Measuring Method, System, Computer Device and Medium for the Spatial Quality of Living Streets

Abstract

The present invention discloses a method, system, computer device and medium for measuring the spatial quality of a living street. The method includes: obtaining street view image data of a target area; establishing a street view image database according to the street view image data, and scoring the street view images through a random forest machine learning model to realize the subjective perception of the street view images; obtaining multiple environmental evaluation indicators of the street through visual image semantic segmentation based on a deep learning fully convolutional network and the K-means clustering algorithm, and conducting an objective environmental evaluation of the street view images; assigning weights to the results of the subjective perception and objective environmental evaluation of the street view images according to the analytic hierarchy process, and calculating the measurement result of the comprehensive evaluation of the street. The present invention can accurately measure the spatial quality of urban living streets, has the advantages of high efficiency and easy popularization and application, and can be widely applied to the field of survey and measurement of the spatial quality of urban planning living streets.

Description

Technical Field

[0001] The present invention relates to a method, a system, a computer device and a medium for measuring the quality of a living street space, and belongs to the field of urban and rural planning. Background Art

[0002] Street space is the "capillary" of a city, touching all aspects of citizens' lives and tourists' travel. Living streets are generally located in residential communities and are the most common type of streets. The width of their carriageways usually does not exceed two-way four lanes, and they have both traffic and the function of carrying residents' daily lives, which is of great importance to people's daily lives and neighborhood interactions. In the study of streets from the perspective of humanism, people and streets are two main bodies that cannot be ignored. People's spatial perception of the street and their activities in the street are two main ways for people to establish connections with the street, and this way is particularly important in living streets.

[0003] The existing measurement and evaluation methods for the quality of street space mainly include those based on field survey data, those based on basic geographic information data, and those based on street view pictures. However, the evaluation systems constructed by a large number of studies on the quality of street space still focus on a single index (such as the green view rate) or are composed of multiple general indicators, without incorporating characteristic indicators under specific regional environments and lacking a certain degree of localness. Summary of the Invention

[0004] In view of this, the present invention provides a method, a system, a computer device and a storage medium for measuring the quality of a living street space. For the built environment of a city, corresponding indicators are selected to construct an evaluation system, and at the same time, methods of subjective perception and objective environment evaluation are combined to achieve an in-depth and comprehensive evaluation of the quality of a living street space, so as to accurately measure the quality of a living street space in a city, which has the advantages of high efficiency and easy popularization and application, and can be widely applied to the field of investigation and measurement of the quality of living street space in urban planning.

[0005] The first object of the present invention is to provide a method for measuring the quality of a living street space

[0006] The second object of the present invention is to provide a system for measuring the quality of a living street space.

[0007] The third object of the present invention is to provide a computer device.

[0008] The fourth object of the present invention is to provide a storage medium.

[0009] The first object of the present invention can be achieved by adopting the following technical solutions:

[0010] A method for measuring the quality of a living street space, the method comprising:

[0011] Obtain the street view image data of the target area;

[0012] According to the street view image data, establish a street view image database, and score the street view images through a trained random forest machine learning model to realize the subjective perception of the street view images;

[0013] Through visual image semantic segmentation based on a deep learning fully convolutional network and the K-means clustering algorithm, obtain multiple environmental evaluation indicators of the street and conduct an objective environmental evaluation of the street view images;

[0014] According to the analytic hierarchy process, assign weights to the results of the subjective perception and objective environmental evaluation of the street view images, and calculate the measure result of the comprehensive evaluation of the street.

[0015] Further, the obtaining of the street view image data of the target area specifically includes:

[0016] Divide the sampling points on the streets in the target area at a preset interval, and obtain the street view images in the four directions of front, back, left, and right of the sampling points.

[0017] Further, the establishing of the street view image database according to the street view image data and scoring the street view images through the random forest machine learning model to realize the subjective perception of the street view images specifically includes:

[0018] According to the street view image data, establish a street view image database;

[0019] Randomly select a preset percentage of street view images from the street view image database as scoring samples, and obtain the multi-level scores of the residents in the location of the target area for the scoring samples;

[0020] Through machine deep learning of the results of the scoring samples by the random forest machine learning model, adjust the parameters of the random forest machine learning model to obtain a trained random forest machine learning model;

[0021] Use the trained random forest machine learning model to score the remaining street view images in the street view image database, and calculate the average value of the scores of the street view images in the four directions of front, back, left, and right of each sampling point as the scoring result.

[0022] Further, the environmental evaluation indicators include green view rate, mountain view rate, enclosure degree, openness, sign density, sky view rate, and color entropy.

[0023] Further, the obtaining methods of the green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate are as follows:

[0024] Through visual image semantic segmentation, conduct street view recognition on the street view images to obtain the proportion results of various street view elements in the street view images;

[0025] Select the proportion results of six types of elements, namely plants, mountains, buildings, roads, signs, and the sky, from the recognition results of various street view elements, and use them as the values of six indicators: green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate respectively;

[0026] Calculate the average values of the six indicators, namely green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate, in the four directions of front, back, left, and right of each sampling point.

[0027] Furthermore, the method for obtaining the color entropy is as follows:

[0028] Use the removebg algorithm to remove the sky and road parts in the street view image after semantic segmentation, and retain the elements that affect the chaos degree of the street facade;

[0029] Use the K-means clustering algorithm to extract seven main colors in the street view image after removing the sky and road parts, and calculate the color proportion of the seven main colors respectively;

[0030] Calculate the average value of the color entropy in the four directions of front, back, left, and right of each sampling point. The calculation formula of the color entropy is as follows:

[0031]

[0032] Among them, S j represents the color chaos degree of the j sampling point, X ij represents the number of pixel points of the i-th dominant color of various building elements at the j sampling point, and X j represents the number of pixel points of building elements in the street view.

[0033] Furthermore, the method for assigning weights to the subjective perception and objective environment evaluation results of the street view image according to the analytic hierarchy process and calculating the measure result of the comprehensive street evaluation specifically includes:

[0034] According to the analytic hierarchy process, assign a weight of 0.5 to the scoring evaluation result of the street view image, assign a weight of 0.1 to the green view rate and color entropy respectively, and assign a weight of 0.06 to the mountain view rate, enclosure degree, openness, sign density, and sky view rate respectively. Calculate the measure result of the comprehensive street evaluation by fusing these indicators and normalizing.

[0035] The second object of the present invention can be achieved by adopting the following technical solutions:

[0036] A measurement system for the quality of a living street space, the system includes:

[0037] An acquisition module for acquiring street view image data of a target area;

[0038] A subjective perception module, which is used to establish a street view image database according to street view image data, score the street view images through a random forest machine learning model, and realize the subjective perception of the street view images;

[0039] An objective environment evaluation module, which is used to obtain multiple environmental evaluation indicators of the street through visual image semantic segmentation and K-means clustering algorithm based on a deep learning fully convolutional network, and conduct an objective environment evaluation on the street view images;

[0040] A measurement module, which is used to assign weights to the results of the subjective perception and objective environment evaluation of the street view images according to the analytic hierarchy process, and calculate the measurement results of the comprehensive evaluation of the street.

[0041] The third object of the present invention can be achieved by adopting the following technical solutions:

[0042] A computer device includes a processor and a memory for storing executable programs of the processor. It is characterized in that when the processor executes the programs stored in the memory, the above-mentioned method for measuring the quality of a living street space is realized.

[0043] The fourth object of the present invention can be achieved by adopting the following technical solutions:

[0044] A storage medium stores a program, and when the program is executed by a processor, the above-mentioned method for measuring the quality of a living street space is realized.

[0045] The present invention has the following beneficial effects compared with the prior art:

[0046] Based on street view image data, the present invention scores street view images through a trained random forest machine learning model to realize the subjective perception of street view images, and obtains multiple environmental evaluation indicators of the street through visual image semantic segmentation and K-means clustering algorithm to conduct an objective environment evaluation on street view images. By combining subjective perception with objective environment indicators, the quality of urban living street space is accurately measured, which has the advantages of high efficiency and easy popularization and application, and can be widely applied to the field of survey and measurement of the quality of urban living street space in urban planning. Description of the Drawings

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on the structures shown in these drawings without creative efforts.

[0048] Figure 1It is a simplified flowchart of the method for measuring the quality of the living street space in Embodiment 1 of the present invention.

[0049] Figure 2 It is a detailed flowchart of the method for measuring the quality of the living street space in Embodiment 1 of the present invention.

[0050] Figure 3 It is an evaluation diagram of the street image in Embodiment 1 of the present invention.

[0051] Figure 4 It is a street green view rate diagram in Embodiment 1 of the present invention.

[0052] Figure 5 It is a street mountain view rate diagram in Embodiment 1 of the present invention.

[0053] Figure 6 It is a street enclosure degree diagram in Embodiment 1 of the present invention.

[0054] Figure 7 It is a street openness diagram in Embodiment 1 of the present invention.

[0055] Figure 8 It is a street sign density diagram in Embodiment 1 of the present invention.

[0056] Figure 9 It is a street sky view rate diagram in Embodiment 1 of the present invention.

[0057] Figure 10 It is a color entropy diagram in Embodiment 1 of the present invention.

[0058] Figure 11 It is a street comprehensive evaluation and measurement result diagram in Embodiment 1 of the present invention.

[0059] Figure 12 It is a structural block diagram of the living street space quality measurement system in Embodiment 2 of the present invention.

[0060] Figure 13 It is a structural block diagram of the computer device in Embodiment 3 of the present invention. Specific implementation mode

[0061] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0062] Embodiment 1:

[0063] Taking the Zhongshan Road area in Siming District, Xiamen as an example, this embodiment provides a method for measuring the quality of a living street space. The Zhongshan Road area is a typical representative of the old urban blocks with modern historical features in Xiamen. It faces the Gulangyu Scenic Area across the sea, has a strong sense of life, and the architectural style is rich in the characteristics and taste of the old Xiamen city. The scope of this embodiment starts from Siming North Road in the east, ends at Lujiang Road in the west, starts from Zhenhai Road in the south, and ends at Xiahe Road in the north. The area of the area is 0.81 square kilometers.

[0064] As Figure 1 and Figure 2 shown, the method for measuring the quality of the living street space in this embodiment includes the following steps:

[0065] S201. Obtain the street view image data of the target area.

[0066] In this embodiment, the network open-source street view image data is obtained in batches. Specifically: write a Python web crawler, and obtain it in batches through the API interface in the Tencent Map Open Platform. Divide the streets at a preset interval of 50m to obtain sampling points, and obtain the street view images in the four directions of front, back, left, and right of the sampling points, with a total of 5,356 images; in addition, in order to display the road data during visualization, the road network vector data within the Zhongshan Road area can also be obtained in batches through the API interface of the openstreetmap website open platform.

[0067] S202. According to the street view image data, establish a street view image database, and score the street view images through a trained random forest machine learning model to realize the subjective perception of the street view images.

[0068] This step S202 specifically includes:

[0069] S2021. According to the street view image data, establish a street view image database.

[0070] S2022. Randomly select a preset percentage of the street view images in the street view image database as scoring samples, and obtain the multi-level scores of the residents in the location of the target area for the scoring samples.

[0071] The preset percentage in this embodiment is 15%. Randomly select 688 street view images from the street view database as scoring samples, and evenly distribute them to eight locations in the target area for the samples to be scored at five levels, from poor to good, from 1 to 5 points.

[0072] S2023. Perform machine deep learning on the results of the scoring samples through the random forest machine learning model, adjust the parameters of the random forest machine learning model, and obtain a trained random forest machine learning model.

[0073] In this embodiment, 80% of the scoring samples are randomly selected as the training set, and 20% are used as the test set. The results of the scoring samples are subjected to machine deep learning through a random forest model based on the Python language, and the parameters of the random forest machine learning model are adjusted to make the random forest machine learning model have a higher model score on the test set, obtaining a more accurate machine scoring value, and finally obtaining a trained random forest machine learning model.

[0074] S2024. Use the trained random forest machine learning model to score the remaining street view images in the street view image database, and calculate the average value of the scores of the street view images in the four directions of front, back, left, and right of each sampling point as the scoring result.

[0075] In this embodiment, the average value of the scores of the street view images in the four directions of front, back, left, and right of each sampling point is calculated to obtain a scoring result in CSV format, and the scoring result is placed in ArcGIS for visualization, as Figure 3 shown.

[0076] S203. Through the visual image semantic segmentation and K-means clustering algorithm based on the deep learning fully convolutional network, multiple environmental evaluation indicators of the street are obtained to conduct an objective environmental evaluation of the street view images.

[0077] The environmental evaluation indicators of this embodiment include green view rate, mountain view rate, enclosure degree, openness, sign density, sky view rate, and color entropy. The acquisition methods of the green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate are as follows:

[0078] 1) Conduct street view recognition on the street view images through visual image semantic segmentation to obtain the proportion results of multiple types of street view elements in the street view images.

[0079] Specifically, through the visual image semantic segmentation software, conduct street view recognition on the street view images obtained in step S201 to obtain the proportion results of 150 types of street view elements in the street view images, obtaining a CSV table file and the street view images after semantic segmentation.

[0080] 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. Based on the deep learning-based scene semantic segmentation method, calculate the semantic category to which each pixel point in the picture belongs, and obtain the visual elements in the scene (such as buildings, vehicles, sky, etc.). Form a multi-dimensional vector, and each dimension in the vector represents the proportion of a specific category of object (such as buildings) in the image.

[0081] 2) Select the proportion results of six types of elements, namely plants, mountains, buildings, roads, signs, and the sky, from the recognition results of various street view elements, and use them as the values of six indicators: green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate. The higher the green view rate, mountain view rate, openness, and sky view rate, and the lower the enclosure degree and sign density, the higher the quality of the street space.

[0082] 3) Calculate the average values of the six indicators of green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate in the four directions of front, back, left, and right of each sampling point, and import them into ArcGIS for visualization, as Figures 4 to 9 shown.

[0083] The method for obtaining color entropy is as follows:

[0084] 1) Remove the sky and road parts that have a greater impact on color clustering in the street view image after semantic segmentation through the removebg algorithm based on Python, and retain elements such as buildings and plants that affect the chaos degree of the street facade.

[0085] 2) Extract seven main colors in the street view image after removing the sky and road parts through the K-means clustering algorithm based on Python, and calculate the color proportions of the seven main colors respectively.

[0086] The principle of the K-Means clustering algorithm is that given the value of K and K initial cluster center points, each point (data record) is assigned to the cluster represented by the nearest cluster center point. After all points are assigned, the center point of the cluster is recalculated (taking the average value) based on all points within the cluster, and then the process of assigning points and updating the cluster center points is iterated until the change in the cluster center point is very small or the specified number of iterations is reached.

[0087] Assume that the given data sample X includes n objects X = {X1, X2, X3,..., Xn}, where each object has attributes with m dimensions. The goal of the K-means clustering algorithm is to cluster the n objects into the specified k clusters based on the similarity between objects, and each object belongs to and only belongs to one cluster with the smallest distance to the cluster center.

[0088] 2.1) Initialize k clustering centers {C1, C2, C3,..., C k}, where 1 < k ≤ n. Further, by calculating the Euclidean distance from each object to the clustering center, the calculation formula is as follows:

[0089]

[0090] where, X i represents the i-th object, 1 ≤ i ≤ n, and C jwhere \(1\leq j\leq k\) represents the \(j\)-th clustering center, and \(X\) it represents the \(t\)-th attribute of the \(i\)-th object, where \(1\leq t\leq m\), and \(C\) jt represents the \(t\)-th attribute of the \(j\)-th clustering center.

[0091] 2.2) Compare the distances between each object and the clustering centers in turn, and assign the object to the cluster with the nearest clustering center, obtaining \(k\) clusters \(\{S_1, S_2, S_3, \ldots, S\) k}\).

[0092] 2.3) The K-means clustering center is the mean of all objects within the cluster, and its calculation formula is as follows:

[0093]

[0094] where \(C\) l represents the center of the \(l\)-th cluster, where \(1\leq l\leq k\), \(|S\) l | represents the number of objects in the \(l\)-th cluster, and \(X\) i represents the \(i\)-th object in the \(l\)-th cluster, where \(1\leq i\leq |S\) l |.

[0095] 3) Calculate the average value of the color entropy in the four directions of front, back, left, and right of each sampling point, and import it into ArcGIS for visualization, as Figure 10 shown.

[0096] The calculation formula of the color entropy is as follows:

[0097]

[0098] where \(S\) j represents the color chaos degree of the \(j\)-th sampling point, \(X\) ij represents the number of pixel points of the \(i\)-th dominant color of various building elements at the \(j\)-th sampling point, and \(X\) j represents the number of pixel points of building elements in the street view; the larger the color entropy value, the more lack of dominant color in the street view picture, that is, the higher the color chaos degree, and the more chaotic the street facade; the smaller the color entropy value, the more dominant color exists in the street view photo, that is, the lower the color chaos degree, and the street facade is more tidy and unified.

[0099] S204. Assign weights to the results of the subjective perception and objective environment evaluation of the street view image according to the Analytic Hierarchy Process (AHP), and calculate the measure result of the comprehensive evaluation of the street.

[0100] According to the analytic hierarchy process, a weight of 0.5 is assigned to the scoring evaluation result of the street view image, a weight of 0.1 is assigned to the green view rate and the color entropy respectively, and a weight of 0.06 is assigned to the mountain view rate, the enclosure degree, the openness, the sign density, and the sky view rate respectively. By fusing these indicators and normalizing them, the measure result of the comprehensive street evaluation is calculated as follows:

[0101] S = ω1S 1+ ω2(S2 + S3) + ω3(S4 + S5 + S6 + S7 + S8) (4)

[0102] Among them, S is the measure result of the street space quality after fusion, S1 is the large-scale scoring evaluation result of the street view image, S2 is the green view rate, S3 is the color entropy, S4 is the mountain view rate, S5 is the enclosure degree, S6 is the openness, S7 is the sign density, S8 is the sky view rate, ω1 = 0.5, which is the weight of the scoring evaluation result of the street view image, ω2 = 0.1, which is the weight of the green view rate and the color entropy, and ω3 = 0.06, which is the weight of the mountain view rate, the enclosure degree, the openness, the sign density, and the sky view rate.

[0103] Put the measure result of the comprehensive street evaluation into ArcGIS for visualization, as Figure 11 shown.

[0104] 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 that specific order, or that all the shown operations must be performed to achieve the desired result. On the contrary, the described steps can be changed in the execution order. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step for execution, and / or one step can be decomposed into multiple steps for execution.

[0105] Embodiment 2:

[0106] As Figure 12 shown, this embodiment provides a system for measuring the space quality of a living street, which includes an acquisition module 1201, a subjective perception module 1202, an objective environment evaluation module 1203, and a measurement module 1204. The specific functions of each module are as follows:

[0107] The acquisition module 1201 is used to acquire the street view image data of the target area.

[0108] The subjective perception module 1202 is used to establish a street view image database according to the street view image data, and score the street view image through a random forest machine learning model to achieve the subjective perception of the street view image.

[0109] The objective environment evaluation module 1203 is used to obtain multiple environmental evaluation indicators of the street through visual image semantic segmentation based on a deep learning fully convolutional network and the K-means clustering algorithm, and conduct an objective environment evaluation on the street view image.

[0110] The measurement module 1204 is used to assign weights to the results of the subjective perception and objective environment evaluation of the street view image according to the analytic hierarchy process, and calculate the measurement result of the comprehensive evaluation of the street.

[0111] It should be noted that the system provided in this embodiment is only illustrated by the above division of each functional module. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure is divided into different functional modules to complete all or part of the functions described above.

[0112] Embodiment 3:

[0113] This embodiment provides a computer device, which can be a computer. As Figure 13 shown, it includes a processor 1302, a memory, an input device 1303, a display 1304, and a network interface 1305 connected through a system bus 1301. The processor is used to provide computing and control capabilities. The memory includes a non-volatile storage medium 1306 and an internal memory 1307. The non-volatile storage medium 1306 stores an operating system, a computer program, and a database. The internal memory 1307 provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. When the processor 1302 executes the computer program stored in the memory, the method for measuring the quality of the living street space in the above Embodiment 1 is implemented as follows:

[0114] Obtain the street view image data of the target area;

[0115] According to the street view image data, establish a street view image database, and score the street view image through a trained random forest machine learning model to achieve the subjective perception of the street view image;

[0116] Through visual image semantic segmentation based on a deep learning fully convolutional network and the K-means clustering algorithm, obtain multiple environmental evaluation indicators of the street, and conduct an objective environment evaluation on the street view image;

[0117] According to the analytic hierarchy process, assign weights to the results of the subjective perception and objective environment evaluation of the street view image, and calculate the measurement result of the comprehensive evaluation of the street.

[0118] Embodiment 4:

[0119] This embodiment provides a storage medium, which is a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method for measuring the quality of a living street space in the above-mentioned Embodiment 1 is implemented as follows:

[0120] Obtain the street view image data of the target area;

[0121] According to the street view image data, establish a street view image database, and score the street view images through a trained random forest machine learning model to achieve the subjective perception of the street view images;

[0122] Through visual image semantic segmentation based on a deep learning fully convolutional network and the K-means clustering algorithm, obtain multiple environmental evaluation indicators of the street and conduct an objective environmental evaluation of the street view images;

[0123] According to the analytic hierarchy process, assign weights to the results of the subjective perception and objective environmental evaluation of the street view images, and calculate the measurement result of the comprehensive street evaluation.

[0124] It should be noted that the computer-readable storage medium in this embodiment can be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0125] In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In this embodiment, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable storage medium other than the computer-readable storage medium, and this computer-readable signal medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable storage medium can be transmitted by any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0126] The above computer-readable storage medium can be written in one or more programming languages or combinations thereof for executing the computer program of this embodiment. The above programming languages include object-oriented programming languages such as Java, Python, and C++, and also include conventional procedural programming languages such as C language or similar programming languages. The program can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0127] In summary, based on street view image data, the present invention scores street view images through a trained random forest machine learning model to achieve the subjective perception of street view images, and obtains multiple environmental evaluation indicators of the street through visual image semantic segmentation and the K-means clustering algorithm to conduct an objective environmental evaluation of street view images. By combining subjective perception with objective environmental indicators, it accurately measures the spatial quality of urban living streets, has the advantages of high efficiency and easy popularization and application, and can be widely applied to the field of investigation and measurement of the spatial quality of urban living streets in urban planning.

[0128] The above is only a preferred embodiment of the present invention patent, but the protection scope of the present invention patent is not limited thereto. Any person skilled in the art within the scope disclosed by the present invention patent, according to the technical solution of the present invention patent and its inventive concept, makes equivalent substitutions or changes, and all belong to the protection scope of the present invention patent.

Claims

1. A method for measuring the quality of a living street space, characterized in that The method includes: Obtaining street view image data of the target area; Based on the street view image data, establishing a street view image database, and scoring the street view images through a trained random forest machine learning model to achieve the subjective perception of the street view images; Through visual image semantic segmentation based on a deep learning fully convolutional network and the K-means clustering algorithm, obtaining multiple environmental evaluation indicators of the street and conducting an objective environmental evaluation of the street view images; Assigning weights to the results of the subjective perception and objective environmental evaluation of the street view images according to the analytic hierarchy process, and calculating the measure result of the comprehensive street evaluation; The step of establishing a street view image database based on the street view image data and scoring the street view images through a random forest machine learning model to achieve the subjective perception of the street view images specifically includes: Establishing a street view image database according to the street view image data; Randomly extracting a preset percentage of street view images from the street view image database as scoring samples, and obtaining multi-level scores of the residents in the location of the target area for the scoring samples; Performing machine deep learning on the results of the scoring samples through a random forest machine learning model, and adjusting the parameters of the random forest machine learning model to obtain a trained random forest machine learning model; Using the trained random forest machine learning model to score the remaining street view images in the street view image database, and calculating the average value of the scores of the street view images in the four directions of front, back, left, and right of each sampling point as the scoring result; The environmental evaluation indicators include green view rate, mountain view rate, enclosure degree, openness, sign density, sky view rate, and color entropy; The acquisition methods of the green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate are as follows: Performing street view recognition on the street view images through visual image semantic segmentation to obtain the proportion results of various street view elements in the street view images; Selecting the proportion results of six types of elements, namely plants, mountains, buildings, roads, signs, and the sky, from the recognition results of various street view elements as the values of the six indicators of the green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate; Calculating the average values of the six indicators of the green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate in the four directions of front, back, left, and right of each sampling point; The acquisition method of the color entropy is as follows: Removing the sky and road parts in the street view images after semantic segmentation through the removebg algorithm, and retaining the elements affecting the chaos degree of the street facades; Extracting seven main colors in the street view images after removing the sky and road parts through the K-means clustering algorithm, and respectively calculating the color proportions of the seven main colors; Calculating the average value of the color entropy in the four directions of front, back, left, and right of each sampling point, and the calculation formula of the color entropy is as follows: Among them, S j represents the color chaos degree of the j-th sampling point, and X ij represents the number of pixel points of the i-th dominant color of various building elements at the j-th sampling point, and X j represents the number of pixel points of building elements in the street view.

2. The method for measuring the quality of the living street space according to claim 1, wherein The step of obtaining the street view image data of the target area specifically includes: Dividing the sampling points on the streets in the target area at a preset interval, and obtaining the street view images in the four directions of front, back, left, and right of the sampling points.

3. The method for measuring the quality of the living street space according to claim 1, wherein The step of assigning weights to the results of the subjective perception and objective environmental evaluation of the street view images according to the analytic hierarchy process and calculating the measure result of the comprehensive street evaluation specifically includes: According to the analytic hierarchy process, a weight of 0.5 is assigned to the scoring evaluation result of the street view image, weights of 0.1 are assigned to the green view rate and the color entropy respectively, and weights of 0.06 are assigned to the mountain view rate, the enclosure degree, the openness, the sign density, and the sky view rate respectively. By fusing these indicators and normalizing them, the measurement result of the comprehensive street evaluation is calculated.

4. A measurement system for the quality of a living street space, characterized in that, The system includes: An acquisition module for acquiring street view image data of a target area; A subjective perception module for establishing a street view image database based on the street view image data, and scoring the street view image through a random forest machine learning model to achieve the subjective perception of the street view image; An objective environment evaluation module for obtaining multiple environmental evaluation indicators of the street through visual image semantic segmentation and K-means clustering algorithm based on a deep learning fully convolutional network, and conducting an objective environment evaluation of the street view image; A measurement module for calculating the measurement result of the comprehensive street evaluation according to the weights of the subjective perception and the objective environment evaluation of the street view image assigned by the analytic hierarchy process; The method of establishing a street view image database based on the street view image data, scoring the street view image through a random forest machine learning model, and achieving the subjective perception of the street view image specifically includes: Establishing a street view image database according to the street view image data; Randomly extracting a preset percentage of street view images from the street view image database as scoring samples, and obtaining multi-level scores of the residents in the location of the target area for the scoring samples; Performing machine deep learning on the results of the scoring samples through a random forest machine learning model, and adjusting the parameters of the random forest machine learning model to obtain a trained random forest machine learning model; Using the trained random forest machine learning model to score the remaining street view images in the street view image database, and calculating the average value of the scores of the street view images in the four directions of the front, back, left, and right of each sampling point as the scoring result; The environmental evaluation indicators include green view rate, mountain view rate, enclosure degree, openness, sign density, sky view rate, and color entropy; The acquisition methods of the green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate are as follows: Performing street view recognition on the street view image through visual image semantic segmentation to obtain the proportion results of various street view elements in the street view image; Selecting the proportion results of six types of elements, namely plants, mountains, buildings, roads, signs, and sky, from the recognition results of various street view elements as the values of the six indicators of the green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate respectively; Calculating the average values of the six indicators of the green view rate, mountain view rate, enclosure degree, openness, sign density, and sky view rate in the four directions of the front, back, left, and right of each sampling point; The acquisition method of the color entropy is as follows: Removing the sky and road parts in the street view image after semantic segmentation through the removebg algorithm, and retaining the elements affecting the chaos degree of the street facade; Extracting seven main colors in the street view image after removing the sky and road parts through the K-means clustering algorithm, and calculating the color proportions of the seven main colors respectively; Calculating the average value of the color entropy in the four directions of the front, back, left, and right of each sampling point. The calculation formula of the color entropy is as follows: Among them, S j represents the color chaos degree of the jth sampling point, and X ij represents the number of pixel points of the ith dominant color of various building elements at the jth sampling point, and X j represents the number of pixel points of building elements in the street view.

5. A computer device, comprising a processor and a memory for storing processor-executable programs, characterized in that, When the processor executes the program stored in the memory, it implements the method for measuring the quality of the living street space according to any one of claims 1-3.

6. A storage medium stores a program, characterized in that, When the program is executed by the processor, it implements the method for measuring the quality of the living street space according to any one of claims 1-3.

Images