A maintenance detection method, system and storage medium for transplanted trees

By acquiring and analyzing the root image of the transplanted tree, automatically determining its variety and age, and generating maintenance instructions based on growth environment information, the problems of untimely maintenance and high cost in the existing technology are solved, and automated and timely detection and cost reduction are achieved effect.

CN117541225BActive Publication Date: 2024-11-15ZHEJIANG ZHEQIN CITY SERVICE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311497631.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-11-11
Publication Date
2024-11-15
Estimated Expiration
2043-11-11

AI Technical Summary

Technical Problem

In the prior art, the maintenance and testing work of the transplanted trees has problems such as labor costs and timeliness, resulting in untimely and high cost.

Method used

By obtaining the root image of the transplanted tree, judging and correcting its integrity, analyzing the images to determine the tree species and age, thereby obtaining reference growth environment information, and generating maintenance instructions based on the actual growth environment information to achieve automated and timely detection.

Benefits of technology

Reliance on labor is reduced, maintenance and testing costs are reduced, timely maintenance and testing of trees after transplantation is realized, and the problem of not being able to take into account both labor costs and timeliness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117541225B_ABST
    Figure CN117541225B_ABST
Patent Text Reader

Abstract

The present application provides a maintenance detection method, system and storage medium for transplanted trees, the method comprising: obtaining a root system image of the transplanted tree, determining whether the root system image is a complete root system image of the transplanted tree, if not, correcting the root system image to obtain a complete root system image of the transplanted tree; analyzing the complete root system image to obtain the variety of the transplanted tree, determining the age of the transplanted tree based on the variety and the complete root system image, obtaining the transplantation time of the transplanted tree, and determining the reference growth environment information range of the transplanted tree from a preset tree library through the age, variety and transplantation time; obtaining the actual growth environment information of the transplanted tree, determining whether the actual growth environment information falls within the reference growth environment information range, if not, determining the maintenance information corresponding to the transplanted tree at the current moment, and generating corresponding maintenance instructions based on the maintenance information. The present application can alleviate the problem that labor costs and timeliness cannot be taken into account at the same time in maintenance detection work.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of maintenance detection technology, and in particular to a maintenance detection method, system and storage medium for transplanted trees. Background Art

[0002] Tree transplantation plays an important role in urban greening projects and public space design. By selecting suitable trees and plants and transplanting them from nurseries or other places to urban streets, parks, gardens and other places, it can provide shade, oxygen and beautiful landscape for urban residents.

[0003] In order to enable the newly transplanted trees to survive, these trees need to be maintained for a period of time after transplantation. At present, the detection work to determine whether the transplanted trees need maintenance is generally carried out by professionals who carry corresponding detection equipment to the location of the above-mentioned trees, so as to complete this round of maintenance detection work on the transplanted trees by using the detection equipment. Although the above-mentioned existing methods can complete the maintenance detection work of transplanted trees, during the entire maintenance stage, not only will a large amount of labor costs be generated due to the participation of a large number of professionals, but also because each professional needs to take into account multiple transplanted trees at the same time, and it is impossible to achieve timeliness in maintenance detection. Therefore, at this stage, there is a problem that labor costs and timeliness cannot be taken into account at the same time in the maintenance detection work of transplanted trees. Summary of the invention

[0004] In order to alleviate the problem that labor costs and timeliness cannot be taken into account at the same time in maintenance and inspection work, the embodiments of the present application provide a maintenance and inspection method, system and storage medium for transplanted trees.

[0005] In a first aspect, this embodiment provides a maintenance detection method for transplanted trees, the method comprising:

[0006] Acquire a root system image of the transplanted tree, determine whether the root system image is a complete root system image of the transplanted tree, and if not, correct the root system image to obtain a complete root system image of the transplanted tree;

[0007] Analyze the complete root system image to obtain the variety of the transplanted tree, determine the age of the transplanted tree based on the variety and the complete root system image, obtain the transplantation time of the transplanted tree, and determine the reference growth environment information range of the transplanted tree from a preset tree library based on the tree age, variety and transplantation time;

[0008] Acquire actual growth environment information of the transplanted trees, determine whether the actual growth environment information falls within the reference growth environment information range, and if not, determine maintenance information corresponding to the transplanted trees at the current moment based on the actual growth environment information and the reference growth environment information range, and generate corresponding maintenance instructions based on the maintenance information.

[0009] In some embodiments, the root system image is a rectangular image of a fixed size, and determining whether the root system image is a complete root system image of the transplanted tree includes:

[0010] Obtain side sub-images corresponding to other side edges of the root system image that are not close to the ground, wherein the side sub-image is in a U-shape with an opening facing upward, and the left side, the lower side, and the right side of the root system image are all outer sides of the side sub-image;

[0011] Determine whether there is a root system in the side sub-image, and if so, the root system image is not a complete root system image of the transplanted tree;

[0012] If not, the root system image is a complete root system image of the transplanted tree.

[0013] In some embodiments, each root system image includes a plurality of first-level roots, and each first-level root includes a plurality of non-first-level roots, and correcting the root system image to obtain a complete root system image of the transplanted tree includes:

[0014] Respectively acquiring each first-level root and the non-first-level roots contained in the root system image to obtain a root group corresponding to each first-level root;

[0015] Using the root system deep learning model to correct each root group separately to obtain a complete root group corresponding to the root group;

[0016] The group number corresponding to each root group in the root system image is obtained, and all complete root groups are sorted according to their corresponding group numbers and placed in the blank area of ​​the fixed size to obtain the complete root system image of the transplanted tree.

[0017] In some embodiments, analyzing the complete root system image to obtain the species of the transplanted tree includes:

[0018] Obtaining the target tree number corresponding to the complete root system image, determining whether the root system image corresponding to the target tree number has been generated, and if not, obtaining a similarity value between the complete root system image and each tree root system image collected and stored in a preset tree library, and determining the variety corresponding to the maximum similarity value among the similarity values ​​as the variety of the transplanted tree;

[0019] If so, the target variety determined by the target tree number is obtained, and the target variety is determined as the variety of the transplanted tree.

[0020] In some embodiments, determining the age of the transplanted tree based on the species and the complete root system image comprises:

[0021] Obtaining the maximum level of the root system in the complete root system image and the average root size corresponding to each level of roots;

[0022] The tree age deep learning model corresponding to the variety is used to process the maximum level root and the average root size corresponding to each level root to obtain the age of the transplanted tree.

[0023] In some embodiments, the actual growth environment information includes but is not limited to actual moisture value, actual nutrient value and actual air permeability value. Two environment detection devices are respectively provided in the south direction and the north direction of the root system of the transplanted tree, and the four environment detection devices constitute four vertices of a rectangle. The actual growth environment information of the transplanted tree is obtained by:

[0024] The moisture value, nutrient value and air permeability value obtained by each environmental detection device are obtained, and the average value of all moisture values ​​is calculated to obtain the actual moisture value of the transplanted tree, the average value of all nutrient values ​​is calculated to obtain the actual nutrient value of the transplanted tree, and the average value of all air permeability values ​​is calculated to obtain the actual air permeability value of the transplanted tree.

[0025] In some embodiments, the method further comprises:

[0026] If the actual growth environment information falls within the reference growth environment information range, a maintenance instruction is generated indicating that the tree does not require maintenance.

[0027] In a second aspect, this embodiment provides a maintenance detection system for transplanted trees, the system comprising: an image acquisition module, an information processing module and a maintenance detection module; wherein:

[0028] The image acquisition module is used to acquire a root system image of the transplanted tree, determine whether the root system image is a complete root system image of the transplanted tree, and if not, correct the root system image to obtain a complete root system image of the transplanted tree;

[0029] The information processing module is used to analyze the complete root system image to obtain the species of the transplanted tree, determine the age of the transplanted tree based on the species and the complete root system image, obtain the transplantation time of the transplanted tree, and determine the reference growth environment information range of the transplanted tree from a preset tree library according to the tree age, species and transplantation time;

[0030] The maintenance detection module is used to obtain the actual growth environment information of the transplanted trees, determine whether the actual growth environment information falls within the reference growth environment information range, and if not, determine the maintenance information corresponding to the transplanted trees at the current moment based on the actual growth environment information and the reference growth environment information range, and generate corresponding maintenance instructions based on the maintenance information.

[0031] In some of the embodiments, the maintenance detection module is further configured to generate a maintenance instruction indicating that the tree does not require maintenance if the actual growth environment information falls within the reference growth environment information range.

[0032] In a third aspect, an embodiment of the present application provides a storage medium on which a computer program that can be run on a processor is stored. When the computer program is executed by the processor, a maintenance and detection method for transplanted trees as described in the first aspect is implemented.

[0033] By adopting the above method, the present application first obtains the root system image of the transplanted tree, and determines whether the root system image is a complete root system image of the transplanted tree. If not, the root system image is modified to obtain a complete root system image of the transplanted tree. After all, only a complete root system image can better, clearly and completely reflect the root system conditions of the transplanted tree, and provide a basis for more accurately determining the reference growth environment information range of the transplanted tree in the future. In the process of obtaining the complete root system image of the transplanted tree, it is not necessary to manually carry the corresponding equipment to the transplanted tree for detection, which reduces the dependence on manual labor. Each transplanted tree is equipped with corresponding equipment, which can obtain the corresponding root system image in time without destroying the existing growth environment of the transplanted tree, so that the transplanted tree can be maintained and inspected in time based on the root system image in the future.

[0034] Then, the complete root system image is analyzed to obtain the variety of the transplanted tree, the age of the transplanted tree is determined based on the variety and the complete root system image, the transplantation time of the transplanted tree is obtained, and the reference growth environment information range of the transplanted tree is determined from the preset tree library by the age, variety and transplantation time. In this way, the reference growth environment information of the transplanted tree can be automatically and timely obtained through the complete root system image and related equipment, which can reduce the dependence on manual labor. Finally, the actual growth environment information of the transplanted tree is obtained, and it is determined whether the actual growth environment information falls within the reference growth environment information range. If not, the maintenance information corresponding to the transplanted tree at the current moment is determined based on the actual growth environment information and the reference growth environment information range, and the corresponding maintenance instructions are generated based on the maintenance information. In this way, the maintenance and detection work of the transplanted data is completed by using the corresponding equipment throughout the whole process, which not only reduces the dependence on manual labor, but also can achieve real-time maintenance and detection of the transplanted trees, thereby alleviating the problem that the labor cost and timeliness in the maintenance and detection work cannot be taken into account at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a block diagram of a maintenance and detection method for transplanted trees provided in this embodiment.

[0036] Figure 2 It is a schematic diagram of a root system image provided by this embodiment.

[0037] Figure 3 This is a frame diagram of the modified root system image provided by this embodiment to obtain a complete root system image of the transplanted tree.

[0038] Figure 4 This embodiment provides a frame diagram for analyzing a complete root system image to obtain the species of the transplanted tree.

[0039] Figure 5 It is a structural schematic diagram of the growth environment information range stored in the preset tree library provided in this embodiment.

[0040] Figure 6 This is a block diagram of a maintenance and detection system for transplanted trees provided in this embodiment. DETAILED DESCRIPTION

[0041] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. However, it should be understood by those of ordinary skill in the art that the present application can be implemented without these details. For those of ordinary skill in the art, it is obvious that various changes can be made to the embodiments disclosed in the present application, and without departing from the principles and scope of the present application, the general principles defined in the present application can be applied to other embodiments and application scenarios. Therefore, the present application is not limited to the embodiments shown, but conforms to the broadest scope consistent with the scope claimed for protection in the present application.

[0042] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0043] Figure 1 FIG. 1 is a block diagram of a maintenance and detection method for transplanted trees provided in this embodiment. Figure 1 As shown, a maintenance and detection method for transplanted trees includes the following steps:

[0044] Step S100, obtaining a root system image of the transplanted tree, determining whether the root system image is a complete root system image of the transplanted tree, and if not, correcting the root system image to obtain a complete root system image of the transplanted tree.

[0045] In this embodiment, the transplanted trees are small trees. Whenever a tree transplanted from a nursery or other place needs to be planted in a city street, park, garden or other place, a pit needs to be dug at the place where the transplanted tree is planted again to obtain a space for the root system of the transplanted tree. The present application uses a signal sending device and a signal receiving device and obtains the root system image of the transplanted tree based on the principle of nuclear magnetic resonance or the principle of X-ray tomography. That is, during the pit digging process, the signal sending device is buried within the horizontal height range of the pit and the distance from the pit is not more than the preset position, such as buried 10 cm from the pit in the horizontal direction and located in the middle of the pit in the vertical direction. A signal receiving device is placed on the ground at the place where the transplanted tree is planted again. Therefore, after the transplanted tree is planted in the pit, the signal receiving device can obtain the root system image of the transplanted tree based on the frequency of the signal sent by the signal sending device. Among them, each pit corresponds to a signal sending device and a signal receiving device.

[0046] After each signal receiving device obtains a root system image, it sends the root system image to the signal processing end so that the signal processing end can obtain the root system image of the transplanted tree. Among them, each root system image obtained by the signal receiving device is a rectangular image of a fixed size. When the signal processing end obtains the root system image of the transplanted tree, it will give priority to determining whether the obtained root system image is a complete root system image of the transplanted tree. After all, only a complete root system image can better, clearly and completely reflect the root system conditions of the transplanted tree, and provide a basis for more accurately determining the reference growth environment information range of the transplanted tree in the future. Among them, judging whether the root system image is a complete root system image of the transplanted tree includes the following steps:

[0047] Step S101, obtaining side sub-images corresponding to other side edges of the root system image that are not close to the ground, wherein the side sub-image is in a U-shape opening upward, and the left side, bottom side and right side of the root system image are all outer sides of the side sub-image.

[0048] Step S102, determining whether there is a root system in the side sub-image, and if so, the root system image is not a complete root system image of the transplanted tree.

[0049] Step S103: if not, the root system image is a complete root system image of the transplanted tree.

[0050] Figure 2 is a schematic diagram of a root system image provided by this embodiment. Figure 2 As shown, the root system image includes a side sub-image and a non-side sub-image, and the root system image includes a left side, a lower side, a right side and an upper side. Wherein, the side sub-image is a fixed-size U-shaped figure with an opening upward, and the left outer side of the side sub-image shares a left side with the root system image, the lower outer side of the side sub-image shares a bottom side with the root system image, and the right outer side of the side sub-image shares a right side with the root system image. The side sub-image is located on other sides that are not close to the ground, and the side sub-image has no intersection with the upper side of the root system image, wherein the other sides refer to the left side, the lower side and the right side of the root system image. Since the sizes of the root system image and the side sub-image are fixed, and the relative position of the side sub-image in the root system image is also fixed, the side sub-image position information of the side sub-image in the root system image can be obtained, so the side sub-image can be obtained by viewing the image at the bottom side sub-image position information in the root system image.

[0051] Feature extraction is performed on the obtained side sub-image to obtain feature information in the side sub-image, and by judging whether the feature information contains feature information representing the root system, it is judged whether the root system image is a complete root system image of the transplanted tree. Wherein, the side sub-image also includes a left inner side, a lower inner side and a right inner side, and the vertical distance between the left inner side and the left outer side of the side sub-image is not greater than 1 cm, preferably 0.5 cm; the vertical distance between the lower inner side and the lower outer side of the side sub-image is not greater than 1 cm, preferably 0.5 cm; the vertical distance between the right inner side and the right outer side of the side sub-image is not greater than 1 cm, preferably 0.5 cm. If there is feature information representing the root system, then the root system image does not fully display all the roots of the transplanted tree, and the root system image is not a complete root system image of the transplanted tree, and is an incomplete root system image. If there is no feature information representing the root system, then the root system image fully displays all the roots of the transplanted tree, and the root system image is a complete root system image of the transplanted tree.

[0052] If the root system image is a complete root system image of a transplanted tree, then the root system image is marked as a complete root system image. If the root system image is not a complete root system image of a transplanted tree, it is necessary to correct the root system image to obtain a complete root system image of the transplanted tree. The roots in each root system image are in a tree-like structure, that is, a root system includes multiple first-level roots, and each first-level root includes several non-first-level roots. The first-level roots refer to the roots that originate from the seeds of the tree, such as Figure 2 As shown, the root system of the transplanted tree has multiple first-level roots. Non-first-level roots include but are not limited to second-level roots, third-level roots and fourth-level roots, where the second-level roots refer to roots branching from the first-level roots, the third-level roots refer to roots branching from the second-level roots, and the fourth-level roots refer to roots branching from the third-level roots. Figure 3 This is a diagram of a modified root system image provided by this embodiment to obtain a complete root system image of a transplanted tree. Figure 3 As shown, correcting the root system image to obtain a complete root system image of the transplanted tree includes the following steps:

[0053] Step S104 , obtaining each first-level root and the non-first-level roots contained therein from the root system image, to obtain a root group corresponding to each first-level root.

[0054] Step S105, using the root system deep learning model to correct each root group respectively to obtain a complete root group corresponding to the root group.

[0055] Step S106, obtaining the group number corresponding to each root group in the root system image, sorting all complete root groups according to their corresponding group numbers and placing them in a blank area of ​​a fixed size to obtain a complete root system image of the transplanted tree.

[0056] In the root system image of the transplanted tree, the root system is divided into a plurality of root groups according to the first-level roots, wherein each root group is essentially also an image, and each root group includes a first-level root and the second-level roots, third-level roots, fourth-level roots and other non-first-level roots contained in the first-level root, and the number of root groups is equal to the number of first-level roots in the root system of the transplanted tree. Thus, by processing the root system image using drawing software that can be used for clipping or segmentation, the same number of root groups as the first-level roots can be obtained.

[0057] After the root group corresponding to the root system image is obtained, each root group is sent to a trained root system deep learning model as input information of the root system deep learning model, so as to obtain the complete root group corresponding to the root group by obtaining the output information of the root system deep learning model. The root system deep learning model is previously trained offline through a large number of known complete root system images, and partial root system images containing only the first few levels of roots but not the last few levels of roots in these complete root system images are obtained by performing operations such as image cropping or cutout on these complete root system images used for training, and each partial root system image is used as input information, and the complete root system image corresponding to the partial root system image is used as output information to train the root system deep learning model, so as to finally obtain a trained root system deep learning model. The root system deep learning model can specifically be one of a convolutional neural network and a generative adversarial network, but is not limited to only one of these two types.

[0058] In this embodiment, each root group corresponding to the root system image has a unique group number. The corresponding root groups can be numbered from small to large or from large to small by checking the order of the first-level root distance signal sending device, so as to obtain the group number corresponding to each root group in the root system image. That is, the complete root group corresponding to each root group has the same group number, and each complete root group is essentially an image. The complete images corresponding to the group numbers are successively reduced in order of the group numbers arranged in the above-mentioned root system image and placed in a blank area of ​​the same size corresponding to the above-mentioned root system image, so as to obtain the complete root system image of the transplanted tree. Among them, in each root system image and each complete root system image, there is a size ratio used to characterize the root system on the image and the actual root system. For example, in the original root system image, the corresponding root groups are numbered in ascending order according to the distance from the signal sending device, so that the root groups in the root system image are numbered 1, 2, 3, and 4 from left to right. Then the obtained complete root system is also placed in the blank area of ​​fixed size in the order of group numbers 1, 2, 3, and 4 from left to right, so as to obtain the complete root system image of the transplanted tree. In this way, whether the root system image is a complete root system image or not, the complete root system image of the same size can be processed subsequently, reducing the complex size correspondence caused by different image sizes. Moreover, in the process of obtaining the complete root system image of the transplanted tree, it is not necessary to manually carry the corresponding equipment to the transplanted tree for detection, which reduces the dependence on manual labor. And each transplanted tree is equipped with corresponding equipment, which can obtain the corresponding root system image in time without destroying the existing growth environment of the transplanted tree, so that the transplanted tree can be maintained and detected in time based on the root system image.

[0059] Step S200, analyze the complete root system image to obtain the variety of the transplanted tree, determine the age of the transplanted tree based on the variety and the complete root system image, obtain the transplantation time of the transplanted tree, and determine the reference growth environment information range of the transplanted tree from a preset tree library through the tree age, variety and transplantation time.

[0060] After the complete root system image is obtained, the growth environment required for the transplanted tree to grow normally can be determined by analyzing the complete root system image. Since different species of trees require different growth environments, it is necessary to first analyze the complete root system image to determine the species of the transplanted tree. Figure 4 This is a diagram of analyzing the complete root system image to obtain the species of the transplanted tree provided by this embodiment. Figure 4 As shown, analyzing the complete root system image to obtain the species of the transplanted tree includes the following steps:

[0061] Step S201, obtain the target tree number corresponding to the complete root system image, determine whether the root system image corresponding to the target tree number has been completed, if not, obtain the similarity value between the complete root system image and each tree root system image collected and stored in the preset tree library, and determine the variety corresponding to the maximum similarity value among the similarity values ​​as the variety of the transplanted tree.

[0062] Step S202: If yes, obtain the target variety whose target tree number has been determined, and determine the target variety as the variety of the transplanted tree.

[0063] Each transplanted tree has its own unique tree number. When the signal receiving device sends the root system image of the transplanted tree, it also sends its own location information or identity information that can prove its identity to the signal processing end, so that the signal processing end can obtain the corresponding location information or identity information while obtaining the root system image of the transplanted tree, and then obtain the target tree number corresponding to the transplanted tree according to the location information or identity information, that is, the target tree number corresponding to the complete root system image. Among them, the signal receiving device will not change its location information or identity information during the process of executing the maintenance and detection method of transplanted trees.

[0064] It can be checked whether it is the first time to obtain the target tree number. If it is the first time to obtain the target tree number, it indicates that no root system image corresponding to the target tree number has been generated. If it is not the first time to obtain the target tree number, it indicates that the root system image corresponding to the target tree number has been generated. If the root system image corresponding to the target tree number has not been generated, then the complete root system image can be compared with each collected tree root system image stored in the preset tree library to obtain the similarity value between the complete root system image and each tree root system image in the tree library. The tree variety corresponding to the maximum similarity value among all the similarity values ​​obtained is the variety of the transplanted tree. Among them, the tree library stores the tree root system images of trees that are daily planted in urban streets, parks, gardens and other places and the varieties corresponding to the trees. If the root system image corresponding to the target tree number has been generated, it means that the target variety corresponding to the target tree number has been known. The target variety determined by the target tree number can be directly obtained by checking the target variety corresponding to the target tree number that has been obtained, and the target variety is determined as the variety of the transplanted tree. In this way, the species of the transplanted tree can be automatically obtained directly through the complete root system graph obtained in the above step S100, without the need to obtain the species of the transplanted tree by being informed by personnel, which can reduce dependence on manual labor; and there is no need to add additional equipment. Only the existing equipment and the information obtained in the previous execution process can be used to determine the species of the transplanted tree, so that the species of the transplanted tree can be obtained in time.

[0065] In addition, transplanted trees of the same species but different ages require different growth environments, so it is also necessary to determine the age of the transplanted trees. Determining the age of the transplanted trees based on the species and the complete root system image includes the following steps:

[0066] Step S203, obtaining the maximum level of the root system in the complete root system image and the average root size corresponding to each level of roots.

[0067] Step S204, using the tree age deep learning model corresponding to the variety to process the maximum level root and the average root size corresponding to each level root to obtain the age of the transplanted tree.

[0068] Check whether there are any roots branching out from the first-level roots in the complete root system image. If so, the roots branching out from the first-level roots are the second-level roots. Then check whether there are any roots branching out from the second-level roots in the complete root system image. If not, the maximum level root contained in the root system in the complete root system image is the second-level root. If so, the roots branching out from the second-level roots are the third-level roots. Then check whether there are any roots branching out from the third-level roots. Stop checking until there are no roots branching out from the n-th level roots, and determine the n-th level root as the maximum level contained in the root system in the complete root system image.

[0069] Through the above process of obtaining the maximum level of the root system in the complete root system image, the level corresponding to each root of the root system in the complete root system image can be obtained. In addition, the size of each root of the root system in the complete root system image can be measured by using a measuring device, and the sizes corresponding to the roots of the same level are averaged to obtain the average root size corresponding to each level root.

[0070] The average root size and maximum level of the roots at each level in the complete root system image are sent to the trained tree age deep learning model as the input information of the tree age deep learning model, so as to obtain the age of the transplanted tree by obtaining the output information of the tree age deep learning model. Among them, the tree age deep learning model is obtained by a large amount of offline training in advance, and the information used for a large amount of training is the maximum level in each complete root system image used by the above-mentioned training root system deep learning model and the corresponding average root size corresponding to each level of roots. But it is not limited to the information used for training the tree age deep learning model. This embodiment can reduce the number of samples for obtaining training information and save the workload brought by obtaining samples by using the same complete root system image or certain information therein in training the root system deep learning model and the tree age deep learning model.

[0071] Once the transplanted trees are planted again, the signal sending device corresponding to the transplanted trees will immediately send out a signal and the corresponding signal receiving device will obtain the root image, and send the time of obtaining the root image to the signal processing end, so that the signal processing end can obtain the image time corresponding to each root image. The transplantation time of the transplanted tree can be obtained by subtracting the image time corresponding to the first root image from the image time corresponding to the most recent root image of the transplanted tree.

[0072] The preset tree database also stores the range of growth environment information required for each tree species at different tree age stages as the transplant time increases. The range of growth environment information required for each tree species at different tree age stages as the transplant time is obtained by integrating the work experience and the latest research results of experienced tree experts. Figure 5 Schematic diagram of the structure of the growth environment information range stored in the preset tree library provided in this embodiment. Figure 5 As shown, the reference growth environment information range of the transplanted tree can be found by comparing the species, tree age and transplantation time in turn, so as to determine the reference growth environment information range of the transplanted tree from the preset tree library by the tree age, species and transplantation time. Determining the reference growth environment information range of the transplanted tree in the above manner can improve the reliability of obtaining the reference growth environment information range, and on the other hand, the storage of the preset tree library in this manner facilitates the subsequent expansion of the tree library information.

[0073] Step S300, obtain the actual growth environment information of the transplanted trees, determine whether the actual growth environment information falls within the reference growth environment information range, if not, determine the maintenance information corresponding to the transplanted trees at the current moment based on the actual growth environment information and the reference growth environment information range, and generate corresponding maintenance instructions based on the maintenance information.

[0074] After obtaining the reference growth environment information range of the transplanted trees, it is necessary to further detect the actual growth environment information of the transplanted trees, so as to determine whether the transplanted trees need maintenance by comparing the actual growth environment information with the reference growth environment information range, thereby completing the maintenance inspection work of the transplanted trees.

[0075] The above-mentioned actual growth environment information includes but is not limited to actual moisture value, actual nutrient value and actual air permeability value. In the process of digging a pit at the place where the transplanted tree is planted again, two environmental detection devices are respectively provided in the south and north directions of the root system of the transplanted tree, and these four environmental detection devices constitute the four vertices of a rectangle. Among them, obtaining the actual growth environment information of the transplanted tree includes: obtaining the moisture value, nutrient value and air permeability value obtained by each environmental detection device, calculating the average value of all moisture values ​​to obtain the actual moisture value of the transplanted tree, calculating the average value of all nutrient values ​​to obtain the actual nutrient value of the transplanted tree, and calculating the average value of all air permeability values ​​to obtain the actual air permeability value of the transplanted tree.

[0076] The south direction is prone to soil erosion, while the north direction is prone to water accumulation. This embodiment places the environmental detection device in the south and north directions, which can take into account both the better and worse conditions of the soil environment where the root system is located. Then, the actual moisture value of the soil where the tree is transplanted is obtained by taking the average value of all the moisture values ​​obtained. Similarly, the actual nutrient value of the soil where the tree is transplanted is obtained by taking the average value of all the nutrient values ​​obtained, and the actual permeability value of the soil where the tree is transplanted is obtained by taking the average value of all the permeability values ​​obtained. In this way, the actual growth environment information of the transplanted tree is obtained by taking the average value, thereby reducing the occurrence of inaccurate actual growth environment information due to the use of one environmental detection device.

[0077] The reference growth environment information range in the above step S200 includes a reference moisture range, a reference nutrient range, and a reference air permeability range. Whether the actual growth environment information falls within the reference growth environment information range obtained in the above step S200 is determined by determining whether the actual moisture value falls within the reference moisture range in the reference growth environment information range obtained in the above step S200, determining whether the actual nutrient value falls within the reference nutrient range in the reference growth environment information range obtained in the above step S200, and determining whether the actual air permeability value falls within the reference air permeability range in the reference growth environment information range obtained in the above step S200. If the actual moisture value falls within the reference moisture range, the actual nutrient value falls within the reference nutrient range, and the actual air permeability value falls within the reference air permeability range, then the actual growth environment information falls within the reference growth environment information range. If at least one of the actual moisture value, the actual nutrient value, and the actual air permeability value does not fall within the corresponding reference range, then the actual growth environment information does not fall within the reference growth environment information range.

[0078] When the actual growth environment information does not fall within the reference growth environment information range, it is possible to check which actual value does not fall within the corresponding reference range, and then compare the corresponding relationship between the actual value and the corresponding reference range to obtain the maintenance information corresponding to the actual value that does not fall within the corresponding reference range. For example, if the actual moisture value is less than the minimum value of the reference moisture range, then the maintenance information that requires watering is generated. After the maintenance information is obtained, the corresponding maintenance instructions are generated based on the maintenance information, and then sent to the corresponding terminal to perform timely maintenance on the transplanted trees. If the actual growth environment information falls within the reference growth environment information range, a maintenance instruction is generated to indicate that the trees do not need maintenance, so as to indicate that the transplanted trees do not need maintenance. In this way, the maintenance and detection work of the transplanted data is completed by using the corresponding equipment throughout the whole process, which not only reduces the dependence on manual labor, but also can achieve real-time maintenance and detection of transplanted trees, thereby alleviating the problem that labor costs and timeliness cannot be taken into account at the same time in the maintenance and detection work.

[0079] Figure 6 FIG. 1 is a block diagram of a maintenance and detection system for transplanted trees provided in this embodiment. Figure 6 As shown, a maintenance detection system for transplanted trees includes: an image acquisition module, an information processing module and a maintenance detection module.

[0080] Among them, the image acquisition module is used to obtain the root system image of the transplanted tree, determine whether the root system image is a complete root system image of the transplanted tree, and if not, correct the root system image to obtain the complete root system image of the transplanted tree. The information processing module is used to analyze the complete root system image to obtain the variety of the transplanted tree, determine the age of the transplanted tree based on the variety and the complete root system image, obtain the transplantation time of the transplanted tree, and determine the reference growth environment information range of the transplanted tree from the preset tree library through the age, variety and transplantation time. The maintenance detection module is used to obtain the actual growth environment information of the transplanted tree, determine whether the actual growth environment information falls within the reference growth environment information range, and if not, determine the maintenance information corresponding to the transplanted tree at the current moment based on the actual growth environment information and the reference growth environment information range, and generate corresponding maintenance instructions based on the maintenance information. The maintenance detection module is also used to generate a maintenance instruction indicating that the tree does not need maintenance if the actual growth environment information falls within the reference growth environment information range.

[0081] The other functions performed by the above-mentioned image acquisition module, information processing module and maintenance detection module and the technical details of each function are the same or similar to the corresponding features in the maintenance detection method for transplanted trees described above, so they are not repeated here.

[0082] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer-readable storage medium is run on a computer, the computer can execute the relevant contents of the aforementioned embodiment of a maintenance and detection method for transplanted trees.

[0083] It should be understood that, although the steps in the flowchart of the accompanying drawings are shown in sequence according to the instructions of the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless explicitly stated in this document, the execution of these steps is not strictly limited in order and can be performed in other orders.

[0084] The above description is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for maintenance and detection of transplanted trees, characterized in that: The method comprises: Acquire a root system image of the transplanted tree, determine whether the root system image is a complete root system image of the transplanted tree, and if not, correct the root system image to obtain a complete root system image of the transplanted tree; Analyze the complete root system image to obtain the variety of the transplanted tree, determine the age of the transplanted tree based on the variety and the complete root system image, obtain the transplantation time of the transplanted tree, and determine the reference growth environment information range of the transplanted tree from a preset tree library based on the tree age, variety and transplantation time; Acquire actual growth environment information of the transplanted tree, determine whether the actual growth environment information falls within the reference growth environment information range, and if not, determine maintenance information corresponding to the transplanted tree at the current moment based on the actual growth environment information and the reference growth environment information range, and generate corresponding maintenance instructions based on the maintenance information; Wherein, the analyzing the complete root system image to obtain the species of the transplanted tree comprises: Obtaining the target tree number corresponding to the complete root system image, determining whether the root system image corresponding to the target tree number has been generated, and if not, obtaining a similarity value between the complete root system image and each tree root system image collected and stored in a preset tree library, and determining the variety corresponding to the maximum similarity value among the similarity values ​​as the variety of the transplanted tree; If so, obtaining the target variety whose target tree number has been determined, and determining the target variety as the variety of the transplanted tree; Determining the age of the transplanted tree based on the variety and the complete root system image comprises: Obtaining the maximum level of the root system in the complete root system image and the average root size corresponding to each level of roots; The tree age deep learning model corresponding to the variety is used to process the maximum level root and the average root size corresponding to each level root to obtain the age of the transplanted tree.

2. The method according to claim 1, characterized in that The root system image is a rectangular image of a fixed size, and the step of determining whether the root system image is a complete root system image of the transplanted tree includes: Obtain side sub-images corresponding to other side edges of the root system image that are not close to the ground, wherein the side sub-image is in a U-shape with an opening facing upward, and the left side, the lower side, and the right side of the root system image are all outer sides of the side sub-image; Determine whether there is a root system in the side sub-image, and if so, the root system image is not a complete root system image of the transplanted tree; If not, the root system image is a complete root system image of the transplanted tree.

3. The method according to claim 2, characterized in that Each root system image includes a plurality of first-level roots, and each first-level root includes a plurality of non-first-level roots. Correcting the root system image to obtain a complete root system image of the transplanted tree includes: Respectively acquiring each first-level root and the non-first-level roots contained in the root system image to obtain a root group corresponding to each first-level root; Using the root system deep learning model to correct each root group separately to obtain a complete root group corresponding to the root group; The group number corresponding to each root group in the root system image is obtained, and all complete root groups are sorted according to their corresponding group numbers and placed in the blank area of ​​the fixed size to obtain the complete root system image of the transplanted tree.

4. The method according to claim 1, characterized in that: The actual growth environment information includes but is not limited to the actual moisture value, the actual nutrient value and the actual air permeability value. Two environment detection devices are respectively provided in the south direction and the north direction of the root system of the transplanted tree, and the four environment detection devices constitute the four vertices of a rectangle. The actual growth environment information of the transplanted tree is obtained by: The moisture value, nutrient value and air permeability value obtained by each environmental detection device are obtained, and the average value of all moisture values ​​is calculated to obtain the actual moisture value of the transplanted tree, the average value of all nutrient values ​​is calculated to obtain the actual nutrient value of the transplanted tree, and the average value of all air permeability values ​​is calculated to obtain the actual air permeability value of the transplanted tree.

5. The method according to claim 1, characterized in that: The method further comprises: If the actual growth environment information falls within the reference growth environment information range, a maintenance instruction is generated indicating that the tree does not require maintenance.

6. A maintenance and detection system for transplanted trees, characterized in that: The system includes: an image acquisition module, an information processing module and a maintenance detection module; wherein, The image acquisition module is used to acquire a root system image of the transplanted tree, determine whether the root system image is a complete root system image of the transplanted tree, and if not, correct the root system image to obtain a complete root system image of the transplanted tree; The information processing module is used to analyze the complete root system image to obtain the species of the transplanted tree, determine the age of the transplanted tree based on the species and the complete root system image, obtain the transplantation time of the transplanted tree, and determine the reference growth environment information range of the transplanted tree from a preset tree library according to the tree age, species and transplantation time; The maintenance detection module is used to obtain the actual growth environment information of the transplanted tree, determine whether the actual growth environment information falls within the reference growth environment information range, and if not, determine the maintenance information corresponding to the transplanted tree at the current moment based on the actual growth environment information and the reference growth environment information range, and generate corresponding maintenance instructions based on the maintenance information; Wherein, the analyzing the complete root system image to obtain the species of the transplanted tree comprises: Obtaining the target tree number corresponding to the complete root system image, determining whether the root system image corresponding to the target tree number has been generated, and if not, obtaining a similarity value between the complete root system image and each tree root system image collected and stored in a preset tree library, and determining the variety corresponding to the maximum similarity value among the similarity values ​​as the variety of the transplanted tree; If so, obtaining the target variety whose target tree number has been determined, and determining the target variety as the variety of the transplanted tree; Determining the age of the transplanted tree based on the variety and the complete root system image comprises: Obtaining the maximum level of the root system in the complete root system image and the average root size corresponding to each level of roots; The tree age deep learning model corresponding to the variety is used to process the maximum level root and the average root size corresponding to each level root to obtain the age of the transplanted tree.

7. The system according to claim 6, characterized in that The maintenance detection module is also used to generate a maintenance instruction indicating that the tree does not need maintenance if the actual growth environment information falls within the reference growth environment information range.

8. A computer-readable storage medium having stored thereon a computer program that can be run on a processor, characterized in that: When the computer program is executed by the processor, the method for maintaining and detecting transplanted trees as claimed in any one of claims 1 to 5 is implemented.