, avoiding the interference of the grain color, realizing the rapid and non-destructive detection of the grain fatty acid, and improving the detection efficiency and accuracy of the grain fatty acid value. Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a schematic flowchart of the method for detecting the fatty acid value of grains based on multi-dimensional spectrum provided by the present invention; Figure 2 It is a segmented result diagram of the multi-dimensional spectrum of paddy rice; Figure 3 It is a segmented result diagram of the multi-dimensional spectrum of corn; Figure 4 It is a detection result diagram of the fatty acid value of paddy rice; Figure 5 It is a detection result diagram of the fatty acid value of corn; Figure 6 It is a schematic structural diagram of a grain fatty acid value detection device based on multi-dimensional spectra provided by the present invention; Figure 7 It is a schematic structural diagram of an electronic device provided by the present invention. Specific embodiments

[0020] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts fall within the scope of protection of the present invention.

[0021] Figure 1 It is a schematic flowchart of a method for detecting the fatty acid value of grains based on multi-dimensional spectra provided by the present invention. As Figure 1 shown, the present invention provides a method for detecting the fatty acid value of grains based on multi-dimensional spectra, including: Step 110, passing multiple light beams through the grain to be detected to obtain the multi-dimensional spectrum corresponding to the grain to be detected.

[0022] Specifically, the grain to be detected is gradually unloaded through structures such as a silo and a wave wheel, and multiple light beams are passed through the grain to be detected. The light energy data of each light beam after passing through the grain to be detected is detected, and this light energy data can reflect the absorption degree of the fatty acid of the grain to be detected to the light beam.

[0023] Obtaining the light energy data of multiple light beams after passing through the grain to be detected is to obtain the multi-dimensional spectrum corresponding to the grain to be detected. The grain to be detected is gradually unloaded through structures such as a silo and a wave wheel to obtain the multi-dimensional spectra of the grain to be detected at different positions.

[0024] Step 120, inputting the multi-dimensional spectrum corresponding to the grain to be detected into a constructed grain fatty acid detection model corresponding to the color of the grain to be detected to obtain the fatty acid value of the grain to be detected; the grain fatty acid detection model is constructed based on the actual fatty acid value of the grain sample and the multi-dimensional spectrum corresponding to the grain sample.

[0025] Specifically, a grain fatty acid detection model is constructed based on the actual fatty acid value of the grain sample and the multi-dimensional spectrum corresponding to the grain sample, that is, the correlation relationship between the fatty acid value and the multi-dimensional spectrum is constructed, so that the corresponding fatty acid value can be determined according to the multi-dimensional spectrum.

[0026] In the scenario of detecting the fatty acid value of grains, the color of the grains will interfere with the detection of the fatty acid value of the grains, resulting in inaccurate detection of the fatty acid value of the grains. Therefore, considering the influence of the grain color, a grain fatty acid detection model corresponding to each grain color is established, that is, using the grain fatty acid detection model to detect the fatty acid value of the grains of the corresponding color, and the accuracy rate of the obtained grain fatty acid value is the highest.

[0027] Input the multi-dimensional spectrum corresponding to the grain to be detected into the established grain fatty acid detection model corresponding to the color of the grain to be detected, so as to obtain the fatty acid value of the grain to be detected.

[0028] The method for detecting the fatty acid value of grains based on multi-dimensional spectrum provided by the present invention constructs a grain fatty acid detection model through the actual fatty acid value of the grain sample and the multi-dimensional spectrum corresponding to the grain sample, inputs the multi-dimensional spectrum corresponding to the grain to be detected into the established grain fatty acid detection model corresponding to the color of the grain to be detected, and obtains the fatty acid value of the grain to be detected. It does not require pretreatment, does not require professional personnel to operate, has a detection time of 1-2 minutes, avoids the interference of grain color, realizes rapid and non-destructive detection of grain fatty acid, and improves the detection efficiency and accuracy of grain fatty acid value.

[0029] In some embodiments, the actual fatty acid value of the grain sample is obtained by separately obtaining multiple titration results for multiple parallel grain samples through manual titration or instrument automatic titration, and averaging the multiple titration results.

[0030] Specifically, multiple parallel grain samples are separately and strictly in accordance with the requirements or steps of manual titration or instrument automatic titration specified by international regulations to obtain multiple titration results, and then the multiple titration results are averaged, and the final average value is regarded as the actual fatty acid value of the grain sample.

[0031] Since obtaining the grain fatty acid value based on manual titration or instrument automatic titration is a prior art, the specific content of obtaining the grain fatty acid value based on manual titration or instrument automatic titration will not be elaborated here.

[0032] In some embodiments, the method for detecting the fatty acid value of grains based on multi-dimensional spectrum provided by the present invention further includes: Using a light wavelength of 680nm-780nm to pass through the grain to be detected to obtain the spectral data of the grain to be detected; Input the spectral data into the established grain color determination model to obtain the color of the grain to be detected; wherein, the grain color determination model is constructed through the spectral data of grain samples of different colors at a light wavelength of 680nm-780nm.

[0033] Specifically, since 680 - 780 nm belongs to the visible light range, it can better identify colors. Therefore, spectral data of cereal samples of different colors (such as yellow cereal samples, white cereal samples, green cereal samples) at a light wavelength of 680 nm - 780 nm are used to train a deep learning model, so that the deep learning model can identify the spectral data of cereal samples of different colors at a light wavelength of 680 nm - 780 nm, realizing the identification of the colors of cereal samples, and thus obtaining a constructed cereal color determination model capable of measuring the color of cereals.

[0034] Optionally, different grades can be further divided according to different shades of the same color. For example, yellow grade 1, yellow grade 2, yellow grade 3, white grade 1, white grade 2, green grade 1, green grade 2, green grade 3, green grade 4.

[0035] Optionally, the deep learning model can be a convolutional neural network model, a recurrent neural network, a Transformer model, etc. The deep learning model can be selected according to the actual situation, and the present invention does not specifically limit the deep learning model.

[0036] Use a light wavelength of 680 nm - 780 nm to pass through the cereal to be detected to obtain the spectral data of the cereal to be detected, and input the spectral data into the constructed cereal color determination model to obtain the color of the cereal to be detected.

[0037] The method for detecting the fatty acid value of cereals based on multi-dimensional spectra provided by the present invention constructs a cereal color determination model through the spectral data of cereal samples of different colors at a light wavelength of 680 nm - 780 nm, uses a light wavelength of 680 nm - 780 nm to pass through the cereal to be detected to obtain the spectral data of the cereal to be detected; inputs the spectral data into the constructed cereal color determination model to accurately measure the color of the cereal to be detected, so as to subsequently select the corresponding cereal fatty acid detection model according to the color of the cereal to be detected, further improving the detection accuracy of the fatty acid value of cereals.

[0038] In some embodiments, the method for detecting the fatty acid value of cereals based on multi-dimensional spectra provided by the present invention further includes: Obtain the multi-dimensional spectra corresponding to cereal samples of different colors; Construct a cereal fatty acid detection model corresponding to each cereal color according to the proportion of cereal samples of different colors, the multi-dimensional spectra corresponding to cereal samples of different colors, and the actual fatty acid values of cereal samples of different colors.

[0039] Specifically, obtain the multi-dimensional spectra corresponding to cereal samples of different colors. Based on the multi-dimensional spectra corresponding to cereal samples of different colors, as well as the actual fatty acid values of cereal samples of different colors, adjust the proportions of cereal samples of different colors to construct a cereal fatty acid detection model corresponding to each cereal color.

[0040] For the construction of a cereal fatty acid detection model corresponding to a certain cereal color, specifically, it can be: constructed using the multi-dimensional spectra corresponding to multiple cereal samples of different colors and the actual fatty acid values of multiple cereal samples of different colors. Among the multiple cereal samples of different colors, it includes cereal samples of this certain cereal color, and the proportion of cereal samples of this certain cereal color in the sample is relatively large, while the proportion of cereal samples of other colors in the sample is relatively small.

[0041] For different cereal fatty acid detection models, the proportions of cereal samples of the same color may be different or the same. The specific proportion of cereal samples of different colors can be adaptively adjusted according to the actual situation.

[0042] Exemplarily, aiming at constructing a cereal fatty acid detection model corresponding to yellow cereals, the specific process can be: First, obtain a large number of yellow cereal samples, as well as a small number of white cereal samples and green cereal samples; Then, based on the multi-dimensional spectra corresponding to a large number of yellow cereal samples, the multi-dimensional spectra corresponding to a small number of white cereal samples, the multi-dimensional spectra corresponding to a small number of green cereal samples, as well as the actual fatty acid values of a large number of yellow cereal samples, the actual fatty acid values of a small number of white cereal samples, and the actual fatty acid values of a small number of green cereal samples, construct a cereal fatty acid detection model to obtain a cereal fatty acid detection model corresponding to yellow cereals.

[0043] The method for detecting the fatty acid value of cereals based on multi-dimensional spectra provided by the present invention accurately constructs a cereal fatty acid detection model corresponding to each cereal color by adjusting the proportions of cereal samples of different colors according to the multi-dimensional spectra corresponding to cereal samples of different colors and the actual fatty acid values of cereal samples of different colors, which further helps to improve the detection accuracy of the fatty acid value of cereals.

[0044] In some embodiments, obtaining the multi-dimensional spectra corresponding to cereal samples of different colors includes: Dividing each color of cereal sample into multiple cereal sub-samples; Passing multiple light beams through the multiple cereal sub-samples respectively to obtain the multi-dimensional spectra corresponding to each cereal sub-sample; Combining the multi-dimensional spectra corresponding to the multiple cereal sub-samples respectively to obtain the multi-dimensional spectra corresponding to each color of cereal sample.

[0045] Specifically, in order to effectively solve the problem of non-uniformity of cereal samples, the following operations are performed on cereal samples of each color to obtain the multi-dimensional spectra corresponding to the cereal samples of each color: For a cereal sample of a certain color, the cereal sample of this color is divided into multiple cereal sub-samples. Multiple light beams are respectively passed through the multiple cereal sub-samples to obtain the multi-dimensional spectra corresponding to each cereal sub-sample, and the multi-dimensional spectra corresponding to the multiple cereal sub-samples are combined to obtain the multi-dimensional spectrum corresponding to the cereal sample of this color.

[0046] The method for detecting the fatty acid value of cereals based on multi-dimensional spectra provided by the present invention divides the cereal samples of each color into multiple cereal sub-samples, passes multiple light beams through the multiple cereal sub-samples respectively to obtain the multi-dimensional spectra corresponding to each cereal sub-sample, and combines the multi-dimensional spectra corresponding to the multiple cereal sub-samples to obtain the multi-dimensional spectra corresponding to the cereal samples of each color, avoiding the problem of non-uniformity of cereal samples, and further facilitating the improvement of the detection accuracy of the fatty acid value of cereals.

[0047] In some embodiments, the method for detecting the fatty acid value of cereals based on multi-dimensional spectra provided by the present invention further includes: When the proportions of cereal samples of different colors are determined, the multi-dimensional spectra corresponding to the cereal samples of each color are divided into multiple segments; The actual fatty acid values of cereal samples of different colors are respectively used to establish multiple regression models with all segments, any segment, and segment combinations of the multi-dimensional spectra corresponding to the cereal samples of different colors; The multiple regression model with the optimal model performance is used as the cereal fatty acid detection model.

[0048] Specifically, when the proportions of cereal samples of different colors are determined, the multi-dimensional spectra corresponding to the cereal samples of different colors are divided into multiple segments according to wavelength points, that is, multiple beams.

[0049] Consider the different effects of different segments of the multi-dimensional spectrum on the detection accuracy of the fatty acid value of cereals. Multiple regression models are established based on the actual fatty acid values of cereal samples of different colors and all segments of the multi-dimensional spectra corresponding to the cereal samples of different colors. Multiple regression models are established based on the actual fatty acid values of cereal samples of different colors and any segment of the multi-dimensional spectra corresponding to the cereal samples of different colors. Multiple regression models are established based on the actual fatty acid values of cereal samples of different colors and segment combinations of the multi-dimensional spectra corresponding to the cereal samples of different colors.

[0050] The multiple regression model with the optimal model performance is used as the grain fatty acid detection model. Among them, the model performance is evaluated by the following methods: 1. Detect the fatty acid values of a batch of grain samples respectively and measure their fatty acid values by the national standard physical and chemical methods; 2. Calculate the correlation of these two groups of data. The higher the correlation, the better the model performance; 3. Calculate the mean square error of these two groups of data. The smaller the mean square error, the better the model performance. Select the multiple regression model with high correlation and small mean square error as the grain fatty acid detection model.

[0051] In some embodiments, when the grain is paddy rice, the paddy rice spectral combinations corresponding to the multiple regression model with the optimal model performance are: 725 - 840 nm, 832 - 946 nm, 960 - 1050 nm.

[0052] Specifically, this solution collected more than 10,000 representative paddy rice samples, specifically including paddy rice samples of multiple varieties from 11 provincial administrative regions such as Guangdong, Guangxi, Sichuan, Hunan, Hubei, Anhui, Jiangsu, Jiangxi, Heilongjiang, Jilin, and Liaoning. The collected samples are highly representative and practical in terms of both region and quantity.

[0053] By collecting multi-dimensional spectra of these samples in large quantities and obtaining fatty acid values through a large number of titration methods, and repeatedly performing multi-dimensional spectral segmentation modeling, the optimal spectral combinations of paddy rice are finally selected as: 725 - 840 nm, 832 - 946 nm, 960 - 1050 nm. Figure 2 It is the segmented result diagram of the multi-dimensional spectrum of paddy rice, as Figure 2 shown, the multi-dimensional spectrum of paddy rice from 680 nm to 1080 nm is segmented. The horizontal axis is the spectrum (unit: nm), the vertical axis is the absorbance. The black curve in the figure is the multi-dimensional spectrum of paddy rice, and the rectangular columns in the figure are the optimal spectral segmentation combinations determined according to the paddy rice fatty acid detection model.

[0054] In some embodiments, when the grain is corn, the corn spectral combinations corresponding to the multiple regression model with the optimal model performance are: 692 - 780 nm, 898 - 1050 nm.

[0055] Specifically, this solution collected more than 5,000 representative corn samples, specifically including corn samples from the three major corn-producing provinces of Heilongjiang, Jilin, and Liaoning, as well as some representative samples from other provinces. The collected samples are highly representative and practical in terms of both region and quantity.

[0056] By collecting multi-dimensional spectra of these samples in large quantities and obtaining fatty acid values through a large number of titration methods, and repeatedly performing multi-dimensional spectral segmentation modeling, the optimal spectral combinations of paddy rice are finally selected as: 692 - 780 nm, 898 - 1050 nm. Figure 3It is the segmented result diagram of the multi-dimensional spectrum of corn. As Figure 3 shown, the multi-dimensional spectrum of corn in the range of 680 - 1080 nm is segmented. The horizontal axis is the spectrum (unit: nm), and the vertical axis is the absorbance. The black curve in the figure is the multi-dimensional spectrum of paddy rice, and the rectangular columns in the figure are the optimal spectral segmentation combinations determined according to the corn fatty acid detection model.

[0057] The method for detecting the fatty acid value of grains based on multi-dimensional spectrum provided by the present invention divides the multi-dimensional spectra corresponding to grain samples of different colors into multiple segments, and respectively establishes multiple regression models between the actual fatty acid values of grain samples of different colors and all segments, any segment, and segment combinations of the multi-dimensional spectra corresponding to grain samples of different colors. The multiple regression model with the optimal model performance is used as the grain fatty acid detection model, further improving the accuracy of detecting the fatty acid value of grains by the grain fatty acid detection model.

[0058] In some embodiments, the method for detecting the fatty acid value of grains based on multi-dimensional spectrum provided by the present invention further includes: Splitting the light wave band from visible light to near-infrared light in the range of 680 nm - 1080 nm to generate multiple light beams.

[0059] Specifically, splitting the light wave band from visible light to near-infrared light to generate multiple light beams. A light splitting device can be used for splitting, and the light splitting device can be a spectroscope, a beam splitter, a grating, a prism, etc.

[0060] Through a large number of experimental verifications and combined with theoretical knowledge, it is concluded that selecting light beams from the visible to near-infrared band has the following advantages: 1. Compared with infrared light, the penetration ability of this part of the light wave is stronger; 2. This part of the light wave carries light beams highly correlated with the fat content.

[0061] The method for detecting the fatty acid value of grains based on multi-dimensional spectrum provided by the present invention splits the light wave from visible light to near-infrared light in the range of 680 nm - 1080 nm to generate multiple light beams, which is further beneficial to improving the detection accuracy of the fatty acid value of grains.

[0062] Applying the method provided by the present invention to the detection of the fatty acid value of paddy rice, Figure 4 It is the detection result diagram of the fatty acid value of paddy rice. As Figure 4 shown, the correlation between the detected value of the fatty acid value of paddy rice and the physical and chemical value of the fatty acid value of paddy rice is 0.9, and the prediction deviation of most samples is within ±3 mg / 100 g (KOH).

[0063] Applying the method provided by the present invention to the detection of the fatty acid value of corn, Figure 5 It is the detection result diagram of the fatty acid value of corn. As Figure 5As shown, the correlation between the detected value of the corn fatty acid value and the physical and chemical value of the corn fatty acid value is 0.92, and the prediction deviation of most samples is within ±6 mg / 100 g (KOH).

[0064] From the above experimental results, it can be seen that the performance difference between this method and the national standard method is small. However, this method does not require sample destruction, reagents, or professional personnel for operation, and the test result can be obtained within only 1 - 2 minutes. It can be used as a rapid detection method for screening the fatty acid value of grains, effectively complementing the national standard method.

[0065] Next, the grain fatty acid value detection device based on multi - dimensional spectroscopy provided by the present invention will be described. The grain fatty acid value detection device based on multi - dimensional spectroscopy described below can be mutually referred to with the grain fatty acid value detection method based on multi - dimensional spectroscopy described above.

[0066] Figure 6 is a schematic structural diagram of the grain fatty acid value detection device based on multi - dimensional spectroscopy provided by the present invention. As Figure 6 shown, the present invention provides a grain fatty acid value detection device based on multi - dimensional spectroscopy, including: A first acquisition module 610, configured to pass multiple light beams through the grain to be detected to obtain the multi - dimensional spectrum corresponding to the grain to be detected; A first input module 620, configured to input the multi - dimensional spectrum corresponding to the grain to be detected into a pre - constructed grain fatty acid detection model corresponding to the color of the grain to be detected to obtain the fatty acid value of the grain to be detected; the grain fatty acid detection model is constructed based on the actual fatty acid value of the grain sample and the multi - dimensional spectrum corresponding to the grain sample.

[0067] In some embodiments, the device further includes: A second acquisition module, configured to pass light with a wavelength of 680 nm - 780 nm through the grain to be detected to obtain the spectral data of the grain to be detected; A second input module, configured to input the spectral data into a pre - constructed grain color determination model to obtain the color of the grain to be detected; wherein, the grain color determination model is constructed based on the spectral data of the grain samples of different colors at a light wavelength of 680 nm - 780 nm.

[0068] In some embodiments, the device further includes: A third acquisition module, configured to acquire the multi - dimensional spectra corresponding to the grain samples of different colors; A construction module, configured to construct the grain fatty acid detection model corresponding to each grain color according to the proportion of the grain samples of different colors, the multi - dimensional spectra corresponding to the grain samples of different colors, and the actual fatty acid values of the grain samples of different colors.

[0069] In some embodiments, the apparatus further comprises: a segmentation module, configured to segment the multi-dimensional spectrum corresponding to each color of the cereal samples into multiple segments when the proportion of the cereal samples of different colors is determined; a model establishment module, configured to establish a multiple regression model by respectively combining the actual fatty acid values of the cereal samples of different colors with all segments, any segment, and segment combinations of the multi-dimensional spectra corresponding to the cereal samples of different colors; a determination module, configured to use the multiple regression model with the optimal model performance as the cereal fatty acid detection model.

[0070] In some embodiments, the apparatus further comprises: a division module, configured to divide each color of the cereal samples into multiple sub-cereal samples; a fourth acquisition module, configured to respectively pass the multiple light beams through the multiple sub-cereal samples, and acquire the multi-dimensional spectra corresponding to each sub-cereal sample; a fifth acquisition module, configured to combine the multi-dimensional spectra respectively corresponding to the multiple sub-cereal samples to obtain the multi-dimensional spectrum corresponding to each color of the cereal samples.

[0071] In some embodiments, the apparatus further comprises: a spectroscope module, configured to split the light band from visible light to near-infrared light in the range of 680 nm - 1080 nm into multiple light beams.

[0072] In some embodiments, when the cereal is paddy rice, the paddy rice spectrum combination corresponding to the multiple regression model with the optimal model performance is: 725 - 840 nm, 832 - 946 nm, 960 - 1050 nm.

[0073] In some embodiments, when the cereal is corn, the corn spectrum combination corresponding to the multiple regression model with the optimal model performance is: 692 - 780 nm, 898 - 1050 nm.

[0074] It should be noted here that the above-mentioned apparatus for detecting the fatty acid value of cereals based on multi-dimensional spectra provided by the present invention can implement all the method steps of the above method embodiments and can achieve the same technical effects. Therefore, the same parts and beneficial effects as those in the method embodiments will not be specifically described in this embodiment.

[0075] Figure 7 is a schematic structural diagram of the electronic device provided by the present invention, as Figure 7As shown in the figure, the electronic device may include: a processor 710, a communications interface 720, a memory 730, and a communication bus 740. Among them, the processor 710, the communications interface 720, and the memory 730 complete communication with each other through the communication bus 740. The processor 710 may call the logical instructions in the memory 730 to execute the method for detecting the fatty acid value of grains based on multi-dimensional spectra. The method includes: passing multiple light beams through the grains to be detected to obtain the multi-dimensional spectra corresponding to the grains to be detected; inputting the multi-dimensional spectra corresponding to the grains to be detected into the constructed grain fatty acid detection model corresponding to the color of the grains to be detected to obtain the fatty acid value of the grains to be detected; the grain fatty acid detection model is constructed by the actual fatty acid value of the grain sample and the multi-dimensional spectra corresponding to the grain sample.

[0076] In addition, when the logical instructions in the above-mentioned memory 730 are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0077] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for detecting the fatty acid value of grains based on multi-dimensional spectra provided by the above-mentioned various methods. The method includes: passing multiple light beams through the grains to be detected to obtain the multi-dimensional spectra corresponding to the grains to be detected; inputting the multi-dimensional spectra corresponding to the grains to be detected into the constructed grain fatty acid detection model corresponding to the color of the grains to be detected to obtain the fatty acid value of the grains to be detected; the grain fatty acid detection model is constructed by the actual fatty acid value of the grain sample and the multi-dimensional spectra corresponding to the grain sample.

[0078] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method for detecting the fatty acid value of grains based on multi-dimensional spectra provided by the above-mentioned various methods. The method includes: passing a plurality of light beams through the grains to be detected to obtain the multi-dimensional spectra corresponding to the grains to be detected; inputting the multi-dimensional spectra corresponding to the grains to be detected into a constructed grain fatty acid detection model corresponding to the color of the grains to be detected to obtain the fatty acid value of the grains to be detected; the grain fatty acid detection model is constructed by the actual fatty acid value of grain samples and the multi-dimensional spectra corresponding to the grain samples.

[0079] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0080] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the fatty acid value of grains based on multi-dimensional spectroscopy, characterized in that, Including: Passing multiple light beams through the grain to be detected to obtain a multi-dimensional spectrum corresponding to the grain to be detected; Inputting the multi-dimensional spectrum corresponding to the grain to be detected into a constructed grain fatty acid detection model corresponding to the color of the grain to be detected to obtain the fatty acid value of the grain to be detected; the grain fatty acid detection model is constructed based on the actual fatty acid value of the grain sample and the multi-dimensional spectrum corresponding to the grain sample.

2. The method for detecting the fatty acid value of grains based on multi-dimensional spectroscopy according to claim 1, characterized in that The method further includes: Using a light wavelength of 680 nm - 780 nm to pass through the grain to be detected to obtain spectral data of the grain to be detected; Inputting the spectral data into a constructed grain color determination model to obtain the color of the grain to be detected; wherein, the grain color determination model is constructed based on the spectral data of the grain samples of different colors at a light wavelength of 680 nm - 780 nm.

3. The method for detecting the fatty acid value of grains based on multi-dimensional spectroscopy according to claim 1, wherein The method further includes: Obtaining multi-dimensional spectra corresponding to the grain samples of different colors; Constructing the grain fatty acid detection model corresponding to each grain color according to the proportion of the grain samples of different colors, the multi-dimensional spectra corresponding to the grain samples of different colors, and the actual fatty acid values of the grain samples of different colors.

4. The method for detecting the fatty acid value of grains based on multi-dimensional spectra according to claim 3, wherein The method further includes: When the proportion of the grain samples of different colors is determined, dividing the multi-dimensional spectra corresponding to the grain samples of each color into multiple segments; Establishing a multiple regression model by respectively combining the actual fatty acid values of the grain samples of different colors with all segments, any segment, and segment combinations of the multi-dimensional spectra corresponding to the grain samples of different colors; Taking the multiple regression model with the optimal model performance as the grain fatty acid detection model.

5. The method for detecting the fatty acid value of grains based on multi-dimensional spectroscopy according to claim 3, characterized in that, The obtaining of the multi-dimensional spectra corresponding to the grain samples of different colors includes: Dividing the grain samples of each color into multiple sub-grain samples; Passing the multiple light beams through the multiple sub-grain samples respectively to obtain the multi-dimensional spectra corresponding to each sub-grain sample; Combining the multi-dimensional spectra corresponding to the multiple sub-grain samples respectively to obtain the multi-dimensional spectra corresponding to the grain samples of each color.

6. The method for detecting the fatty acid value of grains based on multi-dimensional spectroscopy according to claim 1 or 5, wherein The method further includes: Splitting the light waveband from visible light to near-infrared light in the range of 680 nm - 1080 nm to generate the multiple light beams.

7. The method for detecting the fatty acid value of grains based on multi-dimensional spectra according to claim 4, characterized in that, When the grain is paddy rice, the paddy rice spectral combination corresponding to the multiple regression model with the optimal model performance is: 725 - 840 nm, 832 - 946 nm, 960 - 1050 nm.

8. The method for detecting the fatty acid value of grains based on multi-dimensional spectra according to claim 4, wherein When the grain is corn, the corn spectral combination corresponding to the multiple regression model with the optimal model performance is: 692 - 780 nm, 898 - 1050 nm.

9. A device for detecting the fatty acid value of grains based on multi-dimensional spectroscopy, characterized in that, Including: A first obtaining module for passing multiple light beams through the grain to be detected to obtain a multi-dimensional spectrum corresponding to the grain to be detected; The first input module is configured to input the multi-dimensional spectrum corresponding to the grain to be detected into the constructed grain fatty acid detection model corresponding to the color of the grain to be detected, so as to obtain the fatty acid value of the grain to be detected; the grain fatty acid detection model is constructed by using the actual fatty acid value of the grain sample and the multi-dimensional spectrum corresponding to the grain sample.

10. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and running on the processor, wherein, When the processor executes the computer program, it implements the method for detecting the fatty acid value of grains based on multi-dimensional spectrum according to any one of claims 1 to 8.

Images