An artificial intelligence-based building plan generation system and method

Abstract

The present invention discloses an artificial intelligence-based building plan generation system and method, which relates to the field of artificial intelligence technology. The system includes a preprocessing module, a picture classification module, an input analysis module, and a plan generation module. The preprocessing module is used to collect building plans and building pictures, determine the elements of the building plan, and manually label the elements and element values of the pictures. The picture classification module is used to perform preliminary classification and reclassification of the pictures according to the elements, and cluster the pictures of the same classification to determine the representative picture. The input analysis module is used to analyze the user input. The plan generation module is used to generate a building plan according to the user input and feedback, and provide a selection and judgment mechanism to optimize the plan generation. The present invention also proposes a method for implementing the system. The present invention can effectively improve the situation in the prior art where it is difficult to effectively integrate text analysis and picture processing, and the picture and data classification standards are relatively single.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and particularly to an artificial-intelligence-based building plan generation system and method. Background Art

[0002] Artificial intelligence is widely applied in building plan generation. It can quickly analyze factors such as site conditions and functional requirements based on a large amount of data. Through algorithms, it can generate various building layout and form plans, providing rich creative inspirations for designers. For example, it can accurately plan spaces in projects with complex terrains. At the same time, artificial intelligence can simulate building performance, such as lighting, ventilation, and energy consumption, to assist in optimizing the design. It can also combine with virtual reality technology to enable designers to experience the plan effects immersively. In terms of collaborative design, it promotes communication among different professionals, improves design efficiency and quality, and drives the transformation of building plans from traditional manual conceptions to intelligent-assisted generation.

[0003] In the existing building plan generation technologies, when classifying relevant pictures or data, the classification criteria or methods adopted are relatively single, resulting in a relatively one-sided integration of picture and text elements and being unable to meet the classification requirements of complex objects such as building plans in different dimensions; in addition, some existing technologies are based on initial settings or fixed patterns and lack the ability to dynamically adjust and generate plans according to user feedback. Summary of the Invention

[0004] The purpose of the present invention is to provide an artificial-intelligence-based building plan generation system and method to solve the problems raised in the existing technologies.

[0005] To achieve the above purpose, the present invention provides the following technical solution: An artificial-intelligence-based building plan generation system, which includes a preprocessing module, a picture classification module, an input analysis module, and a plan generation module;

[0006] The preprocessing module is used for collecting building plans and building pictures, determining building plan elements, and manually annotating the elements and element values of the pictures; the picture classification module is used for preliminarily classifying and reclassifying pictures according to the elements, storing the classified picture information, and clustering the pictures of the same classification to determine a representative picture; the input analysis module is used for analyzing user input; the plan generation module is used for generating building plans according to user input and feedback and providing a selection and judgment mechanism to optimize plan generation;

[0007] The output end of the preprocessing module is connected to the input end of the picture classification module; the output end of the picture classification module is connected to the input ends of the input analysis module and the plan generation module; the output end of the input analysis module is connected to the input end of the plan generation module.

[0008] The preprocessing module includes a solution collection unit, a factor determination unit, a picture collection unit, and a picture annotation unit;

[0009] The solution collection unit is used to collect building solutions; the factor determination unit is used to determine building solution factors according to the collected building solutions; the picture collection unit is used to collect building pictures; the picture annotation unit is used to manually annotate picture factors and factor values;

[0010] The output end of the solution collection unit is connected to the input end of the factor determination unit; the output end of the factor determination unit is connected to the input end of the picture collection unit; the output end of the picture collection unit is connected to the input end of the picture annotation unit; the output end of the picture annotation unit is connected to the input end of the picture classification module.

[0011] The picture classification module includes a picture classification unit, a type determination unit, a range division unit, a database unit, a picture clustering unit, and a representative picture determination unit;

[0012] The picture classification unit is used to classify pictures according to factors; the type determination unit is used to determine the category of factor values according to the picture annotation situation; the range division unit is used to divide numerical factor values according to the manual annotation situation; the database unit is used to store the picture classification situation classified according to categorical and numerical factor values; the picture clustering unit is used to cluster pictures of the same classification; the representative picture determination unit is used to select a random picture from the category with the largest number of pictures after picture clustering as the representative picture;

[0013] The output end of the picture classification unit is connected to the input ends of the type determination unit and the range division unit; the output end of the type determination unit is connected to the input end of the database unit; the output end of the range division unit is connected to the input end of the database unit; the output end of the database unit is connected to the input ends of the picture clustering unit and the input analysis module; the output end of the picture clustering unit is connected to the input end of the representative picture determination unit; the output end of the representative picture determination unit is connected to the input end of the solution generation module.

[0014] The input analysis module includes a user input unit and an input analysis unit;

[0015] The user input unit is used for the user to input factors and corresponding factor values; the input analysis unit is used to make an analysis according to the matching situation between the user input and the factors and factor values in the database;

[0016] The output end of the user input unit is connected to the input end of the input analysis unit; the output end of the input analysis unit is connected to the input end of the solution generation module.

[0017] The solution generation module includes an output selection unit, an output judgment unit, a user feedback unit, and a solution generation unit;

[0018] The output selection unit is used to output corresponding pictures for the user to judge whether they meet the expectations when the user input matches the classification situation; the output judgment unit is used to output corresponding pictures for the user to make a choice when the user input does not match the classification situation; the user feedback unit is used for the user to give feedback on the output; the solution generation unit is used to generate a building solution by combining the user input, feedback, and selection;

[0019] The output end of the output selection unit is connected to the input end of the user feedback unit; the output end of the output judgment unit is connected to the input end of the user feedback unit; the output end of the user feedback unit is connected to the input end of the solution generation unit.

[0020] An artificial intelligence-based building solution generation method, which includes the following steps:

[0021] Step1. Collect a large number of building solutions and determine the building solution elements; collect a large number of building pictures, manually determine the building solution elements in the pictures and label the element values;

[0022] Step2. Classify and cluster the pictures according to the elements and element value types, determine the representative pictures, and construct a database to store relevant information;

[0023] Step3. Analyze the user input to obtain the matching situation between the user input and the elements and element values;

[0024] Step4. Provide a selection or judgment mechanism according to the matching situation between the user input and the elements and element values, and generate a building solution by combining the user feedback.

[0025] In step Step1, collect a large number of building solutions to ensure sample diversity;

[0026] Use a word segmentation tool to segment each building solution text; create a stop word list and remove the stop words;

[0027] For each sample, calculate the frequency of occurrence of all the segmented words after removing the stop words in all samples: frequency of occurrence = the number of samples with the segmented word / the total number of samples; set a threshold A, and when the frequency of occurrence of the segmented word is greater than A, set the segmented word as a building solution element;

[0028] Calculate the occurrence frequencies of all word segments after removing stop words in all samples, and compare with the threshold A to obtain all building plan elements, expressed as: [E1, E2, …, E a ;

[0029] where a represents the number of building plan elements, and E a represents the a-th building plan element;

[0030] Collect a large number of building pictures and label the pictures manually: including determining the elements in the pictures and labeling the element values: [Pic B :{E i :V(E i ), E j :V(E j ), …, E k :V(E k )}];

[0031] where Pic B represents the B-th picture; E i , E j , …, E k ∈{E1, E2, …, E a}; V(E k ) represents the element value of the picture Pic B .

[0032] In step Step2, initially classify the pictures according to the elements of the pictures to obtain the pictures corresponding to each element: for the element E i , the corresponding pictures are expressed as [Pic B , Pic C , …, Pic D ; where C and D are positive integers;

[0033] Re-classify the pictures according to the type of the element value:

[0034] Elements with categorical element values: for the element E i , summarize the element value categories [T1(E i ), T2(E i ), …, T n (E i )] according to the manual labeling situation;

[0035] where n represents the number of categories of the element value of E i , and T n (E i ) represents the n-th category of the element value of E i ;

[0036] Classify the pictures according to the category of the element value; for the element value category T n (E i ),the corresponding pictures are represented as [Pic F ,Pic G ,…,Pic H ; where F, G, and H are positive integers;

[0037] For the element whose element value is numerical: For the element E j , obtain the maximum value max and the minimum value min of the element value E j according to the manual annotation situation; divide the element values into N parts: {[min, min+(max - min) / N], (min+(max - min) / N, min+(max - min)2 / N], …, (min+(max - min)(N - 1) / N, max]}; classify the pictures according to different intervals;

[0038] where N is a positive integer; (min+(max - min) / N, min+(max - min)2 / N] represents the interval min+(max - min) / N to min+(max - min)2 / N, excluding min+(max - min) / N;

[0039] Construct a database to store different elements, the corresponding different categories of element values, and the element values in different intervals, including the pictures corresponding to different element value categories and the pictures corresponding to different element value intervals;

[0040] For the pictures corresponding to a certain element value category or element value interval, perform clustering processing on them: perform normalization processing on the pictures; use a pre - trained convolutional neural network model to extract picture features and form feature vectors; set the number of clusters K, and use the K - means clustering algorithm to divide the pictures into K classes;

[0041] According to the classification situation, count the number of pictures in each class, and randomly select one picture from the class with the largest number of pictures as the representative picture of the element value category or element value interval.

[0042] In Step 3 and Step 4, analyze the user input: perform word segmentation on the user input; use Word2Vec to assign word vectors to the user input word segmentation, elements, and element type values; calculate the cosine similarity between each user input word segmentation word vector and each element word vector;

[0043] For the element E whose element value is categorical i , when there is a cosine similarity value of 1 between the user input word segmentation word vector and the element E i word vector and there is a user input word segmentation word vector and the element Ei When the cosine similarity of the word vector of a certain element type value is 1, the representative graph corresponding to the element type value is selected and output to the user, and the user makes a judgment whether it meets expectations and gives feedback; when there is a word vector of the user input segmentation and the element E i The word vector cosine similarity value is 1, but there is no user input word vector and element E i When the cosine similarity of any element type value word vector is 1 or when all user input word vectors are similar to element E i When the cosine similarity values of the word vectors are not 1, the output element E i The representative graphs corresponding to all factor value categories are given to the user for selection and feedback;

[0044] For element E whose element value is numeric j , when there is a user input word vector and element E j When the word vector cosine similarity value is 1 and the context analysis of the user input checks that there is numerical data before and after the corresponding word segmentation, select factor E j The image corresponding to the element value with the smallest absolute value of the difference with the checked numerical data is output to the user, and the user makes a judgment whether it meets the expectations and gives feedback; when there is a word vector of the user input segmentation and the element E j When the cosine similarity value of the word vector is 1, but the context analysis of the user input shows that there is no numerical data before and after the corresponding word segmentation, or when all user input word segmentation word vectors are consistent with the element E j When the cosine similarity values of the word vectors are not 1, the output element E j The representative graphs corresponding to all factor value intervals are given to the user for selection and feedback;

[0045] Combine user input, feedback, and choices to generate a building plan: all elements and their corresponding values.

[0046] Compared with the prior art, the beneficial effects of the present invention are: the multi-level classification method of the present invention can classify images more carefully according to different features, and can provide more accurate classification results that are more in line with actual needs, so as to facilitate the subsequent accurate matching of images according to user needs; the present invention extracts feature vectors through a pre-trained convolutional neural network model and uses the K-means clustering algorithm, which can mine the intrinsic similarities between images. The selected representative images can represent the typical characteristics of the element value category or interval to a certain extent, and can better present representative image examples to users, assisting users in understanding and selecting. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A schematic diagram of a process of a building plan generation system based on artificial intelligence according to the present invention;

[0048] Figure 2 Schematic diagram of the steps of a method for generating a building plan based on artificial intelligence according to the present invention. Specific implementation mode

[0049] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0050] Embodiment: As Figure 1 - Figure 2 shown, the present invention provides a technical solution, a building plan generation system based on artificial intelligence, which includes a preprocessing module, a picture classification module, an input analysis module, and a plan generation module;

[0051] The preprocessing module is used to collect building plans and building pictures, determine building plan elements, and manually label the elements and element values of the pictures; the picture classification module is used to preliminarily classify and re-classify pictures according to the elements, store the classified picture information, and cluster the pictures of the same classification to determine the representative picture; the input analysis module is used to analyze the user input; the plan generation module is used to generate a building plan according to the user input and feedback, and provide a selection and judgment mechanism to optimize the plan generation;

[0052] The output end of the preprocessing module is connected to the input end of the picture classification module; the output end of the picture classification module is connected to the input ends of the input analysis module and the plan generation module; the output end of the input analysis module is connected to the input end of the plan generation module.

[0053] The preprocessing module includes a plan collection unit, an element determination unit, a picture collection unit, and a picture annotation unit;

[0054] The plan collection unit is used to collect building plans; the element determination unit is used to determine building plan elements according to the collected building plans; the picture collection unit is used to collect building pictures; the picture annotation unit is used to manually label picture elements and element values;

[0055] The output end of the plan collection unit is connected to the input end of the element determination unit; the output end of the element determination unit is connected to the input end of the picture collection unit; the output end of the picture collection unit is connected to the input end of the picture annotation unit; the output end of the picture annotation unit is connected to the input end of the picture classification module.

[0056] The described picture classification module includes a picture classification unit, a type determination unit, a range division unit, a database unit, a picture clustering unit, and a representative picture determination unit;

[0057] The picture classification unit is used to classify pictures according to elements; the type determination unit is used to determine the element value category according to the picture annotation situation; the range division unit is used to divide the numerical element values according to the manual annotation situation; the database unit is used to store the picture classification situation classified according to the categorical and numerical element values; the picture clustering unit is used to cluster pictures of the same classification; the representative picture determination unit is used to select a random picture from the category with the largest number of pictures after picture clustering as the representative picture;

[0058] The output end of the picture classification unit is connected to the input ends of the type determination unit and the range division unit; the output end of the type determination unit is connected to the input end of the database unit; the output end of the range division unit is connected to the input end of the database unit; the output end of the database unit is connected to the input ends of the picture clustering unit and the input analysis module; the output end of the picture clustering unit is connected to the input end of the representative picture determination unit; the output end of the representative picture determination unit is connected to the input end of the solution generation module.

[0059] The input analysis module includes a user input unit and an input analysis unit;

[0060] The user input unit is used for the user to input elements and corresponding element values; the input analysis unit is used to make an analysis according to the matching situation between the user input and the elements and element values in the database;

[0061] The output end of the user input unit is connected to the input end of the input analysis unit; the output end of the input analysis unit is connected to the input end of the solution generation module.

[0062] The solution generation module includes an output selection unit, an output judgment unit, a user feedback unit, and a solution generation unit;

[0063] The output selection unit is used to output corresponding pictures for the user to judge whether they meet the expectations when the user input matches the classification situation; the output judgment unit is used to output corresponding pictures for the user to make a choice when the user input does not match the classification situation; the user feedback unit is used for the user to give feedback on the output; the solution generation unit is used to generate a building solution by combining the user input, feedback, and selection;

[0064] The output end of the output selection unit is connected to the input end of the user feedback unit; the output end of the output judgment unit is connected to the input end of the user feedback unit; the output end of the user feedback unit is connected to the input end of the solution generation unit.

[0065] An artificial intelligence-based building solution generation method, the method comprising the following steps:

[0066] Step1. Collect a large number of building solutions and determine the building solution elements; collect a large number of building pictures, manually determine the building solution elements in the pictures and label the element values;

[0067] Step2. Classify and cluster the pictures according to the element and element value types, determine the representative pictures, and construct a database to store the relevant information;

[0068] Step3. Analyze the user input to obtain the matching situation between the user input and the elements and element values;

[0069] Step4. Provide a selection or judgment mechanism according to the matching situation between the user input and the elements and element values, and generate a building solution in combination with the user feedback.

[0070] In step Step1, collect a large number of building solutions to ensure sample diversity;

[0071] Use a word segmentation tool to segment each building solution text; create a stop word list and remove the stop words;

[0072] For each sample, calculate the frequency of occurrence of all the words after removing the stop words in all the samples: frequency of occurrence = the number of samples with the word / the total number of samples; set a threshold A, when the word frequency of occurrence is greater than A, set the word as a building solution element;

[0073] Calculate the frequency of occurrence of all the words after removing the stop words in all the samples, and compare it with the threshold A to obtain all the building solution elements, expressed as: [E1, E2, …, E a ;

[0074] where a represents the number of building solution elements, and E a represents the a-th building solution element;

[0075] Collect a large number of building pictures and manually label the pictures: including determining the elements of the pictures and labeling the element values: [Pic B :{E i :V(E i ), E j :V(E j ), …, E k :V(Ek )}];

[0076] Among them, Pic B represents the Bth picture; E i , E j , …, E k ∈{E1, E2, …, E a}; V(E k ) represents the feature value of the picture Pic B .

[0077] In step Step2, the pictures are preliminarily classified according to the features of the pictures to obtain the pictures corresponding to each feature: for the feature E i , the corresponding pictures are represented as [Pic B , Pic C , …, Pic D ; where C and D are positive integers;

[0078] The pictures are re-classified according to the type of the feature value:

[0079] Features with categorical feature values: for the feature E i , summarize the feature value categories [T1(E i ), T2(E i ), …, T n (E i )] according to the manual annotation situation;

[0080] Among them, n represents the number of categories of the feature E i value, and T n (E i ) represents the nth category of the feature E i value;

[0081] Classify the pictures according to the category of the feature value; for the feature value category T n (E i ), the corresponding pictures are represented as [Pic F , Pic G , …, Pic H ; where F, G, and H are positive integers;

[0082] Features with numerical feature values: for the feature E j , obtain the feature E jThe maximum value max and the minimum value min of the values; divide the feature values into N parts: {[min, min + (max - min) / N], (min + (max - min) / N, min + (max - min)2 / N], …, (min + (max - min)(N - 1) / N, max]}; classify the pictures according to different intervals;

[0083] Among them, N is a positive integer; (min + (max - min) / N, min + (max - min)2 / N] represents the interval min + (max - min) / N to min + (max - min)2 / N, excluding min + (max - min) / N;

[0084] Construct a database to store different features, the corresponding feature values of different categories, and the feature values of different intervals, including different feature value categories and their corresponding pictures, and the pictures corresponding to different feature value intervals;

[0085] For the pictures corresponding to a certain feature value category or feature value interval, perform clustering processing on them: perform normalization processing on the pictures; use a pre-trained convolutional neural network model to extract picture features and form feature vectors; set the number of clusters K, and use the K-means clustering algorithm to divide the pictures into K classes;

[0086] According to the classification situation, count the number of pictures in each class, and randomly select a picture from the class with the largest number of pictures as the representative picture of the feature value category or feature value interval.

[0087] In Step 3 and Step 4, analyze the user input: perform word segmentation on the user input; use Word2Vec to assign word vectors to the user input words, features, and feature type values; calculate the cosine similarity between each user input word vector and each feature word vector;

[0088] For the feature E with categorical feature values i , when there is a cosine similarity value of 1 between the user input word vector and the word vector of feature E i and there is a cosine similarity value of 1 between the user input word vector and the word vector of a certain feature type value of feature E i , select the representative picture corresponding to the feature type value and output it to the user, and the user makes a judgment whether it meets the expectation and gives feedback; when there is a cosine similarity value of 1 between the user input word vector and the word vector of feature E i but there is no cosine similarity value of 1 between the user input word vector and the word vector of any feature type value of feature E i or when the cosine similarity values between all user input word vectors and the word vector of feature E i are not 1, output feature Ei Provide the representative diagrams corresponding to all element value categories to the user for the user to select and provide feedback;

[0089] For the element E with numerical element values j when there is a cosine similarity value of 1 between the user input word segmentation word vector and the element E j and numerical data is detected before and after the corresponding word segmentation through context analysis of the user input, select the picture corresponding to the element value with the smallest absolute value of the difference between the element E j and the detected numerical data and output it to the user, and the user makes a judgment on whether it meets the expectation and provides feedback; when there is a cosine similarity value of 1 between the user input word segmentation word vector and the element E j but no numerical data is detected before and after the corresponding word segmentation through context analysis of the user input or when the cosine similarity values between all user input word segmentation word vectors and the element E j are not 1, output the representative diagrams corresponding to all element value intervals of the element E j to the user for the user to select and provide feedback;

[0090] Generate a building plan by combining user input, feedback, and selections: all elements and corresponding element values.

[0091] In this embodiment:

[0092] Step Step1, collect a large number of building plans to ensure sample diversity;

[0093] Use a word segmentation tool to segment the text of each building plan; create a stop word list and remove stop words;

[0094] For each sample, calculate the frequency of occurrence of all word segments after removing stop words in all samples: frequency of occurrence = number of samples with the word segment / total number of samples; set a threshold, and when the frequency of occurrence of the word segment is greater than the threshold, set the word segment as a building plan element;

[0095] Calculate the frequency of occurrence of all word segments after removing stop words in all samples, and compare it with the threshold A to obtain all building plan elements, expressed as: [style, area];

[0096] Collect a large number of building pictures and manually annotate the pictures: including determining the elements of the pictures and annotating the element values: [pic1: {style: Chinese, area: 200}], [pic2: {style: European, area: 150}], [pic3: {style: Chinese, area: 120}], [pic4: {style: European, area: 100}];

[0097] Step 2: Initially classify the pictures according to the elements of the pictures to obtain the pictures corresponding to each element. For the element style, the corresponding pictures are represented as [pic1, pic2, pic3, pic4].

[0098] Re-classify the pictures according to the type of element values:

[0099] Elements with categorical element values: For the element style, summarize the element value categories [Chinese style, European style] according to the manual annotation.

[0100] Classify the pictures according to the categories of element values. For the Chinese style category of element values, the corresponding pictures are represented as [pic1, pic3].

[0101] Elements with numerical element values: For the element area, obtain the maximum value 200 and the minimum value 100 of the element area value according to the manual annotation. Divide the element values into 2 parts: {[100, 150], (150, 200]}. Classify the pictures according to different intervals.

[0102] Build a database to store different elements, the corresponding different categories of element values, and different intervals of element values, including different element value categories and their corresponding pictures, and pictures corresponding to different element value intervals.

[0103] For the pictures corresponding to a certain element value category or element value interval, perform clustering processing on them: Normalize the pictures; Use a pre-trained convolutional neural network model to extract picture features and form feature vectors; Set the number of clusters to 2, and use the K-means clustering algorithm to divide the pictures into 2 classes.

[0104] According to the classification situation, count the number of pictures in each class, and randomly select one picture from the class with the largest number of pictures as the representative picture of the element value category or element value interval.

[0105] Steps Step 3 and Step 4: Analyze the user input. Perform word segmentation on the user input; Use Word2Vec to assign word vectors to the user input word segments, elements, and element type values; Calculate the cosine similarity between each user input word segment word vector and each element word vector.

[0106] For the element style where the element value is of the categorical type, when there is a cosine similarity value of 1 between the user input tokenized word vector and the element style word vector and there is a cosine similarity value of 1 between the user input tokenized word vector and the word vector of a certain element type value of the element style, select the representative graph corresponding to the element type value and output it to the user, and the user makes a judgment on whether it meets the expectation and gives feedback; when there is a cosine similarity value of 1 between the user input tokenized word vector and the element style word vector but there is no cosine similarity value of 1 between the user input tokenized word vector and the word vector of any element type value of the element style or when the cosine similarity values of all user input tokenized word vectors and the element style word vector are not 1, output the representative graphs corresponding to all element value categories of the element style to the user for the user to select and give feedback;

[0107] For the element area where the element value is of the numerical type, when there is a cosine similarity value of 1 between the user input tokenized word vector and the element area word vector and numerical data is detected before and after the corresponding token through context analysis of the user input, select the picture corresponding to the element value with the smallest absolute value of the difference between the element area and the detected numerical data and output it to the user, and the user makes a judgment on whether it meets the expectation and gives feedback; when there is a cosine similarity value of 1 between the user input tokenized word vector and the element area word vector but no numerical data is detected before and after the corresponding token through context analysis of the user input or when the cosine similarity values of all user input tokenized word vectors and the element area word vector are not 1, output the representative graphs corresponding to all element value intervals of the element area to the user for the user to select and give feedback;

[0108] Generate a building plan by combining the user input, feedback, and selection: all elements and corresponding element values.

[0109] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be regarded as limiting the claims involved.

Claims

1. An artificial intelligence-based method for generating building solutions, characterized in that: The method includes the following steps: Step1. Collect a large number of building plans and determine the elements of the building plans; collect a large number of building pictures, manually determine the elements of the building plans in the pictures and label the element values; Step2. Classify and cluster the pictures according to the elements and element value types, determine the representative pictures, and construct a database to store relevant information; Step3. Analyze the user input to obtain the matching situation between the user input and the elements and element values; Step4. Provide a selection or judgment mechanism according to the matching situation between the user input and the elements and element values, and generate a building plan in combination with the user feedback; In Step1, collect a large number of building plans to ensure sample diversity; Use a word segmentation tool to segment each building plan text; create a stop word list and remove the stop words; For each sample, calculate the frequency of occurrence of all the segmented words after removing the stop words in all samples: frequency of occurrence = the number of samples with the segmented word / the total number of samples; set a threshold A, and when the frequency of occurrence of the segmented word is greater than A, set the segmented word as a building plan element; Calculate the occurrence frequencies of all word segments after removing stop words in all samples, and compare them with threshold A to obtain all building plan elements, expressed as: [E1, E2, …, E a ; Among them, a represents the number of building plan elements, and E i represents the i-th building plan element; Collect a large number of building pictures and label the pictures manually, including determining the elements of the pictures and labeling the element values: [Pic B :{E i :V(E i ),E j :V(E j ),…,E k :V(E k )}]; Among them, Pic B represents the B-th picture; E i , E j , …, E k ∈ {E1, E2, …, E a}; V(E k ) represents the element value of the element E B in the picture Pic k . In Step 2, the pictures are preliminarily classified according to the elements of the pictures, and the pictures corresponding to each element are obtained: for element E i , the corresponding pictures are represented as [Pic B , Pic C , …, Pic D ; where C and D are positive integers; Re-classify the pictures according to the type of element value: Element with categorical element values: For element E i , summarize the element value categories [T1(E i ), T2(E i ), …, T n (E i )] according to the manual annotation situation; where n represents the number of categories of the values of element E i the number of value categories, T e (E i ) represents the e-th category of the values of element E i where e ∈ {1, 2,..., n}; Classify the pictures according to the category of the element value; for the element value category T n (E i )), the corresponding pictures are represented as [Pic F , Pic G , …, Pic H ; where F, G, and H are positive integers; Element with a numerical element value: For element E j , obtain the maximum value max and the minimum value min of the element value E according to the manual annotation situation j ; divide the element values into N parts: {[min, min + (max - min) / N], (min + (max - min) / N, min + (max - min)2 / N], …, (min + (max - min)(N - 1) / N, max]}; classify the pictures according to different intervals where N is a positive integer; (min+(max-min) / N,min+(max-min)2 / N] represents the interval min+(max-min) / N to min+(max-min)2 / N, excluding min+(max-min) / N; Construct a database to store different elements, the corresponding element values of different categories, and the element values of different intervals, including the pictures corresponding to different element value categories and the pictures corresponding to different element value intervals; For the pictures corresponding to a certain element value category or element value interval, perform clustering processing on them: perform normalization processing on the pictures; use a pre-trained convolutional neural network model to extract picture features and form feature vectors; set the number of clusters K, and use the K-means clustering algorithm to divide the pictures into K classes; According to the classification situation, count the number of pictures in each of the K classes, and randomly select one picture from the class with the largest number of pictures as the representative picture of the element value category or element value interval.

2. The method for generating an architectural plan based on artificial intelligence according to claim 1, wherein: In Step3 and Step4, analyze the user input: perform word segmentation processing on the user input; Use Word2Vec to assign word vectors to the user input segmentation words, elements, and element type values; calculate the cosine similarity between each user input segmentation word vector and each element word vector; For the element E with categorical element values i , when there is a cosine similarity value of 1 between the user input segmented word vector and the element E i and there is a cosine similarity value of 1 between the user input segmented word vector and the word vector of a certain element type value of the element E i , select the representative graph corresponding to the element type value and output it to the user, and the user makes a judgment on whether it meets the expectation and gives feedback; when there is a cosine similarity value of 1 between the user input segmented word vector and the element E i but there is no cosine similarity value of 1 between the user input segmented word vector and any element type value of the element E i or when the cosine similarity values between all user input segmented word vectors and the element E i are not 1, output the representative graphs corresponding to all element value categories of the element E i to the user for the user to select and give feedback; For the element E with a numerical element value j , when there is a cosine similarity value of 1 between the user input tokenized word vector and the element E j , and numerical data is detected before and after the corresponding token through context analysis of the user input, select the picture corresponding to the element value with the smallest absolute value of the difference between the element E j and the detected numerical data, output it to the user, and the user makes a judgment on whether it meets the expectation and gives feedback; when there is a cosine similarity value of 1 between the user input tokenized word vector and the element E j , but no numerical data is detected before and after the corresponding token through context analysis of the user input, or when the cosine similarity values between all user input tokenized word vectors and the element E j are not 1, output the representative pictures corresponding to all element value intervals of the element E j to the user for the user to select and give feedback; Generate a building plan in combination with the user input, feedback, and selection: all elements and the corresponding element values.

3. An artificial intelligence-based building plan generation system, which is applied to the artificial intelligence-based building plan generation method described in any one of claims 1-2, and is characterized in that: The system includes a preprocessing module, a picture classification module, an input analysis module, and a plan generation module; The preprocessing module is used to collect building plans and building pictures, determine the elements of the building plans, and manually label the elements and element values of the pictures; The picture classification module is used to perform preliminary classification and re-classification of the pictures according to the elements, store the classified picture information, and perform clustering on the pictures of the same classification to determine the representative pictures; The input analysis module is used to analyze the user input; The described solution generation module is used to generate building solutions based on user input and feedback, and to provide selection and judgment mechanisms to optimize solution generation; The output end of the described preprocessing module is connected to the input end of the described picture classification module; The output end of the described picture classification module is connected to the input ends of the described input analysis module and the solution generation module; the output end of the described input analysis module is connected to the input end of the solution generation module.

4. An artificial intelligence-based building plan generation system according to claim 3, characterized in that: The described preprocessing module includes a solution collection unit, a factor determination unit, a picture collection unit, and a picture annotation unit; The described solution collection unit is used to collect building solutions; The described factor determination unit is used to determine building solution factors based on the collected building solutions; The described picture collection unit is used to collect building pictures; The described picture annotation unit is used to manually annotate picture factors and factor values; The output end of the described solution collection unit is connected to the input end of the described factor determination unit; the output end of the described factor determination unit is connected to the input end of the described picture collection unit; the output end of the described picture collection unit is connected to the input end of the described picture annotation unit; the output end of the described picture annotation unit is connected to the input end of the described picture classification module.

5. The building plan generation system based on artificial intelligence according to claim 4, characterized in that: The described picture classification module includes a picture classification unit, a type determination unit, a range division unit, a database unit, a picture clustering unit, and a representative picture determination unit; The described picture classification unit is used to classify pictures according to factors; the described type determination unit is used to determine the category of factor values according to the picture annotation situation; the described range division unit is used to divide numerical factor values according to the manual annotation situation; the described database unit is used to store the picture classification situation classified according to categorical and numerical factor values; the described picture clustering unit is used to cluster pictures of the same classification; the described representative picture determination unit is used to select a random picture from the category with the largest number of pictures after picture clustering as the representative picture; The output end of the described picture classification unit is connected to the input ends of the described type determination unit and the range division unit; The output end of the described type determination unit is connected to the input end of the described database unit; the output end of the described range division unit is connected to the input end of the described database unit; the output end of the described database unit is connected to the input ends of the described picture clustering unit and the input analysis module; The output end of the described picture clustering unit is connected to the input end of the described representative picture determination unit; the output end of the described representative picture determination unit is connected to the input end of the described solution generation module.

6. The building plan generation system based on artificial intelligence according to claim 5, characterized in that: The described input analysis module includes a user input unit and an input analysis unit; The described user input unit is used for users to input factors and corresponding factor values; the described input analysis unit is used to make an analysis based on the matching situation between the user input and the factors and factor values in the database; The output end of the described user input unit is connected to the input end of the described input analysis unit; the output end of the described input analysis unit is connected to the input end of the described solution generation module.

7. An artificial intelligence-based building plan generation system according to claim 6, characterized in that: The described solution generation module includes an output selection unit, an output judgment unit, a user feedback unit, and a solution generation unit; The output selection unit is used to output corresponding pictures for the user to judge whether they meet the expectations when the user input matches the classification situation; the output judgment unit is used to output corresponding pictures for the user to make a choice when the user input does not match the classification situation; The user feedback unit is used for the user to give feedback on the output; The scheme generation unit is used to generate a building scheme by combining the user input, feedback and selection; The output end of the output selection unit is connected to the input end of the user feedback unit; the output end of the output judgment unit is connected to the input end of the user feedback unit; The output end of the user feedback unit is connected to the input end of the scheme generation unit.

Images