CN114298415B - Method and device for predicting pleasure degree of geographic environment - Google Patents
Method and device for predicting pleasure degree of geographic environment Download PDFInfo
- Publication number
- CN114298415B CN114298415B CN202111639205.0A CN202111639205A CN114298415B CN 114298415 B CN114298415 B CN 114298415B CN 202111639205 A CN202111639205 A CN 202111639205A CN 114298415 B CN114298415 B CN 114298415B
- Authority
- CN
- China
- Prior art keywords
- landscape
- pleasure
- image
- sample
- analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of emotion prediction, in particular to a method and a device for predicting the pleasure degree of a geographic environment. Firstly, obtaining visual elements of a sample image, and determining a landscape category to which the sample image belongs according to the proportion of the visual elements; then obtaining an image containing average pleasure scores of different age group and gender combined crowds as an analysis sample, performing significance test and correlation analysis on the analysis sample, determining age group and gender combinations and key visual elements influencing the pleasure scores of the sample image, and finally respectively taking the age group, gender, average pleasure scores and key visual element ratios corresponding to different types of landscape images in the analysis sample as input to construct a mixed linear equation of the pleasure scores of the different types of landscape images; and predicting the pleasure degree of the image to be measured by using a mixed linear equation. The pleasure degree scoring mixed linear equation constructed by the method can accurately predict the pleasure degree of the image to be measured.
Description
Technical Field
The application relates to the technical field of emotion prediction, in particular to a method and a device for predicting the pleasure degree of a geographic environment.
Background
The environmental perception refers to subjective feeling and psychological judgment of people on the surrounding environment and changes of the surrounding environment, is a psychological basis of environmental behaviors of people, and the perception of people on public environments can influence the external behaviors and physical and psychological health of people, so that the method has important significance in researching the relevance of the joyfulness of people and urban environments.
At present, the related art has carried out the research work of the relevance of the pleasure degree and the urban environment, but the research content is relatively single, and the influence of people of different age groups and different sexes and urban visual elements on the pleasure degree is not comprehensively considered. Therefore, it is necessary to provide a new method for accurately predicting the pleasure of the image to be measured.
Disclosure of Invention
The application provides a method and a device for predicting the pleasure degree of a geographic environment, which can accurately predict the pleasure degree of an image to be detected.
In a first aspect, an embodiment of the present application provides a method for predicting a pleasure degree of a geographic environment, including:
performing semantic segmentation on a sample image to obtain at least one visual element of the sample image;
determining the landscape type of the sample image according to the ratio of each visual element of the sample image; wherein the landscape types include urban landscape, transitional landscape and natural landscape;
acquiring the average pleasure degree score of each age group and gender group combined people on each sample image, and taking the image containing the average pleasure degree scores of different age group and gender group combined people as an analysis sample;
performing significance testing and correlation analysis on the analysis sample, and determining age group gender combination and key visual elements which influence the pleasure degree score of the sample image;
respectively taking age groups, sexes, average pleasure degree scores and key visual element ratios corresponding to different types of landscape images in the analysis sample as input, and constructing a mixed linear equation of the pleasure degree scores of the different types of landscape images;
and predicting the pleasure degree of the image to be measured by using the mixed linear equation.
In one possible design, the semantically segmenting the sample image to obtain at least one visual element of the sample image includes:
extracting characteristic information of the sample image by using a ResNet model;
extracting mapping characteristic information of the sample image by using a pyramid model;
and performing feature fusion on the feature information of the sample image and the mapping feature information, and performing convolution operation to obtain the visual elements of each sample image.
In one possible design, before said obtaining an average hedonic score for each age-gender group of people for each of said sample images, comprising:
obtaining an effective questionnaire; wherein each of the available questionnaires comprises gender, age, and a score for pleasure for a predetermined number of the sample images;
and counting age group gender combinations in the effective questionnaire and scoring the pleasure degree of the sample image.
In one possible design, the performing a significance test and a correlation analysis on the analysis sample to determine a gender age combination and key visual elements that affect a pleasure score of the sample image comprises:
performing a mann-whitney U test on the analysis sample to determine age group gender combinations that affect a pleasure score for the sample image;
and performing spearman analysis on the analysis sample, and determining key visual elements influencing the pleasure degree score of the sample image.
In one possible design, the performing the mann-whitney U test on the analysis sample to determine age group gender combinations that affect the enjoyment scores of the sample images comprises:
performing a gender mann-whitney U test on the analysis sample to determine the effect of different genders on the sample image pleasure score;
performing a landscape type mann-whitney U test on the analysis sample to determine the effect of different landscape types on the sample image pleasure score;
an age-bracket mann-whitney U test was performed on the assay sample to determine the effect of different age brackets on the sample image pleasure score.
In one possible design, the spearman analysis of the analysis sample to determine key visual elements that affect the enjoyment score of the sample image comprises:
performing spearman analysis on an analysis sample belonging to the urban landscape to determine key visual elements influencing the pleasure degree score of the urban landscape image;
performing spearman analysis on an analysis sample belonging to the transitional landscape to determine key visual elements influencing the pleasure degree score of the transitional landscape image;
spearman analysis is performed on the analysis samples belonging to the natural landscape to determine key visual elements that affect the pleasure score of the natural landscape image.
In one possible design, the constructing a mixed linear equation of the pleasure scores of the different types of landscape images by respectively taking the age group, the gender, the average pleasure score and the key visual element ratio corresponding to the different types of landscape images in the analysis sample as input comprises:
taking the age bracket, the gender, the average pleasure degree score and the key visual element ratio corresponding to each urban landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the urban landscape images;
taking the age bracket, the sex, the average pleasure degree score and the key visual element ratio corresponding to each transitional landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the transitional landscape images;
and taking the age bracket, the sex, the average pleasure degree score and the key visual element ratio corresponding to each natural landscape image in the analysis sample as input to obtain a mixed linear equation of the pleasure degree scores of the natural landscape images.
In a second aspect, an embodiment of the present application further provides an apparatus for predicting a pleasure degree of a geographic environment, including:
the segmentation module is used for performing semantic segmentation on the sample image to obtain at least one visual element of the sample image;
the classification module is used for determining the landscape type of the sample image according to the proportion of each visual element of the sample image; wherein the landscape types include urban landscape, natural landscape and transitional landscape;
the acquisition module is used for acquiring the average pleasure degree score of each age group and gender group combined person on each sample image and taking the image containing the average pleasure degree scores of different age group and gender group combined persons as an analysis sample;
the analysis module is used for carrying out significance test and correlation analysis on the analysis sample and determining age group gender combination and key visual elements which influence the pleasure degree score of the sample image;
the construction module is used for respectively taking age groups, sexes, average pleasure degree scores and key visual element ratios corresponding to different types of landscape images in the analysis sample as input, and constructing a mixed linear equation of the pleasure degree scores of the different types of landscape images;
and the prediction module is used for predicting the pleasure degree of the image to be measured by utilizing the mixed linear equation.
In a third aspect, an embodiment of the present invention further provides a computing device, including a memory and a processor, where the memory stores a computer program, and the processor, when executing the computer program, implements the method described in any one of the above.
In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed in a computer, the computer program causes the computer to execute any one of the methods described above.
According to the method, firstly, semantic segmentation is carried out on a sample image to obtain at least one visual element of the sample image, secondly, the landscape type of the sample image is determined according to the proportion of each visual element of the sample image, then the average pleasure degree score of each age group and sex combination crowd on each sample image is obtained, the image containing the average pleasure degree scores of different age group and sex combination crowd is used as an analysis sample, then, significance test and correlation analysis are carried out on the analysis sample to determine age group and sex combination and key visual elements influencing the pleasure degree scores of the sample image, and finally, the age group, sex, average pleasure degree scores and key visual element proportions corresponding to different types of landscape images in the analysis sample are used as input to construct a mixed linear equation of the pleasure degree scores of the different types of landscape images; and predicting the pleasure degree of the image to be measured by using a mixed linear equation. The method for predicting the pleasure degree provided by the application considers the influence of age bracket, gender and visual elements on the pleasure degree score, and the mixed linear equation constructed by the method can accurately predict the pleasure degree of the image to be detected.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic diagram of a testing method for the pleasure degree of the geographic environment according to an embodiment of the present invention;
FIG. 2 is a diagram of a hardware architecture of a computing device according to an embodiment of the present invention;
fig. 3 is a block diagram of a testing apparatus for measuring a pleasure degree of a geographic environment according to an embodiment of the present invention.
Detailed Description
The present application will be described in detail below with reference to the drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In the description of the embodiments of the present application, the terms "first", "second", and the like, unless expressly specified or limited otherwise, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance; the term "plurality" means two or more unless specified or indicated otherwise; the terms "connected," "fixed," and the like are to be construed broadly and may, for example, be fixedly connected, detachably connected, integrally connected, or electrically connected; may be directly connected or indirectly connected through an intermediate. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art as appropriate.
In the description of the present application, it should be understood that the terms "upper" and "lower" used in the description of the embodiments of the present application are used in a descriptive sense only and not for purposes of limitation. In addition, in this context, it will also be understood that when an element is referred to as being "on" or "under" another element, it can be directly on "or" under "the other element or be indirectly on" or "under" the other element via an intermediate element.
As described above, when the relevance between the pleasure degree of the population and the urban environment is studied, the content of the study is relatively single, and the influence of the age group, the gender and the visual element on the pleasure degree is not considered, so that the pleasure degree of the image to be measured cannot be accurately predicted.
In order to solve the technical problem, the relevance between the pleasure degree and the age group, the gender and the visual elements can be determined, and then a mixed linear equation containing the combination of the key visual elements and the gender of the age group is constructed to be used for predicting the pleasure degree of the image to be measured.
As shown in fig. 1, an embodiment of the present invention provides a method for predicting a pleasure degree of a geographic environment, including the following steps:
step 100: performing semantic segmentation on the sample image to obtain at least one visual element of the sample image;
step 102: determining the landscape type of the sample image according to the proportion of each visual element of the sample image; wherein, the landscape types comprise urban landscape, transitional landscape and natural landscape;
step 104: acquiring the average pleasure degree score of each age group and gender group combined people on each sample image, and taking the image containing the average pleasure degree scores of different age group and gender group combined people as an analysis sample;
step 106: carrying out significance test and correlation analysis on the analysis sample, and determining age group gender combination and key visual elements which influence the pleasure degree score of the sample image;
step 108: respectively taking age groups, sexes, average pleasure degree scores and key visual element ratios corresponding to different types of landscape images in the analysis sample as input, and constructing a mixed linear equation of the pleasure degree scores of the different types of landscape images;
step 110: and predicting the pleasure degree of the image to be measured by using a mixed linear equation.
In the embodiment of the invention, semantic segmentation is carried out on a sample to be detected to obtain each visual element of the sample image, and the landscape category to which the sample image belongs is determined according to the proportion of each visual element; the method comprises the steps of obtaining an image containing average pleasure scores of people combined with different ages as an analysis sample, determining the relevance between the pleasure and the ages, gender and each visual element by performing significance test and correlation analysis on the sample, and then constructing a mixed linear equation containing key visual elements and age gender combinations for predicting the pleasure of the image to be measured. The pleasure degree prediction method provided by the application is high in accuracy.
It should be noted that, since the present application mainly studies the influence of the urban environment on the pleasure degree of the people, the sample image provided by the present embodiment is a street view image. In the embodiment, 101 street view images are selected as sample images, and in order to ensure the diversity of the sample images, the sample images cover three types of urban landscapes, transitional landscapes and natural landscapes. It will be appreciated that although the sample images of the present application are taken primarily from urban environments, they are equally applicable to field environments, since for any geographic environment image, the visual elements contained therein are constant, but the proportion of visual elements is different, resulting in different types of landscape to which the image belongs. Therefore, the method for testing the pleasure degree is applicable to all geographic environments and is not limited to urban environments.
It should be further noted that the street view image may be obtained by taking a picture, or may be obtained by collecting navigation software such as a Baidu map, a Gaode map, and the like, as long as the visual information in the urban space can be better preserved and the scene repetition is reduced, and the source of the street view image is not specifically limited in the present application.
The implementation of each step is described below.
With respect to step 100, in some embodiments, comprises:
extracting characteristic information of the sample image by using a ResNet model;
extracting mapping characteristic information of the sample image by using the pyramid model;
and performing feature fusion on the feature information of the sample image and the mapping feature information, and performing convolution operation to obtain the visual elements of each sample image.
In the embodiment, all visual elements of each sample image can be accurately obtained through semantic segmentation so as to be used for landscape classification and correlation analysis of subsequent sample images.
In some embodiments, a PSPNet scene analysis network model is selected and combined with an ADE 20K training image data set to perform semantic segmentation on the acquired sample image based on a scene visual element classification method of a PyTorch deep learning framework and a CUDA computing platform. In the segmentation process, a sample image can be firstly input into a PSPNet model, the feature information of the sample image is extracted through a pre-trained ResNet model, the context information of the sample image is collected through a 4-layer pyramid pooling module and is fused into global prior information, finally, the mapping feature information obtained by the pyramid model and the feature information extracted by the ResNet model are fused, and convolution is carried out to obtain all visual elements of each sample image. It should be noted that, this embodiment only provides a preferred semantic segmentation manner, but is not limited thereto.
It should be noted that, in general, each sample image can be divided into nine major visual elements, such as sky, vegetation, water, bare soil, buildings, roads and accessories, sidewalks, vehicles and people. By the processing, on the premise of not influencing the segmentation precision, the segmentation speed can be improved, and the occupation of computer resources is reduced.
With respect to step 102, in some embodiments, comprises:
building elements, roads and accessory elements in the nine types of visual elements are classified as city elements;
dividing the landscape type of each sample image according to the ratio of the vegetation elements and the city elements in the sample image; generally, the landscape can be divided into three categories, namely urban landscape, transitional landscape and natural landscape, wherein,
and (3) urban landscape: the vegetation element proportion is less than 30 percent and the urban element proportion is more than 30 percent;
and (3) natural landscape: the vegetation element proportion is more than 30 percent and the urban element proportion is less than 30 percent;
and (3) transition landscape: transition landscapes are arranged between the urban landscape and the natural landscape.
In this embodiment, by dividing the sample image into different types of landscapes, it is advantageous to subsequently study the influence of the landscape type on the pleasure. In addition, with the classification method of the present embodiment, 26 city landscape images, 32 natural landscape images, and 43 transition landscape images can be obtained from 101 sample images for making a questionnaire and conducting a pleasure analysis.
With respect to step 104, in some embodiments, prior to performing step 104, comprising:
obtaining an effective questionnaire; wherein each valid questionnaire comprises gender, age and a score for the pleasure degree of a preset number of sample images;
age group gender combinations in a statistically valid questionnaire and a score on the pleasure of the sample images.
In this embodiment, by using a questionnaire mode, the gender and age of the person to be investigated and the score of the real pleasure degree of the sample image can be obtained, then the questionnaire with the missing age, gender or partial image score is removed, only the effective questionnaire is retained, finally, the gender combination of the age group in the effective questionnaire and the score of the pleasure degree of the sample image are counted, and the integrity and accuracy of the statistical information can be ensured.
It should be noted that in some embodiments, a web questionnaire may be used, which not only saves time and labor costs, but also makes it easier to obtain more people of different ages and sexes, thereby improving the coverage of the survey results. Of course, paper questionnaires can also be adopted and distributed and collected manually, but the acquisition time and difficulty are increased correspondingly, and the application does not specifically limit the acquisition form of the questionnaires. In addition, in the present embodiment, a five-point system scoring standard is adopted, as shown in table 1, but the present invention is not limited thereto, and in some embodiments, a ten-point system scoring standard may be adopted.
TABLE 1 pleasure rating Scale
| Scoring | 1 | 2 | 3 | 4 | 5 |
| Corresponding emotion | Is very annoying | Bothersome | In general | Pleasure of | Is very pleasant |
It should be noted that the number of sample images in each questionnaire may be 10, and the sample images are randomly extracted from the sample images, so that it is ensured that the sample images in all questionnaires cover all the sample images as much as possible.
It should be noted that, in order to fully study the influence of the age and the gender of the population on the pleasure degree score, the ages can be divided into 6 ages of 10-19, 20-29, 30-39, 40-49, 50-59 and 60 years, and then combined with the gender to form a gender combination of 12 ages, so that the total ages and the genders can be basically covered, and the influence of different people in each age on the pleasure degree score can be determined.
Referring to step 104, in some embodiments, for a sample image, there may be a plurality of pleasure scores of people of the same gender within a specific age group, and then an average pleasure score of the gender group of the age group needs to be taken to ensure that the pleasure scores are representative, and the number of analysis samples is reduced to save computer resources. For example, in the age range of 10 to 19 years, 5 men score the image of the sample A, and the scores are 3 points, 2 points, 3 points and 3 points respectively, so that the average score is 2.8 points, that is, the pleasure degree of the male in the age range of 10 to 19 years on the image of the sample A is 2.8 points. It should be noted that if only male scores and no female scores are given for an image within a certain age range, the sample image is invalid and needs to be removed from the questionnaire. In this way, only the image containing the average pleasure score of the gender combination population in the whole age group is taken as an analysis sample, and the influence of the gender combination in each age group on the pleasure score of the sample image can be accurately determined.
In this embodiment, by using this statistical method, 881 effective analysis samples are obtained from the questionnaire, in which each sample image includes average pleasure scores of people of different genders and different ages, and the influence of the genders, the ages, the landscape types, and the visual elements on the pleasure scores can be determined.
With respect to step 106, in some embodiments, comprises:
step A1: performing a Mann-Whitney U test on the analysis sample, and determining age group gender combinations influencing the pleasure degree score of the sample image;
step A2: and (4) performing spearman test on the analysis sample, and determining key visual elements influencing the pleasure degree score of the sample image.
In this embodiment, by performing the mann-whitney U test and the spearman test on the analysis sample, the influence of different genders, people of different age groups, different types of landscape on the score of the image pleasure of the sample, and key visual elements affecting the score of the image pleasure can be determined. And a mixed linear equation of the pleasure degree score is constructed by using the key factors determined by the analysis result, so that the accuracy is higher.
With respect to step a1, in some embodiments, comprises:
performing a gender Many-Whitney U test on the analysis sample to determine the influence of different genders on the score of the sample image pleasure;
performing landscape type ManWhitney U test on the analysis sample to determine the influence of different landscape types on the score of the sample image pleasure degree;
the age group mann-whitney U test was performed on the analysis samples to determine the effect of different age groups on the sample image pleasure score.
In this example, by performing the mann-whitney U test on 881 analysis samples, it can be determined that: (1) different sexes had no significant effect on the hedonic score; (2) different landscape types have a significant impact on the pleasure score; the overall pleasure degree of the natural landscape image is the highest, the transitional landscape image is the second highest, and the overall pleasure degree of the urban landscape image is the lowest; (3) different age groups have significant effects on the pleasure score, and as age increases, the pleasure score decreases first and then increases; wherein, the pleasure scores of young children aged 10-29 years and middle-aged people aged 30-39 years have significant difference (p <0.05), which indicates that the pleasure perception of the two groups on the sample image is significantly different; and the significant difference (p <0.05) exists between the middle aged people at the age of 30-49 and the middle aged and the elderly people at the age of 50, which shows that the pleasure perception of the middle aged people at the age of 30-49 and the middle aged and the elderly people at the age of 50 is also significantly different.
In some embodiments, the age and gender attributes of the interviewee may also be cross-analyzed with the scene type of the sample image to determine the effect of people of different genders and different age groups on the enjoyment scores of the sample image of different scene types, as shown in table 2:
TABLE 2 pleasure rating for different types of landscape images for different age group gender combinations
| City landscape | Transitional landscape | Natural landscape | |
| Sex | |||
| For male | 2.92 | 3.41 | 3.62 |
| Woman | 2.82 | 3.43 | 3.56 |
| Age group | |||
| 10～19 | 3.03 | 3.50 | 3.72 |
| 20～29 | 3.02 | 3.44 | 3.45 |
| 30～39 | 2.77 | 3.43 | 3.45 |
| 40～49 | 2.70 | 3.34 | 3.57 |
| 50～59 | 2.71 | 3.55 | 3.95 |
| 60 and above | 2.95 | 3.53 | 3.73 |
| Total mean of | 2.90 | 3.34 | 3.58 |
| Total standard deviation of | 0.77 | 0.76 | 0.77 |
As can be seen from table 2: for urban landscapes and natural landscapes, the pleasure degree score of men is slightly higher than that of women, and the pleasure degree scores of the two genders for transitional landscapes are basically consistent; compared with the urban landscape, the transitional landscape and the natural landscape have higher delight degree scores, which indicates that the urban landscape can bring more unpleasant emotions to people; the pleasure degree score of the natural landscape is also obviously higher than that of the transition landscape, which shows that the natural landscape has the highest perception effect on the pleasure degree of people.
With respect to step a2, in some embodiments, comprises:
performing spearman analysis on an analysis sample belonging to the urban landscape to determine key visual elements influencing the pleasure degree score of the urban landscape image;
performing spearman analysis on an analysis sample belonging to the transitional landscape to determine key visual elements influencing the pleasure degree score of the transitional landscape image;
spearman analysis is performed on the analysis samples belonging to the natural landscape to determine key visual elements that affect the pleasure score of the natural landscape image.
In this example, by performing spearman analysis on 881 analysis samples, it can be determined that:
(1) in urban landscapes, the pleasure degree score is obviously related to vegetation elements, building elements and road elements (p is less than 0.01), the vegetation elements and the road elements are positively related to the pleasure degree score, the positive correlation coefficient is 0.36, the building elements are negatively related to the pleasure degree score, and the negative correlation coefficient is-0.37;
(2) in the transitional landscape, the pleasure degree score is obviously related to vegetation elements, building elements and sidewalk elements (p is less than 0.01), the sidewalk elements are positively related to the pleasure degree score, the correlation coefficient is 0.19, the vegetation elements are positively related to the pleasure degree score, the positive correlation coefficient is 0.16, the building elements are negatively related to the pleasure degree score, and the negative correlation coefficient is-0.15;
(3) in natural landscapes, the pleasure degree score is obviously related to sky elements and vegetation elements, the sky elements and the pleasure degree score are in negative correlation, the negative correlation coefficient is-0.17, the vegetation and the pleasure degree score are in positive correlation, and the positive correlation coefficient is 0.15.
With respect to step 108, in some embodiments, comprises:
and taking the age group, the gender, the average pleasure degree score and the key visual element ratio corresponding to each urban landscape image in the analysis sample as input to obtain a mixed linear equation of the pleasure degree scores of the urban landscape images:
and taking the age group, the gender, the average pleasure degree score and the key visual element ratio corresponding to each transition landscape image in the analysis sample as input to obtain a mixed linear equation of the pleasure degree scores of the transition landscape images:
and taking the age group, the gender, the average pleasure degree score and the key visual element ratio corresponding to each natural landscape image in the analysis sample as input to obtain a mixed linear equation of the pleasure degree scores of the natural landscape images:
wherein City Hedonic rating score 、Mix Hedonic rating score 、Nature Hedonic rating score Respectively scoring the pleasure degrees of the urban landscape, the transition landscape and the natural landscape; p is a radical of Vegetation 、p Construction of buildings 、p Road 、p Sidewalk And p Sky Respectively showing the proportions of the visual elements of vegetation elements, building elements, road elements, sidewalk elements and sky elements in the sample image; g Age (age) ：G Sex The combined effect intercept for gender groups of a particular age group is shown in table 3; the significance level is 0, the significance level is 0.01, and the significance level is 0.05.
TABLE 3 Effect intercept for gender specific age groups in combination
In the embodiment, a mixed linear equation containing sex combinations and key visual elements of different age groups is constructed for different types of landscapes by combining the analysis results of the Mann-Whitney U test and the Spireman analysis, and the determined linear equation can better predict the pleasure degree of the image to be measured.
It should be noted that in table 3, the effect of gender on the effect intercept was not considered for the transition and nature landscapes, mainly because the gender has less effect on the pleasure score in the two landscape types, which is consistent with the conclusion that different genders have substantially the same pleasure scores for the transition and nature landscapes as the mann-whitney U test.
As shown in fig. 2 and 3, the embodiment of the invention also provides a testing device for the pleasure degree of the geographic environment. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. From a hardware aspect, as shown in fig. 2, a hardware architecture diagram of a computing device in which a testing apparatus for measuring a pleasure degree of a geographic environment according to an embodiment of the present invention is located is provided, where, in addition to the processor, the memory, the network interface, and the non-volatile storage shown in fig. 2, the computing device in which the apparatus is located may generally include other hardware, such as a forwarding chip responsible for processing a packet, and the like. Taking a software implementation as an example, as shown in fig. 3, as a logical apparatus, a CPU of a computing device in which the apparatus is located reads a corresponding computer program in a non-volatile memory into a memory to run.
As shown in fig. 3, the present embodiment provides a testing apparatus for measuring a pleasure degree of a geographic environment, including:
a segmentation module 300, configured to perform semantic segmentation on the sample image to obtain at least one visual element of the sample image;
the classification module 302 is configured to determine a landscape type to which the sample image belongs according to a ratio of each visual element of the sample image; wherein, the landscape types comprise urban landscape, natural landscape and transition landscape;
an obtaining module 304, configured to obtain an average pleasure score of each age group and gender group for each sample image, and use an image containing the average pleasure score of the age group and gender group as an analysis sample;
the analysis module 306 is used for carrying out significance test and correlation analysis on the analysis sample and determining age group gender combination and key visual elements which influence the pleasure degree score of the sample image;
the construction module 308 is configured to respectively use age groups, genders, average happiness scores and key visual element ratios corresponding to different types of landscape images in the analysis sample as inputs, and construct a mixed linear equation of the happiness scores of the different types of landscape images;
and the prediction module 310 is configured to predict the pleasure degree of the image to be measured by using a mixed linear equation.
In an embodiment of the present invention, the segmentation module 300 may be configured to perform step 100 in the above-described method embodiment, the classification module 302 may be configured to perform step 102 in the above-described method embodiment, the obtaining module 304 may be configured to perform step 104 in the above-described method embodiment, the analysis module 306 may be configured to perform step 106 in the above-described method embodiment, the construction module 308 may be configured to perform step 108 in the above-described method embodiment, and the prediction module 310 may be configured to perform step 110 in the above-described method embodiment.
In one embodiment of the present invention, the segmentation module 300 is configured to perform the following operations:
extracting characteristic information of the sample image by using a ResNet model;
extracting mapping characteristic information of the sample image by using the pyramid model;
and performing feature fusion on the feature information of the sample image and the mapping feature information, and performing convolution operation to obtain the visual elements of each sample image.
In one embodiment of the present invention, the analysis module 306 is configured to perform the following operations:
performing a Mann-Whitney U test on the analysis sample, and determining age group gender combinations influencing the pleasure degree score of the sample image;
and (4) carrying out spearman analysis on the analysis sample, and determining key visual elements influencing the pleasure degree score of the sample image.
In one embodiment of the present invention, the building module 308 is configured to perform the following operations:
taking the age bracket, the gender, the average pleasure degree score and the key visual element ratio corresponding to each urban landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the urban landscape images;
taking the age bracket, the sex, the average pleasure degree score and the key visual element ratio corresponding to each transitional landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the transitional landscape images;
and taking the age group, the sex, the average pleasure degree score and the key visual element ratio corresponding to each natural landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the natural landscape images.
It is to be understood that the illustrated structure of the embodiment of the present invention does not constitute a specific limitation to a pleasure degree test device. In other embodiments of the invention, the pleasure testing means may comprise more or fewer components than shown, or some components may be combined, some components may be separated, or a different arrangement of components. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
Because the content of information interaction, execution process, and the like among the modules in the device is based on the same concept as the method embodiment of the present invention, specific content can be referred to the description in the method embodiment of the present invention, and is not described herein again.
The embodiment of the invention also provides computing equipment which comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the joy testing method in any embodiment of the invention is realized.
An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, causes the processor to execute a method for testing a pleasure degree in any embodiment of the present invention.
Specifically, a system or an apparatus equipped with a storage medium on which software program codes that realize the functions of any of the above-described embodiments are stored may be provided, and a computer (or CPU or GPU) of the system or the apparatus is caused to read out and execute the program codes stored in the storage medium.
In this case, the program code itself read from the storage medium can realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.
Examples of the storage medium for supplying the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (e.g., CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD + RW), a magnetic tape, a nonvolatile memory card, and a ROM. Alternatively, the program code may be downloaded from a server computer via a communications network.
Further, it should be clear that the functions of any one of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform a part or all of the actual operations based on instructions of the program code.
Further, it is to be understood that the program code read out from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion module connected to the computer, and then causes a CPU or the like mounted on the expansion board or the expansion module to perform part or all of the actual operations based on instructions of the program code, thereby realizing the functions of any of the above-described embodiments.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an …" does not exclude the presence of other similar elements in a process, method, article, or apparatus that comprises the element.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (6)
1. A method for predicting a pleasure degree of a geographic environment is characterized by comprising the following steps:
performing semantic segmentation on a sample image to obtain at least one visual element of the sample image; the sample image is a street view image, and the street view image is acquired through a Baidu map or a Gade map; the visual elements comprise nine major visual elements of sky, vegetation, water body, bare soil, buildings, highways and accessories, sidewalks, vehicles and people;
determining the landscape type of the sample image according to the ratio of each visual element of the sample image; wherein the landscape types include urban landscape, transitional landscape and natural landscape;
acquiring the average pleasure degree score of each age group and gender group combined people on each sample image, and taking the image containing the average pleasure degree scores of different age group and gender group combined people as an analysis sample; the age groups are 6 age groups of 10-19 years old, 20-29 years old, 30-39 years old, 40-49 years old, 50-59 years old and over 60 years old, and the age groups and the sexes form a sex combination of 12 age groups;
performing significance testing and correlation analysis on the analysis sample, and determining age group gender combination and key visual elements which influence the pleasure degree score of the sample image;
respectively taking age groups, sexes, average pleasure degree scores and key visual element ratios corresponding to different types of landscape images in the analysis sample as input, and constructing a mixed linear equation of the pleasure degree scores of the different types of landscape images; predicting the pleasure degree of the image to be measured by utilizing the mixed linear equation; the method comprises the following steps of respectively taking age groups, sexes, average pleasure degree scores and key visual element ratios corresponding to different types of landscape images in the analysis sample as input, and constructing a mixed linear equation of the pleasure degree scores of the different types of landscape images, wherein the mixed linear equation comprises the following steps:
taking the age bracket, the gender, the average pleasure degree score and the key visual element ratio corresponding to each urban landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the urban landscape images; the key visual elements influencing the urban landscape are vegetation elements, building elements and road elements, and the mixed linear equation of the joyful degree score of the urban landscape image is as follows:
taking the age bracket, the sex, the average pleasure degree score and the key visual element ratio corresponding to each transitional landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the transitional landscape images; the key visual elements influencing the transitional landscape are vegetation elements, building elements and sidewalk elements, and the mixed linear equation of the joyful degree score of the transitional landscape image is as follows:
taking the age bracket, the sex, the average pleasure degree score and the key visual element ratio corresponding to each natural landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the natural landscape images; wherein, the key visual element that influences natural landscape is sky element and vegetation element, the mixed linear equation of the joyful degree score of natural landscape image is:
the step of conducting significance test and correlation analysis on the analysis sample and determining gender age combination and key visual elements influencing the pleasure degree score of the sample image comprises the following steps:
performing a mann-whitney U test on the analysis sample to determine age group gender combinations that affect a pleasure score for the sample image;
performing spearman analysis on the analysis sample, and determining key visual elements influencing the pleasure degree score of the sample image;
the determining an age group gender combination that affects a pleasure score of the sample image by performing a mann-whitney U test on the analysis sample comprises:
performing a gender mann-whitney U test on the analysis sample to determine the effect of different genders on the sample image pleasure score;
performing a landscape type mann-whitney U test on the analysis sample to determine the effect of different landscape types on the sample image pleasure score;
performing an age-bracket mann-whitney U test on the analysis sample to determine the effect of different age brackets on the sample image pleasure score;
the spearman analysis of the analysis sample to determine key visual elements influencing the pleasure degree score of the sample image comprises the following steps:
performing spearman analysis on an analysis sample belonging to the urban landscape to determine key visual elements influencing the pleasure degree score of the urban landscape image;
performing spearman analysis on an analysis sample belonging to the transitional landscape to determine key visual elements influencing the pleasure degree score of the transitional landscape image;
spearman analysis is performed on the analysis samples belonging to the natural landscape to determine key visual elements that affect the pleasure score of the natural landscape image.
2. The method of claim 1, wherein the semantically segmenting the sample image to obtain at least one visual element of the sample image comprises:
extracting characteristic information of the sample image by using a ResNet model;
extracting mapping characteristic information of the sample image by using a pyramid model;
and performing feature fusion on the feature information of the sample image and the mapping feature information, and performing convolution operation to obtain the visual elements of each sample image.
3. The method of claim 1, prior to said obtaining an average hedonic score for each age-gender group of people for each of said sample images, comprising:
obtaining an effective questionnaire; wherein each of the available questionnaires comprises gender, age, and a score for pleasure for a predetermined number of the sample images;
and counting age group gender combinations in the effective questionnaire and scoring the pleasure degree of the sample image.
4. A prediction apparatus of a degree of pleasure, characterized by comprising:
the segmentation module is used for performing semantic segmentation on the sample image to obtain at least one visual element of the sample image; the sample image is a street view image, and the street view image is acquired through a Baidu map or a Gade map; the visual elements comprise nine major visual elements of sky, vegetation, water body, bare soil, buildings, highways and accessories, sidewalks, vehicles and people;
the classification module is used for determining the landscape type of the sample image according to the proportion of each visual element of the sample image; wherein the landscape types include urban landscape, natural landscape and transitional landscape;
the acquisition module is used for acquiring the average pleasure degree score of each age group and gender group combined person on each sample image and taking the image containing the average pleasure degree scores of different age group and gender group combined persons as an analysis sample; the age groups are 6 age groups of 10-19 years old, 20-29 years old, 30-39 years old, 40-49 years old, 50-59 years old and over 60 years old, and the age groups and the sexes form a sex combination of 12 age groups;
the analysis module is used for carrying out significance test and correlation analysis on the analysis sample and determining age group gender combination and key visual elements which influence the pleasure degree score of the sample image;
the construction module is used for respectively taking age groups, sexes, average pleasure degree scores and key visual element ratios corresponding to different types of landscape images in the analysis sample as input, and constructing a mixed linear equation of the pleasure degree scores of the different types of landscape images;
the prediction module is used for predicting the pleasure degree of the image to be measured by utilizing the mixed linear equation;
the construction module is used for executing the following operations:
taking the age bracket, the gender, the average pleasure degree score and the key visual element ratio corresponding to each urban landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the urban landscape images; the key visual elements influencing the urban landscape are vegetation elements, building elements and road elements, and the mixed linear equation of the joyful degree score of the urban landscape image is as follows:
taking the age bracket, the sex, the average pleasure degree score and the key visual element ratio corresponding to each transitional landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the transitional landscape images; the key visual elements influencing the transitional landscape are vegetation elements, building elements and sidewalk elements, and the mixed linear equation of the joyful degree score of the transitional landscape image is as follows:
taking the age bracket, the sex, the average pleasure degree score and the key visual element ratio corresponding to each natural landscape image in the analysis sample as input, and obtaining a mixed linear equation of the pleasure degree scores of the natural landscape images; wherein, the key visual element that influences natural landscape is sky element and vegetation element, the mixed linear equation of the joyful degree score of natural landscape image is:
the analysis module is used for executing the following operations:
performing a mann-whitney U test on the analysis sample to determine age group gender combinations that affect a pleasure score for the sample image;
performing spearman analysis on the analysis sample, and determining key visual elements influencing the pleasure degree score of the sample image;
the determining an age group gender combination that affects a pleasure score of the sample image by performing a mann-whitney U test on the analysis sample comprises:
performing a gender mann-whitney U test on the analysis sample to determine the effect of different genders on the sample image pleasure score;
performing a landscape type mann-whitney U test on the analysis sample to determine the effect of different landscape types on the sample image pleasure score;
performing an age-bracket mann-whitney U test on the analysis sample to determine the effect of different age brackets on the sample image pleasure score;
the spearman analysis of the analysis sample to determine key visual elements influencing the pleasure degree score of the sample image comprises the following steps:
performing spearman analysis on an analysis sample belonging to the urban landscape to determine key visual elements influencing the pleasure degree score of the urban landscape image;
performing spearman analysis on an analysis sample belonging to the transitional landscape to determine key visual elements influencing the pleasure degree score of the transitional landscape image;
spearman analysis is performed on the analysis samples belonging to the natural landscape to determine key visual elements that affect the pleasure score of the natural landscape image.
5. An electronic device comprising a memory having stored therein a computer program and a processor that, when executing the computer program, implements the method of any of claims 1-3.
6. A computer-readable storage medium, on which a computer program is stored which, when executed in a computer, causes the computer to carry out the method of any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111639205.0A CN114298415B (en) | 2021-12-29 | 2021-12-29 | Method and device for predicting pleasure degree of geographic environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111639205.0A CN114298415B (en) | 2021-12-29 | 2021-12-29 | Method and device for predicting pleasure degree of geographic environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114298415A CN114298415A (en) | 2022-04-08 |
| CN114298415B true CN114298415B (en) | 2022-08-09 |
Family
ID=80972422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111639205.0A Active CN114298415B (en) | 2021-12-29 | 2021-12-29 | Method and device for predicting pleasure degree of geographic environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114298415B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115063653B (en) * | 2022-07-27 | 2023-05-23 | 东南大学 | Method and system for measuring dominant color of built environment based on brain sensitive data of image |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8380558B1 (en) * | 2006-12-21 | 2013-02-19 | Videomining Corporation | Method and system for analyzing shopping behavior in a store by associating RFID data with video-based behavior and segmentation data |
| CN110461060A (en) * | 2019-07-29 | 2019-11-15 | 武汉仕代建筑配套工程有限公司 | A kind of intelligence landscape system and its working method |
| CN111932680A (en) * | 2020-06-16 | 2020-11-13 | 厦门大学 | Urban geomorphic element identification method and system based on VR (virtual reality) depth visual perception |
| CN113469859A (en) * | 2021-07-28 | 2021-10-01 | 国际竹藤中心 | Method and device for improving visual perception of landscape human landscape |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019133849A1 (en) * | 2017-12-29 | 2019-07-04 | Ebay Inc. | Computer vision and image characteristic search |
| CN110458088A (en) * | 2019-08-07 | 2019-11-15 | 南京林业大学 | A kind of forest scenery resources visual quality evaluation method based on image and principal component |
| CN110751362A (en) * | 2019-09-04 | 2020-02-04 | 西安工程大学 | Urban landscape evaluation index calculation method based on artificial intelligence algorithm |
| CN113112450A (en) * | 2021-03-05 | 2021-07-13 | 浙江工业大学 | Image pyramid guided remote sensing image small target detection method |
-
2021
- 2021-12-29 CN CN202111639205.0A patent/CN114298415B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8380558B1 (en) * | 2006-12-21 | 2013-02-19 | Videomining Corporation | Method and system for analyzing shopping behavior in a store by associating RFID data with video-based behavior and segmentation data |
| CN110461060A (en) * | 2019-07-29 | 2019-11-15 | 武汉仕代建筑配套工程有限公司 | A kind of intelligence landscape system and its working method |
| CN111932680A (en) * | 2020-06-16 | 2020-11-13 | 厦门大学 | Urban geomorphic element identification method and system based on VR (virtual reality) depth visual perception |
| CN113469859A (en) * | 2021-07-28 | 2021-10-01 | 国际竹藤中心 | Method and device for improving visual perception of landscape human landscape |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114298415A (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shen et al. | Antecedents of residents’ pro-tourism behavioral intention: Place image, place attachment, and attitude | |
| CN104537027B (en) | Information recommendation method and device | |
| Ko et al. | A structural equation model of residents’ attitudes for tourism development | |
| Tasci | Assessment of factors influencing destination image using a multiple regression model | |
| Harvey et al. | Streetscape skeleton measurement and classification | |
| Huh et al. | Satisfaction with cultural/heritage sites: Virginia historic triangle | |
| CN104794171B (en) | Mark the method and device of picture geographical location information | |
| CN108876475A (en) | A kind of urban function region recognition methods, server and storage medium based on point of interest acquisition | |
| Go et al. | The traces of ecotourism in a digital world: spatial and trend analysis of geotagged photographs on social media and Google search data for sustainable development | |
| CN114298415B (en) | Method and device for predicting pleasure degree of geographic environment | |
| Kim et al. | The role of destination’s built environment on nonmotorized travel behavior: A case of Long Beach, California | |
| Biljecki et al. | Raise the roof: Towards generating LOD2 models without aerial surveys using machine learning | |
| CN112084401A (en) | Tour route customizing device and method | |
| Huang et al. | Sustainable design of running friendly streets: Environmental exposures predict runnability by Volunteered Geographic Information and multilevel model approaches | |
| KR101838573B1 (en) | Place Preference Analysis Method based on Sentimental Analysis using Spatial Sentiment Lexicon | |
| Frew et al. | Investigating geospatial data usability from a health geography perspective using sensitivity analysis: The example of potential accessibility to primary healthcare | |
| Sun et al. | How the natural environment in downtown neighborhood affects physical activity and sentiment: Using social media data and machine learning | |
| CN114528492A (en) | Intelligent city text travel activity recommendation method and system | |
| Xie et al. | Interplay between auditory and visual environments in historic districts: A big data approach based on social media | |
| CN116306661A (en) | Social media data-based ecological system cultural service assessment method | |
| Čaušević et al. | Cognitive image of a tourism destination: The case of Sarajevo | |
| Rizkiyana | The Role of Social Media Benefits in Affecting Purchase Intention to Reserve a Liveaboard Phinisi Trip in Labuan Bajo | |
| Chen et al. | Evaluating implied urban nature vitality in San Francisco: An interdisciplinary approach combining census data, street view images, and social media analysis | |
| Widyawati et al. | Content analysis of tourist opinion based on tourism quality (TOURQUAL) by Text mining online reviews: The case of Borobudur | |
| Blumenfeld-Lieberthal et al. | What makes us think it's a city? |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |