Landscape design identification method, device and equipment based on neural network model

Abstract

The application relates to a landscape design recognition method and device based on a neural network model, computer equipment and a storage medium. The method comprises the following steps: acquiring an outdoor landscape image corresponding to an outdoor landscape three-dimensional model; wherein the outdoor landscape image is an image obtained by conversion of the outdoor landscape three-dimensional model, and each outdoor landscape image comprises at least two kinds of outdoor landscape objects; inputting the outdoor landscape image into a preset neural network model to recognize the correlation of the outdoor landscape objects in the outdoor landscape three-dimensional model, and obtaining a design recognition result of the correlation of the outdoor landscape in the outdoor landscape three-dimensional model. The method can avoid the problems of low adjustment efficiency, low accuracy, incomplete adjustment and high artificial learning cost caused by traditional manual adjustment and optimization of the outdoor landscape three-dimensional model according to experience, improve the recognition efficiency and accuracy, and reduce the artificial learning cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building auxiliary design, and in particular to a landscape design recognition method and device based on a neural network model, equipment and a storage medium. BACKGROUND

[0002] With the development of science and technology, the field of architectural design has achieved more automation.

[0003] Traditional outdoor landscape three-dimensional models are designed by hand, and after the design is completed, the design results are optimized by hand according to years of industry experience.

[0004] However, in the traditional design optimization process, the purely manual experience-based adjustment of outdoor landscapes may have low adjustment efficiency and incomplete optimization, resulting in low-quality design models. SUMMARY

[0005] Therefore, it is necessary to provide a landscape design recognition method and device based on a neural network model, a computer device and a storage medium to solve the above technical problems.

[0006] In a first aspect, an embodiment of the present application provides a landscape design recognition method based on a neural network model, which comprises:

[0007] Obtaining an outdoor landscape image corresponding to an outdoor landscape three-dimensional model, wherein the outdoor landscape image is an image obtained by converting the outdoor landscape three-dimensional model, and each outdoor landscape image includes at least two related outdoor landscape objects;

[0008] Inputting the outdoor landscape image into a preset neural network model to recognize the correlation of outdoor landscape objects in the outdoor landscape three-dimensional model, and obtaining a design recognition result of the correlation of outdoor landscapes in the outdoor landscape three-dimensional model;

[0009] The design recognition result is used to represent the matching degree of the correlation of the outdoor landscape objects and the outdoor landscape design rules, and the neural network model is a neural network obtained by using images carrying outdoor landscape object labels to mine rules.

[0010] In one embodiment, the outdoor landscape image is input into the preset neural network model to recognize the correlation of outdoor landscape objects in the outdoor landscape three-dimensional model, and a design recognition result of the correlation of outdoor landscapes in the outdoor landscape three-dimensional model is obtained, which comprises:

[0011] The outdoor landscape image is input into the preset neural network model to recognize the correlation of outdoor landscape objects, and a correlation list of the outdoor landscape objects is obtained.

[0012] According to each association relationship in the association relationship list and an outdoor landscape design rule, an outdoor landscape object correlation degree design identification result is obtained;

[0013] According to the outdoor landscape object correlation degree design identification result, a prompt is given.

[0014] In one of the embodiments, the association relationship includes at least one of an association relationship between a location area and a water accumulation and leakage prevention measure of a drainage pipe, an association relationship between a tile specification and a water leakage prevention measure when water flows into and out of a pipe, an association relationship between a plant type and a temperature area, an association relationship between a temperature area and a drainage pipe location, an association relationship between a drainage pipe laying and a wall, an association relationship between a drainage pipe location and a drying area, and an association relationship between a balcony structure and an enclosure component.

[0015] In one of the embodiments, before the outdoor landscape three-dimensional model corresponding outdoor landscape image is obtained, the method further includes:

[0016] An outdoor landscape three-dimensional model in a design model is obtained.

[0017] The outdoor landscape three-dimensional model is converted from three-dimensional to two-dimensional to obtain a plurality of outdoor landscape images, and the outdoor landscape images have different representation angles.

[0018] In one of the embodiments, the process of obtaining the neural network model includes:

[0019] An outdoor landscape sample model with an outdoor landscape object pre-labeled is converted from three-dimensional to two-dimensional to obtain a plurality of outdoor landscape initial sample images.

[0020] The outdoor landscape initial sample images are pre-processed to obtain outdoor landscape sample images, and the pre-processing includes at least one of screening and splicing.

[0021] The outdoor landscape sample images are input into an initial identification model, an association relationship is extracted based on a decision tree classification mode, a frequent item set is obtained, and the neural network model including an outdoor landscape design rule is obtained.

[0022] In a second aspect, an embodiment of the present application provides a landscape design identification method based on a neural network model, and the method includes:

[0023] An outdoor landscape sample model with an outdoor landscape object pre-labeled is converted from three-dimensional to two-dimensional to obtain a plurality of outdoor landscape initial sample images.

[0024] The outdoor landscape initial sample images are pre-processed to obtain outdoor landscape sample images, and the pre-processing includes at least one of screening and splicing.

[0025] inputting the outdoor landscape sample image into an initial recognition model, extracting a correlation relationship based on a classification mode of a decision tree, and obtaining a frequent item set to obtain a neural network model including outdoor landscape design rules;

[0026] obtaining an outdoor landscape three-dimensional model in a design model;

[0027] performing three-dimensional to two-dimensional conversion on the outdoor landscape three-dimensional model to obtain a plurality of outdoor landscape images, the outdoor landscape images having different representation angles;

[0028] inputting the outdoor landscape image into the preset neural network model to identify a correlation relationship of an outdoor landscape object to obtain a correlation relationship list of the outdoor landscape object;

[0029] obtaining an outdoor landscape object correlation degree design recognition result according to each correlation relationship in the correlation relationship list and outdoor landscape design rules, wherein the correlation relationship includes at least one of a position area and a water accumulation and leakage prevention measure of a drainage pipe, an outdoor landscape tile specification and a water seepage prevention measure when water flows into and out of a pipe, a flower and plant type and a temperature area, a temperature area and a drainage pipe position, drainage pipe laying and a wall, a drainage pipe position and a drying area, a balcony structure and a surrounding component;

[0030] prompting according to the outdoor landscape object correlation degree design recognition result.

[0031] In a third aspect, an embodiment of the present application provides a landscape design recognition device based on a neural network model, the device comprising:

[0032] an obtaining module configured to obtain outdoor landscape images corresponding to an outdoor landscape three-dimensional model, wherein the outdoor landscape images are images obtained by converting the outdoor landscape three-dimensional model, and each of the outdoor landscape images includes at least two types of outdoor landscape objects;

[0033] an identifying module configured to input the outdoor landscape images into a preset neural network model to identify a correlation relationship of an outdoor landscape object in the outdoor landscape three-dimensional model, and obtain a design recognition result of the correlation relationship of the outdoor landscape in the outdoor landscape three-dimensional model;

[0034] wherein the design recognition result is used to represent a matching degree of the correlation relationship of the outdoor landscape object and outdoor landscape design rules, and the neural network model is a neural network obtained by using images carrying outdoor landscape object labels to mine rules.

[0035] In a fourth aspect, an embodiment of the present application provides a landscape design recognition device based on a neural network model, the device comprising:

[0036] a training module configured to convert an outdoor landscape sample model with pre-labeled outdoor landscape objects from three dimensions to two dimensions to obtain a plurality of outdoor landscape initial sample images, pre-process the outdoor landscape initial sample images to obtain outdoor landscape sample images, and input the outdoor landscape sample images into an initial recognition model to extract correlation based on a decision tree classification method and obtain a frequent item set to obtain a neural network model including outdoor landscape design rules;

[0037] wherein the pre-processing comprises at least one of screening and splicing,

[0038] a processing module configured to obtain an outdoor landscape three-dimensional model in a design model, convert the outdoor landscape three-dimensional model from three dimensions to two dimensions to obtain a plurality of outdoor landscape images, input the outdoor landscape images into the neural network model to recognize the correlation of the outdoor landscape objects, obtain a correlation list of the outdoor landscape objects, and obtain an outdoor landscape object correlation degree design recognition result according to each correlation in the correlation list and the outdoor landscape design rules, and prompt according to the outdoor landscape object correlation degree design recognition result.

[0039] wherein the design recognition result is used to represent the matching degree of the correlation of the outdoor landscape objects and the outdoor landscape design rules; the correlation includes at least one of the following: the correlation of the position area and the water accumulation and leakage prevention measures of the drainage pipeline, the correlation of the outdoor landscape tile specification and the anti-leakage measures when the pipeline is filled with water, the correlation of the flower and plant species and the temperature area, the correlation of the temperature area and the drainage pipeline position, the correlation of the drainage pipeline laying and the wall, the correlation of the drainage pipeline position and the drying area, and the correlation of the balcony structure and the enclosure component.

[0040] In a fifth aspect, an embodiment of the present application provides a computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program:

[0041] obtaining outdoor landscape images corresponding to an outdoor landscape three-dimensional model; wherein the outdoor landscape images are images obtained by converting the outdoor landscape three-dimensional model, and each of the outdoor landscape images includes at least two related outdoor landscape objects;

[0042] inputting the outdoor landscape images into a preset neural network model to recognize the correlation of the outdoor landscape objects in the outdoor landscape three-dimensional model, and obtaining a design recognition result of the correlation of the outdoor landscape in the outdoor landscape three-dimensional model.

[0043] The design recognition result is used to represent a matching degree of the association relationship of the outdoor landscape object and an outdoor landscape design rule, and the neural network model is a neural network obtained by rule mining using an image carrying an outdoor landscape object label.

[0044] In a sixth aspect, an embodiment of the present application provides a computer device, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program:

[0045] An outdoor landscape sample model with pre-labeled outdoor landscape objects is converted from three dimensions to two dimensions to obtain a plurality of outdoor landscape initial sample images;

[0046] The outdoor landscape initial sample images are pre-processed to obtain outdoor landscape sample images; wherein the pre-processing includes at least one of screening and splicing;

[0047] The outdoor landscape sample images are input into an initial recognition model, an association relationship is extracted based on a classification mode of a decision tree, and a frequent item set is obtained to obtain a neural network model including an outdoor landscape design rule;

[0048] An outdoor landscape three-dimensional model in a design model is obtained;

[0049] The outdoor landscape three-dimensional model is converted from three dimensions to two dimensions to obtain a plurality of outdoor landscape images, and the outdoor landscape images have different representation angles;

[0050] The outdoor landscape images are input into the preset neural network model to identify the association relationship of the outdoor landscape objects to obtain an association relationship list of the outdoor landscape objects;

[0051] According to each association relationship in the association relationship list and an outdoor landscape design rule, an outdoor landscape object association degree design recognition result is obtained; wherein the association relationship includes at least one of an association relationship of a position area and a water accumulation and leakage prevention measure of a drainage pipe, an association relationship of an outdoor landscape tile specification and a water seepage prevention measure when water flows into and out of a pipe, an association relationship of a flower and plant type and a temperature area, an association relationship of a temperature area and a drainage pipe position, an association relationship of drainage pipe laying and a wall, an association relationship of a drainage pipe position and a drying area, and an association relationship of a balcony structure and an enclosure component;

[0052] The outdoor landscape object association degree design recognition result is prompted.

[0053] In a seventh aspect, an embodiment of the present application provides a computer readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the following steps:

[0054] obtaining outdoor landscape images corresponding to the outdoor landscape three-dimensional model; wherein the outdoor landscape images are images converted from the outdoor landscape three-dimensional model, and each of the outdoor landscape images includes at least two kinds of outdoor landscape objects;

[0055] inputting the outdoor landscape images into a preset neural network model to identify the association relationship of the outdoor landscape objects in the outdoor landscape three-dimensional model, and obtaining a design identification result of the association relationship of the outdoor landscape in the outdoor landscape three-dimensional model;

[0056] wherein the design identification result is used to represent the matching degree of the association relationship of the outdoor landscape objects and the outdoor landscape design rule, and the neural network model is a neural network obtained by using images carrying outdoor landscape object labels to perform rule mining.

[0057] In an eighth aspect, an embodiment of the present application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the following steps:

[0058] performing three-dimensional to two-dimensional conversion on an outdoor landscape sample model pre-labeled with outdoor landscape objects to obtain a plurality of outdoor landscape initial sample images;

[0059] preprocessing the outdoor landscape initial sample images to obtain outdoor landscape sample images; wherein the preprocessing includes at least one of screening and splicing;

[0060] inputting the outdoor landscape sample images into an initial identification model, performing association relationship extraction based on a decision tree classification method, and obtaining a frequent item set to obtain a neural network model including an outdoor landscape design rule;

[0061] obtaining an outdoor landscape three-dimensional model in a design model;

[0062] performing three-dimensional to two-dimensional conversion on the outdoor landscape three-dimensional model to obtain a plurality of the outdoor landscape images, and the outdoor landscape images have different representation angles;

[0063] inputting the outdoor landscape images into the preset neural network model to identify the association relationship of the outdoor landscape objects, and obtaining an association relationship list of the outdoor landscape objects;

[0064] According to each association relationship in the association relationship list and an outdoor landscape design rule, an outdoor landscape object correlation degree design identification result is obtained; wherein the association relationship includes at least one of the following: an association relationship between a location area and a water accumulation and leakage prevention measure of a drainage pipeline, an association relationship between an outdoor landscape tile specification and a water leakage prevention measure when water is supplied to and drained from a pipeline, an association relationship between a plant type and a temperature area, an association relationship between a temperature area and a drainage pipeline location, an association relationship between drainage pipeline laying and a wall, an association relationship between a drainage pipeline location and a drying area, and an association relationship between a balcony structure and an enclosure component.

[0065] According to the outdoor landscape object correlation degree design identification result, a prompt is given.

[0066] The landscape design identification method, device, computer device, and storage medium based on a neural network model provided in the embodiments of the present application, wherein the computer device acquires an outdoor landscape image corresponding to an outdoor landscape three-dimensional model, inputs the outdoor landscape image into a preset neural network model to identify the association relationship of the outdoor landscape object, and obtains a design identification result of the outdoor landscape three-dimensional model corresponding to the association relationship of the outdoor landscape object. Since the neural network model is a neural network obtained by training an image carrying an outdoor landscape object label, and the neural network model can mine the rules of various association relationships between different outdoor landscape objects as outdoor landscape design rules by identifying the image carrying the outdoor landscape object label, the neural network model can automatically identify the outdoor landscape image to obtain the association relationship of the outdoor landscape object, and compare the association relationship with the outdoor landscape design rules, thereby obtaining the design identification result of the outdoor landscape three-dimensional model representing the matching degree of the association relationship of the outdoor landscape object and the outdoor landscape design rules. Therefore, the method can avoid the problems of low adjustment efficiency, low accuracy, incomplete adjustment, and high artificial learning cost caused by traditional manual adjustment and optimization of the outdoor landscape three-dimensional model. The method can automatically identify the association relationship of the outdoor landscape object by using the neural network model, obtain the design identification result of the outdoor landscape three-dimensional model based on the mined outdoor landscape design rules, greatly improve the identification efficiency, more accurately and comprehensively identify the current association relationship of the outdoor landscape object, greatly improve the accuracy, and reduce the artificial learning cost, thereby greatly saving time and manpower. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 A flowchart of the landscape design identification method based on a neural network model provided in an embodiment is shown;

[0068] Figure 2 A flowchart of the landscape design identification method based on a neural network model provided in another embodiment is shown;

[0069] Figure 3 A flowchart of a landscape design recognition method based on a neural network model is provided for another embodiment;

[0070] Figure 4 A flowchart of a landscape design recognition method based on a neural network model is provided for another embodiment;

[0071] Figure 5 A structural diagram of a landscape design recognition device based on a neural network model is provided for an embodiment;

[0072] Figure 6 A structural diagram of a landscape design recognition device based on a neural network model is provided for another embodiment;

[0073] Figure 7 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0074] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0075] The technical solutions of the present application and how the technical solutions solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present application will be described below in combination with the drawings.

[0076] It should be noted that the execution subject of the following method embodiments can be a landscape design recognition device based on a neural network model, which can be realized by software, hardware or a combination of software and hardware to become part or all of the above computer device. The following method embodiments are described by taking the computer device as an example.

[0077] Figure 1 A flowchart of a landscape design recognition method based on a neural network model is provided for an embodiment. The present embodiment relates to a process in which a computer device uses an artificial neural network to assist in the design of an outdoor landscape three-dimensional model. As shown in Figure 1 , it includes:

[0078] Step S11, obtaining an outdoor landscape image corresponding to an outdoor landscape three-dimensional model.

[0079] Among them, the outdoor landscape image is an image obtained by converting the outdoor landscape three-dimensional model, and each outdoor landscape image includes at least two related outdoor landscape objects.

[0080] The outdoor landscape three-dimensional model can be a terrace model or a balcony model.

[0081] Specifically, the computer device can read a pre-stored two-dimensional outdoor landscape image, can receive an outdoor landscape image sent by another device, and can also obtain an outdoor landscape image by identifying an outdoor landscape three-dimensional model. The present embodiment does not limit this. It should be noted that the above-mentioned outdoor landscape image is obtained by converting the outdoor landscape three-dimensional model from three-dimensional to two-dimensional by the computer device, thereby obtaining two-dimensional outdoor landscape images of different angles and different outdoor landscape objects.

[0082] Optionally, the above-mentioned outdoor landscape three-dimensional model can include a pool, a waterscape, a tile, a flower, a plant, a drain pipe, a water intake, a movable platform, a seat area, a barbecue leisure area, a plant decoration, a window sill plate, a vegetable washing area, and the like outdoor landscape objects.

[0083] Step S12: inputting the outdoor landscape image into a preset neural network model to identify the association relationship of the outdoor landscape objects in the outdoor landscape three-dimensional model, and obtaining a design identification result of the association relationship of the outdoor landscape in the outdoor landscape three-dimensional model.

[0084] The design identification result is used to represent the matching degree of the association relationship of the outdoor landscape objects and the outdoor landscape design rule, and the neural network model is a neural network obtained by using an image carrying an outdoor landscape object label to mine rules.

[0085] Specifically, the computer inputs the outdoor landscape image into a preset neural network model. The neural network model is obtained by training images carrying outdoor landscape labels. The images carrying outdoor landscape labels can include images carrying labels of at least one outdoor landscape object such as a pool, a waterscape, a tile, a flower, a plant, a drainpipe, a water intake, a platform, a seating area, a barbecue leisure area, a plant decoration, a window sill, etc. The images can also mark attribute information of the outdoor landscape objects. For example, the attribute information of the window sill can be marble, solid wood, artificial stone, mosaic, composite acrylic, etc. The neural network model identifies the images carrying outdoor landscape labels, mines the rules of various association relationships between different outdoor landscape objects, and takes the obtained rules as outdoor landscape design rules. Therefore, the neural network model can identify the outdoor landscape image to obtain the association relationship of the outdoor landscape objects, so as to determine the matching degree of the association relationship of the outdoor landscape objects and the mined outdoor landscape design rules (which can be manually set or mined by using a related algorithm), so as to obtain the design identification result of the outdoor landscape three-dimensional model, so as to determine whether the current outdoor landscape three-dimensional model meets the design requirement. If the matching degree of the association relationship of the outdoor landscape objects and the mined outdoor landscape design rules is high, it can be considered that the current outdoor landscape three-dimensional model design meets the design requirement, and the design quality of the outdoor landscape three-dimensional model is high. If the matching degree of the association relationship of the outdoor landscape objects and the mined outdoor landscape design rules is low, it can be considered that the current outdoor landscape three-dimensional model design does not necessarily meet the design requirement, and the design quality of the outdoor landscape three-dimensional model is low.

[0086] Optionally, the association relationship can be a relationship between different outdoor landscape objects, for example, a dependent relationship that a pool and a drainpipe must exist at the same time, or a relationship between attributes or values of different outdoor landscape objects. For example, there is a constraint relationship between the number of seats in a seating area and the shape of a platform. When the shape of the platform is circular or square, the number of seats also changes accordingly. A drainpipe must also be set for a drying area to avoid water accumulation on the ground. The present embodiment is not limited in this regard.

[0087] In this embodiment, the computer device obtains an outdoor landscape image corresponding to the outdoor landscape three-dimensional model, inputs the outdoor landscape image into a preset neural network model to identify the association relationship of the outdoor landscape object, and obtains a design identification result of the outdoor landscape three-dimensional model corresponding to the association relationship of the outdoor landscape object. Since the above neural network model is a neural network obtained by training an image carrying an outdoor landscape label, and the neural network model can mine the rules of various association relationships between different outdoor landscape objects as outdoor landscape design rules by identifying the image carrying the outdoor landscape label, the neural network model can automatically identify the association relationship of the outdoor landscape object from the outdoor landscape image, and compare it with the above outdoor landscape design rules, thereby obtaining the design identification result of the outdoor landscape three-dimensional model representing the matching degree of the association relationship of the outdoor landscape object and the outdoor landscape design rules, thereby avoiding the problems of low adjustment efficiency, low accuracy, incomplete adjustment, and high artificial learning cost caused by traditional manual adjustment and optimization of the outdoor landscape three-dimensional model according to experience. This method can automatically identify the association relationship of the outdoor landscape object by using the identification model, and obtain the design identification result of the outdoor landscape three-dimensional model based on the mined outdoor landscape design rules, greatly improving the identification efficiency, and more accurately and comprehensively identifying the current association relationship of the outdoor landscape object, greatly improving the accuracy, and reducing the artificial learning cost, greatly saving time and manpower.

[0088] Optionally, on the basis of the above embodiment, a possible implementation of the above step S12 can be as shown in the following table: Figure 2

[0089] Step S121, inputting the outdoor landscape image into the preset neural network model to identify the association relationship of the outdoor landscape object, and obtaining an association relationship list of the outdoor landscape object.

[0090] Specifically, the computer device inputs the above outdoor landscape image into the above neural network model, which can identify a plurality of outdoor landscape objects and a plurality of different association relationships of different outdoor landscape objects, and then generates an association relationship list from these association relationships.

[0091] Step S122, according to each association relationship in the association relationship list and the outdoor landscape design rule, obtaining an outdoor landscape object association degree design identification result.

[0092] ​Specifically, the computer device judges each of the association relationships in the association relationship list according to the outdoor landscape three-dimensional model rules one by one, obtains the matching degree between each association relationship and the corresponding outdoor landscape design rule, and takes the matching degree as the outdoor landscape object correlation degree design recognition result. The matching degree can include multiple matching degree levels such as complete matching, partial matching, and complete non-matching. Alternatively, the matching degree can be quantitatively represented, which is not limited in this embodiment. For example, the outdoor landscape design rule includes a seat area on a southern balcony and a sunshade umbrella set in the seat area, the association relationship obtained by the computer device includes a seat area with the attribute of a balcony object, but the seat area does not have a sunshade umbrella, and the computer device can determine that the current seat area design does not meet the use needs, and the obtained outdoor landscape object correlation degree design recognition result can include that the sunshade umbrella state corresponding to the seat area on the southern balcony is default; for example, the outdoor landscape design rule includes that the balcony is a water space, the surface layer is made into a bottom, and the drop plate processing is performed, but the water space is only made into a bottom, and the drop plate processing is not performed, and the computer device can determine that the current balcony design does not meet the use needs, and the obtained outdoor landscape object correlation degree design recognition result can include that the drop plate processing state corresponding to the balcony as a water space is default.

[0093] Step S123, prompting according to the outdoor landscape object correlation degree design recognition result.

[0094] Specifically, when the association relationship in the outdoor landscape object correlation degree design recognition result and the outdoor landscape design rule are completely unmatched, the computer device can output a prompt information of non-matching to prompt the user to pay attention to the outdoor landscape object involved in the association relationship, for example, the outdoor landscape object correlation degree design recognition result can include that the sunshade umbrella corresponding to the seat area on the southern balcony is default or the drop plate processing state corresponding to the balcony as a water space is default, and the computer device outputs that the seat area needs to set a sunshade umbrella or the balcony needs to be made into a drop plate. When the association relationship in the water correlation degree design recognition result and the outdoor landscape design rule are partially matched, the computer device can output different levels of risk prompt information, for example, the current water pipe is a illegal area underground pipe, and the drainage or waterproof measures are not fully prepared, which causes water accumulation when the water pipe is opened or used to water flowers, and therefore, the computer device can output prompt information whether the drop plate processing is needed.

[0095] In this embodiment, the computer device inputs the outdoor landscape image into the neural network model to identify the correlation relationship of the outdoor landscape object, obtains a correlation relationship list of the outdoor landscape object, and obtains each correlation relationship in the correlation relationship list and the outdoor landscape design rule to obtain an outdoor landscape object correlation degree design identification result, and then prompts according to the outdoor landscape object correlation degree design identification result. The method can obtain the outdoor landscape object correlation degree design identification result based on the matching degree of each correlation relationship and the outdoor landscape design rule, and prompt based on the outdoor landscape object correlation degree design identification result, thereby realizing the prompt of the correlation relationship that does not match the outdoor landscape design rule, realizing intelligent auxiliary design, and thus making the design of the outdoor landscape three-dimensional model more accurate, reasonable, and intelligent.

[0096] Optionally, the above-mentioned correlation relationship can include one or more combinations of the following: the correlation relationship between the position area and the water accumulation and leakage prevention measures of the drainage pipeline, the correlation relationship between the tile specification and the anti-leakage measures when the pipeline is filled with water, the correlation relationship between the structure shape and the enclosure component, the correlation relationship between the temperature area and the position of the drainage pipeline, and the correlation relationship between the position of the drainage pipeline and the drying area. The outdoor landscape design rule includes: the correlation relationship support degree rule of the outdoor landscape object, the correlation relationship confidence degree rule of the outdoor landscape object, and the correlation relationship promotion degree rule of the outdoor landscape object. For example, the south and the north are different temperature areas, and the temperature in the south is higher than that in the north. Therefore, by mining the outdoor landscape images of the north and the south, the measures of deep burying, cold prevention, and frost prevention of the drainage pipeline of the outdoor landscape in the north have a correlation relationship, the drainage well and the heating pipeline have a correlation relationship of well merging, and the drainage pipeline well of the outdoor landscape in the south is separately arranged. When the tile specification of the outdoor landscape in the north is floor tile, water seepage and frost cracking of the waterproof layer will occur, resulting in water seepage or leakage when the pipeline is filled with water, so the outdoor landscape tile specification and the anti-leakage measures when the pipeline is filled with water have a correlation relationship. The flower and plant area of the outdoor landscape in the north is suitable for some cold-resistant plants such as camellia and rhododendron, or some non-cold-resistant plants such as succulent plants, asparagus fern, and fortune tree, which need to be set up in the outdoor landscape. The correlation relationship of the cold-proof measures. The various water pools buried in the outdoor landscape in the north have frost prevention measures, the water pool is covered with soil for heat preservation, and the water pool is covered with soil to a depth below the freezing line, the water pool manhole, and the water outlet are set up with a heat preservation well mouth and a wooden heat preservation cover. For example, the diameter of the water pool pipeline is X cm, the material is W, and the drainage flow that can be borne is A, so the water pool pipeline specification, the diameter X cm, the material W, and the drainage flow A have a correlation relationship; or the water appliance, such as the faucet and the drainage pipeline specification, has a correlation relationship. For example, the laying method of the drainage pipeline, the correlation relationship of the drainage pipeline laid along the top, along the wall, and along the ground. The correlation relationship between the water appliance and the drainage flow demand and the drainage pipeline specification.

[0097] Optionally, the total design flow and pressure can also be obtained by aggregating the flow in the water flow direction in sections in the reverse direction, and whether the requirements are met can be determined according to the calculation results. For example, the relationship between the position of the drain pipe and the drying area can be that, when the drying area is arranged on the balcony, in order to avoid ground water, the drain pipe can be arranged corresponding to the drying area, so that the dripping of the clothes in the drying area is not easy to form water accumulation. For another example, the relationship between the balcony structure and the enclosure component can be that, when the balcony is a special-shaped structure, the ceiling of the balcony can be a grape trellis ceiling, a colored glass ceiling, etc.

[0098] Optionally, before the step S11, the method can further include: obtaining an outdoor landscape three-dimensional model in the design model; and performing three-dimensional to two-dimensional conversion on the outdoor landscape three-dimensional model to obtain a plurality of outdoor landscape images, the outdoor landscape images being different in representation angle.

[0099] Specifically, the computer device can traverse the design model, obtain an outdoor landscape three-dimensional model according to the outdoor landscape object obtained by the traversal, and then perform three-dimensional to two-dimensional conversion on the outdoor landscape three-dimensional model to obtain a plurality of two-dimensional outdoor landscape images. Optionally, the computer device can further screen the obtained two-dimensional outdoor landscape images, delete some unclear images, or splice some related outdoor landscape images to obtain a complete image of the outdoor landscape object.

[0100] In this embodiment, the computer device obtains an outdoor landscape three-dimensional model in the design model, performs three-dimensional to two-dimensional conversion on the outdoor landscape three-dimensional model to obtain a plurality of outdoor landscape images, so that the outdoor landscape images of the outdoor landscape three-dimensional model can be automatically obtained from the complete design model, and therefore the degree of automation is higher, and the recognition efficiency and accuracy are further improved.

[0101] Optionally, on the basis of each of the above embodiments, the obtaining process of the neural network model is a learning process of an outdoor landscape design rule in the outdoor landscape three-dimensional model, as shown in Figure 3 , including:

[0102] In step S131, an outdoor landscape sample model with an outdoor landscape object pre-labeled is converted from three-dimensional to two-dimensional to obtain a plurality of outdoor landscape initial sample images.

[0103] Specifically, the computer device can obtain a plurality of outdoor landscape sample models as learning samples, and labels of outdoor landscape objects are labeled in the outdoor landscape sample models, for example, water pools, water features, face bricks, flowers, plants, drain pipes, water intake points, activity platforms, seat areas, barbecue leisure areas, plant decorations, window sill plates, etc. in the outdoor landscape sample models are labeled. The computer device converts the above-mentioned outdoor landscape sample models pre-labeled with outdoor landscape objects from three dimensions to two dimensions to obtain a plurality of outdoor landscape initial sample images. Optionally, labelme or via or other labeling tools can be used to label water pools, flower beds, water features, face bricks, etc. of the outdoor landscape.

[0104] Step S132, pre-processing the outdoor landscape initial sample image to obtain an outdoor landscape sample image.

[0105] The pre-processing includes at least one of screening and splicing.

[0106] Specifically, the computer device can pre-process the above-mentioned outdoor landscape initial sample image, which can include data cleaning, screening, etc., delete unclear or unrecognizable images, for example, images with only one seat leg, images that cannot identify associated objects, can also include splicing, for example, splicing different images of a unified pipeline to obtain a complete drain pipe pipeline, or splicing images of drain pipes and water pools, etc. The initial sample image can also be normalized to obtain outdoor landscape sample images with consistent size or pixels. The specific process of screening and splicing included in the above-mentioned pre-processing is not limited in this embodiment.

[0107] Step S133, inputting the outdoor landscape sample image into an initial recognition model, extracting an associated relationship based on a classification mode of a decision tree, and obtaining a frequent item set to obtain the neural network model including outdoor landscape design rules.

[0108] It should be noted that the above-mentioned outdoor landscape design rules have three measurement indexes: support, confidence, and lift. Support (Support): the support of X→Y represents the probability of {X,Y} appearing in the total item set; Confidence (Confidence): the confidence of X→Y represents the probability of Y being derived from the rule X→Y under the condition that X occurs, i.e. the probability of Y existing under the condition that X exists. Generally, confidence above 95% is considered as a rule that must be met, and confidence between 60-70% is considered as a rule that should be met as much as possible. Lift (Lift), the lift of X→Y represents the probability of containing Y under the condition of containing X, which is the ratio of P(Y|X) / P(Y).

[0109] Optionally, the above-mentioned association relationship can generally include the following four categories: Boolean association rules, quantitative rules, single-dimensional and multi-dimensional rules, single-layer and multi-layer association rules. For example, the Boolean association rule can be that Y must exist in the case of X, that is, if Y does not exist, it is considered not to meet the requirements; the quantitative rule can be a quantitative value rule of a certain index of an outdoor landscape object, for example, the diameter of the drain pipe is more than 3 cm and less than 25 cm; the single-dimensional and multi-dimensional rule can be a two-dimensional association relationship between the drain pipe and the pool and the waterscape respectively; the single-layer and multi-layer association rule can be a single-layer association rule between the total water inlet pipe and the next level drain pipe, and a multi-layer association rule for the end drain pipe.

[0110] Specifically, the computer device inputs the above-mentioned outdoor landscape sample image into an initial recognition model, which can perform hierarchical classification on the outdoor landscape objects and other entity objects in the above-mentioned outdoor landscape sample image based on a decision tree, and extract the association relationship of different outdoor landscape objects based on the hierarchical structure, and then obtain a frequent item set according to the association relationship, and obtain the neural network model including the outdoor landscape design rule.

[0111] Classification using a decision tree can include a tree structure established by a series of rules for classification and prediction. The topmost node of the decision tree is the root node, each node forms a new node downward, and the leaf node has no branch. Each leaf node corresponds to a decision, that is, a possible classification result. When calculating, the root node of the decision tree is traversed from top to bottom, each node corresponds to an attribute, and different branches are selected for different attribute values. Finally, the classification is completed by reaching the leaf node. The decision tree algorithm has a simple structure, high classification accuracy, and good robustness in processing noisy data.

[0112] Optionally, the object detection and segmentation technology of mast rcnn tensorflow can be used for deep learning of the outdoor landscape object. According to the labeled image of the outdoor landscape object, the pool and the drain pipe and the like are identified to obtain a trained model. For example, eighty thousand jpg labeled pictures can be selected for training, and the recognition rate is 98.2%.

[0113] Optionally, the configuration environment for implementing the training process can refer to the following: GPU RTX2080i, memory ddr4 32G, CPU 2.8GHZ, 16 cores, 32 threads, framework: tensorflow-1.12.0, GPU's cuda cuda10.1, cudnn7.3, language: python.

[0114] In this embodiment, a computer device can convert a sample outdoor landscape model pre-labeled with outdoor landscape objects from 3D to 2D, obtaining multiple initial sample images of the outdoor landscape. These initial sample images are then preprocessed to obtain sample images. Finally, the sample images are input into an initial recognition model, where associations are extracted based on a decision tree classification approach, and frequent item sets are obtained to produce a recognition model that includes outdoor landscape design rules. By mining and learning the associations within the 3D outdoor landscape model, this method can extract more implicit patterns and derive more comprehensive outdoor landscape design rules. This method enables comprehensive and effective recognition within the landscape design recognition process based on a neural network model, complementing existing manual design methods and further improving the design quality of 3D outdoor landscape models.

[0115] In order to describe the technical solution provided by this application in more detail, this application is described in detail with a specific embodiment, such as Figure 4 Shown, including:

[0116] Step S21: converting the outdoor landscape sample model pre-labeled with outdoor landscape objects from three dimensions to two dimensions to obtain a plurality of outdoor landscape initial sample images;

[0117] Step S22: pre-processing the outdoor landscape initial sample image to obtain an outdoor landscape sample image; wherein the pre-processing includes at least one of screening and splicing;

[0118] Step S23: inputting the outdoor landscape sample image into the initial recognition model, performing association extraction based on the decision tree classification method, and obtaining frequent item sets to obtain a neural network model including outdoor landscape design rules;

[0119] Step S24: obtaining a three-dimensional model of the outdoor landscape in the design model;

[0120] Step S25: performing a 3D to 2D conversion on the outdoor landscape 3D model to obtain a plurality of outdoor landscape images, wherein the outdoor landscape images are represented from different angles;

[0121] Step S26: inputting the outdoor landscape image into the preset neural network model to identify the association relationship of outdoor landscape objects, and obtaining a list of association relationships of the outdoor landscape objects;

[0122] Step S27, obtaining an outdoor landscape object correlation degree design identification result according to each correlation in the correlation list and outdoor landscape design rules; wherein the correlation includes at least one of a position area and a water accumulation and leakage prevention measure of a drainage pipeline, a face brick specification and a water leakage prevention measure when water flows into and out of a pipeline, a flower and plant type and a temperature area, a temperature area and a drainage pipeline position, a drainage pipeline and a wall, a drainage pipeline position and a drying area, and a balcony structure and a surrounding component;

[0123] Step S28, prompting according to the design identification result of the outdoor landscape object correlation degree.

[0124] The detailed description and technical effects of the steps in the embodiment can be referred to the foregoing embodiments, which will not be described here.

[0125] It should be understood that, although Figures 1-4 the steps in the flowchart are displayed in sequence according to the arrows, these steps are not necessarily executed in sequence according to the arrows. Unless otherwise specified herein, the execution of these steps has no strict sequence limitation, and these steps can be executed in other sequences. Moreover, Figures 1-4 at least part of the steps in the flowchart can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these sub-steps or stages is not necessarily sequential, but can be executed in rotation or alternation with other steps or sub-steps or stages of other steps.

[0126] In one embodiment, as shown in Figure 5 , a landscape design identification device based on a neural network model is provided, comprising:

[0127] A first acquisition module 11 is configured to acquire outdoor landscape images corresponding to an outdoor landscape three-dimensional model; wherein the outdoor landscape images are images obtained by converting the outdoor landscape three-dimensional model, and each of the outdoor landscape images includes at least two kinds of outdoor landscape objects;

[0128] An identification module 12 is configured to input the outdoor landscape images into a preset neural network model to identify the correlation of outdoor landscape objects in the outdoor landscape three-dimensional model, and obtain a design identification result of the correlation of outdoor landscape in the outdoor landscape three-dimensional model;

[0129] The design recognition result is used to represent a matching degree of the association relationship of the outdoor landscape object and an outdoor landscape design rule, and the neural network model is a neural network obtained by rule mining using an image carrying an outdoor landscape object label.

[0130] The identification module 12 specifically can further include a first processing unit, a second processing unit, and a prompting unit.

[0131] Specifically, the first processing unit is configured to input the outdoor landscape image into the neural network model to identify the association relationship of the outdoor landscape object, and obtain an association relationship list of the outdoor landscape object; the second processing unit is configured to obtain an outdoor landscape object association degree design recognition result according to each association relationship in the association relationship list and an outdoor landscape design rule; and the prompting unit is configured to prompt according to the outdoor landscape object association degree design recognition result.

[0132] The association relationship includes at least one of an association relationship of a position area and a water accumulation and leakage prevention measure of a drainage pipeline, an association relationship of a face brick specification and a water leakage prevention measure when water flows into and out of a pipeline, an association relationship of a flower and plant type and a temperature area, an association relationship of a temperature area and a drainage pipeline position, an association relationship of drainage pipeline laying and a wall, an association relationship of a drainage pipeline position and a drying area, and an association relationship of a balcony structure and an enclosure component.

[0133] The landscape design recognition device based on the neural network model can further include a second acquisition model and a third acquisition model.

[0134] Specifically, the second acquisition model is configured to acquire an outdoor landscape three-dimensional model in a design model; and the third acquisition model is configured to perform three-dimensional to two-dimensional conversion on the outdoor landscape three-dimensional model to obtain a plurality of outdoor landscape images, and the outdoor landscape images represent different angles.

[0135] The landscape design recognition device based on the neural network model can further include a fourth acquisition model, a fifth acquisition model, and a sixth acquisition model.

[0136] Specifically, the fourth acquisition model is configured to perform three-dimensional to two-dimensional conversion on an outdoor landscape sample model pre-labeled with an outdoor landscape object to obtain a plurality of outdoor landscape initial sample images; the fifth acquisition model is configured to pre-process the outdoor landscape initial sample images to obtain outdoor landscape sample images; the pre-processing includes at least one of screening and splicing; and the sixth acquisition model is configured to input the outdoor landscape sample images into an initial recognition model, extract an association relationship based on a decision tree classification method, and obtain a frequent item set to obtain the neural network model including an outdoor landscape design rule.

[0137] In one embodiment, as shown in Figure 6 A landscape design recognition device based on a neural network model is provided, comprising:

[0138] The training module 41 is configured to convert an outdoor landscape sample model with pre-labeled outdoor landscape objects into a two-dimensional form to obtain a plurality of outdoor landscape initial sample images, pre-process the outdoor landscape initial sample images to obtain outdoor landscape sample images, and input the outdoor landscape sample images into an initial recognition model to extract correlation based on a decision tree classification method and obtain a frequent item set to obtain a neural network model including outdoor landscape design rules. The pre-processing includes at least one of screening and splicing.

[0139] The processing module 42 is configured to obtain an outdoor landscape three-dimensional model in a design model, convert the outdoor landscape three-dimensional model into a two-dimensional form to obtain a plurality of outdoor landscape images with different representation angles, input the outdoor landscape images into the neural network model to recognize the correlation of the outdoor landscape objects, obtain a correlation list of the outdoor landscape objects, and obtain an outdoor landscape object correlation degree design recognition result according to each correlation in the correlation list and the outdoor landscape design rules, and prompt according to the outdoor landscape object correlation degree design recognition result.

[0140] The design recognition result is used to represent the matching degree of the correlation of the outdoor landscape objects and the outdoor landscape design rules. The correlation includes at least one of the correlation of the location area and the water accumulation and leakage prevention measures of the drainage pipeline, the correlation of the outdoor landscape tile specification and the anti-leakage measures when the pipeline is filled with water, the correlation of the flower and plant species and the temperature area, the correlation of the temperature area and the drainage pipeline position, the correlation of the drainage pipeline laying and the wall, the correlation of the drainage pipeline position and the drying area, and the correlation of the balcony structure and the enclosure component.

[0141] The specific limitations of the landscape design recognition device based on the neural network model can be referred to the limitations of the landscape design recognition method based on the neural network model in the foregoing, which will not be repeated here. Each module in the landscape design recognition device based on the neural network model can be realized by software, hardware, or a combination thereof.

[0142] Each module can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0143] The landscape design recognition method based on the neural network model provided by the embodiments can be applied to Figure 7The computer device shown in the figure. The computer device includes a processor, a memory, a network interface, a database, a display screen and an input device connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium, an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the identification model in the following embodiments, and the specific description of the identification model is described in the following embodiments. The network interface of the computer device can be used to communicate with other devices outside through network connection. Optionally, the computer device can be a server, can be a desktop computer, can be a personal digital assistant, and can also be other terminal devices, such as tablet computers, mobile phones and the like, and can also be a cloud or remote server, and the specific form of the computer device is not limited by the embodiments of the application. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc. Of course, the input device and the display screen can not belong to the computer device, but can be external devices of the computer device.

[0144] Those skilled in the art can understand that, Figure 7 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0145] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the following steps:

[0146] Obtaining an outdoor landscape image corresponding to the outdoor landscape three-dimensional model; wherein the outdoor landscape image is an image obtained by converting the outdoor landscape three-dimensional model, and each outdoor landscape image includes at least two kinds of outdoor landscape objects;

[0147] Inputting the outdoor landscape image into a preset neural network model to identify the association relationship of the outdoor landscape objects in the outdoor landscape three-dimensional model, and obtaining a design identification result of the association relationship of the outdoor landscape in the outdoor landscape three-dimensional model;

[0148] The design recognition result is used to represent the matching degree of the association relationship of the outdoor landscape object and the outdoor landscape design rule, and the neural network model is a neural network obtained by rule mining using an image carrying an outdoor landscape object label.

[0149] It should be clear that the process of the processor executing the computer program in the embodiments of the application is consistent with the execution process of each step in the above method. For details, refer to the description in the above.

[0150] In one embodiment, a computer device is provided, comprising a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the following steps:

[0151] An outdoor landscape sample model pre-labeled with an outdoor landscape object is converted from three dimensions to two dimensions to obtain a plurality of outdoor landscape initial sample images;

[0152] The outdoor landscape initial sample images are pre-processed to obtain outdoor landscape sample images; wherein the pre-processing includes at least one of screening and splicing;

[0153] The outdoor landscape sample images are input into an initial recognition model, the association relationship is extracted based on a decision tree classification method, and a frequent item set is obtained to obtain a neural network model including outdoor landscape design rules;

[0154] An outdoor landscape three-dimensional model in the design model is obtained;

[0155] The outdoor landscape three-dimensional model is converted from three dimensions to two dimensions to obtain a plurality of outdoor landscape images, and the representation angles of the outdoor landscape images are different;

[0156] The outdoor landscape images are input into the preset neural network model to identify the association relationship of the outdoor landscape object, and an association relationship list of the outdoor landscape object is obtained;

[0157] According to each association relationship in the association relationship list and an outdoor landscape design rule, an outdoor landscape object association degree design recognition result is obtained; wherein the association relationship includes at least one of the association relationship of the position area and the water accumulation and leakage prevention measure of the drainage pipe, the association relationship of the outdoor landscape tile specification and the anti-leakage measure when the pipe is filled with water, the association relationship of the flower and plant species and the temperature area, the association relationship of the temperature area and the drainage pipe position, the association relationship of the drainage pipe laying and the wall, the association relationship of the drainage pipe position and the drying area, and the association relationship of the balcony structure and the enclosure component.

[0158] According to the outdoor landscape object association degree design recognition result, a prompt is given.

[0159] It should be clear that the process of the processor executing the computer program in the embodiments of the application is consistent with the execution process of each step in the above method, and specific reference can be made to the description in the above.

[0160] In one embodiment, a computer readable storage medium is provided, and the computer readable storage medium has stored thereon a computer program, and the computer program is executed by a processor to implement the following steps:

[0161] An outdoor landscape image corresponding to the outdoor landscape three-dimensional model is acquired; wherein the outdoor landscape image is an image obtained by conversion of the outdoor landscape three-dimensional model, and each of the outdoor landscape images includes at least two kinds of outdoor landscape objects related;

[0162] The outdoor landscape image is input into a preset neural network model to identify the association relationship of the outdoor landscape objects in the outdoor landscape three-dimensional model, and a design identification result of the association relationship of the outdoor landscape in the outdoor landscape three-dimensional model is obtained;

[0163] The design identification result is used to represent the matching degree of the association relationship of the outdoor landscape objects and the outdoor landscape design rule, and the neural network model is a neural network obtained by rule mining using an image carrying an outdoor landscape object label.

[0164] It should be clear that the process of the processor executing the computer program in the embodiments of the application is consistent with the execution process of each step in the above method, and specific reference can be made to the description in the above.

[0165] In one embodiment, a computer readable storage medium is provided, and the computer readable storage medium has stored thereon a computer program, and the computer program is executed by a processor to implement the following steps:

[0166] An outdoor landscape sample model with pre-labeled outdoor landscape objects is converted from three-dimensional to two-dimensional to obtain a plurality of outdoor landscape initial sample images;

[0167] The outdoor landscape initial sample images are preprocessed to obtain outdoor landscape sample images; wherein the preprocessing includes at least one of screening and splicing;

[0168] The outdoor landscape sample images are input into an initial identification model, the association relationship is extracted based on a decision tree classification method, and a frequent item set is acquired to obtain a neural network model including an outdoor landscape design rule;

[0169] An outdoor landscape three-dimensional model in a design model is acquired;

[0170] The outdoor landscape three-dimensional model is converted from three-dimensional to two-dimensional to obtain a plurality of the outdoor landscape images, and the representation angles of the outdoor landscape images are different;

[0171] Inputting the outdoor landscape image into the preset neural network model to identify the association relationship of outdoor landscape objects, and obtaining a list of association relationships of the outdoor landscape objects;

[0172] Obtaining an outdoor landscape object correlation design recognition result based on each correlation in the correlation list and the outdoor landscape design rules; wherein the correlation includes at least one of: a correlation between a location area and water accumulation and leakage prevention measures for drainage pipes, a correlation between outdoor landscape tile specifications and leakage prevention measures for water inlet and outlet pipes, a correlation between flower and plant species and temperature zones, a correlation between temperature zones and drainage pipe locations, a correlation between drainage pipe laying and walls, a correlation between drainage pipe locations and drying areas, and a correlation between balcony structures and enclosure components;

[0173] Prompts are given based on the design recognition results of the outdoor landscape object relevance.

[0174] It should be clear that the process of executing the computer program by the processor in the embodiment of the present application is consistent with the execution process of each step in the above method. For details, please refer to the description above.

[0175] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0176] Any combination of the technical features in the above embodiments can be made, and for the sake of brevity, not all possible combinations are described above, however, as long as the combination of the technical features does not exist in contradiction, it shall be considered within the scope of the present disclosure.

[0177] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it shall not be understood as a limitation on the patent scope of the present application. It shall be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these shall be within the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A landscape design recognition method based on a neural network model, characterized in that: The method comprises: Acquire an outdoor landscape image corresponding to the outdoor landscape three-dimensional model; wherein the outdoor landscape image is an image obtained by converting the outdoor landscape three-dimensional model, and each of the outdoor landscape images includes at least two related outdoor landscape objects; Inputting the outdoor landscape image into a preset neural network model to perform association relationship recognition of outdoor landscape objects in the outdoor landscape three-dimensional model, and obtaining a design recognition result of the association relationship of the outdoor landscape in the outdoor landscape three-dimensional model; The design recognition result is used to characterize the matching degree between the association relationship of the outdoor landscape objects and the outdoor landscape design rules, and the neural network model is a neural network obtained by mining rules using images carrying outdoor landscape object labels.

2. The method according to claim 1, wherein The step of inputting the outdoor landscape image into a preset neural network model to perform association relationship recognition of outdoor landscape objects in the outdoor landscape three-dimensional model, and obtaining a design recognition result of the association relationship of the outdoor landscape in the outdoor landscape three-dimensional model, includes: Inputting the outdoor landscape image into the preset neural network model to identify the association relationship of outdoor landscape objects, and obtaining a list of association relationships of the outdoor landscape objects; Obtaining an outdoor landscape object correlation design recognition result according to each correlation relationship in the correlation relationship list and the outdoor landscape design rule; Prompts are given based on the design recognition results of the outdoor landscape object relevance.

3. The method according to claim 1, wherein The association relationships include: the association relationship between the location area and the anti-water accumulation and leakage measures of the drainage pipe, the association relationship between the surface brick specifications and the anti-leakage measures during water inlet and outlet of the pipe, the association relationship between the flower and plant species and the temperature zone, the association relationship between the temperature zone and the location of the drainage pipe, the association relationship between the laying of the drainage pipe and the wall, the association relationship between the location of the drainage pipe and the drying area, and the association relationship between the balcony structure and the enclosure components.

4. The method according to claim 1, wherein Before obtaining the outdoor landscape image corresponding to the outdoor landscape three-dimensional model, the method further includes: Obtain the three-dimensional model of the outdoor landscape in the design model; The outdoor landscape three-dimensional model is converted from three-dimensional to two-dimensional to obtain a plurality of outdoor landscape images, where the representation angles of the outdoor landscape images are different.

5. The method according to claim 1, wherein The acquisition process of the neural network model includes: Converting an outdoor landscape sample model with pre-labeled outdoor landscape objects into a three-dimensional to two-dimensional model to obtain a plurality of outdoor landscape initial sample images; Preprocessing the outdoor landscape initial sample image to obtain an outdoor landscape sample image; wherein the preprocessing includes at least one of screening and splicing; The outdoor landscape sample image is input into the initial recognition model, and the association relationship is extracted based on the classification method of the decision tree, and the frequent item sets are obtained to obtain the neural network model including the outdoor landscape design rules.

6. A landscape design recognition method based on a neural network model, characterized in that: The method comprises: Converting an outdoor landscape sample model with pre-labeled outdoor landscape objects into a three-dimensional to two-dimensional model to obtain a plurality of outdoor landscape initial sample images; Preprocessing the outdoor landscape initial sample image to obtain an outdoor landscape sample image; wherein the preprocessing includes at least one of screening and splicing; Inputting the outdoor landscape sample image into the initial recognition model, performing association relationship extraction based on the decision tree classification method, and obtaining frequent item sets to obtain a neural network model including outdoor landscape design rules; Obtain the three-dimensional model of the outdoor landscape in the design model; Performing a three-dimensional to two-dimensional conversion on the three-dimensional model of the outdoor landscape to obtain a plurality of outdoor landscape images, wherein the representation angles of the outdoor landscape images are different; Inputting the outdoor landscape image into the neural network model to identify the association relationship of outdoor landscape objects, and obtaining a list of association relationships of the outdoor landscape objects; Obtaining an outdoor landscape object correlation design recognition result based on each correlation in the correlation list and the outdoor landscape design rules; wherein the correlation includes at least one of: a correlation between a location area and water accumulation and leakage prevention measures for drainage pipes, a correlation between outdoor landscape tile specifications and leakage prevention measures for water inlet and outlet pipes, a correlation between flower and plant species and temperature zones, a correlation between temperature zones and drainage pipe locations, a correlation between drainage pipe laying and walls, a correlation between drainage pipe locations and drying areas, and a correlation between balcony structures and enclosure components; Prompts are given based on the design recognition results of the outdoor landscape object relevance.

7. A landscape design recognition device based on a neural network model, characterized in that: The device comprises: A first acquisition module is configured to acquire an outdoor landscape image corresponding to the outdoor landscape three-dimensional model; wherein the outdoor landscape image is an image obtained by converting the outdoor landscape three-dimensional model, and each outdoor landscape image includes at least two related outdoor landscape objects; a recognition module, configured to input the outdoor landscape image into a preset neural network model to perform association recognition of outdoor landscape objects in the outdoor landscape three-dimensional model, and obtain a design recognition result of the association relationship of the outdoor landscape in the outdoor landscape three-dimensional model; The design recognition result is used to characterize the matching degree between the association relationship of the outdoor landscape objects and the outdoor landscape design rules, and the neural network model is a neural network obtained by mining rules using images carrying outdoor landscape object labels.

8. A landscape design recognition device based on a neural network model, characterized in that: The device comprises: a training module for converting an outdoor landscape sample model pre-labeled with outdoor landscape objects from three dimensions to two dimensions to obtain a plurality of outdoor landscape initial sample images, preprocessing the outdoor landscape initial sample images to obtain outdoor landscape sample images, and inputting the outdoor landscape sample images into an initial recognition model to extract association relationships based on a decision tree classification method, obtain frequent item sets, and obtain a neural network model including outdoor landscape design rules; Wherein, the pre-processing includes at least one of screening and splicing; a processing module configured to obtain a three-dimensional outdoor landscape model in the design model, perform a three-dimensional to two-dimensional conversion on the three-dimensional outdoor landscape model to obtain a plurality of outdoor landscape images, input the outdoor landscape images into the neural network model to perform association recognition of outdoor landscape objects, obtain an association list of the outdoor landscape objects, obtain a design recognition result of an association degree of the outdoor landscape objects based on each association in the association list and an outdoor landscape design rule, and provide a prompt based on the design recognition result of the association degree of the outdoor landscape objects; The design recognition result is used to characterize the matching degree between the association relationship of the outdoor landscape object and the outdoor landscape design rules; the association relationship includes at least one of the following: the association relationship between the location area and the anti-water accumulation and leakage measures of the drainage pipe, the association relationship between the specifications of the outdoor landscape tiles and the anti-leakage measures during the water inlet and outlet of the pipe, the association relationship between the types of flowers and plants and the temperature zone, the association relationship between the temperature zone and the location of the drainage pipe, the association relationship between the laying of the drainage pipe and the wall, the association relationship between the location of the drainage pipe and the drying area, and the association relationship between the balcony structure and the enclosure component.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images