Landscape sequence design method based on human mental health benefits

Abstract

The invention discloses a landscape sequence design method based on human psychological health benefits, and relates to the technical field of landscape planning. The method comprises the following steps: S1, obtaining landscape sequence data, wherein the landscape sequence data comprises visual data, auditory data and sequence data of each scenic spot and a psychological health benefit score; s2, performing standardization processing on the visual data, the auditory data and the sequence data, performing screening through an RFE method based on recursive feature elimination to obtain important features and taking the important features as independent variables, taking psychological health benefit scores as dependent variables, setting hyper-parameters, and obtaining a trained random forest model; s3, recording all scenic spot positions and road positions, and calculating based on a Dijkstra algorithm to obtain N shortest landscape sequence paths which are not repeated; s4, inputting the landscape sequence paths into the trained random forest model to obtain corresponding psychological health benefit scores, wherein the landscape sequence path with the highest score is the best; the method is high in efficiency and good in effect.

Description

Technical Field

[0001] The invention relates to the technical field of landscape planning, and in particular to a landscape sequence design method based on human psychological health benefits. Background Art

[0002] With the acceleration of urbanization, the mental health problems of urban residents are becoming increasingly prominent. Studies have shown that visual and auditory experiences in natural environments have significant positive effects on mental health. However, there is still limited research on how to systematically organize landscape sequences in landscape planning to maximize mental health benefits. Most existing technologies rely on expert experience or simple optimization of a single variable, and lack methods that comprehensively consider visual, auditory and path characteristics. Summary of the invention

[0003] The present invention provides a landscape sequence design method based on human psychological health benefits, which solves the technical problem of low efficiency in obtaining the optimal landscape sequence path.

[0004] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows: A landscape sequence design method based on human psychological health benefits includes the following steps: Step S1: obtaining landscape sequence data, the landscape sequence data including visual data, auditory data and sequence data of each scenic spot and a mental health benefit score, the visual data including panoramic images, green view rate, canopy density and sky visibility factor, the auditory data including sound data, sound type, sound pressure level and sound spectrum, the sequence data including the scenic spot location, road location, distance between scenic spots and the order of each path of each scenic spot, and the mental health benefit score is the pleasure score of all scenic spots under each path; Step S2: Standardizing the visual data, auditory data and sequence data to obtain standardized visual data, auditory data and sequence data; using the RFE method based on recursive feature elimination to screen and obtain important features from the standardized visual data, auditory data and sequence data; using the important features obtained by screening as independent variables, using the mental health benefit score as the dependent variable, setting hyperparameters, training the random forest model, and obtaining a trained random forest model; Step S3: Record the location information of all scenic spots and road locations, including the coordinate information of longitude and latitude, and calculate the shortest N landscape sequence paths that can connect all scenic spots without duplication based on the Dijkstra algorithm; Step S4: Input the data of the N landscape sequence paths obtained into the trained random forest model respectively, obtain the corresponding mental health benefit scores of all landscape sequence paths, compare the mental health benefit scores of each landscape sequence path, and take the corresponding landscape sequence path with the highest score as the best sequence.

[0005] A further technical solution is that: in step S1, the scoring range is 0 to 10 points; in step S3, the location information of all scenic spots and road locations is recorded using GIS software, N=5, to obtain the shortest five landscape sequence paths.

[0006] A further technical solution is that the step S1 is specifically divided into the following steps: Step S101: obtaining visual data of each scenic spot; Step S102: obtaining auditory data of each scenic spot; Step S103: obtaining sequence data; Step S104: Obtain the pleasure scores of all scenic spots on each route by having volunteers fill out questionnaire records, and use the pleasure scores as mental health benefit scores.

[0007] A further technical solution is that the step S101 is specifically divided into the following steps: Step S1011: photographing each scenic spot using a 360-degree panoramic camera to generate a high-resolution panoramic image of each scenic spot; Step S1012: using a U-Net image segmentation algorithm based on deep learning, extracting the green pixel ratio from the panoramic image of each scenic spot to calculate the green viewing rate; Step S1013: Analyze the vegetation coverage area in the panoramic image of each scenic spot, calculate the vegetation coverage area ratio using the Mask R-CNN object detection algorithm, and obtain the canopy density; Step S1014: using a sky area segmentation algorithm, by extracting the proportion of the sky area in the panoramic image of each scenic spot, a sky visibility factor is calculated.

[0008] A further technical solution is that the step S102 is specifically divided into the following steps: Step S1021: Setting a high-sensitivity microphone at each scenic spot to collect and obtain sound data of each scenic spot; Step S1022: input the sound data of each scenic spot obtained in step S1021 into an audio classification model based on a convolutional neural network to identify different sound types; Step S1023: obtaining the sound pressure level of the sound of each scenic spot in real time by using a sound level meter, in dB; Step S1024: Perform Fourier transform FFT on the sound data of each scenic spot obtained in step S1021, extract the sound spectrum data, and analyze to obtain the sound frequency distribution characteristics.

[0009] A further technical solution is that the step S103 is specifically divided into the following steps: Step S1031: Use a GPS device to record and obtain the location of each scenic spot; Step S1032: Mark the roads between scenic spots in the GIS software, obtain the road locations, calculate the distances between scenic spots, and record the order in which each path is obtained.

[0010] A further technical solution is: in step S2, setting the hyper parameters: The number of trees n_estimators: Select from 100 to 500; Maximum depth max_depth: automatically adjusted to achieve optimal model performance; Minimum number of split samples min_samples_split: set to 2 to 5 according to the amount of data.

[0011] The beneficial effects of adopting the above technical solution are: A landscape sequence design method based on human mental health benefits comprises the following steps: step S1: obtaining landscape sequence data, the landscape sequence data comprising visual data, auditory data and sequence data of each scenic spot and a mental health benefit score; step S2: standardizing the visual data, auditory data and sequence data, screening important features through a RFE method based on recursive feature elimination and using them as independent variables, using the mental health benefit score as a dependent variable, setting hyperparameters, and obtaining a trained random forest model; step S3: recording the location information of all scenic spots and road locations, and calculating the shortest N landscape sequence paths that can connect all scenic spots without duplication based on the Dijkstra algorithm; step S4: inputting the data of the obtained N landscape sequence paths into the trained random forest model respectively, obtaining the mental health benefit scores of all corresponding landscape sequence paths, and using the corresponding landscape sequence path with the highest score as the optimal sequence. The method is efficient in obtaining the optimal landscape sequence path and has good effect.

[0012] Please refer to the detailed description of the specific implementation method. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a flow chart of the present invention; Figure 2 yes Figure 1 Flow chart of step S1 in FIG. Figure 3 yes Figure 2 Flow chart of step S101; Figure 4 yes Figure 2 Flowchart of step S102 in FIG. DETAILED DESCRIPTION

[0014] Purpose of this application: The aim is to provide a scientific and easy-to-use method that can optimize the landscape sequence organization based on the visual, auditory and path data of the landscape, thereby maximizing the benefits of human mental health. By combining modeling with algorithms, the scientificity and practicality of landscape planning can be improved.

[0015] It involves the intersection of mental health and landscape planning, aiming to optimize landscape experience to enhance mental health benefits.

[0016] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present application and its application or use. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0017] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0018] like Figure 1 As shown, the present invention discloses a landscape sequence design method based on human psychological health benefits, comprising the following steps: Step S1: Obtain landscape sequence data.

[0019] The landscape sequence data are obtained, which include visual data, auditory data, sequence data and mental health benefit score of each scenic spot. The visual data include panoramic images, green view rate, canopy density and sky visibility factor. The auditory data include sound data, sound type, sound pressure level and sound spectrum. The sequence data include the location of each scenic spot, road location, distance between scenic spots and the order of each path. The mental health benefit score is the pleasure score of all scenic spots under each path, and the score ranges from 0 to 10 points.

[0020] Details are as follows: like Figure 2 , Figure 3 and Figure 4 As shown, step S1 is specifically divided into the following steps: Step S101: Obtain visual data.

[0021] The visual data of each scenic spot is obtained, including panoramic images, green viewing rate, canopy density and sky visibility factor.

[0022] Step S1011: Panoramic shooting.

[0023] A 360-degree panoramic camera is used to shoot each scenic spot to generate a high-resolution panoramic image of each scenic spot.

[0024] Step S1012: calculating the green viewing rate.

[0025] The U-Net image segmentation algorithm based on deep learning is used to extract the proportion of green pixels from the panoramic image of each scenic spot and calculate the green viewing rate.

[0026] Step S1013: calculation of canopy density.

[0027] By analyzing the vegetation coverage area in the panoramic images of each scenic spot, the Mask R-CNN object detection algorithm is used to calculate the vegetation coverage area ratio and obtain the canopy density.

[0028] Step S1014: Sky visibility factor.

[0029] The sky area segmentation algorithm is used to extract the proportion of the sky area in the panoramic image of each scenic spot and calculate the sky visibility factor.

[0030] Step S102: Acquire auditory data.

[0031] Acoustic data of each scenic spot is obtained, and the acoustic data includes sound data, sound type, sound pressure level and sound spectrum.

[0032] Step S1021: sound collection.

[0033] A high-sensitivity microphone is installed at each scenic spot to collect sound data of each scenic spot.

[0034] Step S1022: Sound type.

[0035] The sound data of each scenic spot obtained in step S1021 is input into the audio classification model based on the convolutional neural network to identify different sound types, including bird calls, flowing water, mechanical noise, etc.

[0036] Step S1023: Sound pressure level.

[0037] The sound pressure level of the sound at each scenic spot is measured in real time by a sound level meter in dB.

[0038] Step S1024: Sound spectrum.

[0039] Perform Fourier transform FFT on the sound data of each scenic spot obtained in step S1021, extract the sound spectrum data, and analyze to obtain the sound frequency distribution characteristics.

[0040] Step S103: Obtain sequence data.

[0041] Sequence data are obtained, and the sequence data includes the location of each scenic spot, the location of the road, the distance between the scenic spots, and the order of each path.

[0042] Step S1031: Scenic spot location.

[0043] Use a GPS device to record the location of each attraction.

[0044] Step S1032: Path order.

[0045] Mark the roads between scenic spots in the GIS software, obtain the road locations, calculate the distances between scenic spots, and record the order in which each path is obtained.

[0046] Step S104: Obtain a mental health benefit score.

[0047] The pleasure scores of all scenic spots on each route were obtained by volunteers filling out questionnaires, with scores ranging from 0 to 10 points. The pleasure scores were used as mental health benefit scores.

[0048] Step S2: Construct a random forest model.

[0049] The visual data, auditory data and sequence data obtained in step S1 are standardized to obtain standardized visual data, auditory data and sequence data; the RFE method based on recursive feature elimination is used to screen and obtain important features from the standardized visual data, auditory data and sequence data; the important features obtained by screening are used as independent variables, the mental health benefit score obtained in step S1 is used as the dependent variable, the hyperparameters are set, and the random forest model is trained to obtain a trained random forest model.

[0050] Details are as follows: Step S2 specifically includes the following steps: Step S201: data preprocessing.

[0051] The visual data, auditory data and sequence data are standardized to obtain standardized visual data, auditory data and sequence data to eliminate dimensional differences.

[0052] The RFE method based on recursive feature elimination is used to screen important features from the standardized visual data, auditory data and sequence data.

[0053] Step S202: Random forest model construction.

[0054] The important features screened in step S201 are used as independent variables, the mental health benefit score obtained in step S104 is used as the dependent variable, and the hyperparameters are set to obtain the random forest model.

[0055] Setting hyperparameters: The number of trees n_estimators: Select from 100 to 500.

[0056] Maximum depth max_depth: Automatically adjusted to achieve optimal model performance.

[0057] Minimum number of split samples min_samples_split: set to 2 to 5 according to the amount of data.

[0058] Cross-validation was used to evaluate model performance and ensure the generalization ability of the model.

[0059] Step S3: Generate landscape sequence path.

[0060] GIS software was used to record the location information of all scenic spots and road locations of the case objects, including the coordinate information of longitude and latitude. The five shortest landscape sequence paths that could connect all scenic spots without duplication were calculated based on the Dijkstra algorithm.

[0061] Step S4: Benefit scoring and optimal sequence selection.

[0062] The data of the five generated landscape sequence paths were input into the trained random forest model respectively to obtain the corresponding mental health benefit scores of all landscape sequence paths. The mental health benefit scores of each landscape sequence path were compared, and the corresponding landscape sequence path with the highest score was taken as the optimal sequence.

[0063] Compared with the above embodiment, three landscape sequence paths may be obtained for comparison to obtain the best sequence.

[0064] Compared with the above embodiment, seven landscape sequence paths may be obtained for comparison to obtain the best sequence.

[0065] You can choose how many landscape sequence paths to obtain for comparison as needed.

[0066] advantage: 1. Strong scientificity: It uses machine learning methods to comprehensively consider multi-dimensional data and predict mental health benefits.

[0067] 2. High practicability: The trained model is a universal model that is applicable to different types of scenic spots or parks. The health benefits of the scenic spot can be calculated through a simple landscape data collection process, without the need to gather a large number of volunteers to fill out questionnaires every time.

[0068] 3. Improve health benefits: Provide quantifiable scientific basis for urban planning and landscape design, and help improve the public's mental health level.

[0069] Specific implementation cases: Take a city park as an example: 1. Data collection: Record visual and auditory data of 10 major attractions in the park, namely: Visual data: green viewing rate 0.3~0.8, canopy density 0.2~0.7, sky visibility factor 0.4~0.9.

[0070] Auditory data: sound types include bird songs, running water, mechanical noise, etc., sound pressure level 50-70dB, and sound spectrum distribution.

[0071] The pleasure ratings of 20 volunteers were surveyed and the rating data were recorded at 10 points on each path.

[0072] 2. Feature Engineering and Model Building: The random forest model was trained using the above data and the model prediction accuracy was evaluated through cross validation.

[0073] 3. Landscape sequence path generation: The Dijkstra algorithm was used to generate five landscape sequences containing the above 10 scenic spots.

[0074] 4. Benefit scoring and optimal sequence selection: After inputting into the random forest model, the mental health benefit scores were obtained as follows: 8.5, 7.2, 9.1, 8.0, and 8.7.

[0075] Results: The best landscape sequence with a score of 9.1 was selected for recommendation.

Claims

1. A landscape sequence design method based on human psychological health benefits, characterized by: The following steps are included: Step S1: obtaining landscape sequence data, the landscape sequence data including visual data, auditory data and sequence data of each scenic spot and a mental health benefit score, the visual data including panoramic images, green view rate, canopy density and sky visibility factor, the auditory data including sound data, sound type, sound pressure level and sound spectrum, the sequence data including the scenic spot location, road location, distance between scenic spots and the order of each path of each scenic spot, and the mental health benefit score is the pleasure score of all scenic spots under each path; Step S2: performing standardization processing on the visual data, auditory data and sequence data to obtain standardized visual data, auditory data and sequence data; using the RFE method based on recursive feature elimination to screen and obtain important features from the standardized visual data, auditory data and sequence data; The important features obtained by screening are used as independent variables, the mental health benefit score is used as the dependent variable, the hyperparameters are set, and the random forest model is trained to obtain a trained random forest model; Step S3: Record the location information of all scenic spots and road locations, including the coordinate information of longitude and latitude, and calculate the shortest N landscape sequence paths that can connect all scenic spots without duplication based on the Dijkstra algorithm; Step S4: Input the data of the N landscape sequence paths obtained into the trained random forest model respectively, obtain the mental health benefit scores of all corresponding landscape sequence paths, and take the corresponding landscape sequence path with the highest score as the optimal sequence.

2. A landscape sequence design method based on human psychological health benefits according to claim 1, characterized in that: In step S1, the scoring range is 0 to 10 points; in step S3, the location information of all scenic spots and road locations is recorded using GIS software, N=5, and the shortest five landscape sequence paths are obtained.

3. The method for designing landscape sequences based on the benefits of human psychological health according to claim 1, characterized in that: The step S1 specifically includes the following steps: Step S101: obtaining visual data of each scenic spot; Step S102: obtaining auditory data of each scenic spot; Step S103: obtaining sequence data; Step S104: Obtain the pleasure scores of all scenic spots on each route by having volunteers fill out questionnaire records, and use the pleasure scores as mental health benefit scores.

4. The method for designing landscape sequences based on the benefits of human psychological health according to claim 3, characterized in that: The step S101 specifically includes the following steps: Step S1011: photographing each scenic spot using a 360-degree panoramic camera to generate a high-resolution panoramic image of each scenic spot; Step S1012: using a U-Net image segmentation algorithm based on deep learning, extracting the green pixel ratio from the panoramic image of each scenic spot to calculate the green viewing rate; Step S1013: Analyze the vegetation coverage area in the panoramic image of each scenic spot, calculate the vegetation coverage area ratio using the Mask R-CNN object detection algorithm, and obtain the canopy density; Step S1014: using a sky area segmentation algorithm, by extracting the proportion of the sky area in the panoramic image of each scenic spot, a sky visibility factor is calculated.

5. The method for designing landscape sequences based on the benefits of human psychological health according to claim 4, characterized in that: The step S102 specifically includes the following steps: Step S1021: Setting a high-sensitivity microphone at each scenic spot to collect and obtain sound data of each scenic spot; Step S1022: input the sound data of each scenic spot obtained in step S1021 into an audio classification model based on a convolutional neural network to identify different sound types; Step S1023: obtaining the sound pressure level of the sound of each scenic spot in real time by using a sound level meter, in dB; Step S1024: Perform Fourier transform FFT on the sound data of each scenic spot obtained in step S1021, extract the sound spectrum data, and analyze to obtain the sound frequency distribution characteristics.

6. The method for designing landscape sequences based on the benefits of human psychological health according to claim 3, characterized in that: The step S103 specifically includes the following steps: Step S1031: Use a GPS device to record and obtain the location of each scenic spot; Step S1032: Mark the roads between scenic spots in the GIS software, obtain the road locations, calculate the distances between scenic spots, and record the order in which each path is obtained.

7. The method for designing landscape sequences based on the benefits of human psychological health according to claim 1, characterized in that: In step S2, the hyper parameters are set: The number of trees n_estimators: Select from 100 to 500; Maximum depth max_depth: automatically adjusted to achieve optimal model performance; Minimum number of split samples min_samples_split: set to 2 to 5 according to the amount of data.

Images