CN104933229B - A kind of built environment Landscape Characteristics recognition methods based on network picture - Google Patents

A kind of built environment Landscape Characteristics recognition methods based on network picture Download PDF

Info

Publication number
CN104933229B
CN104933229B CN201510288892.4A CN201510288892A CN104933229B CN 104933229 B CN104933229 B CN 104933229B CN 201510288892 A CN201510288892 A CN 201510288892A CN 104933229 B CN104933229 B CN 104933229B
Authority
CN
China
Prior art keywords
mrow
shooting
picture
environment
msub
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
Application number
CN201510288892.4A
Other languages
Chinese (zh)
Other versions
CN104933229A (en
Inventor
赵渺希
顾沁
贾锐澜
吴江月
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN201510288892.4A priority Critical patent/CN104933229B/en
Publication of CN104933229A publication Critical patent/CN104933229A/en
Priority to PCT/CN2016/083258 priority patent/WO2016192552A1/en
Priority to SG11201709813SA priority patent/SG11201709813SA/en
Application granted granted Critical
Publication of CN104933229B publication Critical patent/CN104933229B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a kind of built environment Landscape Characteristics recognition methods based on network picture, including：The photographs library of target environment and similar built environment is obtained using Baidu's photographic search engine；Shooting standpoint, shooting optic centre point and the shooting sight in shooting behavior of photographing in target environment are recorded, and imports the analysis platform of ArcGIS, using visual acuity area of the ArcGIS software analysis target environment in network picture and to sensitivity analysis；Using target environment and similar built environment webcam picture library, both webcam picture element attributes of record, are examined using independent sample T and filter out target environment visual distinctiveness key element；It is possible thereby to the Landscape Characteristics of target built environment are identified and analyzed.The present invention can carry out visual image analysis to city built environment landscape, have reproducibility and operability, and using photographer's visual analysis of network picture and the regression analysis of shooting behavior, have timeliness and objectivity.

Description

A kind of built environment Landscape Characteristics recognition methods based on network picture
Technical field
It is especially a kind of that ring is built up based on network picture the present invention relates to a kind of built environment Landscape Characteristics recognition methods Border Landscape Characteristics recognition methods, belongs to the urban landscape research field in urban planning field.
Background technology
The Landscape Characteristics of identification urban environment are the premise and basis of urban and rural planning and design, and the shooting behavior of the common people is The Landscape Characteristics of identification abbreviation environment provide a large amount of materials.It is always focus of attention that photographs, which carries out tourist site vision research, Hot spot.
At this stage, there is vision purpose of many scholars by medium expression study city or block, if Liu Jing is by electricity The picture of shadow studies the space image in multiple cities, carries out principal factor analysis (PFA) using Google's picture as Zhao Miao is wished, features wide The Markov Model in eastern 21 cities.But the current country there is no the correlation for carrying out built environment landscape identification using network picture to grind Study carefully.
Yang Jun fetes, history preferably discloses a kind of city horizon in the application for a patent for invention of Application No. 201310097613.7 The quick obtaining and measuring method of contour line facade orthophotoquad, but the not content to image and shooting behavioural information Understood.
The content of the invention
The purpose of the present invention is to solve the defects of the above-mentioned prior art, there is provided a kind of building up based on network picture Environmental landscape characteristic recognition method, this method can carry out network visual image analysis to city built environment landscape, and having can Replicability and operability, and using photographer's visual analysis of network picture and the regression analysis of shooting behavior, there is timeliness Property and objectivity.
The purpose of the present invention can be reached by adopting the following technical scheme that：
A kind of built environment Landscape Characteristics recognition methods based on network picture, comprises the following steps：
S1, obtain target environment and the photographs library similar to built environment using Baidu's photographic search engine；
S2, for target environment, using in the shooting standpoint in AutoCAD chronophotography shooting behaviors, shooting vision Heart point and shooting sight are recorded；
S3, for target environment, standpoint, shooting optic centre point will be shot and shoot the analysis that sight imports ArcGIS Platform, network picture visual acuity area and to sensitivity analysis using ArcGIS software analysis landscapes；
S4, utilize target environment and similar built environment webcam picture library, both webcam picture elements of record Attribute, is examined using the independent sample T of SPSS softwares and filters out target environment visual distinctiveness key element；
S5, for target environment and similar built environment, in summary visual information distribution situation, establishes characteristic element category Property table, will row be defined as characteristic element, by row be defined as shooting standpoint, optic centre point, shooting gaze area, carry out key element The differentiation of importance, instructs the landscape pattern optimizing of target environment.
It is described to obtain target environment and similar using Baidu's photographic search engine in step S1 as a kind of embodiment The photographs library of built environment, specifically includes：
S11, using Baidu's picture query target environment title, and all picture search results are downloaded in order, by picture The picture unrelated with target environment is deleted in search result, preceding 1200 pictures in picture search result is chosen, as target ring Border webcam picture library；
S12, using Baidu's picture inquire about some similar built environment titles respectively, and downloads all pictures in order respectively Search result, picture unrelated with each similar built environment in picture search result is deleted, chooses each picture searching respectively As a result preceding 200 pictures in, totally 1200 pictures are as similar built environment webcam picture library.
As a kind of embodiment, in step S2, the shooting in the chronophotography shooting behavior using AutoCAD is stood Point, shooting optic centre point and shooting sight are recorded, and are specifically included：
S21, the geographic Location Classification by target environment webcam picture library according to shooting；
S22, each pictures in target environment field observation, control target environment webcam picture library, with mesh Environment plan is marked as working base map, using the diagonal intersection point camera site of each pictures as optic centre point, is passed through The mode of scene reduction, CAD work is drawn in CAD plans by shooting standpoint, optic centre point and shooting sight respectively On base map.
As a kind of embodiment, in step S3, for target environment, will shooting standpoint, shooting optic centre point and The analysis platform that sight imports ArcGIS is shot, network picture visual acuity area and to quick using ArcGIS software analysis landscapes Sensitivity analysis, specifically includes：
S31, by shoot standpoint, optic centre point and shoot sight CAD data be directed respectively into ArcGIS softwares into Row GIS space dropping places are rectified a deviation with coordinate；
ArcToolbox in S32, opening ArcGIS softwares, using Spatial Analyze instruments to shooting standpoint Dot density analysis operation is carried out with optic centre point, line density analysis operation is carried out to shooting sight, respectively obtains target environment Shoot standpoint density value, optic centre dot density value and shooting sight density value；
S33, make target environment shooting standpoint density value, optic centre dot density value and shooting sight using INT instruments Density value is changed into shape data, obtains integer shooting standpoint Density Distribution, the distribution of integer optic centre dot density and integer and claps Take the photograph sight linear-density distribution；
S34, calculate shooting standpoint susceptibility, shooting optic centre point susceptibility and shooting sight using ArcGIS softwares Susceptibility；
S35, using ArcGIS softwares to shooting standpoint susceptibility, shooting optic centre point susceptibility and shooting sight it is quick Sensitivity is divided into the visual sensitivity of three ranks from high to low.
As a kind of embodiment, in step S34, the shooting standpoint susceptibility, shooting optic centre point susceptibility With the calculating of shooting sight susceptibility, it is specially：
1) standpoint density value is shot using processed integer, creates field in ArcGIS attribute lists, calculate dot density Average value, dot density average valueCalculation formula be：
Wherein, n be grid point sum, biFor grid point value；
2) result of calculation of dot density value is exported into Excel, standard deviation is calculated in Excel, the calculating of standard deviation is public Formula is：
Wherein, MiFor standard deviation, n is the sum of grid point, biFor grid point value,For dot density average value；
3) using density analysis and raster map layer overlay analysis technology, shooting standpoint sensitivity analysis is carried out, to Mr. Yu The sensitivity analysis calculation formula of any is：
Wherein, kiFor the susceptibility of certain point, MiPoor, the b for shooting sight density criterioniFor grid point value,Regarded for shooting Line density average value；
4) vector format data are created, field is created, an average density value is imparted among attribute list；
5) using vector data is converted to raster data instrument according to field value, above-mentioned vector data is switched into grid respectively Lattice data, obtain representing point average density value raster map layer；
6) according to susceptibility calculation formula, map superposition calculation is carried out using Map Algebra instruments in ArcGIS, is obtained Shoot standpoint sensitivity distribution figure；
7) repeat the above steps 1) to the method for step 6), obtain shooting optic centre point sensitivity distribution figure and shooting regards Line sensitivity distribution figure.
It is described to utilize target environment and similar built environment webcam picture in step S4 as a kind of embodiment Storehouse, both webcam picture element attributes of record, specifically includes：
S41, according to target environment webcam picture library and similar built environment webcam picture library, by all pictures In visible elements be recorded in Excel forms；
S42, to network picture number consecutively of often throwing the net, the alternative elements attribute using visible elements as every numbering picture, 1 is recorded as if occurring certain key element in picture, is otherwise 0；
S43, addition picture ownership attribute, target environment picture record is 1 and similar built environment picture record is 2, knot Fruit saves as target environment webcam picture element attribute list and similar built environment webcam picture element attribute list.
As a kind of embodiment, in step S4, the independent sample T with SPSS softwares is examined and is filtered out target environment Visual distinctiveness key element, specifically includes：
S44, according to target environment webcam picture element attribute list and similar built environment webcam picture element Attribute list, independent sample T check analyses are carried out using SPSS softwares to target environment and similar built environment visible elements, independent Sample T examine formula be：
Wherein, tkFor T test statistics,WithRespectively numbering is k in target environment and similar built environment Photographs component attributes value, n1, kAnd n2, kThe photographs that respectively numbering is k in target environment and similar built environment Sample size, S1, kAnd S2, kThe photographs component attributes value that respectively numbering is k in target environment and similar built environment Sample variance；
S45, T test statistics tkAs each key element significant indexes, if certain element tkValue is more than 0.1, then casts out this yuan Element, if certain element tkValue is not more than 0.1, then retains the element and make target environment visual distinctiveness key element simultaneously its t of corresponding recordkValue, will Obtain target environment visual distinctiveness key element table；
S46, using target environment visual distinctiveness key element table, visual distinctiveness degree point is carried out to target environment Landscape Feature element Level, so as to filter out target environment visual distinctiveness key element.
The present invention has following beneficial effect relative to the prior art：
1st, the present invention based on objective completed region of the city geodata and network opening data, by geoanalysis, The quantitative analysis method such as statistical analysis, correlation analysis, action process analysis city is shot by analyzing the common people in network media The vision hot spot of built environment landscape, characteristic visual element in the environment of Statistical Comparison different cities, aids in the design of urban landscape With planning.
2nd, the present invention is by reducing webcam picture shooting process, compensate for city built environment Landscape character assessment and The deficiency of subjectivity and hysteresis quality in design.
3rd, present invention uses a kind of network visual image analysis method for city built environment landscape, having to answer Property and operability processed, and using photographer's visual analysis of network picture and the regression analysis of shooting behavior, there is timeliness And objectivity.
Brief description of the drawings
Fig. 1 is the flow chart of the built environment Landscape Characteristics recognition methods based on network picture of the embodiment of the present invention 1.
Fig. 2 is the scene reduction schematic diagram of the embodiment of the present invention 1.
Fig. 3 is that the integer of the embodiment of the present invention 2 shoots standpoint density profile.
Fig. 4 is the integer optic centre point distribution map of the embodiment of the present invention 2.
Fig. 5 is that the integer of the embodiment of the present invention 2 shoots sight density profile.
Fig. 6 is the shooting standpoint sensitivity distribution figure of the embodiment of the present invention 2.
Embodiment
With reference to embodiment and attached drawing, the present invention is described in further detail, but embodiments of the present invention are unlimited In this.
Embodiment 1：
As shown in Figure 1, the built environment Landscape Characteristics recognition methods based on network picture of the present embodiment, including following step Suddenly：
1) acquisition of target environment and similar built environment webcam picture library
Built environment refers to the urban environment that completion has been built in city, and target environment refers to need to carry out Landscape Characteristics The built environment of identification, similar built environment refer to target environment it is generic, there are other of similitude and comparability to build Into environment；Using the significance test of similar built environment and target environment, relatively similar built environment in target environment can obtain The more significant element of difference difference, is the characteristic element of target environment.
1.1) Baidu's picture query target environment title is utilized, and downloads all picture search results in order, by picture The picture unrelated with target environment is deleted in search result, preceding 1200 pictures in picture search result is chosen, as target ring Border webcam picture library；
1.1) some similar built environment titles are inquired about respectively using Baidu's picture, and download all pictures in order respectively Search result, picture unrelated with each similar built environment in picture search result is deleted, chooses each picture searching respectively As a result preceding 200 pictures in, totally 1200 pictures are as similar built environment webcam picture library；
2) tourist site visual acuity region recognition
2.1) the target environment scene reduction of photographs shooting behavior
2.1.1) the geographic Location Classification by target environment webcam picture library according to shooting；
2.1.2) the photographing information record of target environment webcam picture
In target environment field observation, each pictures in target environment webcam picture library are compareed, with target ring Border plan, using the diagonal intersection point camera site of each pictures as optic centre point, passes through such as Fig. 2 as working base map Shown live reduction mode, in CAD plans respectively by shooting point, optic centre point and shooting route (i.e. shooting point and Line between optic centre point) it is drawn on CAD working base maps；
2.2) identification in visual acuity region
2.2.1 the processing of behavior restoring data) is shot
2.2.1.1) CAD data for shooting standpoint, shooting focus and shooting route is directed respectively into ArcGIS softwares GIS space dropping places are carried out to rectify a deviation with coordinate；
2.2.1.2 ArcToolbox) is opened, using Spatial Analyze instruments to shooting standpoint and shooting focus Dot density analysis operation is carried out, line density analysis operation is carried out to shooting sight, it is close to respectively obtain target environment shooting standpoint Angle value, optic centre dot density value and shooting sight density value；
2.2.1.3 INT instruments) are utilized so that it becomes shape data, obtains integer shooting standpoint Density Distribution, integer regards Feel central point Density Distribution and integer shooting sight linear-density distribution；
2.2.2) the calculating of visual sensitivity
2.2.2.1 standpoint density value) is shot using processed integer, field is created in ArcGIS attribute lists, calculates Dot density average value, dot density average valueCalculation formula be：
Wherein, n be grid point sum, biFor grid point value；
2.2.2.2 the result of calculation of dot density value) is exported into Excel, standard deviation is calculated in Excel, standard deviation Calculation formula is：
Wherein, MiFor standard deviation, n is the sum of grid point, biFor grid point value,For dot density average value；
2.2.2.3) using density analysis and raster map layer overlay analysis technology, shooting standpoint sensitivity analysis is carried out, Sensitivity analysis calculation formula for certain point is：
Wherein, kiFor the susceptibility of certain point, MiPoor, the b for shooting sight density criterioniFor grid point value,Regarded for shooting Line density average value；
2.2.2.4 vector format data) are created, field is created, an average density value is imparted among attribute list；
2.2.2.5) using vector data is converted to raster data instrument according to field value, above-mentioned vector data is distinguished Switch to raster data, obtain representing point average density value raster map layer；
2.2.2.6) according to susceptibility calculation formula, map superposition meter is carried out using Map Algebra instruments in ArcGIS Calculate, obtain shooting standpoint sensitivity distribution figure；
2.2.2.7 optic centre point sensitivity distribution figure and shooting sight sensitivity distribution diagram data production method) are shot Consistent with the above, shooting optic centre point sensitivity distribution figure and shooting sight sensitivity distribution can be obtained by repeating the above steps Figure；
2.2.3) visual sensitivity classification
Using ArcGIS softwares to shooting standpoint susceptibility, shooting optic centre point susceptibility and shooting sight susceptibility It is divided into the visual sensitivity of three ranks from high to low.
3) target environment landscape visual distinctiveness key element forms analysis
3.1) target environment and similar built environment picture visible elements determine
According to target environment webcam picture library and similar built environment webcam picture library, will go out in all pictures Existing plant, railing, window, pond, roof etc. are as visible elements and are recorded in Excel forms；
3.2) differentiation of target environment and similar built environment webcam picture element attribute
3.2.1) to network picture number consecutively of often throwing the net, the alternative elements category using visible elements as every numbering picture Property, 1 is recorded as if occurring certain key element in picture, is otherwise 0；
3.2.2 picture ownership attribute) is added, target environment picture record is 1 and similar built environment picture record is 2, As a result target environment webcam picture element attribute list and similar built environment webcam picture element attribute list are saved as；
3.3) conspicuousness of target environment landscape visual distinctiveness key element differentiates
3.3.1) will according to target environment webcam picture element attribute list and similar built environment webcam picture Plain attribute list, carries out independent sample T check analyses, solely using SPSS softwares to target environment and similar built environment visible elements Founding the formula that sample T is examined is：
Wherein, tkFor T test statistics,WithRespectively numbering is k in target environment and similar built environment Photographs component attributes value, n1, kAnd n2, kThe photographs that respectively numbering is k in target environment and similar built environment Sample size, S1, kAnd S2, kThe photographs component attributes value that respectively numbering is k in target environment and similar built environment Sample variance；
3.3.2) T test statistics tkEach key element significant indexes are used as, if certain element tkValue is more than 0.1, then casts out this Element, if certain element tkValue is not more than 0.1, then retains the element and make target environment visual distinctiveness key element simultaneously its certain member of corresponding record Plain tkValue, will obtain target environment visual distinctiveness key element table；
3.4) the visual distinctiveness classification of target environment landscape visual distinctiveness key element
Using target environment visual distinctiveness key element table, visual distinctiveness degree classification is carried out to target environment Landscape Feature element, So as to filter out target environment visual distinctiveness key element；
4) overall merit of completed region of the city domain Landscape Characteristics
In summary visual information distribution situation, establishes characteristic element attribute list, and row is defined as characteristic element, and row are fixed Justice carries out the differentiation of key element importance, instructs the scape of target environment to shoot standpoint, optic centre point, shooting gaze area See general layout optimization.
Embodiment 2：
The present embodiment is an application example, chooses Guangzhou Fanyu District Yuyin Garden scenic spot, there is provided one kind is based on network The scenic spot visual information integrated evaluating method of picture, Yuyin Garden scenic spot are located at Guangzhou, Guangdong Fanyu District south villages and small towns southeast corner North Street, is typical Lingnan Gardens.
1) acquisition of Yuyin Garden webcam picture library
1.1) Baidu's picture query keyword " Yuyin Garden " is utilized, and downloads all picture search results in order, will The picture unrelated with Yuyin Garden is deleted in picture search result, preceding 1200 pictures in picture search result is chosen, as remaining Shady mountainous house webcam picture library；
1.2) inquire about Guangzhou respectively similar to built environment keyword using Baidu's picture, this example choose " Yuexiu Park ", " my god River park ", " Li Gulf lake park ", " doctor mountain Forest Park ", " Dongshan Lake park ", " VolksGarden ", and respectively in order under Carry all picture search results.Picture unrelated with park in picture search result is deleted, each park picture is chosen respectively and searches Preceding 200 pictures in hitch fruit, totally 1200 pictures are as whole city park webcam picture library；
2) Yuyin Garden visual acuity region recognition
2.1) the Yuyin Garden scene reduction of photographs shooting behavior
2.1.1) the geographic Location Classification by Yuyin Garden webcam picture library according to shooting
2.1.2) using 3 planning professionals in Yuyin Garden field observation, control Yuyin Garden webcam picture Each pictures in storehouse, the Yuyin Garden scenic spot CAD maps to draw, using AutoCAD softwares, are established as working base map Shoot standpoint, optic centre point, shooting three figure layers of sight；It is each with point orders mark in shooting standpoint figure layer The diagonal intersection point that the shooting standpoint of pictures marks each pictures in optic centre point figure layer with point orders is shot Position is used as shooting in optic centre point figure layer as shooting focus by the use of line orders connection shooting standpoint and shooting focus Sight；
2.2) identification in visual acuity region
2.3.1 the processing of behavior restoring data) is shot
2.3.1.1) CAD data for shooting standpoint, shooting focus and shooting sight is directed respectively into ArcGIS softwares GIS space dropping places are carried out to rectify a deviation with coordinate；
2.3.1.2 ArcToolbox) is opened, using Spatial Analyze instruments to shooting standpoint and shooting focus Dot density analysis operation is carried out, line density analysis operation is carried out to shooting sight, it is close to respectively obtain Yuyin Garden shooting standpoint Angle value, optic centre dot density value and shooting sight density value；
2.3.1.3 INT instruments) are utilized so that it becomes shape data, obtains integer Yuyin Garden integer shooting standpoint point Density Distribution, the distribution of integer optic centre dot density and integer shooting sight Density Distribution, respectively as shown in Fig. 3, Fig. 4 and Fig. 5；
2.3.2) the calculating of visual sensitivity
2.3.2.1 standpoint dot density value) is shot using processed integer, field, meter are created in ArcGIS attribute lists Dot density average value is calculated, the calculation formula of dot density average value is：
Wherein, n be grid point sum, biFor grid point value, dot density average value is obtained as 31.99；
2.3.2.2 the result of calculation of dot density value) is exported into Excel, standard deviation is calculated in Excel, standard deviation Calculation formula is：
Wherein, MiFor standard deviation, n is the sum of grid point, biFor grid point value,For dot density average value, marked Quasi- difference MiFor 103.58；
2.2.2.3) using density analysis and raster map layer overlay analysis technology, shooting standpoint sensitivity analysis is carried out, Sensitivity analysis calculation formula for certain point is：
Wherein, kiFor the susceptibility of certain point, MiPoor, the b for shooting sight density criterioniFor grid point value,Regarded for shooting Line density average value；
2.2.2.4 vector format data) are created, field is created, an average density value is imparted among attribute list；
2.2.2.5) using vector data is converted to raster data instrument according to field value, above-mentioned vector data is distinguished Switch to raster data, obtain representing point average density value raster map layer；
2.2.2.6) according to susceptibility calculation formula, map superposition meter is carried out using Map Algebra instruments in ArcGIS Calculate, obtain shooting standpoint sensitivity distribution figure, as shown in Figure 6；
2.2.2.7 optic centre point sensitivity distribution figure and shooting sight sensitivity distribution diagram data production method) are shot Consistent with the above, shooting optic centre point sensitivity distribution figure and shooting sight sensitivity distribution can be obtained by repeating the above steps Figure；
2.2.3) visual sensitivity classification
Shooting standpoint susceptibility, shooting optic centre point susceptibility and sight susceptibility are carried out using ArcGIS softwares Classification：If ki∈ [0,5) it is then 1 grade of visual sensitivity, if ki∈ [5,9) it is then 2 grades of visual sensitivities, ki∈ [9,13] is then 3 Level visual sensitivity.
3) Yuyin Garden visual distinctiveness key element forms analysis
3.1) visible elements of Yuyin Garden and the picture in whole city park is definite
According to whole city park webcam picture library and Yuyin Garden webcam picture library, by what is occurred in all pictures Plant, railing, window, pond, roof, door, pavilion, mural painting, people, artificial hillock, sculpture, furniture, calligraphy and painting, bridge, vestibule, guideboard, waterfall Cloth, tower, lantern, fish, stair, memorial archway as visible elements and are recorded in Excel forms；
3.2) Yuyin Garden and whole city park webcam picture element attribute are differentiated
3.2.1) to network picture number consecutively of often throwing the net, the alternative elements category using visible elements as every numbering picture Property, 1 is recorded as if occurring certain key element in picture, is otherwise 0, it is as shown in table 1 below；
1 network picture component attributes table of table
3.2.2 picture ownership attribute) is added, Yuyin Garden picture record is 1, and whole city park picture record is 2, by result Save as Yuyin Garden and whole city park webcam picture element attribute list；
3.3) conspicuousness of Yuyin Garden visual distinctiveness key element differentiates
3.3.1) according to Yuyin Garden and whole city park webcam picture element attribute list, using SPSS softwares to the blessings of one's ancestors Mountainous house and whole city park webcam picture element carry out independent sample T check analyses, and the formula that independent sample T is examined is：
Wherein, tkIt is each key element significant indexes for T test statistics,WithRespectively Yuyin Garden and The photographs component attributes value that numbering is k in whole city park, n1, kAnd n2, kRespectively numbered in Yuyin Garden and whole city park For the photographs sample size of k, S1, kAnd S2, kIt is respectively the graph that numbering is k in Yuyin Garden and whole city park The sample variance of piece component attributes value；Independent sample T inspection results, it is as shown in table 2 below；
2 independent sample T examination tables of table
3.3.2 each key element significant indexes t) can be obtained by independent sample T examination tableskIf certain element tkValue is more than 0.2 is cast out the element, if certain element tkValue retains the element no more than 0.2 and makees Yuyin Garden visual distinctiveness key element and correspondence Record its tkValue, will obtain Yuyin Garden visual distinctiveness key element table.
3.4) the visual distinctiveness classification of Yuyin Garden visual distinctiveness key element
It is to feel that characteristic element carries out visual distinctiveness degree classification to Yuyin Garden using Yuyin Garden visual distinctiveness key element table, To the significant indexes p of Mr. Yu's elementiIf pi∈ [0,0.005) it is then 1 grade of visual distinctiveness, if pi∈ [0.005,0.01) be then 2 grades of visual distinctiveness, if pi∈ [0.01,0.1] is then 3 grades of visual distinctiveness, as shown in table 3 below；
Visible elements P value Visual rating scale
Plant 0.000 1
Railing 0.000 1
Window 0.000 1
Pond 0.001 1
Roof 0.005 2
Door 0.000 1
Pavilion 0.000 1
Mural painting 0.000 1
Artificial hillock 0.000 1
Sculpture 0.000 1
Furniture 0.000 1
Calligraphy and painting 0.006 2
Bridge 0.001 1
Vestibule 0.000 1
Waterfall 0.010 3
Lantern 0.000 1
Fish 0.000 1
Stair 0.023 3
Memorial archway 0.105 3
The visual distinctiveness hierarchical table of 3 visual distinctiveness key element of table
4) tourist site visual information overall merit
In summary visual information distribution situation, establishes characteristic element attribute list, and row is defined as characteristic element, and row are fixed For justice to shoot standpoint, optic centre point, shooting gaze area, the differentiation of progress key element importance, instructs target environment, i.e., remaining The landscape pattern optimizing of shady mountainous house；
The frequency of occurrence of lantern is higher in Yuyin Garden, and suggestion is provided to the hanging position of lantern in festival celebration daily lessons；And for example, By the identification to gaze area, suggestion is provided to potted plant put in gardens；For another example, by shooting the identification in standpoint region, Optimize the design of gallery and grandstand in original gardens, correct guidance stream of people tissue.
In conclusion it is of the invention based on objective completed region of the city geodata and network opening data, by ground The quantitative analysis methods such as analysis, statistical analysis, correlation analysis are managed, action process point is shot by analyzing the common people in network media The vision hot spot of city built environment landscape is analysed, characteristic visual element in the environment of Statistical Comparison different cities, aids in urban landscape Design and planning.
The above, is only patent preferred embodiment of the present invention, but the protection domain of patent of the present invention is not limited to This, any one skilled in the art is in the scope disclosed in patent of the present invention, the skill of patent according to the present invention Art scheme and its patent of invention design are subject to equivalent substitution or change, belong to the protection domain of patent of the present invention.

Claims (6)

  1. A kind of 1. built environment Landscape Characteristics recognition methods based on network picture, it is characterised in that：The described method includes following Step：
    S1, obtain target environment and the photographs library similar to built environment using Baidu's photographic search engine；
    S2, for target environment, use the shooting standpoint in AutoCAD chronophotography shooting behaviors, shooting optic centre point Recorded, specifically included with shooting sight：
    S21, the geographic Location Classification by target environment webcam picture library according to shooting；
    S22, each pictures in target environment field observation, control target environment webcam picture library, with target ring Border plan, using the diagonal intersection point camera site of each pictures as optic centre point, passes through scene as working base map The mode of reduction, CAD working base maps are drawn in CAD plans by shooting standpoint, optic centre point and shooting sight respectively On；
    S3, for target environment, standpoint, shooting optic centre point will be shot and shoot sight and import the analysis of ArcGIS and put down Platform, network picture visual acuity area and to sensitivity analysis using ArcGIS software analysis landscapes；
    S4, using target environment and similar built environment webcam picture library, both webcam picture element attributes of record, Examined using the independent sample T of SPSS softwares and filter out target environment visual distinctiveness key element；
    S5, for target environment and similar built environment, in summary visual information distribution situation, establishes characteristic element attribute Table, characteristic element is defined as by row, and row are defined as shooting standpoint, optic centre point, shooting gaze area, carry out key element weight The differentiation for the property wanted, instructs the landscape pattern optimizing of target environment
    Wherein, target environment refers to need the built environment for carrying out Landscape Characteristics identification, and similar built environment refers to and target ring Border is generic, has similitude and other built environments of comparability.
  2. 2. a kind of built environment Landscape Characteristics recognition methods based on network picture according to claim 1, its feature exist In：In step S1, the photographs library that target environment and similar built environment are obtained using Baidu's photographic search engine, tool Body includes：
    S11, using Baidu's picture query target environment title, and all picture search results are downloaded in order, by picture searching As a result the picture unrelated with target environment is deleted in, preceding 1200 pictures in picture search result is chosen, as target environment net Network photographs library；
    S12, using Baidu's picture inquire about some similar built environment titles respectively, and downloads all picture searchings in order respectively As a result, picture unrelated with each similar built environment in picture search result is deleted, each picture search result is chosen respectively In preceding 200 pictures, totally 1200 pictures be used as similar built environment webcam picture library.
  3. 3. a kind of built environment Landscape Characteristics recognition methods based on network picture according to claim 1, its feature exist In：In step S3, for target environment, standpoint, shooting optic centre point and minute for shooting sight importing ArcGIS will be shot Platform is analysed, network picture visual acuity area and to sensitivity analysis using ArcGIS software analysis landscapes, specifically included：
    S31, will shoot standpoint, optic centre point and shoot the CAD data of sight and be directed respectively into ArcGIS softwares and carry out GIS Space dropping place is rectified a deviation with coordinate；
    ArcToolbox in S32, opening ArcGIS softwares, to shooting standpoint and is regarded using Spatial Analyze instruments Feel that central point carries out dot density analysis operation, line density analysis operation is carried out to shooting sight, respectively obtains target environment shooting Standpoint density value, optic centre dot density value and shooting sight density value；
    S33, make target environment shooting standpoint density value, optic centre dot density value and shooting sight density using INT instruments Value is changed into shape data, obtains integer shooting standpoint Density Distribution, the distribution of integer optic centre dot density and integer shooting and regards Line linear-density distribution；
    S34, calculate shooting standpoint susceptibility, shooting optic centre point susceptibility and shooting sight sensitivity using ArcGIS softwares Degree；
    S35, using ArcGIS softwares to shooting standpoint susceptibility, shooting optic centre point susceptibility and shooting sight susceptibility It is divided into the visual sensitivity of three ranks from high to low.
  4. 4. a kind of built environment Landscape Characteristics recognition methods based on network picture according to claim 3, its feature exist In：In step S34, shooting standpoint susceptibility, shooting optic centre point susceptibility and the calculating for shooting sight susceptibility, Specially：
    1) standpoint density value is shot using processed integer, field is created in ArcGIS attribute lists, calculated dot density and be averaged Value, dot density average valueCalculation formula be：
    <mrow> <mover> <mi>b</mi> <mo>&amp;OverBar;</mo> </mover> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&amp;Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>=</mo> <mi>n</mi> </mrow> </msubsup> <msub> <mi>b</mi> <mi>i</mi> </msub> </mrow> <mi>n</mi> </mfrac> </mrow>
    Wherein, n be grid point sum, biFor grid point value；
    2) result of calculation of dot density value is exported into Excel, standard deviation is calculated in Excel, the calculation formula of standard deviation is：
    <mrow> <msub> <mi>M</mi> <mi>i</mi> </msub> <mo>=</mo> <msqrt> <mrow> <mfrac> <mn>1</mn> <mi>n</mi> </mfrac> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>b</mi> <mi>i</mi> </msub> <mo>-</mo> <mover> <mi>b</mi> <mo>&amp;OverBar;</mo> </mover> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow> </msqrt> </mrow>
    Wherein, MiFor standard deviation, n is the sum of grid point, biFor grid point value,For dot density average value；
    3) using density analysis and raster map layer overlay analysis technology, shooting standpoint sensitivity analysis is carried out, for certain point Sensitivity analysis calculation formula be：
    <mrow> <msub> <mi>k</mi> <mi>i</mi> </msub> <mo>=</mo> <mfrac> <mrow> <msub> <mi>b</mi> <mi>i</mi> </msub> <mo>-</mo> <mover> <mi>b</mi> <mo>&amp;OverBar;</mo> </mover> </mrow> <msub> <mi>M</mi> <mi>i</mi> </msub> </mfrac> </mrow>
    Wherein, kiFor the susceptibility of certain point, MiPoor, the b for shooting sight density criterioniFor grid point value,It is close for shooting sight Spend average value；
    4) vector format data are created, field is created, an average density value is imparted among attribute list；
    5) using vector data is converted to raster data instrument according to field value, above-mentioned vector data is switched into grid number respectively According to, obtain represent point an average density value raster map layer；
    6) according to susceptibility calculation formula, map superposition calculation is carried out using Map Algebra instruments in ArcGIS, is shot Standpoint sensitivity distribution figure；
    7) repeat the above steps 1) to the method for step 6), obtain shooting optic centre point sensitivity distribution figure and shooting sight is quick Moves China figure.
  5. 5. a kind of built environment Landscape Characteristics recognition methods based on network picture according to claim 1, its feature exist In：It is described to utilize target environment and similar built environment webcam picture library, both webcam pictures of record in step S4 Component attributes, specifically include：
    S41, according to target environment webcam picture library and similar built environment webcam picture library, by all pictures Visible elements are recorded in Excel forms；
    S42, to network picture number consecutively of often throwing the net, the alternative elements attribute using visible elements as every numbering picture, if figure Occur certain key element in piece and be then recorded as 1, be otherwise 0；
    S43, addition picture ownership attribute, target environment picture record is 1 and similar built environment picture record is 2, is as a result protected Save as target environment webcam picture element attribute list and similar built environment webcam picture element attribute list.
  6. 6. a kind of built environment Landscape Characteristics recognition methods based on network picture according to claim 5, its feature exist In：In step S4, the independent sample T with SPSS softwares is examined and is filtered out target environment visual distinctiveness key element, is specifically included：
    S44, according to target environment webcam picture element attribute list and similar built environment webcam picture element attribute Table, independent sample T check analyses, independent sample are carried out using SPSS softwares to target environment and similar built environment visible elements T examine formula be：
    <mrow> <msub> <mi>t</mi> <mi>k</mi> </msub> <mo>=</mo> <mfrac> <mrow> <mover> <msub> <mi>X</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>&amp;OverBar;</mo> </mover> <mo>-</mo> <mover> <msub> <mi>X</mi> <mrow> <mn>2</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>&amp;OverBar;</mo> </mover> </mrow> <msqrt> <mrow> <mfrac> <mrow> <mo>(</mo> <msub> <mi>n</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>-</mo> <mn>1</mn> <mo>)</mo> <msup> <msub> <mi>S</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mn>2</mn> </msup> <mo>+</mo> <mo>(</mo> <msub> <mi>n</mi> <mrow> <mn>2</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>-</mo> <mn>1</mn> <mo>)</mo> <msup> <msub> <mi>S</mi> <mrow> <mn>2</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mn>2</mn> </msup> </mrow> <mrow> <msub> <mi>n</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>n</mi> <mrow> <mn>2</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> <mo>-</mo> <mn>2</mn> </mrow> </mfrac> <mrow> <mo>(</mo> <mfrac> <mn>1</mn> <msub> <mi>n</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> </mfrac> <mo>+</mo> <mfrac> <mn>1</mn> <msub> <mi>n</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>k</mi> </mrow> </msub> </mfrac> <mo>)</mo> </mrow> </mrow> </msqrt> </mfrac> </mrow>
    Wherein, tkFor T test statistics,WithThe photography that respectively numbering is k in target environment and similar built environment Picture element property value, n1,kAnd n2,kThe photographs sample number that respectively numbering is k in target environment and similar built environment Amount, S1,kAnd S2,kThe sample side for the photographs component attributes value that numbering is k respectively in target environment and similar built environment Difference；
    S45, T test statistics tkAs each key element significant indexes, if certain element tkValue is more than 0.1, then casts out the element, if Certain element tkValue is not more than 0.1, then retains the element and make target environment visual distinctiveness key element simultaneously its t of corresponding recordkValue, will obtain Target environment visual distinctiveness key element table；
    S46, using target environment visual distinctiveness key element table, visual distinctiveness degree classification is carried out to target environment Landscape Feature element, So as to filter out target environment visual distinctiveness key element.
CN201510288892.4A 2015-05-29 2015-05-29 A kind of built environment Landscape Characteristics recognition methods based on network picture Active CN104933229B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201510288892.4A CN104933229B (en) 2015-05-29 2015-05-29 A kind of built environment Landscape Characteristics recognition methods based on network picture
PCT/CN2016/083258 WO2016192552A1 (en) 2015-05-29 2016-05-25 Built environment landscape characteristic recognition method based on network picture
SG11201709813SA SG11201709813SA (en) 2015-05-29 2016-05-25 Built environment landscape characteristic recognition method based on network picture

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510288892.4A CN104933229B (en) 2015-05-29 2015-05-29 A kind of built environment Landscape Characteristics recognition methods based on network picture

Publications (2)

Publication Number Publication Date
CN104933229A CN104933229A (en) 2015-09-23
CN104933229B true CN104933229B (en) 2018-04-13

Family

ID=54120395

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510288892.4A Active CN104933229B (en) 2015-05-29 2015-05-29 A kind of built environment Landscape Characteristics recognition methods based on network picture

Country Status (3)

Country Link
CN (1) CN104933229B (en)
SG (1) SG11201709813SA (en)
WO (1) WO2016192552A1 (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104933229B (en) * 2015-05-29 2018-04-13 华南理工大学 A kind of built environment Landscape Characteristics recognition methods based on network picture
CN108053458A (en) * 2017-12-08 2018-05-18 河海大学 A kind of watershed generalization figure based on GIS-Geographic Information System makes and methods of exhibiting
CN110175215B (en) * 2019-04-29 2022-11-11 河南大学 Method for dividing geographical transition zone
CN112117431B (en) * 2019-06-21 2021-12-03 无锡先导智能装备股份有限公司 Deviation rectifying method and deviation rectifying system
CN111008368A (en) * 2019-11-26 2020-04-14 武汉大学 Grid-based collaborative design safety product data exchange method and system
CN113159122B (en) * 2021-03-16 2022-03-15 重庆市地理信息和遥感应用中心 City geomorphology analysis method based on social media image data
CN114626966A (en) * 2022-03-18 2022-06-14 杭州师范大学 Method for quantizing large-scale urban blue-green space cooling scale based on landscape pattern
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 (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5631970A (en) * 1993-05-21 1997-05-20 Hsu; Shin-Yi Process for identifying simple and complex objects from fused images and map data
CN103196429A (en) * 2013-03-25 2013-07-10 东南大学 Method for quickly obtaining and measuring orthophotoquad of city skyline contour line facede
CN103823914A (en) * 2013-05-24 2014-05-28 东南大学 Touring line planning method

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100668159B1 (en) * 2005-07-01 2007-01-11 김문정 Landscape Architecture Design Method Based On Real Image and System thereof
DE202012013403U1 (en) * 2011-08-26 2016-10-27 Terra Bella Technologies Inc. Capture and process adaptive images with image analysis feedback
EP2797011A1 (en) * 2013-04-25 2014-10-29 Thomson Licensing Method of and apparatus for providing access to content in a database based on environment-derived contextual information
CN104077447B (en) * 2014-07-01 2017-02-22 东南大学 Urban three-dimensional space vector modeling method based on paper plane data
CN104933229B (en) * 2015-05-29 2018-04-13 华南理工大学 A kind of built environment Landscape Characteristics recognition methods based on network picture

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5631970A (en) * 1993-05-21 1997-05-20 Hsu; Shin-Yi Process for identifying simple and complex objects from fused images and map data
CN103196429A (en) * 2013-03-25 2013-07-10 东南大学 Method for quickly obtaining and measuring orthophotoquad of city skyline contour line facede
CN103823914A (en) * 2013-05-24 2014-05-28 东南大学 Touring line planning method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Exploring place through user-generated content:Using Flickr to describe city cores;Hollenstein,L等;《Journal of Spatial Information Science》;20101231(第1期);第21-48页 *
Mining flickr landmarks by modeling reconstruction sparsity;RONGRONG JI等;《ACM Transactions on Multimedia Computing,Communications,and Applications》;20111031;第7S卷(第1期);第1-22页 *
上海市中心城空间意象的媒介表征;赵渺希等;《人文地理》;20121015(第5期);第36-41页 *
互联网媒介中的城市意象图景-以广东21个城市为例;赵渺希等;《建筑学报》;20150220(第02期);第44-49页 *

Also Published As

Publication number Publication date
CN104933229A (en) 2015-09-23
WO2016192552A1 (en) 2016-12-08
SG11201709813SA (en) 2017-12-28

Similar Documents

Publication Publication Date Title
CN104933229B (en) A kind of built environment Landscape Characteristics recognition methods based on network picture
Banzhaf et al. Monitoring urban structure types as spatial indicators with CIR aerial photographs for a more effective urban environmental management
KR20160048255A (en) A gis-based thematic mapping method to use in estimating the annual amount of non-point source pollutants using partially updated land-cover map
CĂLINA et al. Research on the use of aerial scanning for completing a GIS database
CN107796414A (en) A kind of most U.S. method for path navigation and system based on the aesthetics scoring of streetscape figure
Beiranvand et al. Evaluation of changes in per capita green space through remote sensing data
Nduka et al. Classifying urban climate field sites by “thermal climate zones” the case of Onitsha metropolis
CN113506371A (en) Method and device for drawing street scale climate map, electronic equipment and storage medium
Zhou et al. Shadow pattern-enhanced building height extraction using very-high-resolution image
Guo et al. Combining Luojia1-01 nighttime light and points-of-interest data for fine mapping of population spatialization based on the zonal classification method
Xiao et al. A review of remote sensing applications in urban planning and management in China
Wibowo et al. Mapping and grouping of farm land with Graham scan algorithm on convex hull method
Schaller et al. Geodesign apps and 3D modelling with CityEngine for the city of tomorrow
Zhang et al. The color analysis of building façades: Based on the panoramic street view images
Navas Carrillo et al. The Urban Heritage characterization using 3D Geographic Information Systems. The system of medium-sized cities in Andalusia
Schaffer et al. Challenges and possible approaches for using GIS as a tool in historical geography landscape research: A meta-analysis review
Li et al. A novel outdoor thermal comfort simulation model for heritage environments (OTC-SM-HE): Verify the effectiveness in Gulangyu, China
Adão et al. Bringing together UAS-based land surveying and procedural modelling of buildings to set up enhanced VR environments for cultural heritage
Kowalczyk The iconic model of landscape aesthetic value
Erving et al. Data integration from different sources to create 3D virtual model
Lee Case study on emerging trends in geospatial technologies for study of urban form
Dhaoui Urban sprawl: The GIS and remote sensing data assessments
Zarbakhsh et al. Points-of-interest from mapillary street-level imagery: a dataset for neighborhood analytics
Shao Baselining UBC Vancouver Campus Urban Forest and Land Use: Developing and Validating Up-to-date Tree Inventory and Land Classification Map of the UBC Vancouver Campus
Tsang Do You See What I See? Assessing the Relationships between Demographics, Street Trees and Visual Recognition of Urban Buildings

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant