Gardening flower maintenance method and system based on deep learning and storage medium

Abstract

The invention discloses a gardening flower maintenance method and system based on deep learning, and a storage medium, and relates to the technical field of gardening maintenance, and the method comprises the steps: obtaining gardening flower information, obtaining gardening flower state data based on flower state detection according to the gardening flower information, and obtaining gardening flower state data based on flower state evaluation according to the gardening flower state data. And judging whether the gardening flowers are abnormal in health or not. The flower leaf health abnormality index is obtained through the leaf area situation information and the leaf area surface feature information, the health condition of the flower leaf is accurately evaluated, the flower health abnormality index is obtained through the flower state data, the health abnormality of the flower is accurately evaluated, and the health abnormality of the flower is accurately evaluated according to the flower health abnormality index. According to the method, the situation that the gardening flowers cannot be maintained in time is avoided, the flower maintenance efficiency is improved, the flower health abnormality types are accurately recognized through the flower health abnormality difference coefficient, and the stability and efficiency of the gardening flowers are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of horticultural maintenance, and in particular to a horticultural flower maintenance method, system and storage medium based on deep learning. Background Art

[0002] 11With the development of urbanization, more and more people are pouring into cities, and the urban environment is deteriorating. The buildings everywhere make people living in cities feel bored. The rapid increase in urban population and the sharp decline in urban environmental quality have made people yearn for green rural life. People grow and cultivate flowers and plants in gardens, courtyards or other planting environments to decorate and beautify the environment, add to the mood of life, and to a certain extent improve their mood and quality of life.

[0003] At present, there are still problems in the maintenance of horticultural flowers, such as the inability to accurately evaluate the status of horticultural flowers based on the horticultural flower status data, the inability to avoid the impact of different growth stages of horticultural flowers on the health status assessment of horticultural flowers, and the inability to accurately identify abnormal flower health information based on abnormal horticultural flower health data, resulting in the inability to maintain flowers in time. Summary of the invention

[0004] In order to solve the above technical problems, a horticultural flower maintenance method, system and storage medium based on deep learning are provided. The technical solution solves the problems raised in the above background technology that the status of horticultural flowers cannot be accurately evaluated based on the horticultural flower status data, the influence of different growth stages of horticultural flowers on the health status assessment of horticultural flowers cannot be avoided, and the abnormal health information of flowers cannot be accurately identified based on the abnormal health data of horticultural flowers, resulting in the inability to maintain flowers in time.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] Gardening flower maintenance methods based on deep learning, including:

[0007] Acquire horticultural flower information, wherein the horticultural flower information includes horticultural flower species information and flower growth stage information, wherein the flower growth stage includes a flower development stage, a flower bud stage, and a flower blooming stage;

[0008] According to the horticultural flower information, based on the flower state detection, the horticultural flower state data is obtained, wherein the horticultural flower state data includes flower leaf state data and flower flower state data;

[0009] According to the horticultural flower status data, based on the flower status assessment, determine whether the horticultural flowers have health abnormalities, if not, record the horticultural flower status data, if yes, obtain horticultural flower health abnormality data according to the horticultural flower status data;

[0010] According to the horticultural flower type information, the flower historical health abnormality information is obtained, wherein the flower historical health abnormality information includes the flower historical health abnormality type information and the flower historical health abnormality data corresponding to each flower health abnormality;

[0011] According to the historical health abnormality information of flowers, a flower health abnormality recognition model is obtained;

[0012] According to the abnormal health data of horticultural flowers and the abnormal health recognition model of flowers, the abnormal health information of horticultural flowers is obtained;

[0013] Maintain horticultural flowers based on abnormal health information of horticultural flowers.

[0014] Preferably, judging whether the horticultural flowers have health abnormalities based on the horticultural flower status data and the flower status assessment specifically includes:

[0015] According to the horticultural flower status data, the flower leaf status data and the flower flower status data are obtained;

[0016] According to the flower leaf state data, obtaining flower leaf region state data, wherein the flower leaf region includes a leaf base region, a leaf edge region, a leaf tip region and a leaf middle region;

[0017] According to the flower leaf area status data, the leaf area situation information and the leaf area surface feature information are obtained;

[0018] Obtain the abnormal health index of flower leaves according to the leaf area situation information and the leaf area surface feature information;

[0019] Based on the maintenance needs of horticultural flowers, obtain the threshold of abnormal health index of flower leaves;

[0020] According to the flower leaf health abnormality index and the flower leaf health abnormality index threshold, determine whether the horticultural flower leaves have health abnormalities, and if so, obtain the flower leaf health abnormality data according to the leaf area situation information and the leaf area surface feature information;

[0021] Among them, the calculation formula of flower leaf health abnormality index is:

[0022]

[0023] Where Q is the abnormal health index of flower leaves, is the morphological change area of ​​the jth region of the i-th flower leaf, ∠A is the morphological change angle of the jth region of the i-th flower leaf, ω ij is the surface characteristic abnormal area of ​​the jth region of the i-th flower leaf, S ijis the total area of ​​the jth region of the i-th flower leaf, and n is the total number of leaves of horticultural flowers.

[0024] Preferably, judging whether the horticultural flower leaves have health abnormalities according to the flower leaf health abnormality index and the flower leaf health abnormality index threshold value further includes:

[0025] If the flower leaf health abnormality index is lower than the flower leaf health index threshold, the flower leaf has no health abnormality;

[0026] According to the flower status data, obtain the flower health abnormality index;

[0027] Based on the maintenance needs of horticultural flowers, obtain the threshold of abnormal health index of flowers and plants;

[0028] According to the flower health abnormality index and the flower health abnormality index threshold, it is judged whether the horticultural flower has health abnormality; if the flower health abnormality index is higher than the flower health abnormality index threshold, the horticultural flower has health abnormality;

[0029] According to the flower status data, obtain the abnormal health data of the flower;

[0030] According to the abnormal health data of flower flowers and the abnormal health data of flower leaves, the abnormal health data of horticultural flowers are obtained;

[0031] Among them, the calculation formula for the abnormal health index of flowers is:

[0032]

[0033] Where, E is the abnormal health index of flowers, f kz is the area of ​​the zth abnormal health feature of the kth flower, h z is the influence coefficient of the zth abnormal health characteristic of flowers, S k is the surface area of ​​the kth flower, a is the total number of flowers, a min is the minimum number of standard flowers, a max is the maximum number of standard flowers of flowers, and c is the total number of types of abnormal health characteristics of flowers.

[0034] Preferably, judging whether a horticultural flower has a health abnormality according to the flower health abnormality index and the flower health abnormality index threshold value further includes:

[0035] If the flower health abnormality index is lower than the flower health abnormality index threshold, the horticultural flower has no health abnormality;

[0036] According to the horticultural flower information, obtain the flower growth stage information;

[0037] According to the horticultural flower status data, obtain the flower root and stem status information;

[0038] Obtain the abnormal health index of flowers according to the abnormal health index of flowers and leaves, the abnormal health index of flowers and rhizomes, and the information of the growth stage of flowers;

[0039] Based on the maintenance requirements of horticultural flowers, obtain the threshold of abnormal flower health index;

[0040] According to the flower health abnormality index and the flower health abnormality index threshold, it is judged whether the horticultural flowers have health abnormalities. If the flower health abnormality index is lower than the flower health abnormality index threshold, the horticultural flowers have no health abnormalities.

[0041] If the flower health abnormality index is higher than the flower health abnormality index threshold, the horticultural flower has health abnormality, and the horticultural flower health abnormality data is obtained according to the flower health abnormality index;

[0042] The calculation formula of flower health abnormality index is:

[0043]

[0044] Wherein, R is the abnormal health index of flowers, μ is the abnormal coefficient of the rhizome of horticultural flowers, δ is the rhizome state index of horticultural flowers, δ0 is the standard state index of the rhizome of horticultural flowers, τ(α, β) represents the influence coefficient of the growth stage of flowers, wherein, if the growth stage of flowers is the development stage of flowers, then τ(α, β)=0; if the growth stage of flowers is the bud stage of flowers, then τ(α, β)=α; if the growth stage of flowers is the blooming stage of flowers, then τ(α, β)=β.

[0045] Preferably, the obtaining of horticultural flower health abnormality information based on horticultural flower health abnormality data and a flower health abnormality recognition model specifically includes:

[0046] According to the abnormal health data of horticultural flowers and the abnormal health recognition model of flowers, the abnormal health difference coefficient of flowers is obtained;

[0047] Based on the abnormal analysis of horticultural flower health, the threshold of the coefficient of difference of abnormal flower health is obtained;

[0048] According to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold, the horticultural flower health abnormality information is obtained, wherein if the flower health abnormality difference coefficient exceeds the flower health abnormality difference coefficient threshold, the flower has flower health abnormality;

[0049] Among them, the calculation formula for the coefficient of abnormal difference in flower health is:

[0050]

[0051] Where W(t) is the coefficient of difference of abnormal health of the tth abnormal flower, σ g is the characteristic coefficient of the gth health abnormality manifestation of the tth health abnormality, ρ i is the position coefficient of the i-th flower leaf, B gi is the characteristic area of ​​the gth health abnormality in the tth health abnormality of the i-th flower leaf, D ga is the characteristic area of ​​the g-th health abnormality of the t-th health abnormality in the i-th flower, ε(0,1) is the flower health abnormality compensation coefficient, wherein, if the flower health abnormality index is lower than the flower health abnormality index threshold and the flower health abnormality index is higher than the flower health abnormality index threshold, then ε(0,1)=1, if the flower leaf health abnormality index is higher than the flower leaf health index threshold or the flower flower health abnormality index is higher than the flower flower health abnormality index threshold, then ε(0,1)=0, G is the total number of characteristic types of the t-th health abnormality.

[0052] Furthermore, a horticultural flower maintenance system based on deep learning is proposed, which is used to implement the flower maintenance method as described above, including:

[0053] A main control module, wherein the main control module is used to determine whether a health abnormality occurs in a horticultural flower leaf according to a flower leaf health abnormality index and a flower leaf health abnormality index threshold, determine whether a horticultural flower has a health abnormality according to a flower flower health abnormality index and a flower flower health abnormality index threshold, determine whether a horticultural flower has a health abnormality according to a flower health abnormality index and a flower health abnormality index threshold, obtain a flower health abnormality difference coefficient according to horticultural flower health abnormality data and a flower health abnormality identification model, obtain horticultural flower health abnormality information according to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold, and maintain the horticultural flower according to the horticultural flower health abnormality information;

[0054] An information acquisition module, the information acquisition module is used to acquire horticultural flower information, horticultural flower species information and flower growth stage information, acquire horticultural flower status data based on the horticultural flower information and flower status detection, acquire flower historical health abnormality information based on the horticultural flower species information, acquire flower root and stem status information based on the horticultural flower status data, and transmit the information to the evaluation module;

[0055] An evaluation module, the evaluation module is used to obtain leaf area situation information and leaf area surface feature information according to flower leaf area status data, obtain flower leaf health abnormality index according to the leaf area situation information and the leaf area surface feature information, obtain flower flower health abnormality index according to flower flower status data, and obtain flower health abnormality index according to the flower flower health abnormality index, flower leaf health abnormality index, flower root and stem status information and flower growth stage information;

[0056] The display module interacts with the main control module and is used to display the abnormal health index of the flower flowers and the abnormal health index of the flower leaves, and to obtain the abnormal health index of the flower and the abnormal health information of the horticultural flowers.

[0057] Optionally, the main control module specifically includes:

[0058] A control unit, the control unit is used to obtain a flower health abnormality difference coefficient according to the horticultural flower health abnormality data and the flower health abnormality identification model, obtain horticultural flower health abnormality information according to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold, and maintain the horticultural flowers according to the horticultural flower health abnormality information;

[0059] An information receiving unit, which interacts with the information acquisition module and the evaluation module to receive data and transmit it to the judgment unit;

[0060] A judgment unit, wherein the judgment unit is used to judge whether the leaves of horticultural flowers have health abnormalities according to the flower leaf health abnormality index and the flower leaf health abnormality index threshold, judge whether the flowers of horticultural flowers have health abnormalities according to the flower flower health abnormality index and the flower flower health abnormality index threshold, and judge whether the horticultural flowers have health abnormalities according to the flower health abnormality index and the flower health abnormality index threshold.

[0061] Optionally, the information acquisition module specifically includes:

[0062] A first acquisition unit, the first acquisition unit is used to acquire horticultural flower information, horticultural flower species information and flower growth stage information, and acquire horticultural flower status data based on the horticultural flower information and flower status detection;

[0063] The second acquisition unit is used to acquire historical abnormal health information of flowers according to the type information of horticultural flowers, acquire root and stem status information of flowers according to the status data of horticultural flowers, and transmit the information to the evaluation module.

[0064] Optionally, the evaluation module specifically includes:

[0065] A flower leaf evaluation unit, the flower leaf evaluation unit is used to obtain leaf area situation information and leaf area surface feature information according to the flower leaf area status data, and obtain the flower leaf health abnormality index according to the leaf area situation information and the leaf area surface feature information;

[0066] A flower evaluation unit, the flower evaluation unit is used to obtain a flower health abnormality index according to the flower state data;

[0067] The flower abnormality assessment unit is used to obtain the flower health abnormality index according to the flower flower health abnormality index, the flower leaf health abnormality index, the flower root and stem status information and the flower growth stage information.

[0068] Furthermore, a computer-readable storage medium is proposed, on which a computer-readable program is stored, and when the computer-readable program is called, the flower maintenance method as described above is executed.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] The present invention proposes a horticultural flower maintenance method, system and storage medium based on deep learning. The flower leaf health abnormality index is obtained through leaf area situation information and leaf area surface feature information, so as to accurately evaluate the health status of the flower leaves. The flower flower health abnormality index is obtained through flower flower status data, and the health abnormality of the flower flower is accurately evaluated. According to the flower health abnormality index, the failure to maintain the horticultural flowers in time is avoided, and the flower maintenance efficiency is improved. The flower health abnormality difference coefficient is used to accurately identify the flower health abnormality type, thereby improving the stability and efficiency of the horticultural flowers. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 A flow chart of the horticultural flower maintenance method based on deep learning proposed by the present invention;

[0072] Figure 2 A flow chart for obtaining the abnormal health index of flower leaves in the present invention;

[0073] Figure 3 This is a flow chart for obtaining abnormal data on flower health in the present invention;

[0074] Figure 4 A flow chart for obtaining the abnormal health index of horticultural flowers in the present invention;

[0075] Figure 5 This is a structural block diagram of the horticultural flower maintenance system based on deep learning proposed in the present invention. DETAILED DESCRIPTION

[0076] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0077] Reference Figure 1-4 As shown, the horticultural flower maintenance method based on deep learning in the embodiment of the present invention includes:

[0078] S100: Acquire horticultural flower information, wherein the horticultural flower information includes horticultural flower species information and flower growth stage information, wherein the flower growth stage includes a flower development stage, a flower bud stage, and a flower blooming stage;

[0079] S200: Acquire horticultural flower status data based on flower status detection according to horticultural flower information, wherein the horticultural flower status data includes flower leaf status data and flower flower status data;

[0080] S300: judging whether the horticultural flowers have health abnormalities based on the horticultural flower status data and the flower status assessment, and if not, recording the horticultural flower status data; and if so, obtaining horticultural flower health abnormality data based on the horticultural flower status data;

[0081] Specifically, according to the horticultural flower status data and based on the flower status assessment, it is judged whether the horticultural flowers have health abnormalities, including:

[0082] According to the horticultural flower status data, the flower leaf status data and the flower flower status data are obtained;

[0083] According to the flower leaf state data, obtaining flower leaf region state data, wherein the flower leaf region includes a leaf base region, a leaf edge region, a leaf tip region and a leaf middle region;

[0084] According to the flower leaf area status data, the leaf area situation information and the leaf area surface feature information are obtained;

[0085] Obtain the abnormal health index of flower leaves according to the leaf area situation information and the leaf area surface feature information;

[0086] Based on the maintenance needs of horticultural flowers, obtain the threshold of abnormal health index of flower leaves;

[0087] According to the flower leaf health abnormality index and the flower leaf health abnormality index threshold, determine whether the horticultural flower leaves have health abnormalities, and if so, obtain the flower leaf health abnormality data according to the leaf area situation information and the leaf area surface feature information;

[0088] Among them, the calculation formula of flower leaf health abnormality index is:

[0089]

[0090] Where Q is the abnormal health index of flower leaves, is the morphological change area of ​​the jth region of the i-th flower leaf, ∠A is the morphological change angle of the jth region of the i-th flower leaf, ω ij is the surface characteristic abnormal area of ​​the jth region of the i-th flower leaf, S ij is the total area of ​​the jth region of the i-th flower leaf, and n is the total number of leaves of horticultural flowers.

[0091] In this scheme, the flower leaf state data is used to obtain the flower leaf area state data, and the leaf area situation information and the leaf area surface feature information are obtained according to the flower leaf area state data. The flower leaf health abnormality index is obtained according to the leaf area situation information and the leaf area surface feature information. According to the flower leaf health abnormality index and the flower leaf health abnormality index threshold, it is judged whether the horticultural flower leaves have health abnormalities, and the health status of the flower leaves is accurately evaluated to avoid the failure to maintain the horticultural flowers in time.

[0092] It is understandable that for flower leaves, different abnormal health manifestations, such as yellowing, withering or spotting of leaves, often spread from the edges, tips and bases of the leaves to the middle of the leaves. Abnormal health characteristics such as atrophy, edema or bending also often start from the edges of the leaves. By dividing the leaves into regions, it is easier to assess the health of the leaves, thereby improving the efficiency and accuracy of flower maintenance.

[0093] Specifically, judging whether the leaves of horticultural flowers have health abnormalities according to the flower leaf health abnormality index and the flower leaf health abnormality index threshold value also includes:

[0094] If the flower leaf health abnormality index is lower than the flower leaf health index threshold, the flower leaf has no health abnormality;

[0095] According to the flower status data, obtain the flower health abnormality index;

[0096] Based on the maintenance needs of horticultural flowers, obtain the threshold of abnormal health index of flowers and plants;

[0097] According to the flower health abnormality index and the flower health abnormality index threshold, it is judged whether the horticultural flower has health abnormality; if the flower health abnormality index is higher than the flower health abnormality index threshold, the horticultural flower has health abnormality;

[0098] According to the flower status data, obtain the abnormal health data of the flower;

[0099] According to the abnormal health data of flower flowers and the abnormal health data of flower leaves, the abnormal health data of horticultural flowers are obtained;

[0100] Among them, the calculation formula for the abnormal health index of flowers is:

[0101]

[0102] Where, E is the abnormal health index of flowers, f kz is the area of ​​the zth abnormal health feature of the kth flower, h z is the influence coefficient of the zth abnormal health characteristic of flowers, S k is the surface area of ​​the kth flower, a is the total number of flowers, a min is the minimum number of standard flowers, a max is the maximum number of standard flowers of flowers, and c is the total number of types of abnormal health characteristics of flowers.

[0103] In this scheme, the flower health abnormality index is obtained through the flower status data. According to the flower health abnormality index and the flower health abnormality index threshold, it is judged whether the horticultural flower has health abnormalities. According to the flower status data, the flower health abnormality data is obtained. According to the flower health abnormality data and the flower leaf health abnormality data, the horticultural flower health abnormality data is obtained. The flower surface abnormality characteristics and the number of flowers are used to analyze and evaluate the flowers, thereby improving the reliability of the data. It is understandable that some flower health abnormalities may only be discovered by observing the flowers. Therefore, when evaluating the health status of flowers, it is necessary to comprehensively consider the status of flowers and leaves. For example, when plants are in a drought or nutrient-deficient environment, the flowers may wither prematurely or wither abnormally, but the leaves may not show obvious yellowing or shrinkage. This is because plants will give priority to protecting the leaves to ensure the conduct of photosynthesis, and sacrificing flowers is an adaptive response. The health status of flowers is sometimes not directly reflected on the leaves, but this does not mean that the leaves are not affected. It is just that plants may have different coping strategies, which makes the leaves relatively stable while the flowers are affected. Therefore, a comprehensive consideration of the status of flowers and leaves can more comprehensively assess the health of flowers. It should be noted that, in this embodiment, if the horticultural flowers are in the bud stage, the health status of the buds is detected and recorded as flower status data.

[0104] Specifically, judging whether a horticultural flower has a health abnormality according to the flower health abnormality index and the flower health abnormality index threshold value also includes:

[0105] If the flower health abnormality index is lower than the flower health abnormality index threshold, the horticultural flower has no health abnormality;

[0106] According to the horticultural flower information, obtain the flower growth stage information;

[0107] According to the horticultural flower status data, obtain the flower root and stem status information;

[0108] Obtain the abnormal health index of flowers according to the abnormal health index of flowers and leaves, the abnormal health index of flowers and rhizomes, and the information of the growth stage of flowers;

[0109] Based on the maintenance requirements of horticultural flowers, obtain the threshold of abnormal flower health index;

[0110] According to the flower health abnormality index and the flower health abnormality index threshold, it is judged whether the horticultural flowers have health abnormalities. If the flower health abnormality index is lower than the flower health abnormality index threshold, the horticultural flowers have no health abnormalities.

[0111] If the flower health abnormality index is higher than the flower health abnormality index threshold, the horticultural flower has health abnormality, and the horticultural flower health abnormality data is obtained according to the flower health abnormality index;

[0112] The calculation formula of flower health abnormality index is:

[0113]

[0114] Wherein, R is the abnormal health index of flowers, μ is the abnormal coefficient of the rhizome of horticultural flowers, δ is the rhizome state index of horticultural flowers, δ0 is the standard state index of the rhizome of horticultural flowers, τ(α, β) represents the influence coefficient of the growth stage of flowers, wherein, if the growth stage of flowers is the development stage of flowers, then τ(α, β)=0; if the growth stage of flowers is the bud stage of flowers, then τ(α, β)=α; if the growth stage of flowers is the blooming stage of flowers, then τ(α, β)=β.

[0115] In this solution, the flower growth stage information is obtained through the horticultural flower information, the flower rhizome status information is obtained according to the horticultural flower status data, the flower health abnormality index is obtained according to the flower flower health abnormality index, the flower leaf health abnormality index, the flower rhizome status information and the flower growth stage information, and the flower health abnormality index and the flower health abnormality index threshold are used to determine whether the horticultural flowers have health abnormalities. It can be understood that in this embodiment, the flower leaves and flowers are monitored for health abnormalities through the threshold to ensure timely maintenance of the horticultural flowers, but it is inevitable that when the flower leaves and flowers are not identified as health abnormalities, the horticultural flowers may have certain health abnormalities, but the individual leaf health abnormalities and flower health abnormalities do not reach the health abnormality threshold. The flower health abnormality index is used to evaluate the overall health abnormality of the flowers to ensure timely maintenance of the horticultural flowers.

[0116] Among them, different proportions of the flower health abnormality index and the flower leaf health abnormality index are set according to the different growth stages of horticultural flowers, which improves the accuracy of flower health abnormality assessment.

[0117] S400: acquiring historical health abnormality information of flowers according to the type information of horticultural flowers, wherein the historical health abnormality information of flowers includes historical health abnormality type information of flowers and historical health abnormality data of flowers corresponding to each health abnormality of flowers;

[0118] S500: Obtaining a flower health abnormality recognition model according to the flower health abnormality historical information;

[0119] S600: Acquire horticultural flower health abnormality information according to horticultural flower health abnormality data and a flower health abnormality recognition model;

[0120] Specifically, according to the abnormal health data of horticultural flowers and the abnormal health recognition model of flowers, the abnormal health information of horticultural flowers is obtained, including:

[0121] According to the abnormal health data of horticultural flowers and the abnormal health recognition model of flowers, the abnormal health difference coefficient of flowers is obtained;

[0122] Based on the abnormal analysis of horticultural flower health, the threshold of the coefficient of difference of abnormal flower health is obtained;

[0123] According to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold, the horticultural flower health abnormality information is obtained, wherein if the flower health abnormality difference coefficient exceeds the flower health abnormality difference coefficient threshold, the flower has flower health abnormality;

[0124] Among them, the calculation formula for the coefficient of abnormal difference in flower health is:

[0125]

[0126] Where W(t) is the coefficient of difference of abnormal health of the tth abnormal flower, σ g is the characteristic coefficient of the gth health abnormality manifestation of the tth health abnormality, ρ i is the position coefficient of the i-th flower leaf, B gi is the characteristic area of ​​the gth health abnormality in the tth health abnormality of the i-th flower leaf, D gais the characteristic area of ​​the g-th health abnormality of the t-th health abnormality in the i-th flower, ε(0,1) is the flower health abnormality compensation coefficient, wherein, if the flower health abnormality index is lower than the flower health abnormality index threshold and the flower health abnormality index is higher than the flower health abnormality index threshold, then ε(0,1)=1, if the flower leaf health abnormality index is higher than the flower leaf health index threshold or the flower flower health abnormality index is higher than the flower flower health abnormality index threshold, then ε(0,1)=0, G is the total number of characteristic types of the t-th health abnormality.

[0127] In this scheme, the flower health abnormality difference coefficient is obtained through the horticultural flower health abnormality data and the flower health abnormality identification model, and the horticultural flower health abnormality information is obtained according to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold. The horticultural flowers are maintained through the horticultural flower health abnormality information, and the problems in the flower growth process are discovered and solved in time, so as to improve the timeliness and accuracy of maintenance, so as to meet the growth needs of flowers to the greatest extent and improve the maintenance efficiency;

[0128] It is understandable that it is more convenient and faster to directly analyze the health abnormality information of horticultural flowers according to the possible types of health abnormalities of horticultural flowers and obtain the causes of flower health abnormalities. However, it should be noted that when horticultural flowers have health abnormalities, they do not only show health abnormality characteristics related to the cause, but also have health abnormality characteristics caused by other complications. Directly identifying the health abnormalities of flowers cannot accurately identify the health abnormalities of horticultural flowers. By using the flower health abnormality difference coefficient, the errors caused by single health abnormality identification are avoided, and the efficiency of horticultural flower maintenance is improved.

[0129] S700: Maintain the horticultural flowers according to abnormal health information of the horticultural flowers.

[0130] Reference Figure 5 As shown, further, in combination with the above-mentioned horticultural flower maintenance method based on deep learning, a horticultural flower maintenance system based on deep learning is proposed, including:

[0131] A main control module, wherein the main control module is used to determine whether a health abnormality occurs in a horticultural flower leaf according to a flower leaf health abnormality index and a flower leaf health abnormality index threshold, determine whether a horticultural flower has a health abnormality according to a flower flower health abnormality index and a flower flower health abnormality index threshold, determine whether a horticultural flower has a health abnormality according to a flower health abnormality index and a flower health abnormality index threshold, obtain a flower health abnormality difference coefficient according to horticultural flower health abnormality data and a flower health abnormality identification model, obtain horticultural flower health abnormality information according to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold, and maintain the horticultural flower according to the horticultural flower health abnormality information;

[0132] An information acquisition module, the information acquisition module is used to acquire horticultural flower information, horticultural flower species information and flower growth stage information, acquire horticultural flower status data based on the horticultural flower information and flower status detection, acquire flower historical health abnormality information based on the horticultural flower species information, acquire flower root and stem status information based on the horticultural flower status data, and transmit the information to the evaluation module;

[0133] An evaluation module, the evaluation module is used to obtain leaf area situation information and leaf area surface feature information according to flower leaf area status data, obtain flower leaf health abnormality index according to the leaf area situation information and the leaf area surface feature information, obtain flower flower health abnormality index according to flower flower status data, and obtain flower health abnormality index according to the flower flower health abnormality index, flower leaf health abnormality index, flower root and stem status information and flower growth stage information;

[0134] The display module interacts with the main control module and is used to display the abnormal health index of the flower flowers and the abnormal health index of the flower leaves, and to obtain the abnormal health index of the flower and the abnormal health information of the horticultural flowers.

[0135] Main control module, specifically including:

[0136] A control unit, the control unit is used to obtain a flower health abnormality difference coefficient according to the horticultural flower health abnormality data and the flower health abnormality identification model, obtain horticultural flower health abnormality information according to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold, and maintain the horticultural flowers according to the horticultural flower health abnormality information;

[0137] An information receiving unit, which interacts with the information acquisition module and the evaluation module to receive data and transmit it to the judgment unit;

[0138] A judgment unit, wherein the judgment unit is used to judge whether the leaves of horticultural flowers have health abnormalities according to the flower leaf health abnormality index and the flower leaf health abnormality index threshold, judge whether the flowers of horticultural flowers have health abnormalities according to the flower flower health abnormality index and the flower flower health abnormality index threshold, and judge whether the horticultural flowers have health abnormalities according to the flower health abnormality index and the flower health abnormality index threshold.

[0139] Information acquisition module, specifically including:

[0140] A first acquisition unit, the first acquisition unit is used to acquire horticultural flower information, horticultural flower species information and flower growth stage information, and acquire horticultural flower status data based on the horticultural flower information and flower status detection;

[0141] The second acquisition unit is used to acquire historical abnormal health information of flowers according to the type information of horticultural flowers, acquire root and stem status information of flowers according to the status data of horticultural flowers, and transmit the information to the evaluation module.

[0142] Assessment modules include:

[0143] A flower leaf evaluation unit, the flower leaf evaluation unit is used to obtain leaf area situation information and leaf area surface feature information according to the flower leaf area status data, and obtain the flower leaf health abnormality index according to the leaf area situation information and the leaf area surface feature information;

[0144] A flower evaluation unit, the flower evaluation unit is used to obtain a flower health abnormality index according to the flower state data;

[0145] The flower abnormality assessment unit is used to obtain the flower health abnormality index according to the flower flower health abnormality index, the flower leaf health abnormality index, the flower root and stem status information and the flower growth stage information.

[0146] Furthermore, the present solution also proposes a computer-readable storage medium, on which a computer-readable program is stored, and when the computer-readable program is called, the flower maintenance method as described above is executed;

[0147] It is understandable that the computer-readable storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state disk (SSD).

[0148] In summary, the advantages of the present invention are: through horticultural flower information, based on flower status detection, horticultural flower status data is obtained, through leaf area situation information and leaf area surface feature information, the flower leaf health abnormality index is obtained, the health status of flower leaves is accurately evaluated through the flower leaf health abnormality index, through flower flower status data, the flower flower health abnormality index is obtained, and the health abnormality of flower flowers is accurately evaluated, and the flower health abnormality index is obtained according to the flower flower health abnormality index, the flower leaf health abnormality index, the flower root and stem status information and the flower growth stage information, so as to avoid the failure to maintain horticultural flowers in time and improve the flower maintenance efficiency, and the flower health abnormality difference coefficient is obtained through the horticultural flower health abnormality data and the flower health abnormality recognition model, so as to improve the stability and efficiency of horticultural flowers.

[0149] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A horticultural flower maintenance method based on deep learning, characterized in that: include: Acquire horticultural flower information, wherein the horticultural flower information includes horticultural flower species information and flower growth stage information, wherein the flower growth stage includes a flower development stage, a flower bud stage, and a flower blooming stage; According to the horticultural flower information, based on the flower state detection, the horticultural flower state data is obtained, wherein the horticultural flower state data includes flower leaf state data and flower flower state data; According to the horticultural flower status data, based on the flower status assessment, determine whether the horticultural flowers have health abnormalities, if not, record the horticultural flower status data, if yes, obtain horticultural flower health abnormality data according to the horticultural flower status data; According to the horticultural flower type information, the flower historical health abnormality information is obtained, wherein the flower historical health abnormality information includes the flower historical health abnormality type information and the flower historical health abnormality data corresponding to each flower health abnormality; According to the historical health abnormality information of flowers, a flower health abnormality recognition model is obtained; According to the abnormal health data of horticultural flowers and the abnormal health recognition model of flowers, the abnormal health information of horticultural flowers is obtained; Maintain horticultural flowers based on abnormal health information of horticultural flowers.

2. The horticultural flower maintenance method based on deep learning according to claim 1, characterized in that: The step of judging whether the horticultural flowers have health abnormalities based on the horticultural flower status data and the flower status assessment specifically includes: According to the horticultural flower status data, the flower leaf status data and the flower flower status data are obtained; According to the flower leaf state data, obtaining flower leaf region state data, wherein the flower leaf region includes a leaf base region, a leaf edge region, a leaf tip region and a leaf middle region; According to the flower leaf area status data, the leaf area situation information and the leaf area surface feature information are obtained; Obtain the abnormal health index of flower leaves according to the leaf area situation information and the leaf area surface feature information; Based on the maintenance needs of horticultural flowers, obtain the threshold of abnormal health index of flower leaves; According to the flower leaf health abnormality index and the flower leaf health abnormality index threshold, determine whether the horticultural flower leaves have health abnormalities, and if so, obtain the flower leaf health abnormality data according to the leaf area situation information and the leaf area surface feature information; Among them, the calculation formula of flower leaf health abnormality index is: Where Q is the abnormal health index of flower leaves, is the morphological change area of ​​the jth region of the i-th flower leaf, ∠A is the morphological change angle of the jth region of the i-th flower leaf, ω ij is the surface characteristic abnormal area of ​​the jth region of the i-th flower leaf, S ij is the total area of ​​the jth region of the i-th flower leaf, and n is the total number of leaves of horticultural flowers.

3. The horticultural flower maintenance method based on deep learning according to claim 2, characterized in that: The step of judging whether the horticultural flower leaves have health abnormalities according to the flower leaf health abnormality index and the flower leaf health abnormality index threshold value further includes: If the flower leaf health abnormality index is lower than the flower leaf health index threshold, the flower leaf has no health abnormality; According to the flower status data, obtain the flower health abnormality index; Based on the maintenance needs of horticultural flowers, obtain the threshold of abnormal health index of flowers and plants; According to the flower health abnormality index and the flower health abnormality index threshold, it is judged whether the horticultural flower has health abnormality; if the flower health abnormality index is higher than the flower health abnormality index threshold, the horticultural flower has health abnormality; According to the flower status data, obtain the abnormal health data of the flower; According to the abnormal health data of flower flowers and the abnormal health data of flower leaves, the abnormal health data of horticultural flowers are obtained; Among them, the calculation formula for the abnormal health index of flowers is: Where, E is the abnormal health index of flowers, f kz is the area of ​​the zth abnormal health feature of the kth flower, h z is the influence coefficient of the zth abnormal health characteristic of flowers, S k is the surface area of ​​the kth flower, a is the total number of flowers, a min is the minimum number of standard flowers, a max is the maximum number of standard flowers of flowers, and c is the total number of types of abnormal health characteristics of flowers.

4. The horticultural flower maintenance method based on deep learning according to claim 3, characterized in that: The step of judging whether a horticultural flower has a health abnormality according to the flower health abnormality index and the flower health abnormality index threshold may further include: If the flower health abnormality index is lower than the flower health abnormality index threshold, the horticultural flower has no health abnormality; According to the horticultural flower information, obtain the flower growth stage information; According to the horticultural flower status data, obtain the flower root and stem status information; Obtain the abnormal health index of flowers according to the abnormal health index of flowers and leaves, the abnormal health index of flowers and rhizomes, and the information of the growth stage of flowers; Based on the maintenance requirements of horticultural flowers, obtain the threshold of abnormal flower health index; According to the flower health abnormality index and the flower health abnormality index threshold, it is judged whether the horticultural flowers have health abnormalities. If the flower health abnormality index is lower than the flower health abnormality index threshold, the horticultural flowers have no health abnormalities. If the flower health abnormality index is higher than the flower health abnormality index threshold, the horticultural flower has health abnormality, and the horticultural flower health abnormality data is obtained according to the flower health abnormality index; The calculation formula of flower health abnormality index is: Wherein, R is the abnormal health index of flowers, μ is the abnormal coefficient of the rhizome of horticultural flowers, δ is the rhizome state index of horticultural flowers, δ0 is the standard state index of the rhizome of horticultural flowers, τ(α, β) represents the influence coefficient of the growth stage of flowers, wherein, if the growth stage of flowers is the development stage of flowers, then τ(α, β)=0; if the growth stage of flowers is the bud stage of flowers, then τ(α, β)=α; if the growth stage of flowers is the blooming stage of flowers, then τ(α, β)=β.

5. The horticultural flower maintenance method based on deep learning according to claim 1, characterized in that: The step of obtaining the abnormal health information of horticultural flowers based on the abnormal health data of horticultural flowers and the abnormal health recognition model of flowers specifically includes: According to the abnormal health data of horticultural flowers and the abnormal health recognition model of flowers, the abnormal health difference coefficient of flowers is obtained; Based on the abnormal analysis of horticultural flower health, the threshold of the coefficient of difference of abnormal flower health is obtained; According to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold, the horticultural flower health abnormality information is obtained, wherein if the flower health abnormality difference coefficient exceeds the flower health abnormality difference coefficient threshold, the flower has flower health abnormality; Among them, the calculation formula for the coefficient of abnormal difference in flower health is: Where W(t) is the coefficient of difference of abnormal health of the tth abnormal flower, σ g is the characteristic coefficient of the gth health abnormality manifestation of the tth health abnormality, ρ i is the position coefficient of the i-th flower leaf, B gi is the characteristic area of ​​the gth health abnormality in the tth health abnormality of the i-th flower leaf, D ga is the characteristic area of ​​the g-th health abnormality of the t-th health abnormality in the i-th flower, ε(0,1) is the flower health abnormality compensation coefficient, wherein, if the flower health abnormality index is lower than the flower health abnormality index threshold and the flower health abnormality index is higher than the flower health abnormality index threshold, then ε(0,1)=1, if the flower leaf health abnormality index is higher than the flower leaf health index threshold or the flower flower health abnormality index is higher than the flower flower health abnormality index threshold, then ε(0,1)=0, G is the total number of characteristic types of the t-th health abnormality.

6. A horticultural flower maintenance system based on deep learning, used to implement the flower maintenance method according to any one of claims 1 to 5, characterized in that: include: A main control module, wherein the main control module is used to determine whether a health abnormality occurs in a horticultural flower leaf according to a flower leaf health abnormality index and a flower leaf health abnormality index threshold, determine whether a horticultural flower has a health abnormality according to a flower flower health abnormality index and a flower flower health abnormality index threshold, determine whether a horticultural flower has a health abnormality according to a flower health abnormality index and a flower health abnormality index threshold, obtain a flower health abnormality difference coefficient according to horticultural flower health abnormality data and a flower health abnormality identification model, obtain horticultural flower health abnormality information according to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold, and maintain the horticultural flower according to the horticultural flower health abnormality information; An information acquisition module, the information acquisition module is used to acquire horticultural flower information, horticultural flower species information and flower growth stage information, acquire horticultural flower status data based on the horticultural flower information and flower status detection, acquire flower historical health abnormality information based on the horticultural flower species information, acquire flower root and stem status information based on the horticultural flower status data, and transmit the information to the evaluation module; An evaluation module, the evaluation module is used to obtain leaf area situation information and leaf area surface feature information according to flower leaf area status data, obtain flower leaf health abnormality index according to the leaf area situation information and the leaf area surface feature information, obtain flower flower health abnormality index according to flower flower status data, and obtain flower health abnormality index according to the flower flower health abnormality index, flower leaf health abnormality index, flower root and stem status information and flower growth stage information; The display module interacts with the main control module and is used to display the abnormal health index of the flower flowers and the abnormal health index of the flower leaves, and to obtain the abnormal health index of the flower and the abnormal health information of the horticultural flowers.

7. The gardening flower maintenance system based on deep learning according to claim 6, characterized in that: The main control module specifically includes: A control unit, the control unit is used to obtain a flower health abnormality difference coefficient according to the horticultural flower health abnormality data and the flower health abnormality identification model, obtain horticultural flower health abnormality information according to the flower health abnormality difference coefficient and the flower health abnormality difference coefficient threshold, and maintain the horticultural flowers according to the horticultural flower health abnormality information; An information receiving unit, which interacts with the information acquisition module and the evaluation module to receive data and transmit it to the judgment unit; A judgment unit, wherein the judgment unit is used to judge whether the leaves of horticultural flowers have health abnormalities according to the flower leaf health abnormality index and the flower leaf health abnormality index threshold, judge whether the flowers of horticultural flowers have health abnormalities according to the flower flower health abnormality index and the flower flower health abnormality index threshold, and judge whether the horticultural flowers have health abnormalities according to the flower health abnormality index and the flower health abnormality index threshold.

8. The gardening flower maintenance system based on deep learning according to claim 6, characterized in that: The information acquisition module specifically includes: A first acquisition unit, the first acquisition unit is used to acquire horticultural flower information, horticultural flower species information and flower growth stage information, and acquire horticultural flower status data based on the horticultural flower information and flower status detection; The second acquisition unit is used to acquire historical abnormal health information of flowers according to the type information of horticultural flowers, acquire root and stem status information of flowers according to the status data of horticultural flowers, and transmit the information to the evaluation module.

9. The gardening flower maintenance system based on deep learning according to claim 6, characterized in that: The evaluation module specifically includes: A flower leaf evaluation unit, the flower leaf evaluation unit is used to obtain leaf area situation information and leaf area surface feature information according to the flower leaf area status data, and obtain the flower leaf health abnormality index according to the leaf area situation information and the leaf area surface feature information; A flower evaluation unit, the flower evaluation unit is used to obtain a flower health abnormality index according to the flower state data; The flower abnormality assessment unit is used to obtain the flower health abnormality index according to the flower flower health abnormality index, the flower leaf health abnormality index, the flower root and stem status information and the flower growth stage information.

10. A computer-readable storage medium having a computer-readable program stored thereon, characterized in that: When the computer-readable program is called, the flower maintenance method according to any one of claims 1 to 5 is executed.

Images