A method and system for detecting the quality of flour based on optical analysis

Abstract

The present invention relates to the field of optical detection technology, and particularly to a method and system for detecting the quality of flour based on optical analysis. First, the present invention uses a Raman spectrometer including a plurality of intelligent sensors to obtain Raman spectral images corresponding to each local detection area of a flour sample; according to the morphological distribution characteristics of the fluctuation peaks to be analyzed in the Raman spectral images, and the coincidence degree between the fluctuation peaks to be analyzed and the preset frequency shift standard ranges corresponding to each flour component, the component fluctuation peaks and their corresponding flour components are obtained; according to the impurity possibility of the component fluctuation peaks, the impurity fluctuation peaks are screened out from the component fluctuation peaks; according to the overall level and difference of the impurity metrics corresponding to all local detection areas of the flour sample, the flour quality of the flour sample is obtained. By deeply analyzing the characteristics of the fluctuation peaks corresponding to the impurities in the flour, the present invention can more accurately detect the fluctuation peaks corresponding to the impurities and improve the detection effect of the flour quality.

Description

Technical Field

[0001] The present invention relates to the technical field of optical detection, and particularly relates to a method and system for detecting the quality of flour based on optical analysis. Background Art

[0002] As a basic raw material in the food industry, the quality of flour is directly related to the safety of the final product. In order to ensure the quality and safety of the product, it is necessary to detect the quality of flour. By detecting the quality of flour in a timely manner, problems in the process of producing flour can be discovered and solved in a timely manner, helping the factory to optimize the production process to produce products that better meet the market demand.

[0003] In the detection of flour quality, Raman spectroscopy imaging technology, as an advanced non-destructive detection technology, the Raman spectroscopy image of flour can reflect the presence and distribution state of various chemical components in the flour. The prior art uses the Raman spectroscopy image of flour to detect the quality of flour. However, in the process of using the Raman spectroscopy image of flour to detect the quality of flour, due to the influence that the molecular structures of flour and impurities are similar, the recognition effect of the fluctuation peaks corresponding to the impurities is not good, resulting in difficulty in ensuring the accuracy of flour quality detection. Summary of the Invention

[0004] In order to solve the technical problem that it is difficult to ensure the accuracy of flour quality detection in the prior art, the purpose of the present invention is to provide a method and system for detecting the quality of flour based on optical analysis, and the specific technical solutions adopted are as follows:

[0005] A method for detecting the quality of flour based on optical analysis, the method includes:

[0006] Obtain the Raman spectroscopy images corresponding to each local detection area of the flour sample;

[0007] In the Raman spectroscopy image, according to the fluctuation situation of each spectral line, obtain all the to-be-analyzed fluctuation peaks of each spectral line; according to the morphological distribution characteristics of the to-be-analyzed fluctuation peaks and the coincidence degree between the to-be-analyzed fluctuation peaks and the preset frequency shift standard ranges corresponding to each flour component, screen all the to-be-analyzed fluctuation peaks to obtain the component fluctuation peaks and their corresponding flour components;

[0008] Take any one of the component fluctuation peaks in the Raman spectroscopy image as the to-be-analyzed impurity peak; according to the coincidence difference situation between each of the component fluctuation peaks in the Raman spectroscopy image to which the to-be-analyzed impurity peak belongs and the corresponding preset frequency shift standard range, and the area difference situation between the to-be-analyzed impurity peak and all its corresponding component fluctuation peaks, obtain the impurity possibility of the to-be-analyzed impurity peak; according to the impurity possibility of the component fluctuation peaks, screen out the impurity fluctuation peaks from the component fluctuation peaks;

[0009] Obtain the impurity metric of the local detection area according to the proportion of the number of impurity fluctuation peaks in the Raman spectrum image; obtain the flour quality index of the flour sample according to the overall level and difference of the impurity metrics corresponding to all the local detection areas of the flour sample.

[0010] Further, the method for obtaining the to-be-analyzed fluctuation peaks includes:

[0011] Take any spectral line as the target spectral line, and obtain all the peak points and all the valley points of the target spectral line; take the target spectral line corresponding to the previous valley point to the next valley point of each peak point as each to-be-analyzed fluctuation peak of the target spectral line.

[0012] Further, the method for obtaining the component fluctuation peaks includes:

[0013] Obtain the first possible degree of the components of the to-be-analyzed fluctuation peaks according to the distribution characteristics of the peak points and the valley points in the to-be-analyzed fluctuation peaks;

[0014] Obtain the second possible degree of the components of the to-be-analyzed fluctuation peaks and the corresponding flour components according to the coincidence degree between the to-be-analyzed fluctuation peaks and the preset frequency shift standard ranges corresponding to each flour component;

[0015] Obtain the component confidence degree of the to-be-analyzed fluctuation peaks according to the first possible degree of the components and the second possible degree of the components;

[0016] Screen all the to-be-analyzed fluctuation peaks according to the component confidence degree of the to-be-analyzed fluctuation peaks and the corresponding flour components, and obtain the component fluctuation peaks and the corresponding flour components.

[0017] Further, the method for obtaining the first possible degree of the components includes:

[0018] Take the interval between the abscissa values of the two valley points corresponding to the to-be-analyzed fluctuation peak as the total width of the to-be-analyzed fluctuation peak;

[0019] Take the interval between the maximum ordinate value and the minimum ordinate value corresponding to the to-be-analyzed fluctuation peak as the total height of the to-be-analyzed fluctuation peak;

[0020] Take the interval between the abscissa value of the peak point and the abscissa value of the previous valley point as the first width of the to-be-analyzed fluctuation peak;

[0021] Take the interval between the abscissa value of the peak point and the abscissa value of the next valley point as the second width of the to-be-analyzed fluctuation peak;

[0022] Calculate the absolute value of the difference between the first width and the second width to obtain the asymmetry parameter;

[0023] Calculate the sum value of the asymmetry parameter and the preset denominator adjustment factor, calculate the product of the total width and the total height of the fluctuating peak to be analyzed, calculate the ratio of the product to the sum value, and obtain the first possible degree of the component of the fluctuating peak to be analyzed.

[0024] Further, the method for obtaining the second possible degree of the component of the fluctuating peak to be analyzed and its corresponding flour component includes:

[0025] Take the interval corresponding to the minimum abscissa value and the maximum abscissa value of the fluctuating peak to be analyzed as the displacement interval of the fluctuating peak to be analyzed;

[0026] Obtain the overlapping interval between the displacement interval of the fluctuating peak to be analyzed and the preset frequency shift standard range corresponding to each flour component; take the interval between the minimum abscissa value and the maximum abscissa value corresponding to the largest overlapping interval as the second possible degree of the component of the fluctuating peak to be analyzed; take the flour component corresponding to the largest overlapping interval as the flour component corresponding to the component fluctuating peak.

[0027] Further, the method for obtaining the impurity possibility includes:

[0028] In the Raman spectrum image where the impurity peak to be analyzed is located, obtain the overall parameter of the coincidence difference of the impurity peak to be analyzed according to the difference in the coincidence degree between each component fluctuating peak and the corresponding preset frequency shift standard range;

[0029] In all Raman spectrum images, take the component fluctuating peak with the same flour component as the impurity peak to be analyzed as the reference component peak of the impurity peak to be analyzed; obtain the overall parameter of the area difference of the impurity peak to be analyzed according to the area difference between the impurity peak to be analyzed and all its corresponding reference component peaks;

[0030] According to the overall parameter of the coincidence difference and the overall parameter of the area difference, obtain the impurity possibility of the impurity peak to be analyzed; both the overall parameter of the coincidence difference and the overall parameter of the area difference are positively correlated with the impurity possibility.

[0031] Further, the method for obtaining the overall parameter of the coincidence difference includes:

[0032] In the Raman spectrum image to which the impurity peak to be analyzed belongs, the range corresponding to the minimum abscissa value and the maximum abscissa value of the component fluctuation peak is used as the first range of the component fluctuation peak; the preset frequency shift standard range of the flour component corresponding to the component fluctuation peak is used as the second range of the component fluctuation peak; the overlapping range corresponding to the first range and the second range is used as the third range; calculate the ratio of the abscissa interval of the third range to the abscissa interval of the first range to obtain the standard coincidence degree of the component fluctuation peak; calculate the absolute value of the difference between the standard coincidence degrees of the impurity peak to be analyzed and each component fluctuation peak to obtain the local coincidence difference parameter of the component fluctuation peak; calculate the mean value of all local coincidence difference parameters to obtain the overall coincidence difference parameter.

[0033] Further, the method for obtaining the overall area difference parameter includes:

[0034] Calculate the absolute value of the difference between the area of the impurity peak to be analyzed and the area of each component fluctuation peak to obtain the local area difference parameter of each component fluctuation peak; calculate the mean value of all the local area difference parameters to obtain the overall area difference parameter.

[0035] Further, the method for obtaining the flour quality index of the flour sample includes:

[0036] Calculate the mean value of the impurity metrics corresponding to all the local detection regions of the flour sample to obtain the overall impurity parameter;

[0037] Calculate the absolute value of the difference between the impurity metrics corresponding to every two local detection regions to obtain the impurity difference parameter corresponding to every two local detection regions; calculate the mean value of all impurity difference parameters to obtain the overall impurity fluctuation parameter;

[0038] Calculate the product of the overall impurity parameter and the overall impurity fluctuation parameter, and perform inverse normalization on the product to obtain the flour quality index.

[0039] The present invention provides a flour quality detection system based on optical analysis, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the flour quality detection method based on optical analysis are implemented.

[0040] The present invention has the following beneficial effects:

[0041] First, a Raman spectrometer equipped with multiple intelligent sensors is used to obtain Raman spectral images corresponding to each local detection area of the flour sample, providing data support for subsequent analysis. Considering that the fluctuations in the spectral lines reflect different flour components in the flour sample, all the fluctuating peaks to be analyzed for each spectral line are obtained, providing data support for subsequent analysis. First, considering the morphological distribution characteristics of the fluctuating peaks to be analyzed corresponding to the flour components, that is, the fluctuating peaks to be analyzed corresponding to the flour components generally have larger amplitudes, wider wavelength ranges, and symmetry in shape. Also, considering that different flour components have specific frequency shift ranges in the Raman spectrum, by analyzing the overlap degree between the fluctuating peaks to be analyzed and the preset frequency shift standard ranges corresponding to each flour component, the fluctuating peaks generated by the flour components can be obtained and the corresponding flour components can be determined. In order to more accurately screen out the fluctuating peaks corresponding to impurities from the Raman spectral image, considering that the offset of the fluctuating peaks caused by impurities is more local and specific, impurities will cause the corresponding fluctuating peaks to shift and change the shape of the fluctuating peaks. First, any component fluctuating peak in the Raman spectral image is used as the impurity peak to be analyzed. By analyzing whether the impurity peak to be analyzed conforms to the characteristics of the fluctuating peaks caused by impurities, the impurity fluctuating peaks are screened out from the component fluctuating peaks, so that the impurity fluctuating peaks can more accurately reflect the fluctuating peaks corresponding to the impurities.

[0042] Considering that the number of impurity fluctuating peaks directly reflects the impurity content in this area, the impurity degree of the local area is reflected by constructing the impurity metric of the local detection area. The larger the value of the impurity metric, the higher the impurity degree of the local detection area; considering that the larger the impurity metric corresponding to all local detection areas of the flour sample, the higher the overall impurity content of the impurity flour sample; at the same time, considering that large flour particles often sink and aggregate under the action of gravity, making the impurity distribution uneven corresponding to different local detection areas of the flour sample, resulting in differences in the impurity metrics corresponding to the local detection areas, and finally obtaining the flour quality index of the flour sample. The higher the flour quality index, the less impurities there are in the flour and the better the flour quality. Description of the Drawings

[0043] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only 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.

[0044] Figure 1 It is a flowchart of a method for detecting the quality of flour based on optical analysis provided by an embodiment of the present invention;

[0045] Figure 2 It is a flowchart of a method for obtaining component fluctuating peaks provided by an embodiment of the present invention;

[0046] Figure 3 Flow chart of a method for obtaining impurity possibility provided by an embodiment of the present invention;

[0047] Figure 4 Structural diagram of a flour quality detection system based on optical analysis provided by an embodiment of the present invention. Specific embodiments

[0048] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following combines the accompanying drawings and preferred embodiments to detail the specific embodiments, structures, features and their effects of a flour quality detection method and system based on optical analysis proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0050] The following specifically describes the specific solutions of a flour quality detection method and system based on optical analysis provided by the present invention with reference to the accompanying drawings.

[0051] The embodiments of the present invention provide a flour quality detection method and system based on optical analysis. Please refer to Figure 1 , which shows a flow chart of a flour quality detection method based on optical analysis provided by an embodiment of the present invention. The method includes the following steps:

[0052] Step S1: Obtain Raman spectrum images corresponding to each local detection area of the flour sample.

[0053] In order to detect the quality of flour samples subsequently, it is first necessary to obtain Raman spectral images corresponding to each local detection area of the flour samples. The specific steps include: First, evenly place the flour samples in each local detection area on the sample stage of the Raman spectrometer. Before performing Raman spectral detection, reasonably set the parameters of the Raman spectrometer containing multiple intelligent sensors to obtain the best Raman spectral images. Perform Raman spectral detection on each local detection area, record the spectral data of each local detection area through the Raman spectrometer, and generate Raman spectral images corresponding to each local detection area of the flour sample. It should be noted that the method for obtaining Raman spectral images is a well-known technical means to those skilled in the art, and only a brief description is given here. It should be noted that the horizontal axis of the Raman spectral image represents the Raman shift, that is, the wavenumber difference between the scattered light and the incident light, and different flour components in the flour sample will produce different Raman shifts. The vertical axis of the Raman spectral image represents the intensity of Raman scattered light. The Raman scattering intensity reflects the scattering ability of a specific flour component in the flour sample to the incident light and is related to the flour component concentration.

[0054] It should be noted that, for the convenience of calculation, all the index data involved in the operations in the embodiments of the present invention have undergone data preprocessing, thereby eliminating the influence of dimensions. The specific means for eliminating dimension influence are well-known technical means to those skilled in the art and are not limited here.

[0055] In order to accurately detect the quality of flour, the present invention combines the characteristics of impurities, more accurately screens out the fluctuation peaks corresponding to impurities from the Raman spectral images to determine the impurity content in the local area, and then combines the distribution characteristics of the flour impurity content in different local detection areas of the flour samples with higher impurity content to more accurately detect the quality of the flour samples.

[0056] Step S2: In the Raman spectral image, according to the fluctuation situation of each spectral line, obtain all the to-be-analyzed fluctuation peaks of each spectral line; according to the morphological distribution characteristics of the to-be-analyzed fluctuation peaks and the coincidence degree between the to-be-analyzed fluctuation peaks and the preset frequency shift standard ranges corresponding to each flour component, screen all the to-be-analyzed fluctuation peaks to obtain the component fluctuation peaks and their corresponding flour components.

[0057] Considering that the fluctuation situation of the spectral line reflects different flour components in the flour sample, obtaining all the to-be-analyzed fluctuation peaks of each spectral line provides data support for subsequent analysis. First, considering the morphological distribution characteristics of the to-be-analyzed fluctuation peaks corresponding to the flour components, that is, the to-be-analyzed fluctuation peaks corresponding to the flour components often have a larger amplitude, a wider wavelength range and symmetry in shape, and also considering that different flour components have their specific frequency shift ranges in the Raman spectrum, by analyzing the coincidence degree between the to-be-analyzed fluctuation peaks and the preset frequency shift standard ranges corresponding to each flour component, the fluctuation peaks generated by the flour components can be obtained and their corresponding flour components can be determined.

[0058] In order to initially identify the suspected fluctuation peaks corresponding to the flour components, preferably, in an embodiment of the present invention, the method for obtaining the fluctuation peaks to be analyzed includes:

[0059] Taking any spectral line as the target spectral line, obtaining all the peak points and all the valley points of the target spectral line; taking the target spectral line corresponding to the previous valley point to the next valley point of each peak point as each fluctuation peak to be analyzed of the target spectral line.

[0060] Regarding the above steps, considering that the Raman spectrum image contains a large number of data points, these data points constitute the spectral lines in the Raman spectrum image, and the flour components contained in the flour sample will cause the spectral lines to produce fluctuation peaks, that is, the molecular structure contained in the flour will cause the spectral lines to produce fluctuation peaks. The fluctuation situation of the spectral lines reflects the vibrations of different flour components in the flour sample. In order to extract the fluctuation peaks that may represent the flour components from the spectral lines, it is necessary to analyze each spectral line in detail, identify all the peak points and valley points, and take the target spectral line corresponding to the previous valley point to the next valley point of each peak point as each fluctuation peak to be analyzed of the target spectral line. The fluctuation peaks to be analyzed initially reflect the suspected fluctuation peaks corresponding to the flour components.

[0061] In an embodiment of the present invention, taking any spectral line as the target spectral line, using the AMDP peak detection algorithm (Amplitude Modulation Depth Peak Detection Algorithm) to obtain all the peak points in the target spectral line, taking the data point corresponding to the minimum value between adjacent peak points as the valley point; for any peak point in the target spectral line, taking the valley point adjacent to the front side of the peak point as the previous valley point of the peak point; taking the valley point adjacent to the rear side of the peak point as the next valley point of the peak point; taking the target spectral line corresponding to the previous valley point to the next valley point as the fluctuation peak to be analyzed corresponding to the peak point; counting the fluctuation peaks to be analyzed corresponding to all the peak points in the target spectral line to obtain all the fluctuation peaks to be analyzed of the target spectral line. It should be noted that the AMDP peak detection algorithm is a well-known existing technology to those skilled in the art and will not be elaborated here. It should be noted that taking the minimum value before the first peak point in the target spectral line as the first valley point; taking the minimum value after the last peak point in the target spectral line as the last valley point.

[0062] In order to more accurately identify the fluctuation peaks of the flour components, please refer to Figure 2 , which shows a flowchart of a method for obtaining the fluctuation peaks of a component in an embodiment of the present invention. Preferably, in an embodiment of the present invention, the method for obtaining the fluctuation peaks of the component includes:

[0063] Step S201: Obtain the first possibility degree of the components of the fluctuation peak to be analyzed according to the distribution characteristics of the peak points and valley points in the fluctuation peak to be analyzed.

[0064] Considering that the fluctuation peaks corresponding to the flour components often have large amplitudes, wide wavelength ranges, and symmetry in shape, obtain the first possibility degree of the components of the fluctuation peak to be analyzed. The larger the value of the first possibility degree of the components, the greater the possibility that the fluctuation peak to be analyzed corresponds to the flour component.

[0065] Preferably, in an embodiment of the present invention, the method for obtaining the first possibility degree of the components includes:

[0066] Take the interval between the abscissa values of the two valley points corresponding to the fluctuation peak to be analyzed as the total width of the fluctuation peak to be analyzed;

[0067] Take the interval between the maximum ordinate value and the minimum ordinate value corresponding to the fluctuation peak to be analyzed as the total height of the fluctuation peak to be analyzed;

[0068] Take the interval between the abscissa value of the peak point and the abscissa value of the previous valley point as the first width of the fluctuation peak to be analyzed;

[0069] Take the interval between the abscissa value of the peak point and the abscissa value of the next valley point as the second width of the fluctuation peak to be analyzed;

[0070] Calculate the absolute value of the difference between the first width and the second width to obtain the asymmetry parameter;

[0071] Calculate the sum value of the asymmetry parameter and the preset denominator adjustment factor, calculate the product of the total width of the fluctuation peak to be analyzed and the total height of the fluctuation peak to be analyzed, and calculate the ratio of the product to the sum value to obtain the first possibility degree of the components of the fluctuation peak to be analyzed. Among them, in an embodiment of the present invention, the preset denominator adjustment factor is used to ensure that the denominator is not 0, and the preset denominator adjustment factor is a number greater than 0. In this embodiment, it is set to 0.001, and the implementer can set it according to the implementation scenario.

[0072] For the above steps, the total width of the fluctuation peak to be analyzed reflects the wavelength range occupied by the fluctuation peak in the Raman spectrum image, and the total height of the fluctuation peak to be analyzed represents the amplitude of the fluctuation peak. The first width of the fluctuation peak to be analyzed and the second width of the fluctuation peak to be analyzed are used to evaluate the symmetry of the fluctuation peak. By calculating the absolute value of the difference between the first width and the second width, the asymmetry parameter is obtained. The asymmetry parameter quantifies the degree of asymmetry of the fluctuation peak. The smaller the degree of asymmetry, the better the symmetry of the fluctuation peak; finally, the first possibility degree of the components of the fluctuation peak to be analyzed is obtained. The larger the value of the first possibility degree of the components, the stronger the correlation between the fluctuation peak and the flour component, and the greater the possibility that the fluctuation peak to be analyzed corresponds to the flour component.

[0073] Step S202: Obtain the second possibility degree of the component of the peak to be analyzed and its corresponding flour component according to the overlap degree between the peak to be analyzed and the preset frequency shift standard ranges corresponding to each flour component.

[0074] Considering that different flour components have their specific frequency shift ranges in the Raman spectrum, by analyzing the overlap degree between the peak to be analyzed and the preset frequency shift standard ranges corresponding to each flour component, the second possibility degree of the component of the peak to be analyzed is obtained. The larger the value of the second possibility degree of the component, the greater the possibility that the peak to be analyzed corresponds to the flour component.

[0075] Preferably, in one embodiment of the present invention, the method for obtaining the second possibility degree of the component of the peak to be analyzed and its corresponding flour component includes:

[0076] Take the interval corresponding to the minimum abscissa value and the maximum abscissa value corresponding to the peak to be analyzed as the displacement interval of the peak to be analyzed;

[0077] Obtain the overlapping intervals between the displacement interval of the peak to be analyzed and the preset frequency shift standard ranges corresponding to each flour component; take the interval between the minimum abscissa value and the maximum abscissa value corresponding to the largest overlapping interval as the second possibility degree of the component of the peak to be analyzed; take the flour component corresponding to the largest overlapping interval as the flour component corresponding to the peak of the component. Wherein, in one implementation of the present invention, the method for obtaining the preset frequency shift standard ranges corresponding to each flour component is as follows: Since different components of flour will generate Raman peaks in different frequency shift ranges, obtain the frequency shift ranges corresponding to each flour component in each flour sample from the flour detection system, and take the frequency shift range corresponding to the flour component as the preset frequency shift standard range corresponding to the flour component. For example, the preset frequency shift standard range of the C-H bond of the flour component is 2800 - 3100, and the preset frequency shift standard range of the C=O bond of the flour component is 1600 - 1800.

[0078] For the above steps, the displacement interval of the peak to be analyzed reflects the frequency shift range corresponding to the peak to be analyzed; the preset frequency shift standard range corresponding to the flour component reflects the unique frequency shift range of each flour component in the Raman spectrum. By comparison, find out the overlapping intervals between the displacement interval of the peak to be analyzed and the preset frequency shift standard ranges corresponding to each flour component. The larger the overlapping interval, the higher the matching degree, and the greater the possibility that the peak to be analyzed corresponds to the flour component. Take the interval between the minimum abscissa value and the maximum abscissa value corresponding to the largest overlapping interval as the second possibility degree of the component of the peak to be analyzed; the larger the value of the second possibility degree of the component, the greater the possibility that the peak to be analyzed corresponds to the flour component. Take the flour component corresponding to the largest overlapping interval as the flour component corresponding to the peak of the component. This means that according to the analysis result of the Raman spectrum, the peak of the component is most likely to come from this specific flour component.

[0079] Step S203: Obtain the component confidence of the fluctuating peak to be analyzed according to the first component possibility and the second component possibility.

[0080] By integrating the first component possibility and the second component possibility, and comprehensively considering the morphological distribution characteristics and position information of the fluctuating peak to be analyzed, the component confidence of the fluctuating peak to be analyzed is obtained. The component confidence can more accurately reflect the possibility of the flour component corresponding to the fluctuating peak to be analyzed. The greater the component confidence, the greater the possibility of the flour component corresponding to the fluctuating peak to be analyzed.

[0081] In one embodiment of the present invention, the product of the first component possibility and the second component possibility is calculated, and the product is normalized to obtain the component confidence of the fluctuating peak to be analyzed. In one embodiment of the present invention, the normalization method is as follows: normalization is performed using the norm normalization function, and the numerical range is restricted between 0 and 1.

[0082] Step S204: Screen all the fluctuating peaks to be analyzed according to the component confidence of the fluctuating peak to be analyzed and the corresponding flour component, and obtain the component fluctuating peak and its corresponding flour component.

[0083] Considering that the greater the component confidence, the greater the possibility of the flour component corresponding to the fluctuating peak to be analyzed, all the fluctuating peaks to be analyzed are screened according to the component confidence of the fluctuating peak to be analyzed, and the component fluctuating peaks are screened out and their corresponding flour components are obtained.

[0084] In one embodiment of the present invention, the fluctuating peak to be analyzed with a component confidence greater than the preset confidence threshold is used as the component fluctuating peak and the corresponding flour component of the component fluctuating peak is determined. In one embodiment of the present invention, the preset confidence threshold is 0.87, and the implementer can set it according to the implementation scenario.

[0085] Considering that there are similar molecular structures between the flour and the impurities, making it difficult to determine whether the flour component belongs to the impurities. For example, there are organic brightening agent impurities in the flour, and the molecular structure of the organic brightening agent impurities is similar to that of the natural pigments in the flour. In Raman spectroscopy analysis, the organic brightening agent impurities and natural pigments may produce similar or overlapping Raman peaks. Due to their similar molecular structures, it is difficult to distinguish whether the component fluctuating peak comes from the organic brightening agent impurities or the natural pigments in the flour only by the flour component corresponding to the component fluctuating peak. It is necessary to further determine the fluctuating peak corresponding to the impurities according to the characteristics of the fluctuating peak caused by the impurities.

[0086] Step S3: Take any component fluctuation peak in the Raman spectrum image as the impurity peak to be analyzed; according to the coincidence of each component fluctuation peak in the Raman spectrum image to which the impurity peak to be analyzed belongs with the corresponding preset frequency shift standard range, and the area difference between the impurity peak to be analyzed and all its corresponding component fluctuation peaks, obtain the impurity possibility of the impurity peak to be analyzed; screen out the impurity fluctuation peaks from the component fluctuation peaks according to the impurity possibility of the component fluctuation peaks.

[0087] Considering that the shift of the fluctuation peak caused by the impurity is more local and specific, and the impurity will cause the corresponding fluctuation peak to shift and change the shape of the fluctuation peak. First, take any component fluctuation peak in the Raman spectrum image as the impurity peak to be analyzed, and screen out the impurity fluctuation peaks from the component fluctuation peaks by analyzing that the impurity peak to be analyzed conforms to the characteristics of the fluctuation peak caused by the impurity, so that the impurity fluctuation peak can more accurately reflect the corresponding fluctuation peak of the impurity.

[0088] To analyze the possibility that the impurity peak to be analyzed is the fluctuation peak caused by the impurity, please refer to Figure 3 , which shows a flowchart of a method for obtaining impurity possibility in an embodiment of the present invention. Preferably, in an embodiment of the present invention, the method for obtaining impurity possibility includes:

[0089] Step S301: In the Raman spectrum image to which the impurity peak to be analyzed belongs, obtain the overall coincidence difference parameter of the impurity peak to be analyzed according to the difference in the coincidence degree of each component fluctuation peak with the corresponding preset frequency shift standard range.

[0090] To more accurately screen out the fluctuation peak corresponding to the impurity from the Raman spectrum image, considering that both the impurity and the baseline drift will cause the shift of the fluctuation peak, the baseline drift will cause all the fluctuation peaks in the Raman spectrum image to shift as a whole without changing the shape of a single fluctuation peak; while the shift caused by the impurity is more local and specific, that is, the impurity will cause the corresponding fluctuation peak to shift, and obtain the overall coincidence difference parameter of the impurity peak to be analyzed. The larger the overall coincidence difference parameter, the more different the shift of the impurity peak to be analyzed is from the shift of all the fluctuation peaks in the Raman spectrum image, and the more likely the impurity peak to be analyzed is the fluctuation peak caused by the impurity.

[0091] Preferably, in an embodiment of the present invention, the method for obtaining the overall coincidence difference parameter includes:

[0092] In the Raman spectrum image where the impurity peak to be analyzed is located, the range corresponding to the minimum abscissa value and the maximum abscissa value of the component fluctuation peak is taken as the first range of the component fluctuation peak; the preset frequency shift standard range of the flour component corresponding to the component fluctuation peak is taken as the second range of the component fluctuation peak; the overlapping range corresponding to the first range and the second range is taken as the third range; the ratio of the abscissa interval of the third range to the abscissa interval of the first range is calculated to obtain the standard coincidence degree of the component fluctuation peak; the absolute value of the difference between the standard coincidence degree of the impurity peak to be analyzed and each component fluctuation peak is calculated to obtain the local parameter of the coincidence difference of the component fluctuation peak; the mean value of all local parameters of the coincidence difference is calculated to obtain the overall parameter of the coincidence difference. It should be noted that the abscissa interval of the first range is the maximum abscissa value of the first range minus the minimum abscissa value, and the abscissa interval of the third range is the maximum abscissa value of the third range minus the minimum abscissa value.

[0093] Regarding the above steps, the larger the third range, the greater the degree of coincidence between the actual frequency shift range and the standard frequency shift range of the component fluctuation peak, and the smaller the component fluctuation peak compared to the standard shift range; by calculating the ratio of the abscissa interval of the third range to the abscissa interval of the first range and normalizing the abscissa interval of the third range using the abscissa interval of the first range, the standard coincidence degree of the component fluctuation peak is obtained. The larger the standard coincidence degree, the smaller the component fluctuation peak compared to the standard shift range. The local parameter of the coincidence difference reflects the degree of movement difference between the impurity peak to be analyzed and the component fluctuation peak, and the overall parameter of the coincidence difference comprehensively reflects the overall degree of movement difference between the impurity peak to be analyzed and all component fluctuation peaks. The larger the overall parameter of the coincidence difference, the more distinct the shift of the impurity peak to be analyzed is from the shifts of all the fluctuation peaks in the Raman spectrum image, and the more likely the impurity peak to be analyzed is the fluctuation peak caused by impurities.

[0094] Step S302: In all Raman spectrum images, the component fluctuation peaks with the same flour component as the impurity peak to be analyzed are taken as the reference component peaks of the impurity peak to be analyzed; according to the area difference situation between the impurity peak to be analyzed and all its corresponding reference component peaks, the overall parameter of the area difference of the impurity peak to be analyzed is obtained.

[0095] Considering that large flour particles tend to sink and aggregate under the action of gravity, resulting in uneven impurity distribution in different local detection areas of the flour sample, and thus the shape change degree of the fluctuation peaks generated by the same impurity in different local detection areas is different. By using the reference component peaks to reflect the fluctuation peaks of the same flour component as the impurity peak to be analyzed, and according to the area difference situation between the impurity peak to be analyzed and all its corresponding reference component peaks, the overall parameter of the area difference of the impurity peak to be analyzed is obtained. The larger the overall parameter of the area difference, the more distinct the area of the fluctuation peak generated by the flour component corresponding to the impurity peak to be analyzed is from the area of the fluctuation peak generated by the same flour component, and the more likely the impurity peak to be analyzed is the fluctuation peak caused by impurities.

[0096] Preferably, in one embodiment of the present invention, the method for obtaining the overall parameter of area difference includes:

[0097] Calculate the absolute value of the difference between the area of the impurity peak to be analyzed and the area of each component fluctuation peak to obtain the local area difference parameter of each component fluctuation peak; calculate the mean value of all local area difference parameters to obtain the overall area difference parameter. It should be noted that, in one embodiment of the present invention, the method for obtaining the area of the component fluctuation peak includes: calculate the interval between the maximum ordinate value and the minimum ordinate value of the component fluctuation peak to obtain the height of the component fluctuation peak; calculate the interval between the maximum abscissa value and the minimum abscissa value of the component fluctuation peak to obtain the width of the component fluctuation peak; calculate the product of the height of the component fluctuation peak and the width of the component fluctuation peak to obtain the area of the component fluctuation peak. In other embodiments of the present invention, the method for obtaining the area of the component fluctuation peak includes: connect the two endpoints of the component fluctuation peak to obtain a reference line, and use the corresponding area of the region enclosed by the reference line and the corresponding spectral line of the component fluctuation peak as the area of the component fluctuation peak.

[0098] For the above steps, the local area difference parameter reflects the degree of area difference between the impurity peak to be analyzed and the component fluctuation peak. Through comprehensive analysis, the overall area difference parameter is obtained. The larger the overall area difference parameter, it represents that the area of the fluctuation peak generated by the flour component corresponding to the impurity peak to be analyzed is different from the area of the fluctuation peak generated by the same flour component, and the impurity peak to be analyzed is more likely to be the fluctuation peak caused by impurities.

[0099] Step S303: According to the overall coincidence difference parameter and the overall area difference parameter, obtain the impurity possibility of the impurity peak to be analyzed; both the overall coincidence difference parameter and the overall area difference parameter are positively correlated with the impurity possibility.

[0100] Based on the overall situation of the coincidence difference and the area difference, obtain the impurity possibility of the impurity peak to be analyzed. The impurity possibility comprehensively reflects the possibility of the peak to be analyzed being an impurity peak. The larger the impurity possibility, it indicates that the impurity peak to be analyzed is more likely to be the fluctuation peak caused by impurities.

[0101] Preferably, in one embodiment of the present invention, calculate the product of the overall coincidence difference parameter and the overall area difference parameter, and normalize the product to obtain the impurity possibility of the impurity peak to be analyzed. In one embodiment of the present invention, the method of normalization is: use the norm normalization function for normalization, and limit the numerical range to between 0 and 1.

[0102] Considering that the larger the impurity possibility, it indicates that the impurity peak to be analyzed is more likely to be the fluctuation peak caused by impurities. Based on the impurity possibility of the component fluctuation peak, screen out the impurity fluctuation peaks from the component fluctuation peaks. Preferably, in one embodiment of the present invention, the method for obtaining the impurity fluctuation peak includes:

[0103] Mark the component fluctuation peaks with impurity possibility greater than the preset impurity measurement threshold as impurity fluctuation peaks. In an embodiment of the present invention, the preset impurity measurement threshold is 0.68, and the implementer can set it according to the implementation scenario.

[0104] Step S4: Obtain the impurity measurement of the local detection area according to the proportion of the number of impurity fluctuation peaks in the Raman spectrum image; obtain the flour quality index of the flour sample according to the overall level and difference of the impurity measurements corresponding to all local detection areas of the flour sample.

[0105] Through the above steps, the fluctuation peaks corresponding to impurities can be screened out more accurately. Considering that the number of impurity fluctuation peaks directly reflects the impurity content in this area, the impurity measurement of the local detection area is constructed to reflect the impurity degree of the local area. The larger the value of the impurity measurement, the higher the impurity degree of the local detection area; considering that the larger the impurity measurement corresponding to all local detection areas of the flour sample, the higher the overall impurity content of the impurity flour sample; at the same time, considering that large-particle impurities in the flour tend to sink and aggregate under the action of gravity, resulting in uneven impurity distribution corresponding to different local detection areas of the flour sample, that is, there are differences in the impurity measurements corresponding to the local detection areas, and finally the flour quality index of the flour sample is obtained. The higher the flour quality index, the less impurities there are in the flour and the better the flour quality.

[0106] In order to analyze the impurity degree of the local detection area, considering that the local area with a higher impurity degree has a higher number of corresponding impurity fluctuation peaks, the impurity measurement of the local detection area is constructed to reflect the impurity degree of the local area. Preferably, in an embodiment of the present invention, the method for obtaining the impurity measurement includes:

[0107] In the Raman spectrum image corresponding to the local detection area, calculate the ratio of the total number of impurity fluctuation peaks to the total number of all fluctuation peaks to be analyzed to obtain the impurity measurement of the local detection area.

[0108] For the above steps, considering that the number of impurity fluctuation peaks directly reflects the impurity content in this area, by calculating the ratio of the total number of impurity fluctuation peaks to the total number of all fluctuation peaks to be analyzed, the impurity measurement of the local detection area is constructed to reflect the impurity degree of the local area. The larger the value of the impurity measurement, the higher the impurity degree of the local detection area.

[0109] In order to analyze the quality of the flour, preferably, in an embodiment of the present invention, the method for obtaining the flour quality index of the flour sample includes:

[0110] Calculate the mean value of the impurity measurements corresponding to all local detection areas of the flour sample to obtain the overall impurity parameter;

[0111] Calculate the absolute value of the difference between the impurity metrics corresponding to every two local detection regions to obtain the impurity difference parameters corresponding to every two local detection regions; calculate the mean value of all the impurity difference parameters to obtain the overall impurity fluctuation parameter;

[0112] Calculate the product of the overall impurity parameter and the overall impurity fluctuation parameter, and perform inverse normalization on the product to obtain the flour quality index. Among them, the overall level is reflected by the mean value, and the difference situation is reflected by the absolute value of the difference. In one embodiment of the present invention, the method of inverse normalization is: use the opposite number of the product as the power of the exponential function with the natural constant as the base to obtain the negatively correlated mapping result.

[0113] For the above steps, the overall impurity parameter represents the overall content level of impurities in the flour sample. The larger the overall impurity parameter, the higher the average content of impurities in the flour sample; in order to evaluate the uneven distribution of impurities in the flour sample, calculate the absolute value of the difference between the impurity metrics corresponding to every two local detection regions. The absolute value of the difference reflects the difference in impurity content between different regions. Take the mean value of all the absolute values of the differences to obtain the overall impurity fluctuation parameter. The larger the overall impurity fluctuation parameter, the more uneven the distribution of impurities in the flour sample, that is, there are cases where the impurity content in some regions is different from that in other regions, indicating a higher possibility of impurities in the flour sample. By calculating the product of the overall impurity parameter and the overall impurity fluctuation parameter, the product takes into account both the overall content and the uneven distribution of impurities, reflecting the corresponding impurity content of the flour. The larger the product, the higher the impurity content of the flour. Perform inverse normalization on the product to obtain the flour quality index. The flour quality index can comprehensively and accurately reflect the quality of the flour sample. The higher the flour quality index, the less impurities there are in the flour and the better the flour quality.

[0114] Specifically, mark the flour samples with a flour quality index greater than the preset quality parameter as high-quality flour, and mark the flour samples with a flour quality index not greater than the preset first quality parameter as unqualified flour. The product engineer analyzes the reasons for product unqualifiedness and optimizes the production process according to the reasons. In one embodiment of the present invention, the preset quality parameter is set to 0.73, and the implementer can set it according to the implementation scenario.

[0115] The present invention also proposes a flour quality detection system based on optical analysis. Please refer to Figure 4 , which shows the structural diagram of a flour quality detection system based on optical analysis provided by one embodiment of the present invention. The system includes: a data acquisition module 101, a component fluctuation peak analysis module 102, an impurity fluctuation peak analysis module 103, and a quality analysis module 104.

[0116] The data acquisition module 101 is used to acquire the Raman spectroscopy images corresponding to each local detection region of the flour sample.

[0117] The component fluctuation peak analysis module 102 is used to obtain all the to-be-analyzed fluctuation peaks of each spectral line in the Raman spectrum image according to the fluctuation conditions of each spectral line; and screen all the to-be-analyzed fluctuation peaks according to the morphological distribution characteristics of the to-be-analyzed fluctuation peaks and the coincidence degree between the to-be-analyzed fluctuation peaks and the preset frequency shift standard ranges corresponding to each flour component, so as to obtain the component fluctuation peaks and their corresponding flour components.

[0118] The impurity fluctuation peak analysis module 103 is used to take any component fluctuation peak in the Raman spectrum image as the to-be-analyzed impurity peak; obtain the impurity possibility of the to-be-analyzed impurity peak according to the coincidence difference between each component fluctuation peak in the Raman spectrum image to which the to-be-analyzed impurity peak belongs and the corresponding preset frequency shift standard range, and the area difference between the to-be-analyzed impurity peak and all its corresponding component fluctuation peaks; and screen out the impurity fluctuation peaks from the component fluctuation peaks according to the impurity possibility of the component fluctuation peaks.

[0119] The quality analysis module 104 is used to obtain the impurity metric of the local detection area according to the proportion of the number of impurity fluctuation peaks in the Raman spectrum image; and obtain the flour quality index of the flour sample according to the overall level and difference of the impurity metrics corresponding to all the local detection areas of the flour sample.

[0120] It should be noted that: for the system provided in the above embodiment, only the above division of each functional module is used for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the above embodiment of a flour quality detection system based on optical analysis and the embodiment of a flour quality detection method based on optical analysis belong to the same concept, and the specific implementation process can be seen in the method embodiment, which will not be elaborated here.

[0121] In summary, the embodiment of the present invention provides a flour quality detection method and system based on optical analysis. First, all the to-be-analyzed fluctuation peaks are screened according to the morphological distribution characteristics of the to-be-analyzed fluctuation peaks and the coincidence degree between the to-be-analyzed fluctuation peaks and the preset frequency shift standard ranges corresponding to each flour component, so as to obtain the component fluctuation peaks and their corresponding flour components; the impurity fluctuation peaks are screened out from the component fluctuation peaks according to the impurity possibility of the component fluctuation peaks; and the flour quality of the flour sample is obtained according to the overall level and difference of the impurity metrics corresponding to all the local detection areas of the flour sample. In the embodiment of the present invention, by deeply analyzing the influence of flour impurities on the fluctuation peaks, the flour quality can be analyzed more accurately.

[0122] It should be noted that the above order of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0123] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for detecting the quality of flour based on optical analysis, characterized in that, The method includes: Obtaining Raman spectral images corresponding to respective local detection regions of a flour sample; In the Raman spectral image, according to the fluctuation condition of each spectral line, obtaining all to-be-analyzed fluctuation peaks of each spectral line; according to the morphological distribution characteristics of the to-be-analyzed fluctuation peaks and the coincidence degree between the to-be-analyzed fluctuation peaks and the preset frequency shift standard ranges corresponding to each flour component, screening all the to-be-analyzed fluctuation peaks to obtain component fluctuation peaks and their corresponding flour components; Taking any one component fluctuation peak in the Raman spectral image as a to-be-analyzed impurity peak; according to the coincidence difference condition between each component fluctuation peak in the Raman spectral image to which the to-be-analyzed impurity peak belongs and the corresponding preset frequency shift standard range, and the area difference condition between the to-be-analyzed impurity peak and all its corresponding component fluctuation peaks, obtaining the impurity possibility of the to-be-analyzed impurity peak; screening out impurity fluctuation peaks from the component fluctuation peaks according to the impurity possibility of the component fluctuation peaks; According to the proportion of the number of the impurity fluctuation peaks in the Raman spectral image, obtaining the impurity measure of the local detection region; according to the overall level condition and the difference condition of the impurity measures corresponding to all the local detection regions of the flour sample, obtaining the flour quality index of the flour sample; The method for obtaining the impurity possibility includes: In the Raman spectral image to which the to-be-analyzed impurity peak belongs, according to the difference condition of the coincidence degree between each component fluctuation peak and the corresponding preset frequency shift standard range, obtaining the overall coincidence difference parameter of the to-be-analyzed impurity peak; In all Raman spectral images, taking the component fluctuation peaks with the same flour component as the to-be-analyzed impurity peak as the reference component peaks of the to-be-analyzed impurity peak; according to the area difference condition between the to-be-analyzed impurity peak and all its corresponding reference component peaks, obtaining the overall area difference parameter of the to-be-analyzed impurity peak; According to the overall coincidence difference parameter and the overall area difference parameter, obtaining the impurity possibility of the to-be-analyzed impurity peak; both the overall coincidence difference parameter and the overall area difference parameter are positively correlated with the impurity possibility; The method for obtaining the impurity measure includes: In the Raman spectral image corresponding to the local detection region, calculating the ratio of the total number of the impurity fluctuation peaks to the total number of all the to-be-analyzed fluctuation peaks to obtain the impurity measure of the local detection region.

2. The method for detecting the quality of flour based on optical analysis according to claim 1, wherein The method for obtaining the to-be-analyzed fluctuation peaks includes: Taking any one spectral line as the target spectral line, obtaining all the peak points and all the valley points of the target spectral line; taking the target spectral line corresponding to the previous valley point to the next valley point of each peak point as each to-be-analyzed fluctuation peak of the target spectral line.

3. The method for detecting the quality of flour based on optical analysis according to claim 2, characterized in that, The method for obtaining the component fluctuation peaks includes: According to the distribution characteristics of the peak points and the valley points in the to-be-analyzed fluctuation peaks, obtaining the first component possibility of the to-be-analyzed fluctuation peaks; According to the coincidence degree between the to-be-analyzed fluctuation peaks and the preset frequency shift standard ranges corresponding to each flour component, obtaining the second component possibility of the to-be-analyzed fluctuation peaks and their corresponding flour components; According to the first component possibility and the second component possibility, obtaining the component confidence degree of the to-be-analyzed fluctuation peaks; Screen all the to-be-analyzed fluctuation peaks according to the component confidence of the to-be-analyzed fluctuation peaks and their corresponding flour components, and obtain the component fluctuation peaks and their corresponding flour components.

4. The method for detecting the quality of flour based on optical analysis according to claim 3, wherein The method for obtaining the first possibility degree of the component includes: Take the interval between the abscissa values of the two valley points corresponding to the to-be-analyzed fluctuation peak as the total width of the to-be-analyzed fluctuation peak; Take the interval between the maximum ordinate value and the minimum ordinate value corresponding to the to-be-analyzed fluctuation peak as the total height of the to-be-analyzed fluctuation peak; Take the interval between the abscissa value of the peak point and the abscissa value of the previous valley point as the first width of the to-be-analyzed fluctuation peak; Take the interval between the abscissa value of the peak point and the abscissa value of the next valley point as the second width of the to-be-analyzed fluctuation peak; Calculate the absolute value of the difference between the first width and the second width to obtain the asymmetry parameter; Calculate the sum value of the asymmetry parameter and the preset denominator adjustment factor, calculate the product of the total width of the to-be-analyzed fluctuation peak and the total height of the to-be-analyzed fluctuation peak, and calculate the ratio of the product to the sum value to obtain the first possibility degree of the component of the to-be-analyzed fluctuation peak.

5. The method for detecting the quality of flour based on optical analysis according to claim 3, wherein, The method for obtaining the second possibility degree of the component of the to-be-analyzed fluctuation peak and its corresponding flour component includes: Take the interval corresponding to the minimum abscissa value and the maximum abscissa value of the to-be-analyzed fluctuation peak as the displacement interval of the to-be-analyzed fluctuation peak; Obtain the overlapping interval corresponding to the displacement interval of the to-be-analyzed fluctuation peak and the preset frequency shift standard range corresponding to each flour component; take the interval between the minimum abscissa value and the maximum abscissa value corresponding to the largest overlapping interval as the second possibility degree of the component of the to-be-analyzed fluctuation peak; take the flour component corresponding to the largest overlapping interval as the flour component corresponding to the component fluctuation peak.

6. The method for detecting the quality of flour based on optical analysis according to claim 1, wherein, The method for obtaining the overall parameter of coincidence difference includes: In the Raman spectrum image where the to-be-analyzed impurity peak is located, take the range corresponding to the minimum abscissa value and the maximum abscissa value of the component fluctuation peak as the first range of the component fluctuation peak; take the preset frequency shift standard range of the flour component corresponding to the component fluctuation peak as the second range of the component fluctuation peak; take the overlapping range corresponding to the first range and the second range as the third range; calculate the ratio of the abscissa interval of the third range to the abscissa interval of the first range to obtain the standard coincidence degree of the component fluctuation peak; calculate the absolute value of the difference between the standard coincidence degree of the to-be-analyzed impurity peak and each component fluctuation peak to obtain the local parameter of coincidence difference of the component fluctuation peak; calculate the mean value of all the local parameters of coincidence difference to obtain the overall parameter of coincidence difference.

7. The method for detecting the quality of flour based on optical analysis according to claim 1, wherein, The method for obtaining the overall parameter of area difference includes: Calculate the absolute value of the difference between the area of the to-be-analyzed impurity peak and the area of each component fluctuation peak to obtain the local parameter of area difference of each component fluctuation peak; calculate the mean value of all the local parameters of area difference to obtain the overall parameter of area difference.

8. The method for detecting the quality of flour based on optical analysis according to claim 1, characterized in that, The method for obtaining the flour quality index of the flour sample includes: Calculate the mean value of the impurity metrics corresponding to all the local detection regions of the flour sample to obtain the overall impurity parameter; Calculate the absolute value of the difference between the impurity metrics corresponding to every two local detection regions to obtain the impurity difference parameter corresponding to every two local detection regions; calculate the mean value of all the impurity difference parameters to obtain the overall impurity fluctuation parameter; Calculate the product of the overall impurity parameter and the overall impurity fluctuation parameter, and perform inverse normalization on the product to obtain the flour quality index.

9. A flour quality detection system based on optical analysis, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for detecting the quality of flour based on optical analysis according to any one of claims 1 to 7.

Images