AI-based expressway greening engineering disease intelligent identification, management and maintenance system

Abstract

The invention relates to an AI-based highway greening engineering disease intelligent identification and maintenance management system, which mainly comprises a central control system, AI intelligent identification software, machine learning algorithm software, mobile equipment, a digital camera, maintenance construction equipment and an engineering acceptance system, intelligent management and control of the whole process of highway greening engineering from disease discovery to maintenance construction acceptance are achieved, and the maintenance construction efficiency and the maintenance engineering quality are improved. The central control system is used for controlling expressway greening engineering image acquisition, greening engineering disease AI identification software to determine the type and area of the expressway greening engineering, and generating a maintenance task notice; the maintenance construction unit receives the task book and then carries out maintenance construction, and the supervision unit accepts the maintenance project quality and the project number. AI identification software and machine learning algorithm software for various diseases of the highway greening engineering are researched and developed, and the functions of automatically determining the types and areas of the diseases in the greening engineering, automatically generating the maintenance engineering amount and the engineering cost and the like are achieved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of highways, and in particular to an AI-based intelligent identification and maintenance system for highway greening project diseases. Background Art

[0002] The greening project of expressways plays an important role in beautifying the traffic environment, relieving fatigue of drivers and passengers, and preventing the night lights of oncoming vehicles from affecting the driver's vision. Due to the high speed and large traffic flow of expressways, once the driver's vision is affected, it will inevitably affect the driving behavior and endanger traffic safety. Therefore, timely, scientific and reasonable maintenance is urgently needed. Timely and rapid discovery of hidden dangers of diseases and timely maintenance can effectively ensure the traffic safety of expressways and the safety of life and property of the people. At present, in the relevant technologies of expressway inspection, it mainly relies on manual inspection of greening trees, finding trees with diseases, and determining the type and area of ​​the diseases. Not only is the work efficiency low and the accuracy low, but there is also a large safety risk in the inspection process. The present invention uses a car or a drone to inspect the diseases of expressway greening projects, uses AI recognition software for greening project diseases and machine learning algorithm software for disease identification, determines the type and area of ​​diseases, and issues maintenance construction instructions to the maintenance construction unit through the central control system, and the maintenance construction unit performs maintenance construction to eliminate safety hazards in time. The present invention can significantly improve the inspection efficiency and the accuracy of disease identification, and effectively reduce safety risks. Summary of the invention

[0003] In order to solve the problems existing in the current related technologies, the present invention provides an AI-based intelligent identification and maintenance system for highway greening project diseases.

[0004] According to an embodiment of the present disclosure, there is provided an AI-based highway greening project disease intelligent identification and management system, including: a mobile device, a digital camera, a central control system, greening project disease AI identification software and machine learning algorithm software;

[0005] The mobile device is a vehicle or a drone; the digital camera is arranged on the mobile device and is used to collect digital images of the highway greening project;

[0006] The central control system is used to control the mobile device to move along the target road, and adjust the movement trajectory and direction of the mobile device in real time according to the inspection results, so as to carry out more targeted data collection; at the same time, the central control system is also used to control the camera to collect images, call the AI ​​recognition software and the machine learning algorithm software to perform image analysis and processing and disease identification, so as to determine the type of disease and the area where it occurs, optimize the maintenance plan, issue maintenance construction instructions, calculate the maintenance project volume and project costs, and other procedures, so as to realize the management and control of the whole process of greening project maintenance.

[0007] The digital camera is used to photograph the greening project on the target road and obtain the inspection digital image of the greening project;

[0008] The AI ​​recognition software for greening project diseases and the machine learning algorithm software are used to analyze and process the inspection digital images to automatically determine the type and area of ​​the diseases in the greening project;

[0009] The machine learning algorithm software is used to provide samples of trees in normal state to the AI ​​recognition software of the greening project diseases, so as to determine the type and area of ​​the diseases in the greening project;

[0010] The greening project includes: greening projects in the central dividing strip of expressways and on the sides of roads.

[0011] The technical solution provided by the embodiment of the present disclosure may include the following beneficial effects: The AI-based intelligent identification and maintenance system for diseases of highway greening projects of the present application can replace manual work through mobile devices, which can be vehicles or drones, and can significantly improve the inspection efficiency. By setting a digital camera on a mobile device (including vehicles and drones), the digital camera digitally photographs the greening project on the target road and collects the inspection digital image of the greening project. The central control system receives the inspection digital image of the greening project taken by the digital camera; the inspection digital image is analyzed by the greening project disease digital image AI recognition software and machine learning algorithm software to determine the type and area of ​​the disease in the greening project, list the disease list, and select targeted maintenance plans. The central control system pushes the disease list and maintenance plan to the construction unit and the supervision unit, and issues maintenance construction instructions and supervision instructions; the maintenance construction unit performs maintenance construction according to the maintenance plan, and the supervision unit inspects and accepts the quality of the maintenance construction, and verifies the engineering quantity and maintenance project cost of the maintenance construction. The present invention discloses a management and control method for the entire process of highway greening project defect inspection, AI intelligent identification of defects, issuance of maintenance task orders, maintenance construction by construction units, maintenance quality inspection and acceptance, and calculation of maintenance project quantities and maintenance project costs, which can significantly improve the efficiency and accuracy of greening project maintenance construction management.

[0012] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0014] Figure 1 is a schematic diagram of a highway greening project disease intelligent identification and management system according to an exemplary embodiment;

[0015] Figure 2 is a structural block diagram of a whole process according to an exemplary embodiment;

[0016] Figure 3 is a schematic diagram of a leaf according to an exemplary embodiment;

[0017] Figure 4 The figure is a schematic diagram showing a curling leaf according to an exemplary embodiment. DETAILED DESCRIPTION

[0018] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0019] This application proposes an intelligent identification and maintenance system for highway greening project diseases based on artificial intelligence (AI). Figure 1 The structure diagram of a highway greening project disease intelligent identification and management system 1 based on AI is shown; the system includes: a mobile device 4, a digital camera 3 and a central control system 2, greening project disease AI identification software and machine learning algorithm software; the digital camera 3 and the central control system 2 are arranged on the mobile device 4. The greening project disease AI identification software and machine learning algorithm software are arranged in the central control system 2.

[0020] The central control system 2 includes a processor 21 and a memory 22 .

[0021] Specifically, the memory 22 is used to store the independently developed AI recognition software and machine learning algorithm software for the above-mentioned greening project diseases.

[0022] The processor 21 controls and runs the above-mentioned AI recognition software and machine learning algorithm software for greening project diseases.

[0023] The mobile device 4 is a vehicle or a drone.

[0024] The central control system 2 is used to control the mobile device 4 to move along the target road, and adjust the movement trajectory and movement direction of the mobile device 4 in real time according to the detection results to obtain more targeted detection. At the same time, it is also used to control the greening project image acquisition, disease identification, determination of disease type and area, formulation of targeted maintenance plans, calculation of maintenance project volume and project costs and other maintenance work procedures and methods.

[0025] In this embodiment, the processor 21 of the central control system 2 includes a motion control module 211; the motion control module 211 is used to control the mobile device 4 to move along the target road, and adjust the motion track and motion direction of the mobile device 4 in real time according to the detection results to perform more targeted detection. For example, the motion control module 211 can control the specific movement of a vehicle or a drone.

[0026] The digital camera 3 is used to photograph the greening project on the target road and obtain the inspection digital image of the greening project.

[0027] The central control system 2 is used to receive the inspection digital image of the greening project taken by the digital camera, and use the AI ​​recognition software for greening project diseases and the machine learning algorithm software to analyze the inspection digital image to determine the type and area of ​​the disease in the greening project.

[0028] Among them, the machine learning algorithm software is trained through repeated learning of the normal state of highway greening projects, mastering the shape, color and chromaticity of leaves, and storing them as samples, providing standardized samples of greening trees in normal state for AI recognition software of various diseases, so as to determine the type and area of ​​diseases in the greening projects.

[0029] The leaf morphology refers to the leaf's appearance, shape, geometric dimensions and natural extension state.

[0030] The AI ​​intelligent recognition software for greening project diseases analyzes and processes the inspection digital images, identifies the shape, color and chromaticity of leaves through digital images, and measures the leaf size through pixels.

[0031] The dimensions include length, width and leaf spacing parameters, and the type and area of ​​​​the greening project disease are determined by comparing them with the images of normal leaves stored in the machine learning algorithm.

[0032] Specifically, the leaf abnormality judgment module 212 is used to analyze the inspection digital image to determine whether the leaves are in an abnormal state, and then determine the target trees and their locations of the disease, thereby determining the trees and areas of the disease.

[0033] The sending module 213 is used to send construction instructions to the maintenance construction unit, and the maintenance construction unit performs greening engineering maintenance construction on the diseased trees.

[0034] The greening project includes: greening projects on the central dividing strip and the roadside.

[0035] The above-mentioned technical solution of the present application is that in the inspection system of the highway greening project, a mobile device is provided, and the mobile device is a vehicle or a drone. It can move quickly. A digital camera and a central control system are also provided on the mobile device. The central control system receives the inspection digital image of the greening project taken by the digital camera, and controls the identification of greening project diseases, determines the type and area of ​​the disease, automatically optimizes the maintenance plan, calculates the maintenance project volume and the maintenance project cost. The greening project disease AI recognition software and machine learning algorithm software are used to analyze the inspection digital image to determine the type and area of ​​the disease in the greening project. The disease types include: leaf shrinkage, yellowing of leaves, sparse branches and leaves, or invasive branches and leaves. Compared with the prior art, which uses artificial naked eye judgment to identify disease warnings, the present disclosure significantly improves the inspection efficiency and accuracy.

[0036] See attached Figure 2 In some embodiments, the central control system is used to control the inspection vehicle or inspection drone to collect digital images of greening projects, perform AI intelligent identification of diseases, automatically generate maintenance plans, issue maintenance construction instructions, calculate the number of maintenance projects and project costs, and manage the entire maintenance process and data flow.

[0037] In some embodiments, a machine learning algorithm software for highway greening projects is developed. Through repeated learning and training on the normal status of highway greening projects, the shape, color and chromaticity of leaves are mastered and stored as samples to provide standard samples for AI intelligent recognition software of various diseases to determine the type and area of ​​diseases in the greening project.

[0038] In some embodiments, AI intelligent recognition software for greening project diseases is developed, which can analyze and process the inspection digital images of the greening project, including identifying the morphology, color and chromaticity, and leaf spacing of the leaves of the greening project, identifying leaves in abnormal states of shrinkage, curling, yellowing or discoloration, as well as sparse branches and leaves and branch growth limiting diseases, and recording the location and area where they occur;

[0039] Wherein, the position includes a mileage post position or a Beidou coordinate position;

[0040] Statistics are collected from the inspection digital images of the greening project to determine the total number of trees with shrunken leaves, yellowing leaves, sparse branches and leaves, and branches and leaves infected with diseases, as well as the geographical locations where the diseases occurred.

[0041] In some embodiments, targeted maintenance plans are adopted for different types of diseases, including watering for trees with shrunken leaves; fertilizing for trees with yellowing leaves; watering and fertilizing for trees with sparse leaves; and pruning and trimming for trees with limited branches and leaves.

[0042] In some embodiments, a maintenance construction task list is issued to the maintenance construction unit, and the maintenance construction task list includes: the location where the disease of the greening project occurs, the type and area of ​​the disease, the maintenance plan to be adopted, and the estimated maintenance project volume and project cost;

[0043] The location includes a mileage post number or Beidou coordinates.

[0044] In some embodiments, the central control system is also used for the maintenance supervision unit to issue supervision instructions and maintenance construction task orders to the maintenance construction unit, conduct on-site supervision of construction work, inspect and accept the quality of maintenance projects, verify the maintenance project volume and project costs on site, and sign for confirmation; if the maintenance standards are not met, record the existing quality problems and their locations, issue rework or improvement supervision instructions to the maintenance construction unit, and supervise the construction quality of the maintenance construction unit.

[0045] In some embodiments, the central control system is also used to calculate a maintenance project cost list according to the completed greening project maintenance work volume and the approved unit price, and to calculate the overall project cost.

[0046] See attached Figure 2 In the highway greening project disease AI intelligent recognition, the disease type and area are identified. After image acquisition, the disease is identified based on the image. Image acquisition is performed during the intelligent inspection of the highway greening project. Image acquisition is performed under the control of the central control system.

[0047] The maintenance plan includes: watering, fertilizing, fertilizing and filling soil, pruning, and pruning.

[0048] Among them, for page shrinkage and page yellowing, the maintenance plan of watering and fertilizing is adopted.

[0049] For plants with sparse branches and leaves, adopt a maintenance plan of fertilizing and filling the soil, and calculate the cost.

[0050] For foliage encroachment, adopt a maintenance plan of pruning and pruning, and calculate the cost.

[0051] The calculation of maintenance work volume includes the calculation of watering amount, fertilization amount, fertilization and soil filling amount, pruning and pruning amount, and cost calculation.

[0052] The project cost calculation includes the calculation of watering costs, fertilization costs, fertilization and soil filling costs, pruning and trimming costs, and other costs.

[0053] In some embodiments, the central control system is further used to analyze and process the digital images of leaves and branches of the greening project based on the inspection digital images of the greening project using AI recognition software for greening project diseases, and determine the trees with diseases so that the maintenance construction unit can organize maintenance construction in a timely manner;

[0054] Through the AI ​​recognition software for greening project diseases, the number and area of ​​trees with abnormal conditions in the inspection digital images of the greening project can be automatically found, and the total number of trees that need maintenance can be determined accordingly.

[0055] For example, to determine whether the leaves are in an abnormal state, the color factor can be considered. In some embodiments, the central control system is also used to determine the color of the edge of each leaf using the AI ​​recognition software for greening project diseases. If the color of the edge is yellow, the leaf is in an abnormal color state. The possible reason is drought, lack of water, or lack of fertilizer. At this time, watering and fertilizing are needed for maintenance.

[0056] Specifically, the area of ​​the yellow area in the edge area can be further counted, and the ratio of the area of ​​the yellow area to the area of ​​the leaf can be calculated. If the ratio is greater than a predetermined area ratio threshold, it is determined that the leaf is in an extremely abnormal state.

[0057] For example, the area ratio threshold is 0.5, and the ratio of the area of ​​the yellow area to the area of ​​the leaves is 0.6. If it is greater than 0.5, it can be determined that the leaves are in an extremely abnormal state and need to be maintained in time, including watering, fertilizing and soiling.

[0058] In addition to the color factors mentioned above, size factors can also be considered. In some embodiments, the central control system is also used to determine the size and shape of each leaf using the AI ​​recognition software for greening project diseases. If the size of the leaf is reduced and the shape curls, the leaf is determined to be in an abnormal state.

[0059] The abnormal state of the leaves shrinking in size and curling in shape is one of the abnormal states. By judging whether the leaves are in the abnormal curling state, it can be determined whether the tree has begun to develop a disease. Because when the weather is dry, the trees lack water, which often causes the leaves to be in a curled or wrinkled state, and the size of the leaves is displayed in the digital image.

[0060] In some embodiments, it may also include the situation of weeding. In greening projects, other types of trees may be mixed among the specified tree species. In order to remove the weeds, it is necessary to determine the type of tree based on the leaves.

[0061] AI intelligent identification software for greening engineering diseases can identify the type of tree to which the current leaves belong. Machine learning algorithm software can determine the type of specified trees in samples under normal conditions. By comparing the two types, impurity removal can be performed.

[0062] For example, the tree species specified in the sample under normal conditions is Sophora japonica, and among the many Sophora japonica trees, there is a poplar. Because the leaves of the poplar and the Sophora japonica are quite different in appearance, the AI ​​intelligent recognition software for greening engineering diseases can distinguish the two species. Therefore, the poplar can be removed to play a role in removing impurities.

[0063] See attached Figure 3 and Figure 4 In order to determine whether the leaves are in an abnormal curling state, in some embodiments, the central control system is also used to determine the width dimension L1 of the outline of each leaf using the machine learning algorithm software.

[0064] exist Figure 3 In FIG. 1 , the shape and size of the leaves in a normal condition without any disease are shown. The size of the leaves in a normal condition without any disease can be obtained by learning the size of a large number of leaves and obtaining a machine learning algorithm as a reference comparison standard.

[0065] The above-mentioned AI recognition software for greening engineering diseases is used to determine the edge position of each leaf; and the actual distance value L2 of the leaf edge is calculated based on the edge position of each leaf.

[0066] exist Figure 4In the figure, the edge of the leaf is shown to be curled inward when the leaf is actually curled. As shown in the figure, the edge 42 in the curled state is obviously contracted inward compared with the edge 41 in the normal state. The degree of curling is determined by the contraction distance.

[0067] The AI ​​recognition software for greening engineering diseases can be used to determine the actual distance value L2 of the leaf edge. If the difference between the width dimension value L1 and the actual distance value L2 is greater than a predetermined difference threshold, the leaf is determined to be in a suspected curling abnormal state. This indicates that the tree lacks water or fertilizer, causing the leaves to shrink and curl. The tree needs maintenance.

[0068] In some embodiments, in order to improve the accuracy of detection, multiple image frames may be collected, and in each image frame, the number of leaves in a curled state may be determined, and the total number of leaves in the image frame may be calculated;

[0069] Calculate the ratio of the number of leaves in a curled state to the total number of leaves;

[0070] If the ratio is greater than or equal to a predetermined leaf quantity ratio threshold, the image frame is determined to be a curled image frame.

[0071] If the ratio is less than a predetermined leaf quantity ratio threshold, the image frame is determined to be a normal image frame.

[0072] Here is an example to illustrate:

[0073] Assuming that in image frame A, the total number of leaves is 10, and the number of leaves in a curled state is 7, which is greater than the ratio threshold 0.6, image frame A is determined to be a curled image frame.

[0074] Only one curled image frame A cannot prove whether the target tree needs watering or fertilizing. It is also necessary to collect image frames of the target tree at different positions and angles from multiple angles to make a comprehensive judgment.

[0075] Therefore, in some embodiments, multiple image frames at different positions of the tree crown are collected at multiple angles;

[0076] Count the number of multiple image frames;

[0077] Counting the number of curled image frames in the multiple image frames;

[0078] calculating a ratio of the number of the warped image frames to the number of the plurality of image frames;

[0079] If the ratio is greater than or equal to a predetermined image quantity ratio threshold, it is determined that the target tree needs maintenance.

[0080] If the ratio is less than a predetermined image quantity ratio threshold, it is determined that the target tree does not require maintenance.

[0081] Here is an example: suppose multiple image frames of different positions of the tree crown are collected from multiple angles; the number of multiple image frames is 10. Among the multiple image frames, the number of curled image frames is 7. The image number ratio is 0.7, which is greater than the image number ratio threshold of 0.6, and it is determined that the target tree needs watering and fertilization.

[0082] If the number of curled image frames is 4, and the image number ratio is 0.4, which is less than a predetermined image number ratio threshold of 0.6, it is determined that the target tree does not need watering or fertilizing.

[0083] In some embodiments, to determine whether branches and leaves are sparse, the following method may be used:

[0084] The number of branches and leaves per unit area in the current image frame can be determined;

[0085] Calculate the number of branches and leaves per unit area and the number of branches and leaves per unit area under normal conditions;

[0086] Calculate the ratio of the number of branches and leaves per unit area mentioned above to the number of branches and leaves per unit area under normal conditions.

[0087] If the ratio is less than a predetermined ratio threshold, it is determined that the branches and leaves are sparse.

[0088] The following example illustrates that: under normal conditions, the number of leaves per unit area is 20; in the captured image frame, the number of leaves per unit area is 10; the ratio threshold is 0.6; the actual ratio value is 10 / 20, which is equal to 0.5, which indicates that the leaves are sparse.

[0089] AI recognition software is used to measure the crown diameter of greening trees. If the growth of branches exceeds the scope of the highway green belt, it means that the tree growth exceeds the limit, which may cause traffic safety hazards. In order to ensure highway traffic safety, the width H of the central dividing strip of the highway is used as the threshold of the tree crown diameter. When the growth of branches exceeds the threshold, they should be pruned in time.

[0090] In some embodiments, the central control system is further used to determine the width distance H by the following method;

[0091] Step 1: Use a drone to inspect the greening trees in the central dividing strip of the expressway, with the digital camera lens pointing vertically downward to collect digital images of the tree crown diameters, and use the AI ​​recognition software for greening project diseases to measure the number of pixels N1 of the tree crown diameter R in the digital image of the greening project inspection;

[0092] Step 2: The maximum permissible transverse width of trees in highway greening projects is the distance L between the curbstones on both sides of the highway central divider as the standard distance. If the tree crown diameter R is greater than L, it means that the branches of the tree have exceeded the curbstones of the highway central divider, occupying the space of the highway lanes, which is likely to cause traffic accidents.

[0093] Step 3: Using the AI ​​recognition software for greening engineering diseases to determine the pixel N2 corresponding to the standard distance L;

[0094] Step 4: The calculation formula for the tree crown diameter of the greening project is:

[0095] R = N1 / N2 × L;

[0096] Step 5: Compare the crown diameter R of the tree and the distance L between the curbstones. If R>L, it means that the tree has grown beyond the limit and needs to be pruned and maintained in time. If R≦L, it means that the tree has not grown beyond the limit and is in normal condition.

[0097] In some embodiments, after determining the abnormal area, the central control system 2 is also used to send an alarm message to the central control console, and the alarm message includes the specific location information of the disease. The specific location information can further include: the identification of the digital image frame, the location of the target tree that needs maintenance in the digital image frame, so as to facilitate the subsequent maintenance work on the target trees that need maintenance. The platform can determine the distance from the actual position of the target tree photographed by the digital image frame to the starting point on the actual highway through distance conversion based on the identification of the above-mentioned digital image frame. When converting the distance, the distance from the actual position of the target tree photographed by the digital image frame to the starting point can be determined based on the actual distance corresponding to each digital image frame and the serial number of the digital image frame in the digital image frame sequence. Among them, the identification of the above-mentioned digital image frame includes the serial number of the above-mentioned digital image frame.

[0098] After receiving the above alarm information, the platform can issue construction instructions to the maintenance construction unit, notifying it to carry out greening project maintenance in a timely manner.

[0099] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the present disclosure. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims of the present disclosure.

[0100] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. The AI-based intelligent identification and maintenance system for highway greening project diseases is characterized by: include: Central control systems, mobile devices, digital cameras, AI intelligent identification software for greening engineering diseases, and machine learning algorithm software; The mobile device is a vehicle or a drone; the digital camera is arranged on the mobile device, and the driving route of the mobile device is planned through the central control system to ensure the consistency and accuracy of image acquisition; The machine learning algorithm software is trained repeatedly on the normal state of the highway greening project to master the morphology, color and chromaticity of leaves, and store them as samples, providing standard samples for the AI ​​intelligent recognition software of various diseases, so as to determine the type and area of ​​the disease of the greening project; The leaf morphology refers to the appearance, geometric size and natural extension state of the leaf. The AI ​​intelligent recognition software for greening engineering diseases analyzes and processes the inspection digital images, identifies the shape, color and chromaticity of leaves through digital images, and measures the leaf size through pixels; The dimensions include length, width and leaf spacing parameters, and the type and area of ​​the greening project disease are determined by comparing with the image of normal leaves stored by the machine learning algorithm; The greening project includes: greening projects in the central dividing strip of expressways and on the sides of roads.

2. The AI-based intelligent identification and maintenance system for highway greening engineering diseases as claimed in claim 1 is characterized in that: The central control system is used to control the inspection vehicle or inspection drone to collect digital images of greening projects, intelligently identify diseases through AI, automatically generate maintenance plans, issue maintenance construction instructions, calculate the number of maintenance projects and project costs, and manage the entire maintenance process and data flow.

3. The AI-based highway greening project disease intelligent identification and management system as claimed in claim 1 is characterized in that: Develop machine learning algorithm software for highway greening projects. Through repeated learning and training on the normal status of highway greening projects, the shape, color and chroma of leaves are mastered and stored as samples. Standard samples are provided for AI intelligent recognition software of various diseases to determine the type and area of ​​diseases in the greening projects.

4. The AI-based highway greening project disease intelligent identification and management system as claimed in claim 1 is characterized in that: Develop AI intelligent recognition software for greening project diseases, which can analyze and process the digital images of the greening project inspections, including identifying the shape, color and chromaticity of the leaves of the greening project, leaf spacing, and identifying leaves that are shrunken, curled, and yellowed, as well as sparse branches and leaves and branch growth-limiting diseases, and record the location and area where they occur; Wherein, the position includes a mileage post position or a Beidou coordinate position; Statistics are collected from the inspection digital images of the greening project to determine the total number of trees with shrunken leaves, yellowing leaves, sparse branches and leaves, and branches and leaves infected with diseases, as well as the geographical locations where the diseases occurred.

5. The AI-based intelligent identification and maintenance system for highway greening engineering diseases according to claim 2 is characterized in that: Targeted maintenance plans are adopted for different types of diseases, including watering for trees with shrunken leaves; fertilizing for trees with yellowing leaves; watering and fertilizing for trees with sparse leaves; and pruning and trimming for trees with limited branches and leaves.

6. The AI-based highway greening project disease intelligent identification and management system according to claim 2 is characterized in that: Issuing a maintenance construction task list to the maintenance construction unit, the maintenance construction task list includes: the location where the disease of the greening project occurs, the type and area of ​​the disease, the maintenance plan to be adopted, and the estimated maintenance project volume and project cost; The location includes a mileage post number or Beidou coordinates.

7. The AI-based highway greening project disease intelligent identification and management system according to claim 2 is characterized in that: The central control system is also used for the maintenance supervision unit to issue supervision instructions and maintenance construction task orders to the maintenance construction unit, conduct on-site supervision of construction work, inspect and accept the quality of maintenance projects, verify the maintenance project volume and project costs on site, and sign for confirmation; if the maintenance standards are not met, record the existing quality problems and their locations, issue rework or improvement supervision instructions to the maintenance construction unit, and supervise the construction quality of the maintenance construction unit.

8. The AI-based highway greening project disease intelligent identification and management system according to claim 2 is characterized in that: The central control system is also used to calculate the maintenance project cost list according to the completed greening project maintenance project volume and the approved unit price, and to calculate the overall project cost.

9. The AI-based highway greening project disease intelligent identification and management system according to claim 1 is characterized in that: The digital camera has a pixel of 9 million or more and a frame rate of collecting 30 or 60 frames of digital images per second. When the mobile device travels at a speed of 60 to 100 km / h, the greening project on the target road can be digitally photographed to obtain digital images of the greening project inspection.

Images