Crop germplasm resource evaluation system and method based on image analysis

Abstract

The invention relates to the technical field of germplasm resource management, and discloses a crop germplasm resource evaluation system and method based on image analysis, and the evaluation method comprises the following steps: S1, planting crop germplasm resources in a test field; s2, a plurality of standard color cards with the same color are fixedly arranged in the test field; the standard color card information is input into the standard color card information module; s3, randomly shooting a plurality of pieces of crop germplasm resource growth image information at specified heights of different positions of the test field through an image acquisition module; s4, analyzing the standard color card information and the image information through an analysis module, and judging whether the quality of the image information is qualified or not according to an analysis result; and the crop germplasm resource growth state is evaluated by analyzing the information of a plurality of images with qualified quality, so that the accuracy of the evaluation result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of germplasm resource management, and in particular to a crop germplasm resource evaluation system and method based on image analysis. Background Art

[0002] Germplasm resources contain rich genetic diversity, which is the basis of breeding work. By utilizing the excellent genes and traits in germplasm resources, new crop varieties with higher yield, better quality and stronger stress resistance can be bred. In order to prevent the permanent loss of germplasm resources due to human or environmental factors, it is necessary to collect the collected germplasm resources. After the germplasm resources are collected, they need to be distributed to breeding centers for breeding in order to study the characteristics of the crop germplasm resources.

[0003] At present, in order to monitor the growth status of crop germplasm resources for breeding in real time, each breeding center will install image acquisition modules to monitor the crop germplasm resources in real time, and automatically evaluate the growth status of crop germplasm resources through the obtained image information.

[0004] However, due to different shooting parameters set by the image acquisition module and different lighting and positions during shooting, the collected image information will show different degrees of color difference from the actual image color. Therefore, if the growth status of crop germplasm resources is evaluated directly based on the image information obtained by the image acquisition module, the evaluation results may be biased and the reliability of the evaluation results is low. Summary of the invention

[0005] The purpose of the present invention is to provide a crop germplasm resource evaluation system and method based on image analysis to solve the following technical problems:

[0006] How to improve the reliability of crop germplasm resource assessment.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A method for evaluating crop germplasm resources based on image analysis, the method comprising the following steps:

[0009] S1: Planting crop germplasm resources in experimental fields;

[0010] S2: multiple standard color cards of the same color are fixedly set in the test field; and the standard color card information is input into the standard color card information module;

[0011] S3: randomly taking a number of images of crop germplasm resources growth information at designated heights at different locations in the experimental field through an image acquisition module;

[0012] S4: Analyze the standard color card information and image information through the analysis module, and judge whether the image information quality is qualified according to the analysis result; and evaluate the growth status of crop germplasm resources by analyzing multiple images of qualified quality.

[0013] As a further solution of the present invention: the analysis module includes an identification unit and an analysis unit;

[0014] The recognition unit is a trained neural network model, which is used to recognize image information and mark the standard color cards and complete leaves in the image information; obtain the number of standard color cards, the LAB color space of each standard color card number, and the number of complete leaves;

[0015] The analysis unit analyzes the number of standard color cards, the LAB color space of each standard color card number and the number of complete leaves, and determines whether the image information quality is qualified according to the analysis results; and evaluates the growth status of crop germplasm resources by analyzing multiple pieces of qualified image information.

[0016] As a further solution of the present invention: the analysis process of the analysis module is:

[0017] S10: randomly moving the image acquisition module to a specified height above any position of the test field to obtain the nth image information;

[0018] S20: Identify and annotate the standard color cards and complete leaves in the image information through the recognition unit; obtain the number of standard color cards, the LAB color space of each standard color card number, the number of complete leaves, the LAB color space of each complete leaf, and the damage point of each complete leaf;

[0019] S30: Analyze the number of standard color cards, the LAB color space of each standard color card, and the number of complete leaves, and evaluate whether the quality of the nth image information is qualified according to the analysis results; if qualified, proceed to step S40; if unqualified, the shooting parameters of the image acquisition module are automatically corrected, and step S10 is repeated;

[0020] S40: Set n=n+1, repeat S10; until n>N, N is the preset number of image information, and enter step S50:

[0021] S50: Evaluate the growth status of crop germplasm resources by analyzing N pieces of image information of qualified quality.

[0022] As a further solution of the present invention: the evaluation process of whether the quality of the n-th image information is qualified is performed by the formula:

[0023]

[0024] Where f(X) is the judgment function. When X>0, f(X)=1; when X≤0, f(X)=0; Q n is the number of standard color cards in the nth image information, q n is the qth standard color card in the nth image information; Q0 is the number of preset standard color cards; P n is the judgment index of the nth image information; is the color difference between the qth standard color card in the nth image information and the actual color of the standard color card; W1 is the first allowable error value; W2 is the second allowable error value; is the average color difference in the nth image information; S n is the number of complete leaves in the nth image information; S0 is the preset number of complete leaves.

[0025] As a further solution of the present invention: the standard color card information includes the actual brightness value, the actual red-green axis value and the actual yellow-blue axis value of the standard color card.

[0026] As a further solution of the present invention: the color difference between the qth standard color card in the nth image information and the actual color of the standard color card To pass the formula:

[0027]

[0028] in, is the brightness value of the qth standard color card in the nth image information; L0 is the actual brightness value of the standard color card; is the red-green axis value of the qth standard color card in the nth image information; A0 is the actual red-green axis value of the standard color card; is the yellow-blue axis value of the qth standard color card in the nth image information; B0 is the actual yellow-blue axis value of the standard color card.

[0029] As a further solution of the present invention: the process of determining whether the quality of the n-th image information is qualified further includes:

[0030] When P n =1, the image information quality is unqualified;

[0031] When P n When =e, the image information quality is qualified.

[0032] A crop germplasm resource evaluation system based on image analysis, the evaluation system comprising:

[0033] Image acquisition module, used to monitor the growth status of germplasm resources and obtain image information of germplasm resources growth;

[0034] Standard color card information module, used to store standard color card information;

[0035] The analysis module is used to analyze the standard color card information and image information to determine whether the image information quality is qualified; and by analyzing multiple pieces of qualified image information, the growth status of crop germplasm resources is evaluated.

[0036] Beneficial effects of the present invention:

[0037] (1) The present invention first plants crop germplasm resources in an experimental field; then, a plurality of standard color cards of the same color are fixedly arranged in the experimental field; and the standard color card information is input into a standard color card information module; and then, an image acquisition module is used to randomly capture a plurality of image information of the growth of crop germplasm resources at a specified height at different locations in the experimental field; specifically, the image acquisition module can be captured by an unmanned aerial vehicle; finally, an analysis module is used to analyze the standard color card information and the image information, and whether the image information quality is qualified is determined according to the analysis result; and by analyzing a plurality of qualified image information, the growth status of the crop germplasm resources is evaluated, thereby improving the reliability of the evaluation result. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The present invention will be further described below in conjunction with the accompanying drawings.

[0039] Figure 1 A method flow chart of an embodiment of the present invention;

[0040] Figure 2 A system module framework diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1 As shown, in one embodiment, a method for evaluating crop germplasm resources based on image analysis is provided, and the evaluation method comprises the following steps:

[0043] S1: Planting crop germplasm resources in experimental fields;

[0044] S2: multiple standard color cards of the same color are fixedly set in the test field; and the standard color card information is input into the standard color card information module;

[0045] S3: randomly taking a number of images of crop germplasm resources growth information at designated heights at different locations in the experimental field through an image acquisition module;

[0046] S4: Analyze the standard color card information and image information through the analysis module, and judge whether the image information quality is qualified according to the analysis result; and evaluate the growth status of crop germplasm resources by analyzing multiple images of qualified quality.

[0047] Through the above technical scheme, this embodiment first plants crop germplasm resources in the experimental field; then fixes a plurality of standard color cards of the same color in the experimental field; and inputs the standard color card information into the standard color card information module; then randomly shoots a plurality of image information of the growth of crop germplasm resources at a specified height at different positions in the experimental field through the image acquisition module; specifically, the image acquisition module can use a drone for acquisition; finally, the standard color card information and image information are analyzed through the analysis module, and whether the quality of the image information is qualified is determined according to the analysis results; and by analyzing a plurality of qualified image information, the growth status of the crop germplasm resources is evaluated, thereby improving the reliability of the evaluation results.

[0048] As an embodiment of the present invention, the analysis module includes an identification unit and an analysis unit;

[0049] The recognition unit is a trained neural network model, which is used to recognize image information and mark the standard color cards and complete leaves in the image information; obtain the number of standard color cards, the LAB color space of each standard color card number, and the number of complete leaves;

[0050] The analysis unit analyzes the number of standard color cards, the LAB color space of each standard color card number, and the number of complete leaves, and determines whether the image information quality is qualified according to the analysis results; and evaluates the growth status of crop germplasm resources by analyzing multiple pieces of qualified image information;

[0051] The analysis process of the analysis module is as follows:

[0052] S10: randomly moving the image acquisition module to a specified height above any position of the test field to obtain the nth image information;

[0053] S20: Identify and annotate the standard color cards and complete leaves in the image information through the recognition unit; obtain the number of standard color cards, the LAB color space of each standard color card number, the number of complete leaves, the LAB color space of each complete leaf, and the damage point of each complete leaf;

[0054] S30: Analyze the number of standard color cards, the LAB color space of each standard color card, and the number of complete leaves, and evaluate whether the quality of the nth image information is qualified according to the analysis results; if qualified, proceed to step S40; if unqualified, the shooting parameters of the image acquisition module are automatically corrected, and step S10 is repeated;

[0055] S40: Set n=n+1, repeat S10; until n>N, N is the preset number of image information, and enter step S50:

[0056] S50: evaluating the growth status of crop germplasm resources by analyzing N pieces of image information with qualified quality;

[0057] Through the above technical scheme, this embodiment first randomly moves the image acquisition module to a specified height above any position of the test field to obtain the nth image information; then identifies the image information through the recognition unit to mark the standard color card and the complete leaf in the image information; obtains the number of standard color cards, the LAB color space of each standard color card number, the number of complete leaves, the LAB color space of each complete leaf and the damage point of each complete leaf; then analyzes the number of standard color cards, the LAB color space of each standard color card number, and the number of complete leaves, and evaluates whether the quality of the nth image information is qualified according to the analysis result; if qualified, n=n+1 is set, and the image acquisition module is randomly moved to a specified height above any position of the test field to obtain the nth image information again; if unqualified, after the shooting parameters of the image acquisition module are automatically corrected, the image acquisition module is randomly moved to a specified height above any position of the test field to obtain the nth image information and then judged again until n>N, and the growth status of crop germplasm resources is evaluated by analyzing N qualified image information; thereby improving the reliability of the evaluation result.

[0058] It should be noted that obtaining the number of standard color cards, the LAB color space of each standard color card and the number of complete leaves through the neural network model, the specific training process of the neural network model and the automatic correction of the shooting parameters of the image acquisition module are existing technologies, and the preset number N of image information is a preset value obtained based on experience, which will not be described in detail here.

[0059] As an implementation manner of the present invention, the process of determining whether the quality of the n-th image information is qualified includes the following formula:

[0060]

[0061] Where f(X) is the judgment function. When X>0, f(X)=1; when X≤0, f(X)=0; Q n is the number of standard color cards in the nth image information, q nis the qth standard color card in the nth image information; Q0 is the number of preset standard color cards; P n is the judgment index of the nth image information; is the color difference between the qth standard color card in the nth image information and the actual color of the standard color card; W1 is the first allowable error value; W2 is the second allowable error value; is the average color difference in the nth image information; S n is the number of complete leaves in the nth image information; S0 is the preset number of complete leaves;

[0062] The process of judging whether the quality of the nth image information is qualified also includes:

[0063] When P n =1, the image information quality is unqualified;

[0064] When P n =e, the image information quality is qualified;

[0065] Through the above technical solution, this embodiment uses Q n -Q0 is the difference between the number of standard color cards in the nth image information and the number of preset standard color cards. n When -Q0≤0, it means that the number of standard color cards in the nth image information does not reach the preset number of standard color cards, and the image information cannot obtain a sufficient number of standard color card color spaces to compare with the actual color spaces of the standard color cards, which may easily lead to inaccurate image information quality assessment. Therefore, the judgment index P in the nth image information n =1; when Q n - When Q0>0, it means that the number of standard color cards in the nth image information exceeds the preset number of standard color cards, so that a sufficient number of standard color card color spaces can be obtained in the image information to compare with the actual color space of the standard color cards, thereby improving the accuracy of image information quality assessment; is the accumulated value of color difference of all standard color cards in the nth image information; is the difference between the first allowable error value and the accumulated value of the color difference of all standard color cards in the nth image information; when When , it means that the cumulative value of the color difference of all standard color cards in the nth image information exceeds the first allowable error value, the image quality in the nth image information is unqualified, and the judgment index P in the nth image information is n =1; when When , it indicates that the accumulated value of the color difference of all the standard color cards in the nth image information is within the first allowable error value, indicating that the color of the standard color card in the image information is very close to the actual color of the standard color card, and the color difference of the image information is small, so it can be used as image information for evaluating the growth status of crops; is the standard deviation of the color difference values ​​of all standard color cards in the nth image information, standard deviation The bigger, is the difference between the second allowable error value and the standard value of the color difference values ​​of all standard color cards in the nth image information; when When , it means that the standard value of the color difference of all standard color cards in the nth image information exceeds the second allowable error value, the color difference of each standard color card in the nth image information is large, the image quality in the nth image information is unqualified, and the judgment index P in the nth image information is n =1; when When , it means that the cumulative value of the color difference of all standard color cards in the nth image information is within the first allowable error value, which means that the standard value of the color difference of all standard color cards in the nth image information is within the second allowable error value, and the color difference of the image information is small, so it can be used as image information for evaluating the growth status of crops; the color difference of each standard color card in the nth image information is small, and the image quality in the nth image information is unqualified, so this image information can be used as image information for evaluating the growth status of crops to improve the accuracy of evaluating the growth status of germplasm resources; S n -S0 is the difference between the number of complete leaves in the nth image information and the preset number of complete leaves. n When -S0≤0, it means that the number of complete leaves in the nth image information does not reach the preset number of complete leaves, and the image information cannot obtain a sufficient number of complete leaves to evaluate the growth status of germplasm resources, so it cannot be used as image information for evaluating the growth status of crops. Therefore, the judgment index P in the nth image information n =1; when S n When -S0>0, it means that the number of complete leaves in the nth image information exceeds the preset number of complete leaves, so that a sufficient number of complete leaves can be obtained in the image information to evaluate the growth status of germplasm resources, thereby improving the accuracy of the evaluation of the growth status of germplasm resources; therefore, when P n = 1, the quality of the image information is unqualified; when Q n -Q0>0 and and And S n - When S0>0,

[0066] Therefore, the judgment index P in the nth image information n =e;

[0067] It should be noted that the preset number of standard color cards Q0, the first allowable error value W1, the second allowable error value W2 and the preset number of complete leaves S0 are preset values, which are obtained based on experience and will not be described in detail here.

[0068] As an embodiment of the present invention, the standard color card information includes the actual brightness value, the actual red-green axis value and the actual yellow-blue axis value of the standard color card; the color difference between the qth standard color card in the nth image information and the actual color of the standard color card To pass the formula:

[0069]

[0070] in, is the brightness value of the qth standard color card in the nth image information; L0 is the actual brightness value of the standard color card; is the red-green axis value of the qth standard color card in the nth image information; A0 is the actual red-green axis value of the standard color card; is the yellow-blue axis value of the qth standard color card in the nth image information; B0 is the actual yellow-blue axis value of the standard color card;

[0071] Through the above technical solution, this embodiment is the brightness difference between the brightness value of the qth standard color card in the nth image information and the actual brightness value of the standard color card; the larger the brightness difference, the greater the color difference The bigger; The red-green axis difference between the red-green axis value of the qth standard color card in the nth image information and the actual red-green axis value of the standard color card; the larger the red-green axis difference, the greater the color difference The bigger; The yellow-blue axis difference between the yellow-blue axis value of the qth standard color card in the nth image information and the actual yellow-blue axis value of the standard color card; the larger the yellow-blue axis difference, the greater the color difference The bigger;

[0072] It should be noted that the actual lightness value L0 of the standard color card, the actual red-green axis value A0 of the standard color card, and the actual yellow-blue axis value B0 of the standard color card are obtained according to the standard color card and are preset values, which will not be described in detail here.

[0073] See also Figure 2 As shown, a crop germplasm resource evaluation system based on image analysis, the evaluation system comprises:

[0074] Image acquisition module, used to monitor the growth status of germplasm resources and obtain image information of germplasm resources growth;

[0075] Standard color card information module, used to store standard color card information;

[0076] The analysis module is used to analyze the standard color card information and image information to determine whether the image information quality is qualified; and by analyzing multiple pieces of qualified image information, the growth status of crop germplasm resources is evaluated.

[0077] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for evaluating crop germplasm resources based on image analysis, characterized in that: The evaluation method comprises the following steps: S1: Planting crop germplasm resources in experimental fields; S2: multiple standard color cards of the same color are fixedly set in the test field; and the standard color card information is input into the standard color card information module; S3: randomly taking a number of images of crop germplasm resources growth information at designated heights at different locations in the experimental field through an image acquisition module; S4: Analyze the standard color card information and image information through the analysis module, and judge whether the image information quality is qualified according to the analysis result; and evaluate the growth status of crop germplasm resources by analyzing multiple images of qualified quality.

2. The method for evaluating crop germplasm resources based on image analysis according to claim 1, characterized in that: The analysis module includes an identification unit and an analysis unit; The recognition unit is a trained neural network model, which is used to recognize image information and mark the standard color cards and complete leaves in the image information; obtain the number of standard color cards, the LAB color space of each standard color card number, and the number of complete leaves; The analysis unit analyzes the number of standard color cards, the LAB color space of each standard color card number and the number of complete leaves, and determines whether the image information quality is qualified according to the analysis results; and evaluates the growth status of crop germplasm resources by analyzing multiple pieces of qualified image information.

3. The method for evaluating crop germplasm resources based on image analysis according to claim 2, characterized in that: The analysis process of the analysis module is as follows: S10: randomly moving the image acquisition module to a specified height above any position of the test field to obtain the nth image information; S20: Identify and annotate the standard color cards and complete leaves in the image information through the recognition unit; obtain the number of standard color cards, the LAB color space of each standard color card number, the number of complete leaves, the LAB color space of each complete leaf, and the damage point of each complete leaf; S30: Analyze the number of standard color cards, the LAB color space of each standard color card, and the number of complete leaves, and evaluate whether the quality of the nth image information is qualified according to the analysis results; if qualified, proceed to step S40; if unqualified, the shooting parameters of the image acquisition module are automatically corrected, and step S10 is repeated; S40: Set n=n+1, repeat S10; until n>N, N is the preset number of image information, and enter step S50: S50: Evaluate the growth status of crop germplasm resources by analyzing N pieces of image information of qualified quality.

4. The method for evaluating crop germplasm resources based on image analysis according to claim 3, characterized in that: The evaluation process of whether the quality of the nth image information is qualified is through the formula: Where f(X) is the judgment function. When X>0, f(X)=1; when X≤0, f(X)=0; Q n is the number of standard color cards in the nth image information, q n is the qth standard color card in the nth image information; Q0 is the number of preset standard color cards; P n is the judgment index of the nth image information; is the color difference between the qth standard color card in the nth image information and the actual color of the standard color card; W1 is the first allowable error value; W2 is the second allowable error value; is the average color difference in the nth image information; S n is the number of complete leaves in the nth image information; S0 is the preset number of complete leaves.

5. The method for evaluating crop germplasm resources based on image analysis according to claim 4, characterized in that: The standard color card information includes the actual brightness value, the actual red-green axis value and the actual yellow-blue axis value of the standard color card.

6. The method for evaluating crop germplasm resources based on image analysis according to claim 5, characterized in that: The color difference between the qth standard color card in the nth image information and the actual color of the standard color card To pass the formula: in, is the brightness value of the qth standard color card in the nth image information; L0 is the actual brightness value of the standard color card; is the red-green axis value of the qth standard color card in the nth image information; A0 is the actual red-green axis value of the standard color card; is the yellow-blue axis value of the qth standard color card in the nth image information; B0 is the actual yellow-blue axis value of the standard color card.

7. The method for evaluating crop germplasm resources based on image analysis according to claim 6, characterized in that: The process of judging whether the quality of the nth image information is qualified also includes: When P n =1, the image information quality is unqualified; When P n When =e, the image information quality is qualified.

8. A crop germplasm resource evaluation system based on image analysis, applicable to the crop germplasm resource evaluation method based on image analysis according to any one of claims 1 to 7, characterized in that: The evaluation system comprises: Image acquisition module, used to monitor the growth status of germplasm resources and obtain image information of germplasm resources growth; Standard color card information module, used to store standard color card information; The analysis module is used to analyze the standard color card information and image information to determine whether the image information quality is qualified; and by analyzing multiple pieces of qualified image information, the growth status of crop germplasm resources is evaluated.

Images