Cold region city emotion perception evaluation and measurement method

Abstract

According to the cold region city emotion perception evaluation and measurement method, a cold region city is selected as a research object, and physiological signals and subjective scores are comprehensively considered; street view images are scored by training the street view perception model, perception of the street view images is realized, physiological signals and subjective scores are combined to realize comprehensive evaluation of corresponding areas of the cold region city, and a comprehensive evaluation result is less subject to subjective influence and is stable; the method can analyze the influence of the winter street space of the cold region city on the residential emotion perception, accurately measure the quality of the cold region city street space influencing the residential emotion, has the advantages of high efficiency and easy popularization and application, can be widely applied to the field of urban planning living street space quality investigation and measurement, and has a wide application prospect. And a method and a basis are provided for cold region city planning street space optimization practice.

Description

Technical Field

[0001] The present application relates to emotion perception evaluation technology, and in particular to a cold-region city emotion perception evaluation measurement method. Background Art

[0002] Traditional urban renewal emphasizes the improvement of infrastructure and buildings. With the rise of humanism, the focus of urban development has begun to shift to improving the quality of life of residents and enhancing emotional experience. Based on the development of crowdsourcing technology, map services and volunteer geographic information, a large number of geo-tagged images have been generated, covering every corner of the city. As a result, urban researchers have conducted a large number of "environment-perception" correlation studies. Existing studies have used machine learning to create a global urban perception dataset and established a research framework for the impact of urban space on place perception. However, the impact of the regional characteristics and humanistic environmental characteristics of the city on the research results still needs to be studied in a targeted manner according to local characteristics. If targeted differentiation is not carried out, it is difficult for the prediction results to accurately express the emotional perception of the crowd. In terms of regional environmental characteristics, current research mainly uses spring and summer street view images obtained from data sources such as Google Maps, but these research results are difficult to extend beyond spring and autumn. For example, in some cold cities, due to the influence of snowfall and low temperature, the winter street environment gives people a sharp contrast with the spring and summer. Regarding the study of humanistic environmental characteristics, existing studies have taken Beijing, Guangzhou, Hong Kong and other cities as examples. However, my country has a vast territory, and the humanistic characteristics of various cities have diverse characteristics and significant differences. Therefore, the existing research results have limited applicability to cold-region cities.

[0003] Traditional street view evaluation methods rely on subjective judgment, and there are problems such as inconsistent evaluation standards and unstable results. In recent years, many studies have begun to introduce physiological detection technologies such as EEG to assist in urban environment perception research. However, there is still a lack of research on the correlation between physiological emotions and subjective emotions in urban environments. Summary of the invention

[0004] In order to solve the problem that the street scene evaluation method in the existing technology relies on subjective judgment and leads to unstable evaluation results, a cold-region city emotion perception evaluation and measurement method is proposed.

[0005] A cold-region city emotion perception evaluation and measurement method, comprising:

[0006] Step 1: Acquire street view image data of the target area to establish a street view image data sample set;

[0007] Step 2: Local residents subjectively score each street view image data in the street view image data sample set to obtain a subjective perception result corresponding to each street view image data in the street view image data sample set; when local residents subjectively score each street view image data in the street view image data sample set, physiological signals of residents are obtained through electroencephalographic equipment, and objective perception results corresponding to each street view image data in the street view image data sample set are obtained based on the obtained physiological signals;

[0008] Step 3: taking the street view image data sample set as input, and taking the subjective perception results and objective perception results corresponding to the street view image data sample set as output to train the street view perception model, thereby obtaining a trained street view perception model;

[0009] Step 4: obtaining multiple street view images from different directions of the area to be evaluated, and inputting the multiple street view image data from different directions of the area to be evaluated into the trained street view perception model respectively, to obtain subjective perception results and objective perception results corresponding to each street view image data from different directions of the area to be evaluated, and to obtain a total of multiple subjective perception results and multiple objective perception results, and taking the average value of the multiple subjective perception results as the subjective perception score of the area to be evaluated, and taking the average value of the multiple objective perception results as the objective perception score of the area to be evaluated;

[0010] Step 5: Set the subjective perception weight and the objective perception weight of the area to be evaluated, and obtain the comprehensive emotion evaluation result of the area to be evaluated according to the subjective perception score, the objective perception score, the subjective perception weight and the objective perception weight of the area to be evaluated.

[0011] Beneficial Effects

[0012] The present application discloses a cold-region city emotion perception evaluation and measurement method, which selects cold-region cities as research objects, comprehensively considers physiological signals and subjective scores as perception results, scores street scene images by training street scene perception models, realizes the perception of street scene images, and combines physiological signals with subjective scores to realize comprehensive evaluation of corresponding areas of cold-region cities. The comprehensive evaluation results are less affected by subjective factors and the results are stable. The method can analyze the impact of winter street space in cold-region cities on residents' emotional perception and accurately measure the quality of street space in cold-region cities that affects residents' emotions. It has the advantages of high efficiency and easy promotion and application, and can be widely used in the field of urban planning living street space quality investigation and measurement, and provides a method and basis for the practice of optimizing street space planning in cold-region cities. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A flow chart of a method for evaluating and measuring emotional perception in cold-region cities according to a specific implementation method of this application;

[0014] Figure 2 A regional map of a cold-region city emotion perception evaluation measurement method for a specific implementation method of this application;

[0015] Figure 3 A schematic diagram of shooting angles of street view image data according to a specific implementation mode of the present application;

[0016] Figure 4 A schematic diagram of street view image data for a specific implementation of the present application;

[0017] Figure 5 This is a physiological signal monitoring page for a specific implementation of this application;

[0018] Figure 6 This is an objective environmental evaluation score diagram for a specific implementation of this application;

[0019] Figure 7 This is a perception score diagram of a specific implementation of the present application. DETAILED DESCRIPTION

[0020] Specific implementation method 1: The following will be combined with the attached embodiment of the present invention Figure 1 To Attachment Figure 7 , illustrate this implementation mode, and clearly and completely describe the technical solutions in the embodiments of the present invention:

[0021] A cold-region city emotion perception evaluation and measurement method, comprising:

[0022] Step 1: Acquire street view image data of the target area to establish a street view image data sample set;

[0023] Step 2: Local residents subjectively score each street view image data in the street view image data sample set to obtain a subjective perception result corresponding to each street view image data in the street view image data sample set; when local residents subjectively score each street view image data in the street view image data sample set, physiological signals of residents are obtained through electroencephalographic equipment, and objective perception results corresponding to each street view image data in the street view image data sample set are obtained based on the obtained physiological signals;

[0024] Step 3: taking the street view image data sample set as input, and taking the subjective perception results and objective perception results corresponding to the street view image data sample set as output to train the street view perception model, thereby obtaining a trained street view perception model;

[0025] Step 4: obtaining multiple street view images from different directions of the area to be evaluated, and inputting the multiple street view image data from different directions of the area to be evaluated into the trained street view perception model respectively, to obtain subjective perception results and objective perception results corresponding to each street view image data from different directions of the area to be evaluated, and to obtain a total of multiple subjective perception results and multiple objective perception results, and taking the average value of the multiple subjective perception results as the subjective perception score of the area to be evaluated, and taking the average value of the multiple objective perception results as the objective perception score of the area to be evaluated;

[0026] Step 5: Set the subjective perception weight and the objective perception weight of the area to be evaluated, and obtain the comprehensive emotion evaluation result of the area to be evaluated according to the subjective perception score, the objective perception score, the subjective perception weight and the objective perception weight of the area to be evaluated.

[0027] Specifically, the sampling points of the streets in the target area are divided into preset intervals of 25 m, and street view images in four directions of front, back, left, and right of the sampling points are obtained.

[0028] The winter environment simulation is carried out in a laboratory under settable and controlled constant temperature conditions. There are fixed measurement points to track the physical environment of the room (i.e. air temperature, wind speed, relative humidity, radiant temperature) to ensure the stability of the thermal environment. The temperature is set to -10℃ based on the average daily temperature from December to January in cold cities.

[0029] A preset percentage of street view images are randomly selected from the street view image database as a street view image data sample set, and subjective scores of the scoring samples by residents in the target area are obtained in the laboratory;

[0030] Furthermore, the method for local residents to subjectively score each street view image data in the street view image data sample set to obtain a subjective perception result corresponding to each street view image data in the street view image data sample set is:

[0031] Local residents give subjective perception scores to each street view image data in the street view image data sample set based on pleasure, boredom, depression and relaxation, and obtain subjective perception results;

[0032] The subjective perception results include: pleasure score, boredom score, depression score and relaxation score.

[0033] Specifically, based on the four-category emotion theory and related emotion measurement scales, four indicators, namely pleasure, boredom, depression, and relaxation, were selected to measure residents' subjective evaluation of the scene.

[0034] Furthermore, the method of obtaining the objective perception result corresponding to each street view image data in the street view image data sample set according to the acquired physiological signal is as follows: analyzing and scoring the acquired physiological signal by using EEG analysis software and a performance measurement algorithm to obtain the objective perception result corresponding to each street view image data in the street view image data sample set;

[0035] The objective perception results include: engagement score, excitement score, stress score, relaxation score, interest score and concentration score.

[0036] Furthermore, the EEG device is a 14-channel portable EEG device of model EMOTIV EPOC X.

[0037] Furthermore, the EEG analysis software is EMOTIVpro analysis software.

[0038] Specifically, residents use the EEG device EMOTIV EPOC X for monitoring during the scoring process, and EMOTIVpro generates objective emotion index scores in real time based on physiological signals. The physiological emotion perception index uses EEG data provided by the 14-channel mobile EEG headset EMOTIVEPOC X, and uses the official performance measurement algorithm to calculate real-time emotion metrics. These metrics are output at a frequency of 0.1Hz and cover six physiological emotion perception indicators: EEG for engagement (En), EEG for excitement (Ex), EEG for stress (St), EEG for relaxation (Re), EEG for interest (In), and EEG for concentration (At). To enhance the comparability between samples, the numerical range of the six physiological emotion indicators is adjusted to 0-1.

[0039] Furthermore, the street view perception model is a ResNet50 neural network.

[0040] The street view perception model is a random forest machine learning model.

[0041] Furthermore, the pleasure score, boredom score, depression score and relaxation score range from 1 to 100, respectively.

[0042] Furthermore, the engagement score, excitement score, stress score, relaxation score, interest score and concentration score are respectively in the range of 0 to 1.

[0043] Embodiment 1:

[0044] This embodiment takes the Zhongshan Road area in Nangang District, Harbin as an example to provide a measurement method for evaluating the emotional perception of cold-region cities. The urban center area surrounded by Zhongshan Road Street, Wenchang Street, and Dacheng Street is selected, with a total of 161 streets. The street types are rich and are typical representatives of cold-region urban space, such as Figure 2 shown.

[0045] This embodiment obtains winter urban street view image data through autonomous collection. Since the existing street view data does not include winter street views, this study uses PinSurvey software to autonomously collect street view data. Sampling points are selected every 25 meters along the road in the study area. Each sampling point captures street view images from four directions (0 degrees, 90 degrees, 180 degrees, and 270 degrees) at a fixed height, such as Figure 3 and Figure 4 shown.

[0046] According to the street view image data, a street view image database is established, and a street view image data sample set is divided;

[0047] A preset percentage of street view images are randomly extracted from the street view image database as scoring samples, and multi-level scores of residents in the target area on the sample set are obtained; the preset percentage in this embodiment is 15%, and 2580 street view images are randomly extracted from the street view database as the sample set. Based on the four-category theory of emotions and related emotion measurement scales, four indicators of pleasure, boredom, depression, and relaxation are selected to measure the participants' subjective evaluation of the scene, with scores ranging from 1 to 100 from bad to good.

[0048] A preset percentage of street view images are randomly selected from the street view image database as scoring samples. In the laboratory, the subjective perception results of residents in the target area for each street view in the sample set are obtained. During the scoring process, the EEG device EMOTIV EPOC X is used for monitoring, and EMOTIV pro simultaneously generates objective perception results in real time based on physiological signals.

[0049] The physiological emotion perception index uses EEG data provided by the 14-channel mobile EEG headset EMOTIV EPOC X to calculate real-time emotion metrics using the official performance measurement algorithm. These metrics are output at a frequency of 0.1Hz and cover six physiological emotion scores: engagement EEG (En) score, excitement EEG (Ex) score, stress EEG (St) score, relaxation EEG (Re) score, interest EEG (In) score, and concentration EEG (At) score, such as Figure 5 As shown; to enhance the comparability among samples, the numerical range of the six physiological emotion indicators was adjusted to 0-1.

[0050] Taking a street view image data sample set as input, and taking a subjective perception result and an objective perception result corresponding to the street view image data sample set as output to train a street view perception model, thereby obtaining a trained street view perception model;

[0051] In this embodiment, 80% of the street view image data sample set is randomly selected as a training set and 20% as a test set. Machine deep learning is performed through a random forest model based on the Python language, and parameters of the street view perception model are adjusted to make the street view perception model have a higher model score on the test set, obtain a more accurate machine score value, and finally obtain a trained street view perception model.

[0052] The street view images of the area to be evaluated obtained from four directions (0 degrees, 90 degrees, 180 degrees and 270 degrees) at a fixed height are respectively input into the trained street view perception model to obtain the subjective perception result of the 0 degree direction, the objective perception result of the 0 degree direction, the subjective perception result of the 90 degree direction, the objective perception result of the 90 degree direction, the subjective perception result of the 180 degree direction, the objective perception result of the 180 degree direction, the subjective perception result of the 270 degree direction, and the objective perception result of the 270 degree direction;

[0053] The subjective perception score of the area to be evaluated is obtained by taking the average of the subjective perception results in the 0-degree direction, the 90-degree direction, the 180-degree direction, and the 270-degree direction;

[0054] The objective perception score of the area to be evaluated is obtained by taking the average of the objective perception results in the 0-degree direction, the 90-degree direction, the 180-degree direction, and the 270-degree direction;

[0055] The subjective perception scores of the evaluated area include the mean scores of pleasure, boredom, depression and relaxation. The mean score of pleasure is calculated by the scores of pleasure in the 0-degree direction, pleasure in the 90-degree direction, pleasure in the 180-degree direction and pleasure in the 270-degree direction. Similarly, the mean scores of boredom, depression and relaxation can be calculated.

[0056] The objective perception scores of the area to be evaluated include the mean engagement score, the mean excitement score, the mean stress score, the mean relaxation score, the mean interest score and the mean concentration score; the mean engagement score is calculated by averaging the engagement score in the 0-degree direction, the engagement score in the 90-degree direction, the engagement score in the 180-degree direction and the engagement score in the 270-degree direction; similarly, the mean excitement score, the mean stress score, the mean relaxation score, the mean interest score and the mean concentration score can be calculated.

[0057] According to the Analytic Hierarchy Process (AHP), subjective perception weights and objective perception weights are assigned to street view images, and the comprehensive emotion evaluation result of the area to be evaluated is obtained according to the subjective perception score of the area to be evaluated, the objective perception score of the area to be evaluated, the subjective perception weight of the area to be evaluated and the objective perception weight of the area to be evaluated.

[0058] In this embodiment, according to the hierarchical analysis method, a weight of 0.7 is assigned to the subjective perception result and a weight of 0.3 is assigned to the objective perception result to obtain the perception score in CSV format. The comprehensive street sentiment evaluation results are placed in ArcGIS for visualization, as shown in Figure 7 shown.

[0059] It should be noted that although the method operations of the above embodiments are described in a specific order, this does not require or imply that these operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. On the contrary, the steps depicted can be performed in a different order. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps.

[0060] Embodiment 2:

[0061] With the help of deeplabv3 and ade-20k annotation datasets, 151 elements such as trees, buildings, and cars in winter street scene images are identified. Semantic recognition is analyzed and street space element indicators are calculated. Objective environmental evaluation is performed on street scene images. The environmental evaluation indicators of this embodiment include six indicators: external enclosure, sky openness, street green view rate, sidewalk width, roadway width, and vehicle density;

[0062] Street view recognition is performed on street view images through visual image semantic segmentation to obtain the proportion of multiple types of street view elements in the street view images.

[0063] Specifically, the obtained street view images are subjected to street view recognition through visual image semantic segmentation software, and the proportion of 150 types of street view elements in the street view images is obtained, and a CSV table file and a street view image after semantic segmentation are obtained. The principle of semantic segmentation is based on the scene expression vector analysis framework. The scene expression vector is composed of visual elements and is used to measure and express specific urban scenes. The scene semantic segmentation method based on deep learning calculates the semantic category to which each pixel in the image belongs, and obtains the visual elements in the scene (such as buildings, vehicles, sky, etc.). A multidimensional vector is formed, and each dimension in the vector represents the proportion of a specific category of objects (such as buildings) in the image.

[0064] The proportion of plants, mountains, buildings, roads, signs, and the sky in the identification results of various street scene elements were selected, and the external enclosure, sky openness, street green view rate, sidewalk width, lane width, and vehicle density were calculated respectively; the average values ​​of the six indicators of external enclosure, sky openness, street green view rate, sidewalk width, lane width, and vehicle density in the four directions of front, back, left, and right of each sampling point were calculated, and placed in ArcGIS for visualization, such as Figure 6 shown.

[0065] Furthermore, the method for obtaining the environmental factor index corresponding to each street view image data in different directions of the area to be evaluated through the visual image semantic segmentation model is as follows:

[0066] Environmental factor indicators include external enclosure, sidewalk width, roadway width, sky openness, street green view rate and vehicle density;

[0067] The total number of pixels in the face area, the number of pixels occupied by buildings, the number of pixels occupied by walls, the number of pixels occupied by sidewalks, the number of pixels occupied by roadways, the number of pixels occupied by the sky, the number of pixels occupied by trees, the number of pixels occupied by grass, the number of pixels occupied by shrubs, and the number of pixels occupied by motor vehicles in each street view image data in different directions of the area to be evaluated are obtained through the visual image semantic segmentation model;

[0068] The external surrounding degree of the street view image data is calculated according to the total number of surface pixels, the number of pixels occupied by walls and the number of pixels occupied by buildings in each street view image data in different directions; the sidewalk width of the street view image data is calculated according to the total number of surface pixels and the number of pixels occupied by the sidewalk in each street view image data in different directions; the roadway width of the street view image data is calculated according to the total number of surface pixels and the number of pixels occupied by the roadway in each street view image data in different directions; the sky openness of the street view image data is calculated according to the total number of surface pixels and the number of pixels occupied by the sky in each street view image data in different directions; the street green view rate of the street view image data is calculated according to the total number of surface pixels, the number of pixels occupied by trees, the number of pixels occupied by grass and the number of pixels occupied by shrubs in each street view image data in different directions; the vehicle density of the street view image data is calculated according to the total number of surface pixels and the number of pixels occupied by vehicles in each street view image data in different directions.

[0069] Furthermore, the calculation method of taking the average of the obtained multiple environmental factor indicators as the objective environmental evaluation score of the area to be evaluated is as follows:

[0070] The objective perception scores include average external enclosure, average sidewalk width, average road width, average sky openness, average street green view rate, and average vehicle density;

[0071] The calculation method of the average external surrounding OB is: Among them B iis the number of pixels occupied by buildings in the i-th street view image data, Sun i is the total number of pixels in the area of ​​the ith street view image data, W i is the number of pixels occupied by the wall in the i-th street view image data, n is the total number of street view image data, i∈[1,2,3,···,n];

[0072] The average sidewalk width WP is calculated as follows: Where P i is the number of pixels occupied by the sidewalk in the i-th street view image data;

[0073] The calculation method of average lane width WV is: Where R i is the number of pixels occupied by the road in the i-th street view image data;

[0074] The average sky openness is calculated as: Where S i is the number of pixels occupied by the sky in the i-th street view image data;

[0075] The calculation method of average street green view rate is: where t i is the number of pixels occupied by trees in the i-th street view image data, g i is the number of pixels occupied by grass in the i-th street view image data, p i is the number of pixels occupied by shrubs in the i-th street view image data;

[0076] The average vehicle density is calculated as: Among them C i is the number of pixels occupied by motor vehicles in the i-th street view image data.

[0077] The correlation analysis between residents' emotional perception evaluation and spatial factor indicators was conducted. From the goodness of fit, it can be seen that the overall fitting effect of the GWR (spatial analysis technology) model is better than that of the OLS (ordinary least squares) model. In the OLS model, the relationship between some spatial factor indicators and emotional perception is not significant, but in the GWR model, it still shows a significant local correlation. This proves that the GWR model can detect more details. The variance inflation coefficient (VIF) of the six types of street spatial factor indicators is in the range of [1.995990, 6.174060], which is much less than 10 (Table 1), indicating that there is no obvious collinearity between the elements and no need to eliminate them.

[0078] Because the subjective emotion perception data differ significantly, a more detailed description of the correlation and influence between it and street space elements is given. By comparing the OLS coefficient and GWR coefficient in Table 1, the most robust influencing factors of different emotion perceptions can be found. The most robust influencing factors of boredom are sky openness, external enclosure, and street green view rate. Among them, sky openness, street green view rate and boredom are significantly negatively correlated. The most robust influences on pleasure are sky openness and external enclosure. Pleasure is significantly negatively correlated with external enclosure, and significantly positively correlated with sky openness. That is, too many or too high surrounding buildings blocking the sky can cause people to feel depressed. The influence of street green view rate on pleasure perception is weaker than the previous indicators, which should be related to the state of plants in winter. In the street view images collected in winter, most trees and shrubs are in the state of dead branches, which is difficult to arouse a high degree of pleasure perception. Indicators such as external enclosure, sky openness, street greening rate, and vehicle density have a strong impact on the sense of depression. Except for external enclosure and vehicle density, the others are significantly negatively correlated, indicating that the wider the view on the street, the higher the sky openness, and the more street plants, the lower the sense of depression. On the contrary, the denser the vehicles on the street and the higher the external enclosure, the more it will lead to an increase in the sense of depression. In addition, the more robust influencing factors of relaxation include street greening rate, sky openness, external enclosure, and road width. Except for external enclosure, other environmental factors are significantly positively correlated with relaxation. This shows that the lower the external enclosure, the higher the openness and greening rate, the wider the road width, and the stronger the sense of relaxation.

[0079] In the analysis of physiological perception and street space elements, it was found that large-area elements such as the sky and the road surface have a particularly obvious impact on physiological perception. For example, the openness of the sky can significantly affect EEG (Ex), EEG (In) and EEG (At). Vehicle density is strongly positively correlated with EEG (En) and EEG (At), and significantly negatively correlated with EEG (Re) and EEG (In), indicating that vehicles can significantly reduce the relaxation and interest of the crowd, and increase concentration and excitement. A small part of the data has a small correlation. This may be because the monitoring sensitivity of the equipment is low and there are errors in the calculation of software indicators. In the future, deeper processing of physiological signal data may significantly improve its correlation.

[0080] The results of the correlation, OLS, and GWR model analysis between winter street space elements and emotional perception are as follows:

[0081]

[0082]

[0083] Note: **The correlation is significant at the 0.01 level (two-tailed). *The correlation is significant at the 0.05 level (two-tailed).

[0084] In the analysis of subjective perception and street space elements, it was found that subjective emotional perception and various street space elements were significantly correlated, and external enclosure and sky openness had the greatest impact on emotional perception. The same spatial element had significant differences in the impact on emotional perception in different types of spaces. For example, large areas of greenery and plants in vast landscape parks can significantly relieve people's depressed emotions and make people feel relaxed, while in some scenic spots or pedestrian streets, streets with high green view rates make residents feel stressed. In most study areas, the increase in sky openness can make residents feel more positive, but in streets with single facilities and deserted streets, high sky openness may lead to a stronger sense of depression. This may mean that the combination of environmental elements can also affect users' emotional perception. These findings help designers to better understand the psychological impact of the physical environment of urban streets on people.

[0085] In summary, the present invention is based on street view image data, and through the trained street view perception model, street view images are scored to achieve subjective and objective perception of street view images, and the subjective and objective perceptions are combined to accurately measure the quality of urban living street space. It has the advantages of high efficiency and easy promotion and application, and can be widely used in the field of urban planning living street space quality survey and measurement.

[0086] Specific implementation method 2: The technical solution in the embodiment of the present invention will be clearly and completely described below:

[0087] A cold-region city emotion perception evaluation and measurement system, characterized by comprising a collection module, a subjective perception module, an objective perception module, a model training module and a measurement module;

[0088] The acquisition module is used to obtain street view image data of the target area to establish a street view image data sample set; the street view image data sample set is sent to the subjective perception module, the objective perception module and the model training module;

[0089] The subjective perception module is used to obtain a subjective perception result corresponding to each street view image data in the street view image data sample set according to the received street view image data sample set; and send the subjective perception result to the model training module;

[0090] The objective perception module is used to obtain an objective perception result corresponding to each street view image data in the street view image data sample set according to the received street view image data sample set; and send the objective perception result to the model training module;

[0091] The model training module is used to take the received street view image data sample set as input, and take the received subjective perception results and objective perception results corresponding to the street view image data sample set as output to train the street view perception model, so as to obtain the trained street view perception model;

[0092] The acquisition module is also used to obtain multiple street view images in different directions of the area to be evaluated, and input the multiple street view image data in different directions of the area to be evaluated into the trained street view perception model;

[0093] The trained street view perception model outputs a subjective perception result and an objective perception result corresponding to each street view image data of the area to be evaluated based on a plurality of street view images of different directions of the area to be evaluated, and obtains a plurality of subjective perception results and a plurality of objective perception results in total, and sends the obtained plurality of subjective perception results and a plurality of objective perception results to the measurement module;

[0094] The measurement module is used to calculate the average value of multiple received subjective perception results to obtain the subjective perception score of the area to be evaluated, and calculate the average value of multiple received objective perception results to obtain the objective perception score of the area to be evaluated;

[0095] The measurement module is also used to calculate the comprehensive emotion evaluation result of the area to be evaluated based on the calculated subjective perception score, objective perception score and the set subjective perception weight of the area to be evaluated and the objective perception weight of the area to be evaluated.

[0096] Specific implementation method three: A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described in specific implementation method one when executing the computer program.

[0097] Although the present invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. It should therefore be understood that many modifications may be made to the exemplary embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the various dependent claims and features described herein may be combined in a manner different from that described in the original claims. It should also be understood that features described in conjunction with individual embodiments may be used in other described embodiments.

Claims

1. A method for evaluating and measuring emotional perception in cold-region cities, characterized by: include: Step 1: Acquire street view image data of the target area to establish a street view image data sample set; Step 2: Local residents subjectively score each street view image data in the street view image data sample set to obtain a subjective perception result corresponding to each street view image data in the street view image data sample set; when local residents subjectively score each street view image data in the street view image data sample set, physiological signals of residents are obtained through electroencephalographic equipment, and objective perception results corresponding to each street view image data in the street view image data sample set are obtained based on the obtained physiological signals; Step 3: taking the street view image data sample set as input, and taking the subjective perception results and objective perception results corresponding to the street view image data sample set as output to train the street view perception model, thereby obtaining a trained street view perception model; Step 4: obtaining multiple street view images from different directions of the area to be evaluated, and inputting the multiple street view image data from different directions of the area to be evaluated into the trained street view perception model respectively, to obtain subjective perception results and objective perception results corresponding to each street view image data from different directions of the area to be evaluated, and to obtain a total of multiple subjective perception results and multiple objective perception results, and taking the average value of the multiple subjective perception results as the subjective perception score of the area to be evaluated, and taking the average value of the multiple objective perception results as the objective perception score of the area to be evaluated; Step 5: Set the subjective perception weight and the objective perception weight of the area to be evaluated, and obtain the comprehensive emotion evaluation result of the area to be evaluated according to the subjective perception score, the objective perception score, the subjective perception weight and the objective perception weight of the area to be evaluated.

2. The cold-region city emotion perception evaluation and measurement method according to claim 1 is characterized by: The method for local residents to subjectively score each street view image data in the street view image data sample set to obtain the subjective perception result corresponding to each street view image data in the street view image data sample set is: Local residents give subjective perception scores to each street view image data in the street view image data sample set based on pleasure, boredom, depression and relaxation, and obtain subjective perception results; The subjective perception results include: pleasure score, boredom score, depression score and relaxation score.

3. The method for evaluating and measuring emotional perception in cold regions according to claim 2, characterized in that: The method for obtaining the objective perception result corresponding to each street view image data in the street view image data sample set according to the acquired physiological signal is as follows: analyzing and scoring the acquired physiological signal by using EEG analysis software and a performance measurement algorithm to obtain the objective perception result corresponding to each street view image data in the street view image data sample set; The objective perception results include: engagement score, excitement score, stress score, relaxation score, interest score and concentration score.

4. The method for evaluating and measuring emotional perception in cold regions according to claim 1, characterized in that: The EEG device is a 14-channel portable EEG device, model EMOTIV EPOC X.

5. The method for evaluating and measuring emotional perception in cold regions according to claim 3 is characterized by: The EEG analysis software is EMOTIVpro analysis software.

6. The method for evaluating and measuring emotional perception in cold regions cities according to claim 1, characterized in that: The street view perception model is the ResNet50 neural network.

7. The method for evaluating and measuring emotional perception in cold regions according to claim 2, characterized in that: The scores for pleasure, boredom, depression, and relaxation ranged from 1 to 100, respectively.

8. The method for evaluating and measuring emotional perception in cold regions according to claim 3 is characterized by: The engagement score, excitement score, stress score, relaxation score, interest score and concentration score range from 0 to 1 respectively.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images