CN117808900B - Method and device for classifying color development intensity of maize anthocyanin - Google Patents

Abstract

The invention discloses a method and device for grading the color intensity of anthocyanins in corn silks. The method divides a sample set consisting of corn silk images into an analysis sample and a verification sample, establishes a relationship model between the color intensity of anthocyanins in corn silks and the CIELAB value by visually grading the color intensity of anthocyanins in the analysis sample and the measured CIELAB value, and then verifies and optimizes the relationship model by using the verification sample, so that the relationship model has high detection accuracy, and implements the grading of the color intensity of anthocyanins in corn silks to be tested by using a comparison relationship table of each anthocyanin color intensity corresponding to the relationship model and the CIELAB value. Thus, while ensuring the detection accuracy of the color intensity of anthocyanins in corn silks, the calculation complexity is reduced, the operation steps are reduced, and detection convenience is provided, thereby providing a method for grading the color intensity of anthocyanins in corn silks with a more effective boundary in actual application scenarios.

Description

A method and device for grading the color intensity of corn silk anthocyanin

Technical Field

The invention relates to the technical field of corn variety identification, and in particular to a method, device, equipment and storage medium for grading the color intensity of corn filament anthocyanin.

Background technique

The color intensity of anthocyanins refers to the color intensity or depth presented under specific test conditions. In practical applications, the detection of the color intensity of anthocyanins is mainly used to evaluate and compare the content of anthocyanins in different samples. Anthocyanins are a natural pigment that is widely present in plants and has multiple biological activities such as antioxidant and anti-inflammatory. Therefore, the detection of the color intensity of anthocyanins has certain practical significance for the identification and screening of plant varieties. The anthocyanin content in corn silk is one of the important indicators for measuring corn quality. By detecting the color intensity of anthocyanins, the quality of corn can be quickly and accurately evaluated, providing a reference for growers, processing companies and consumers.

The existing color intensity detection of anthocyanins in corn silk is mainly carried out by spectrophotometry and high performance liquid chromatography; among them, the spectrophotometry method is to scan the spectrum of the corn silk sample extract in the visible light region to find the maximum absorption wavelength of anthocyanins, and measure the absorbance at this wavelength. According to the absorbance and the standard curve, the content of anthocyanins can be calculated; the high performance liquid chromatography method is to extract anthocyanins from corn silk samples, separate and detect them using a high performance liquid chromatograph, and compare the retention time and peak area of different anthocyanins to qualitatively and quantitatively analyze the content of anthocyanins. It can be seen that the existing color intensity detection scheme of anthocyanins in corn silk has the defects of complex calculation and cumbersome operation.

Therefore, how to reduce the calculation complexity, reduce the operation steps and provide detection convenience while ensuring the detection accuracy of the color intensity of corn silk anthocyanins is a technical problem that needs to be solved urgently.

Summary of the invention

The main purpose of the present invention is to provide a method, device, equipment and storage medium for grading the color intensity of corn filament anthocyanins, aiming to solve the problems that the existing color intensity detection scheme of corn filament anthocyanins has defects such as complex calculation and cumbersome operation.

To achieve the above object, the present invention provides a method for grading the color intensity of anthocyanins in corn silk, comprising the following steps:

Acquire a corn silk sample set, and divide the corn silk sample set into an analysis sample and a verification sample; wherein the corn silk sample set includes a plurality of corn silk images;

placing the analysis sample and the verification sample into an image acquisition chamber, acquiring a color measurement image in the image acquisition chamber using an image acquisition device, and determining CIELAB values of the analysis sample and the verification sample based on the color measurement image;

Using the CIELAB values of the analysis samples and visual grading of anthocyanin color intensity, a relationship model between anthocyanin color intensity and CIELAB value of corn silk is established;

Verifying the relationship model using the CIELAB values of the verification samples and visual grading of anthocyanin color intensity, and optimizing the relationship model according to the verification results;

Based on the optimized relationship model, a comparison relationship table between corn silk at each anthocyanin color intensity and CIELAB value is generated, and the anthocyanin color intensity of the corn silk to be tested is graded using the comparison relationship table.

Optionally, obtaining a corn silk sample set and dividing the corn silk sample set into analysis samples and verification samples comprises:

Acquire a corn silk sample set; wherein the corn silk sample set includes a plurality of corn silk images whose anthocyanin color intensity is visually graded and evenly distributed in 9 anthocyanin color intensity levels;

Dividing the corn silk images of each anthocyanin color intensity level in the corn silk sample set into analysis sample images and verification sample images according to a preset ratio;

The analysis sample images of each anthocyanin color intensity level are used as analysis samples of the corn silk sample set, and the verification sample images of each anthocyanin color intensity level are used as verification samples of the corn silk sample set.

Optionally, the step of placing the analysis sample and the verification sample into an image acquisition room, using an image acquisition device to acquire a color measurement image in the image acquisition room, and determining the CIELAB values of the analysis sample and the verification sample based on the color measurement image specifically includes:

The analysis sample and the verification sample are placed in an image acquisition room together with a standard color card, and an image acquisition device is used to acquire a first color measurement image and a second color measurement image including the analysis sample and the standard color card or the verification sample and the standard color card at two opposite positions in the image acquisition room;

Establishing a color value conversion matrix according to the first RGB value of each color block of the standard color card in the first color measurement image and the second color measurement image and the CIEXYZ value of the color block itself;

Determine the CIEXYZ value of the analysis sample or the verification sample according to the second RGB value of the analysis sample or the verification sample in the first color measurement image and the second color measurement image and the color value conversion matrix, and convert the CIEXYZ value of the analysis sample or the verification sample into a CIELAB value.

Optionally, the step of establishing a relationship model between the anthocyanin color intensity and the CIELAB value of corn silk by visually grading the CIELAB value of the analyzed sample specifically comprises:

Obtaining visual grading of anthocyanin color intensity and corresponding CIELAB values of several corn silk images in the analysis sample;

The least squares univariate regression analysis was performed using the visual classification of anthocyanin color intensity and the corresponding color value L in the CIELAB value to establish a relationship model between the anthocyanin color intensity of corn silk and the CIELAB value, and generate a univariate regression line between the anthocyanin color intensity and the CIELAB value.

Optionally, the relationship model is verified using the CIELAB value of the verification sample and the visual grading of anthocyanin color intensity, and the relationship model is optimized according to the verification result, specifically comprising:

Inputting CIELAB values of several corn silk images in the verification sample into the relationship model, calculating the numerical difference between the anthocyanin color intensity corresponding to the corn silk image output by the relationship model and the visual classification of the corn silk image and the anthocyanin color intensity;

Determine whether there are corn silk images in the verification sample whose value difference exceeds the preset error allowable value. If so, divide all color values L of the corn silk images in the analysis sample into intervals, and use the color values L of the corn silk images in the two intervals after the division and the corresponding anthocyanin color intensity visual grading to perform least squares univariate regression analysis in the intervals respectively, and obtain a univariate regression line for each interval;

The above verification process is repeated until there is no corn silk image in the verification sample whose numerical difference exceeds the preset error allowable value, and a relationship model between the anthocyanin color intensity of corn silk and the CIELAB value is generated based on the univariate regression line in each interval.

Optionally, the division position is each color value L of the corn silk image in the analysis sample; and the step of performing an interval division on all the color values L of the corn silk image in the analysis sample specifically includes:

Through repeated verification, if the interval after the current color value L is used as the division position for interval division cannot meet the requirements of the numerical difference and the preset error allowable value, the next color value L is used as the division position for interval division;

If the interval corresponding to each division position cannot meet the requirements of the numerical difference and the preset error tolerance value, then the interval division with the least number of occurrences of the numerical difference exceeding the preset error tolerance value is selected from all the results of the first interval division, and each interval obtained from the result of the interval division is divided again, and the interval verification process is repeated.

Optionally, based on the optimized relationship model, a comparison relationship table between corn silk at each anthocyanin color intensity and CIELAB value is generated, and the step of grading the anthocyanin color intensity of the corn silk to be tested by using the comparison relationship table specifically includes:

Obtain a univariate regression line for each interval corresponding to the optimized relationship model, input the middle value of the color intensity of each two adjacent anthocyanins into the univariate regression line, and output the color value L corresponding to the middle value;

The color value L of two adjacent intermediate values corresponding to each anthocyanin color intensity is used as the color value L range corresponding to the anthocyanin color intensity, and a comparison relationship table between corn silk at each anthocyanin color intensity and CIELAB value is generated;

The control relationship table is used to grade the anthocyanin color intensity of the corn silk to be tested.

In addition, in order to achieve the above-mentioned purpose, the present invention also provides a corn silk anthocyanin color intensity grading device, comprising:

A division module is used to obtain a corn silk sample set and divide the corn silk sample set into an analysis sample and a verification sample; wherein the corn silk sample set includes a plurality of corn silk images;

A determination module, configured to place the analysis sample and the verification sample into an image acquisition chamber, acquire a color measurement image in the image acquisition chamber using an image acquisition device, and determine CIELAB values of the analysis sample and the verification sample based on the color measurement image;

Establishing a module for visually grading the anthocyanin color intensity of the analysis sample and establishing a relationship model between the anthocyanin color intensity of corn silk and the CIELAB value;

An optimization module, used to verify the relationship model using the CIELAB value of the verification sample and the visual grading of anthocyanin color intensity, and optimize the relationship model according to the verification result;

The grading module is used to generate a comparison relationship table between corn silk at each anthocyanin color intensity and CIELAB value based on the optimized relationship model, and use the comparison relationship table to grade the anthocyanin color intensity of the corn silk to be tested.

In addition, in order to achieve the above-mentioned purpose, the present invention also provides a corn filament anthocyanin color intensity grading device, which includes: a memory, a processor, and a corn filament anthocyanin color intensity grading program stored in the memory and executable on the processor, and when the corn filament anthocyanin color intensity grading program is executed by the processor, the steps of the corn filament anthocyanin color intensity grading method as described above are implemented.

In addition, in order to achieve the above-mentioned purpose, the present invention also provides a storage medium, on which is stored a corn filament anthocyanin color intensity grading program, and when the corn filament anthocyanin color intensity grading program is executed by a processor, the steps of the above-mentioned corn filament anthocyanin color intensity grading method are implemented.

The beneficial effects of the present invention are as follows: a method, device, equipment and storage medium for grading the color intensity of anthocyanins in corn silks are proposed. The present invention divides a sample set consisting of corn silk images into analysis samples and verification samples, establishes a relationship model between the color intensity of anthocyanins in corn silks and the CIELAB value by visually grading the color intensity of anthocyanins in the analysis samples and the measured CIELAB value, and then verifies and optimizes the relationship model by using the verification sample, so that the relationship model has high detection accuracy, and uses a comparison relationship table of each anthocyanin color intensity corresponding to the relationship model and the CIELAB value to achieve the grading of the color intensity of anthocyanins in the tested corn silks. Therefore, while ensuring the detection accuracy of the color intensity of anthocyanins in corn silks, the calculation complexity is reduced, the operation steps are reduced, and detection convenience is provided, thereby providing a more effective boundary grading method for the color intensity of anthocyanins in corn silks in actual application scenarios.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a device structure of a hardware operating environment involved in an embodiment of the present invention;

FIG2 is a schematic flow diagram of an embodiment of a method for grading the color intensity of anthocyanins in corn filaments according to the present invention;

FIG. 3 is a structural block diagram of a device for grading the color intensity of anthocyanins in corn silk according to an embodiment of the present invention.

The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings.

Detailed ways

In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

As shown in FIG. 1 , FIG. 1 is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.

As shown in Figure 1, the device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

Those skilled in the art will appreciate that the structure of the device shown in FIG. 1 does not limit the device, and may include more or fewer components than shown, or a combination of certain components, or a different arrangement of components.

As shown in FIG. 1 , the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a corn silk anthocyanin color intensity grading program.

In the terminal shown in FIG1 , the network interface 1004 is mainly used to connect to the backend server and perform data communication with the backend server; the user interface 1003 is mainly used to connect to the client (user end) and perform data communication with the client; and the processor 1001 can be used to call the corn silk anthocyanin color intensity grading program stored in the memory 1005 and perform the following operations:

Acquire a corn silk sample set, and divide the corn silk sample set into an analysis sample and a verification sample; wherein the corn silk sample set includes a plurality of corn silk images;

placing the analysis sample and the verification sample into an image acquisition chamber, acquiring a color measurement image in the image acquisition chamber using an image acquisition device, and determining CIELAB values of the analysis sample and the verification sample based on the color measurement image;

Using the CIELAB values of the analysis samples and visual grading of anthocyanin color intensity, a relationship model between anthocyanin color intensity and CIELAB value of corn silk is established;

Verifying the relationship model using the CIELAB values of the verification samples and visual grading of anthocyanin color intensity, and optimizing the relationship model according to the verification results;

Based on the optimized relationship model, a comparison relationship table between corn silk at each anthocyanin color intensity and CIELAB value is generated, and the anthocyanin color intensity of the corn silk to be tested is graded using the comparison relationship table.

The specific embodiments of the present invention applied to the device are basically the same as the embodiments of the following method for grading the color intensity of corn silk anthocyanins, and will not be described in detail here.

The embodiment of the present invention provides a method for grading the color intensity of anthocyanins in corn filaments. Referring to FIG. 2 , FIG. 2 is a schematic flow chart of an embodiment of the method for grading the color intensity of anthocyanins in corn filaments of the present invention.

In this embodiment, the method for grading the color intensity of corn filament anthocyanins comprises the following steps:

S100: Obtain a corn silk sample set, and divide the corn silk sample set into an analysis sample and a verification sample; wherein the corn silk sample set includes a plurality of corn silk images;

S200: placing the analysis sample and the verification sample into an image acquisition room, using an image acquisition device to acquire a color measurement image in the image acquisition room, and determining CIELAB values of the analysis sample and the verification sample based on the color measurement image;

S300: using the CIELAB value and anthocyanin color intensity of the analysis sample for visual grading, to establish a relationship model between anthocyanin color intensity and CIELAB value of corn silk;

S400: verifying the relationship model by using the CIELAB value of the verification sample and visual grading of anthocyanin color intensity, and optimizing the relationship model according to the verification result;

S500: Based on the optimized relationship model, a comparison relationship table between corn silk at each anthocyanin color intensity and CIELAB value is generated, and the anthocyanin color intensity of the corn silk to be tested is graded using the comparison relationship table.

It should be noted that the existing corn filament anthocyanin color intensity detection scheme has the problems of complex calculation and cumbersome operation. In order to solve the above problems, this embodiment divides the sample set composed of corn filament images into analysis samples and verification samples, and uses the visual classification of the anthocyanin color intensity of the analysis sample and the measured CIELAB value to establish a relationship model between the anthocyanin color intensity of the corn filament and the CIELAB value, and then uses the verification sample to verify and optimize the relationship model, so that the relationship model has a high detection accuracy, and uses the relationship table corresponding to each anthocyanin color intensity of the relationship model and the CIELAB value to realize the classification of the anthocyanin color intensity of the corn filament to be tested. Therefore, while ensuring the detection accuracy of the color intensity of the anthocyanin of the corn filament, the calculation complexity is reduced, the operation steps are reduced, and the detection convenience is provided, providing a more effective boundary corn filament anthocyanin color intensity classification method in actual application scenarios.

In a preferred embodiment, a corn silk sample set is obtained, and the steps of dividing the corn silk sample set into analysis samples and verification samples specifically include: obtaining a corn silk sample set; wherein the corn silk sample set includes a plurality of corn silk images whose anthocyanin color intensity is visually graded and evenly distributed in 9 anthocyanin color intensity levels; dividing the corn silk images of each anthocyanin color intensity level in the corn silk sample set into analysis sample images and verification sample images according to a preset ratio; using the analysis sample images of each anthocyanin color intensity level as the analysis samples of the corn silk sample set, and using the verification sample images of each anthocyanin color intensity level as the verification samples of the corn silk sample set.

In this embodiment, corn silk images including several anthocyanin color intensity visually graded uniformly distributed in 9 anthocyanin color intensity levels are obtained as corn silk sample sets, and the corn silk sample sets are divided into analysis samples and verification samples. Specifically, in practical applications, standard silk images of 225 corn varieties can be collected as research materials, and the anthocyanin color intensity is 25 varieties from level 1 to level 9. Among them, 20 varieties are selected for regression analysis at each level, and a total of 180 varieties are used for analysis; 5 varieties are selected for verification at each level, and a total of 45 varieties are used for verification.

In a preferred embodiment, the steps of placing the analysis sample and the verification sample in an image acquisition room, using an image acquisition device to acquire a color measurement image in the image acquisition room, and determining the CIELAB values of the analysis sample and the verification sample based on the color measurement image specifically include: placing the analysis sample and the verification sample in an image acquisition room together with a standard color card, respectively, and using an image acquisition device to acquire a first color measurement image and a second color measurement image containing the analysis sample and the standard color card or the verification sample and the standard color card at two opposite positions in the image acquisition room; establishing a color value conversion matrix according to the first RGB value of each color block of the standard color card in the first color measurement image and the second color measurement image and the CIEXYZ value of the color block itself; determining the CIEXYZ value of the analysis sample or the verification sample according to the second RGB value of the analysis sample or the verification sample in the first color measurement image and the second color measurement image and the color value conversion matrix, and converting the CIEXYZ value of the analysis sample or the verification sample into a CIELAB value.

In this embodiment, the method for determining the anthocyanin color of the corn silk image corresponding to the analysis sample and the verification sample is as follows: the corn silk image and the standard color card are placed in an image acquisition room, the image acquisition room is provided with a shooting light source, and the overall sealing design prevents the external light source from entering the image acquisition room, thereby reducing the influence of environmental changes during the anthocyanin color determination of the corn silk image; the image acquisition room contains a corn silk image photo and a standard color card and is located at two opposite positions in the image acquisition room; the first color measurement image and the second color measurement image are collected in the image acquisition room; the first RGB value of each color block in the first color measurement image and the first RGB value of each color block in the second color measurement image are taken as the average value of the RGB value of each color block in the standard color card after removing environmental interference; and the RGB value is calculated based on the RGB value and the standard color card. The CIEXYZ value of the color block itself is used to establish a color value conversion matrix, and the CIEXYZ value of the corn silk image is determined according to the second RGB value of the corn silk image in the first color measurement image and the second color measurement image and the color value conversion matrix, and the CIEXYZ value of the corn silk image is converted into the CIELAB value of the corn silk image; thereby, high-precision detection of the anthocyanin color of the corn silk image corresponding to the analysis sample and the verification sample is achieved, avoiding the problem of low detection accuracy caused by human eye observation, and then a relationship model can be established according to the anthocyanin color value of the corn silk image detected with high precision and the corresponding visual grading of the anthocyanin color intensity, providing data support for the anthocyanin color value determination of the corn silk to be tested and the fine classification of corn quality.

In a preferred embodiment, the step of establishing a relationship model between the anthocyanin color intensity and the CIELAB value of corn silk using the CIELAB value and the visual grading of the anthocyanin color intensity of the analysis sample specifically includes: obtaining the visual grading of the anthocyanin color intensity and the corresponding CIELAB value of several corn silk images in the analysis sample; performing a least squares univariate regression analysis using the visual grading of the anthocyanin color intensity and the color value L in the corresponding CIELAB value to establish a relationship model between the anthocyanin color intensity and the CIELAB value of corn silk, and generating a univariate regression line between the anthocyanin color intensity and the CIELAB value.

In this embodiment, after obtaining the CIELAB value and the visual grade of anthocyanin color intensity of the analysis sample, the CIELAB value and the visual grade of anthocyanin color intensity of each corn silk image in the analysis sample can be used as a set of data pairs. According to several sets of data pairs of several corn silk images, a least squares univariate regression analysis is performed on the visual grade of anthocyanin color intensity of corn silk and the color value L of the corn silk image to generate a univariate regression line for representing the relationship between the anthocyanin color intensity of corn silk and the CIELAB value. The univariate regression line is used to preliminarily characterize the relationship model between the anthocyanin color intensity of corn silk and the color value L.

Furthermore, the relationship model is verified using the CIELAB values and the visual grading of anthocyanin color intensity of the verification sample, and the relationship model is optimized according to the verification result, specifically including: inputting the CIELAB values of several corn silk images in the verification sample into the relationship model, calculating the numerical difference between the anthocyanin color intensity corresponding to the corn silk image output by the relationship model and the corn silk image and the visual grading of anthocyanin color intensity; judging whether there are corn silk images in the verification sample whose numerical difference exceeds the preset error allowable value, if so, dividing all the color values L of the corn silk images in the analysis sample into intervals, and using the color values L of the corn silk images in the two divided intervals and the corresponding visual grading of anthocyanin color intensity to perform least squares univariate regression analysis in the intervals, respectively, to obtain a univariate regression line for each interval; repeating the above verification process until there are no corn silk images in the verification sample whose numerical difference exceeds the preset error allowable value, and generating a relationship model between the anthocyanin color intensity of corn silk and the CIELAB value based on the univariate regression line of each interval.

Among them, the division position is each color value L of the corn silk image in the analysis sample; all the color values L of the corn silk image in the analysis sample are divided into intervals once, specifically including: through repeated verification, if the interval after the current color value L is used as the division position for interval division cannot meet the requirements of the numerical difference and the preset error allowable value, then the next color value L is used as the division position for interval division; if the interval corresponding to each division position cannot meet the requirements of the numerical difference and the preset error allowable value, then the interval division with the least number of occurrences of the numerical difference exceeding the preset error allowable value is selected from all the interval division results, and each interval obtained from the result of the interval division is divided again, and the interval verification process is repeated.

Considering that the univariate regression line obtained by least squares univariate regression analysis based on the visual grading of corn silk anthocyanin color intensity and the color value L of corn silk images may have a large error in practical applications, in this embodiment, the relationship model is verified and optimized using the CIELAB value of the verification sample and the visual grading of anthocyanin color intensity. The specific verification and optimization process is as follows: first, the relationship model is used to generate the anthocyanin color intensity of several corn silk images in the verification sample, and the numerical difference between the anthocyanin color intensity and the visual grading of the anthocyanin color intensity is calculated. If there is a corn silk image in the verification sample with a numerical difference exceeding the preset error allowable value, it indicates that the univariate regression line is not good. The error of the straight line within a certain range is relatively large. At this time, the color value L is selected as the dividing position from all the color values L of the corn silk image in the analysis sample in turn to divide the entire color value L into intervals, and then the least squares univariate regression and verification process is performed on each divided interval until there is no corn silk image with a numerical difference exceeding the preset error allowable value in the verification sample. At this time, the relationship formed by all the univariate regression lines in each interval is a relationship model between the anthocyanin color intensity and the CIELAB value with a smaller error and higher detection accuracy. The relationship model is used for the anthocyanin color intensity grading of the corn silk to be tested, which has higher detection accuracy and detection efficiency.

In the specific example of practical application, after using the analysis sample for least squares univariate regression and verification optimization, the analysis results obtained are as follows:

L≧30 L<30 Multiple R 0.996684515 0.932113420 R Square 0.993380023 0.868835428 Adjusted R Square 0.993265885 0.867723864

Table 1: Regression effect parameters

Among them, Multiple R, R Square and Adjusted R Square are three evaluation indicators of the regression equation. The three values are between 0 and 1. The closer to 1, the better the effect.

Table 2: Regression analysis results

Among them, Intercept is the constant term and L is the coefficient of color value L. Among them, the P values of the constant term and L are less than 0.05, and the difference is significant.

Thus, the relationship model between the anthocyanin color intensity and the CIELAB value of corn silk in the specific example was obtained = -0.0829L + 5.5711 (L value ≧30); anthocyanin color intensity = -0.6038L + 21.728 (L value <30).

In a preferred embodiment, based on the optimized relationship model, a comparison relationship table between corn silks at each anthocyanin color intensity and CIELAB values is generated, and the anthocyanin color intensity grading step of the corn silks to be tested is performed using the comparison relationship table, which specifically includes: obtaining a univariate regression line for each interval corresponding to the optimized relationship model, inputting the middle value of each two adjacent anthocyanin color intensities into the univariate regression line, and outputting a color value L corresponding to the middle value; using the color value L of the two adjacent middle values corresponding to each anthocyanin color intensity as the color value L range corresponding to the anthocyanin color intensity, generating a comparison relationship table between corn silks at each anthocyanin color intensity and CIELAB values; and using the comparison relationship table to grade the anthocyanin color intensity of the corn silks to be tested.

In this embodiment, the comparison relationship table is used to grade the anthocyanin color intensity of the corn silk to be tested. The comparison relationship table between the corn silk at each anthocyanin color intensity and the CIELAB value is as follows:

Classification Code 1 2 3 4 5 6 7 8 9 L value L>49.10 37.04<L≦49.10 30≦L≦37.04 28.53<L<30 26.87<L≦28.53 25.22<L≦26.87 23.56<L≦25.22 21.90<L≦23.56 ≦21.90

Table 3: Comparison table

In this embodiment, a sample set consisting of corn silk images is divided into analysis samples and verification samples, and a relationship model between the anthocyanin color intensity of corn silk and the CIELAB value is established by using the visual grading of the anthocyanin color intensity of the analysis samples and the measured CIELAB value. The verification sample is then used to verify and optimize the relationship model, so that the relationship model has a high detection accuracy. The anthocyanin color intensity grading of the corn silk to be tested is realized by using a comparison relationship table of each anthocyanin color intensity corresponding to the relationship model and the CIELAB value. Thus, while ensuring the detection accuracy of the anthocyanin color intensity of corn silk, the calculation complexity is reduced, the operation steps are reduced, and the detection convenience is provided, thereby providing a more effective boundary corn silk anthocyanin color intensity grading method in actual application scenarios.

Refer to FIG. 3 , which is a structural block diagram of an embodiment of a device for grading the color intensity of anthocyanins in corn filaments according to the present invention.

As shown in FIG3 , the corn silk anthocyanin color intensity grading device proposed in the embodiment of the present invention comprises:

A division module 10 is used to obtain a corn silk sample set and divide the corn silk sample set into an analysis sample and a verification sample; wherein the corn silk sample set includes a plurality of corn silk images;

A determination module 20, configured to place the analysis sample and the verification sample into an image acquisition chamber, acquire a color measurement image in the image acquisition chamber using an image acquisition device, and determine CIELAB values of the analysis sample and the verification sample based on the color measurement image;

Establishing module 30, for establishing a relationship model between anthocyanin color intensity and CIELAB value of corn silk by visually grading the CIELAB value and anthocyanin color intensity of the analysis sample;

An optimization module 40, for verifying the relationship model using the CIELAB value of the verification sample and the visual grading of anthocyanin color intensity, and optimizing the relationship model according to the verification result;

The grading module 50 is used to generate a comparison relationship table between corn silk at each anthocyanin color intensity and CIELAB value based on the optimized relationship model, and use the comparison relationship table to grade the anthocyanin color intensity of the corn silk to be tested.

Other embodiments or specific implementations of the corn silk anthocyanin color intensity grading device of the present invention can refer to the above-mentioned method embodiments, which will not be repeated here.

In addition, the present invention also proposes a corn filament anthocyanin color intensity grading device, which includes: a memory, a processor, and a corn filament anthocyanin color intensity grading program stored in the memory and executable on the processor, wherein the corn filament anthocyanin color intensity grading program implements the steps of the corn filament anthocyanin color intensity grading method as described above when executed by the processor.

The specific implementation of the corn filament anthocyanin color intensity grading device of the present application is basically the same as the various embodiments of the above-mentioned corn filament anthocyanin color intensity grading method, and will not be repeated here.

In addition, the present invention also proposes a readable storage medium, the readable storage medium includes a computer-readable storage medium, on which a corn filament anthocyanin color intensity grading program is stored. The readable storage medium can be the memory 1005 in the terminal of FIG1 , or can be at least one of a ROM (Read-Only Memory)/RAM (Random Access Memory), a magnetic disk, and an optical disk, and the readable storage medium includes a number of instructions for enabling a corn filament anthocyanin color intensity grading device having a processor to execute the corn filament anthocyanin color intensity grading method described in each embodiment of the present invention.

The specific implementation in the readable storage medium of the present application is basically the same as the embodiments of the above-mentioned corn silk anthocyanin color intensity classification method, and will not be repeated here.

It is understood that, in the description of this specification, the description with reference to the terms "one embodiment", "another embodiment", "other embodiments", or "first to Nth embodiments" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims (6)

1. The method for classifying the color development intensity of the maize anthocyanin is characterized by comprising the following steps of:

obtaining a maize filament sample set, and dividing the maize filament sample set into an analysis sample and a verification sample; the corn silk sample set comprises a plurality of corn silk images;

Placing the analysis sample and the verification sample into an image acquisition chamber, acquiring a color measurement image in the image acquisition chamber by using an image acquisition device, and determining CIELAB values of the analysis sample and the verification sample based on the color measurement image;

visual classification is carried out by utilizing the CIELAB value and anthocyanin color development intensity of the analysis sample, and a relation model of the anthocyanin color development intensity and the CIELAB value of the maize filament is established;

Verifying the relation model by visual classification of CIELAB values and anthocyanin color development intensity of the verification samples, and optimizing the relation model according to verification results; the method specifically comprises the following steps:

Inputting CIELAB values of a plurality of corn silk images in the verification sample into the relation model, and calculating the numerical difference between anthocyanin chromogenic intensity corresponding to the corn silk images output by the relation model and visual classification of the corn silk images and the anthocyanin chromogenic intensity;

Judging whether a maize filament image with a numerical value difference exceeding a preset error allowable value exists in the verification sample, if so, dividing all color values L of the maize filament image in the analysis sample into sections once, and carrying out least square unitary regression analysis in the sections by utilizing the color values L of the maize filament image in the two sections after division and the corresponding anthocyanin color development intensity visual classification respectively to obtain a unitary regression line of each section;

Repeating the verification process until no maize filament image with the numerical value difference exceeding the preset error allowable value exists in the verification sample, and generating a relation model of anthocyanin color development intensity and CIELAB value of the maize filament based on a unitary regression line of each interval;

The dividing position of the interval division is each color value L of the corn silk image in the analysis sample; performing a primary interval division step on all color values L of the maize filament image in the analysis sample, wherein the method specifically comprises the following steps:

Through repeated verification, if the interval after the interval division by taking the current color value L as the dividing position cannot meet the requirements of the numerical value difference and the preset error allowable value, the interval division is carried out by taking the next color value L as the dividing position;

If the interval corresponding to each dividing position cannot meet the requirements of the numerical value difference and the preset error allowable value, selecting interval division with the minimum number of times that the numerical value difference exceeds the preset error allowable value from all interval division results, respectively carrying out interval division again on each interval obtained by the interval division result, and repeating the interval verification process;

Based on the optimized relation model, a comparison relation table of the anthocyanin color development intensity of the maize filaments and CIELAB values is generated, and the comparison relation table is utilized to classify the anthocyanin color development intensity of the maize filaments to be detected.

2. The method for classifying the color development intensity of the anthocyanin in maize filaments according to claim 1, wherein the step of obtaining a maize filament sample set and dividing the maize filament sample set into an analysis sample and a verification sample comprises the steps of:

Obtaining a maize filament sample set; the corn silk sample set comprises a plurality of corn silk images with anthocyanin color development intensities which are visually classified and uniformly distributed in 9 anthocyanin color development intensity levels;

dividing a maize filament image of each anthocyanin color development intensity level in the maize filament sample set into an analysis sample image and a verification sample image according to a preset proportion;

Taking the analysis sample image of each anthocyanin color development intensity level as an analysis sample of the corn silk sample set, and taking the verification sample image of each anthocyanin color development intensity level as a verification sample of the corn silk sample set.

3. The method for classifying the color development intensity of maize silk anthocyanin according to claim 2, wherein the step of placing the analysis sample and the verification sample into an image collection room, collecting a color measurement image in the image collection room by using an image collection device, and determining the CIELAB values of the analysis sample and the verification sample based on the color measurement image specifically comprises:

respectively placing the analysis sample and the verification sample into an image acquisition chamber together with a standard color card, and acquiring a first color measurement image and a second color measurement image which comprise two opposite positions of the analysis sample and the standard color card or the verification sample and the standard color card in the image acquisition chamber by using an image acquisition device;

Establishing a color value conversion matrix according to a first RGB value of each color block of a standard color card and a CIEXYZ value of the color block in the first color measurement image and the second color measurement image;

And determining the CIEXYZ value of the analysis sample or the verification sample according to the second RGB value of the analysis sample or the verification sample in the first color measurement image and the second color measurement image and the color value conversion matrix, and converting the CIEXYZ value of the analysis sample or the verification sample into the CIELAB value.

4. The method for classifying anthocyanin color development intensity of maize filaments according to claim 3, wherein the step of establishing a model of the relationship between anthocyanin color development intensity and CIELAB value of maize filaments by visual classification using CIELAB value and anthocyanin color development intensity of the analysis sample comprises:

Obtaining anthocyanin color development intensity visual classification and corresponding CIELAB values of a plurality of maize filament images in the analysis sample;

And carrying out least square unitary regression analysis on the anthocyanin color development intensity visual classification and the color value L in the corresponding CIELAB value, and establishing a relationship model of the anthocyanin color development intensity and the CIELAB value of the maize filament to generate a unitary regression line of the anthocyanin color development intensity and the CIELAB value.

5. The method for classifying anthocyanin color intensity of maize filaments according to claim 4, wherein based on the optimized relation model, a comparison relation table of anthocyanin color intensity of maize filaments and CIELAB value is generated, and the comparison relation table is utilized to classify the anthocyanin color intensity of maize filaments to be detected, which specifically comprises the following steps:

Acquiring a unitary regression line of each interval corresponding to the optimized relation model, inputting the intermediate value of the color development intensity of each two adjacent anthocyanin into the unitary regression line, and outputting a color value L corresponding to the intermediate value;

Taking the color value L of two adjacent intermediate values corresponding to each anthocyanin color development intensity as the range of the color value L corresponding to the anthocyanin color development intensity, and generating a comparison relation table of the maize filaments and CIELAB values under each anthocyanin color development intensity;

and classifying anthocyanin color development intensity of the maize filaments to be detected by using the comparison relation table.

6. The utility model provides a maize silk anthocyanin color development intensity grading plant which characterized in that includes:

The dividing module is used for obtaining a corn silk sample set and dividing the corn silk sample set into an analysis sample and a verification sample; the corn silk sample set comprises a plurality of corn silk images;

The determining module is used for placing the analysis sample and the verification sample into an image acquisition chamber, acquiring a color measurement image in the image acquisition chamber by using an image acquisition device, and determining CIELAB values of the analysis sample and the verification sample based on the color measurement image;

The establishing module is used for visually classifying by utilizing the CIELAB value and anthocyanin color development intensity of the analysis sample and establishing a relation model of the anthocyanin color development intensity and the CIELAB value of the maize filament;

The optimizing module is used for verifying the relation model by visual classification of CIELAB values and anthocyanin color development intensity of the verification sample, and optimizing the relation model according to verification results; the method specifically comprises the following steps:

Inputting CIELAB values of a plurality of corn silk images in the verification sample into the relation model, and calculating the numerical difference between anthocyanin chromogenic intensity corresponding to the corn silk images output by the relation model and visual classification of the corn silk images and the anthocyanin chromogenic intensity;

Judging whether a maize filament image with a numerical value difference exceeding a preset error allowable value exists in the verification sample, if so, dividing all color values L of the maize filament image in the analysis sample into sections once, and carrying out least square unitary regression analysis in the sections by utilizing the color values L of the maize filament image in the two sections after division and the corresponding anthocyanin color development intensity visual classification respectively to obtain a unitary regression line of each section;

Repeating the verification process until no maize filament image with the numerical value difference exceeding the preset error allowable value exists in the verification sample, and generating a relation model of anthocyanin color development intensity and CIELAB value of the maize filament based on a unitary regression line of each interval;

the dividing position of the interval division is each color value L of the corn silk image in the analysis sample; dividing all color values L of the maize filament image in the analysis sample into intervals, wherein the method specifically comprises the following steps:

Through repeated verification, if the interval after the interval division by taking the current color value L as the dividing position cannot meet the requirements of the numerical value difference and the preset error allowable value, the interval division is carried out by taking the next color value L as the dividing position;

If the interval corresponding to each dividing position cannot meet the requirements of the numerical value difference and the preset error allowable value, selecting interval division with the minimum number of times that the numerical value difference exceeds the preset error allowable value from all interval division results, respectively carrying out interval division again on each interval obtained by the interval division result, and repeating the interval verification process;

and the grading module is used for generating a comparison relation table of the anthocyanin color development intensity of the maize filaments and the CIELAB value under each anthocyanin color development intensity based on the optimized relation model, and grading the anthocyanin color development intensity of the maize filaments to be tested by utilizing the comparison relation table.