CN110570470A - ghost community identification and housing vacancy rate estimation method based on multi-source remote sensing data - Google Patents
ghost community identification and housing vacancy rate estimation method based on multi-source remote sensing data Download PDFInfo
- Publication number
- CN110570470A CN110570470A CN201910840231.6A CN201910840231A CN110570470A CN 110570470 A CN110570470 A CN 110570470A CN 201910840231 A CN201910840231 A CN 201910840231A CN 110570470 A CN110570470 A CN 110570470A
- Authority
- CN
- China
- Prior art keywords
- ghost
- data
- remote sensing
- sensing data
- community
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000007781 pre-processing Methods 0.000 claims description 21
- 241000222511 Coprinus Species 0.000 claims description 11
- 238000011160 research Methods 0.000 claims description 11
- 238000003708 edge detection Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 claims description 2
- 238000000605 extraction Methods 0.000 claims description 2
- 238000011835 investigation Methods 0.000 abstract description 4
- 238000011156 evaluation Methods 0.000 abstract description 2
- 238000011161 development Methods 0.000 description 5
- 230000018109 developmental process Effects 0.000 description 5
- 240000001606 Adenanthera pavonina Species 0.000 description 2
- 235000011470 Adenanthera pavonina Nutrition 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000009440 infrastructure construction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 244000234623 Coprinus comatus Species 0.000 description 1
- 235000004439 Coprinus comatus Nutrition 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Traffic Control Systems (AREA)
Abstract
the invention discloses a ghost community identification and housing vacancy rate estimation method based on multi-source remote sensing data, which comprehensively utilizes a large-scale multi-source remote sensing data product with the advantage of spatial resolution to construct a ghost phenomenon evaluation index applicable to a block scale, seeks a threshold setting method with regional difference adaptivity, and realizes the block scale ghost identification in the national range; on the basis, a housing vacancy rate estimation model with a suitable street scale is constructed by combining real-time video recording of a network street view and on-site investigation data, and the nationwide housing vacancy rate is estimated.
Description
Technical Field
the invention relates to the field of image processing, in particular to ghost community identification and housing vacancy rate estimation based on multi-source remote sensing data.
background
The local planning and construction of new city new districts is driven to enable the formation risk of the 'ghost city' phenomenon to be remarkably increased, the research of national development and reform committee cities and small town reform development centers shows that the number of new city new districts in counties and counties reaches more than 3500 and 34 hundred million population can be accommodated by 2016 years of 5 months. Comparing projected population capacity (34 billion) of a new city new region with the current population count nationwide (about 13.9 billion by 2018), it can be seen that the new city new region can accommodate a population number far exceeding the population count nationwide. In this context, there are some new urban areas with very low population density and very high housing vacancy rates. This phenomenon has attracted international media attention around 2010, and these new areas are labeled "ghost" in the city. According to the approval of the national Committee for the approval of scientific and technical terms, "ghost city" belongs to the geographic term and refers to a city in which resources are exhausted and abandoned. This conventional definition clearly does not correspond to the current situation and a new definition of "ghost" is sought. This "Coprinus" was defined in the book "Ghost Cities of China" published by He 2015 by Wedders Shepard (Wadesshepard) in the United states as: existing population and business activities are far below the newly developed areas that they can accommodate and carry, characterized as: the population density is extremely low, the housing vacancy rate is extremely high, and weak illumination intensity and extremely small illumination area are shown at night. This "Coprinus" definition and characterization is widely cited and recognized in the present study.
although the definition and the characteristics of the 'ghost city' have more uniform cognition, the 'ghost city' judgment and identification face a plurality of difficulties from theory to practice. For example: the population migration of the new city area needs a buffer period, no population or only few population migrates after a sufficient time, the development of the new city area exceeds the actual market demand, a large possibility of becoming the 'ghost', and the determination of the buffer period is the premise of 'ghost' identification; the core of the 'ghost' identification is that subjective experience is converted into objective indexes, threshold setting is carried out on the calculation results of the objective indexes, and how to realize scientific conversion of the indexes and objective setting of the thresholds are the key points of the 'ghost' identification; the 'ghost city' does not refer to a real city, the actual space coverage range of the 'ghost city' is greatly different, the coverage range is as small as a contiguous residential area, the coverage range is as large as a whole new city new area, and how to process the difference of the scale is directly related to the credibility of the 'ghost city' identification. The housing vacancy rate is the most intuitive evaluation index of the severity of the 'ghost' phenomenon, and the estimation of the 'ghost' housing vacancy rate is limited by the recognition accuracy of the current research on the 'ghost'. On the other hand, the current study on the housing vacancy rate is also limited in that the data precision cannot accurately identify the 'ghost city', so that the housing vacancy rate of the 'ghost city' cannot be acquired.
the current "ghost city" identification faces the following challenges: 1) the coverage of the current research area is generally small, most of the current research area is a typical city and a typical area, and the research of nationwide 'ghost city' identification is less in occurrence; 2) limited by the spatial resolution of basic data, the identification precision of the 'ghost city' stays in administrative units in counties/regions and above, and the 'ghost city' coverage area which is accurate cannot be identified deeply in cities; 3) the threshold setting of the current 'ghost' index is based on ranking, standard deviation and the like, has certain subjectivity, and can influence the accuracy and the reliability of the 'ghost' identification result without considering the regional difference of cities.
the housing vacancy rate is the most fundamental quantitative index for identifying the 'ghost city', however, the current research does not consider identifying the 'ghost city' by the housing vacancy rate for the following two reasons: firstly, from the viewpoint of statistical data, currently, there is no official data of the housing vacancy rate for reference, and there is no authoritative threshold setting standard, that is, how much the housing vacancy rate is higher than what belongs to the "ghost city"; secondly, considering from the aspect of the technical method, the demand of the residential space rate estimation on data is higher than that of the 'ghost city' identification, the 'ghost city' identification is difficult to construct the method, and the 'ghost city' identification by estimating the residential space rate faces the problems of high difficulty, low efficiency, high instability and the like.
disclosure of Invention
Based on the problems of the existing ghost city identification and the housing vacancy rate, the invention provides a method and a system for identifying a Chinese ghost community and estimating the housing vacancy rate based on multi-source remote sensing data.
A method for identifying a ghost community based on multi-source remote sensing data is characterized by comprising the following steps:
s01, acquiring high-resolution luminous remote sensing data and town land data with the same coverage area, and respectively preprocessing the data;
step S02, extracting road information based on the noctilucent remote sensing data, and dividing the noctilucent remote sensing data into patch images of a block scale by combining town land data;
step S03, calculating the average noctilucence intensity, the internal difference of the noctilucence intensity and the proportion of the noctilucence intensity-free area of the block scale based on the block image of the block scale, and linearly fitting the three parameters to be used as a ghost index;
Step S04, setting a reasonable threshold value based on the Coprinus city index to identify potential ghost community plaques;
and S05, introducing building height data, calculating the average building height of the block scale patches, and removing patches with the average building height smaller than a threshold value in the potential ghost community patches identified in the step S04 to obtain a final ghost community distribution map.
A house vacancy rate inversion method based on multi-source remote sensing data is characterized by comprising the following steps:
S01, acquiring high-resolution luminous remote sensing data and town land data with the same coverage area, and respectively preprocessing the data;
Step S02, extracting road information based on the noctilucent remote sensing data, and dividing the noctilucent remote sensing data into patch images of a block scale by combining town land data;
Step S03, calculating the average noctilucence intensity, the internal difference of the noctilucence intensity and the proportion of the noctilucence intensity-free area of the block scale based on the block image of the block scale, and linearly fitting the three parameters to be used as a ghost index;
Step S04, introducing building height data, calculating the average building height of the block scale patch, and calculating the luminous intensity in unit volume based on the Coprinus city index, the average building height and the patch area;
And S05, acquiring actual housing vacancy rate data of the block scale based on a survey method, and acquiring a housing vacancy rate estimation model by adopting the actual housing vacancy rate data and the noctilucence intensity data in unit volume and adopting a linear fitting method.
the utility model provides a ghost community recognition device based on multisource remote sensing data which characterized in that:
The preprocessing module is used for acquiring high-resolution luminous remote sensing data and town land data with the same coverage area and respectively preprocessing the data;
the patch segmentation module extracts road information based on the noctilucent remote sensing data, and segments the noctilucent remote sensing data into patch images of a block scale by combining town land data;
the ghost index building module is used for calculating the average luminous intensity, the internal difference of the luminous intensity and the proportion of a non-luminous intensity area of the block scale plaque based on the block image of the block scale, and linearly fitting the three parameters to be used as a ghost index;
And the ghost community extraction module is used for setting a reasonable threshold value based on the ghost index to identify potential ghost community plaques.
The utility model provides a house vacancy rate inversion device based on multisource remote sensing data which characterized in that:
The preprocessing module is used for acquiring high-resolution luminous remote sensing data and town land data with the same coverage area and respectively preprocessing the data;
The patch segmentation module extracts road information based on the noctilucent remote sensing data, and segments the noctilucent remote sensing data into patch images of a block scale by combining town land data;
The ghost index building module is used for calculating the average luminous intensity, the internal difference of the luminous intensity and the proportion of a non-luminous intensity area of the block scale plaque based on the block image of the block scale, and linearly fitting the three parameters to be used as a ghost index;
The optimization module is used for introducing building height data, calculating the average building height of the block scale patch, and calculating the luminous intensity in unit volume based on the average building height of the ghost index and the patch area;
The building module of the housing vacancy rate model is used for acquiring actual housing vacancy rate data of a block scale based on a survey method, acquiring the housing vacancy rate estimation model by adopting the actual housing vacancy rate data and noctilucence intensity data in unit volume and adopting a linear fitting method.
the application has the following advantages: (1) at present, the precision of ghost city identification and housing vacancy rate estimation is limited by the spatial resolution of basic data, and ghost city identification or housing vacancy rate estimation of a large area only stays at a county/district level and cannot go deep into the interior of a city. According to the invention, an international advanced multi-source remote sensing data product is adopted, the dilemma faced by the current research is broken through, and the precision of estimating the housing vacancy rate of the 'ghost city' is improved to the level of a block; (2) in the previous research, the regional difference of the luminous intensity is not considered in the model construction process, the difference of the urban luminous intensity is fully considered, the interference of a background pixel is fully considered by taking a city as a unit, and the method creatively proposes that the ghost index is constructed by adopting the average luminous intensity of the plaque, the internal difference of the luminous intensity and the proportion of the region without the luminous intensity; (3) the method fully considers the influence caused by the housing height in the housing vacancy rate estimation, and creatively constructs the housing vacancy rate estimation model which is tried in the size of the block.
drawings
FIG. 1 is a schematic view of the height inversion of the present invention.
fig. 2 is a diagram corresponding to ghost community identification of the present invention.
Detailed Description
to facilitate an understanding of the invention, the invention is described more fully below with reference to the accompanying drawings. Preferred embodiments are shown in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
the method and the system for ghost community identification and house vacancy rate estimation based on multi-source remote sensing data are disclosed by the embodiment.
A method for identifying a ghost community based on multi-source remote sensing data is characterized by comprising the following steps:
s01, acquiring high-resolution luminous remote sensing data and town land data with the same coverage area, and respectively preprocessing the data;
The high-resolution luminous remote sensing data is Lopa gamma first-number luminous data; the Lopa Jia I is the first professional luminous remote sensing satellite in the world, is researched and developed by Wuhan university team and related institutions together, and is successfully launched in No. 6/month 2 in 2018. The image space resolution is 130 m, the breadth is 250km multiplied by 250km, and the global luminous image drawing can be completed in 15 days under ideal conditions. The invention adopts 'Lopa one number' night light data between 2018 and 2019. The pretreatment comprises the following steps: and converting the DN value into one or more of a radiance value, splicing, geometric correction and desaturation treatment of the luminous image. The Lopa I-Ha 01 star product conforms to the brightness conversion formula as follows:
wherein L is the absolute radiation corrected radiance value in W/(m)2sr μm), DN is the image gray value.
The town land data adopts Global Urban Fountain (GUF) town land classification data, GUF the town land classification data is created by the Germany aerospace center, data based on Germany radar satellites TerrraSAR-X and TanDEM-X are automatically extracted, time nodes are 2011-2012, and the spatial resolution is as high as 12 meters. Because the town land data comprises a plurality of sparsely distributed rural residential points, the rural residential points are generally not strong in light intensity at night and are easily identified as the ghost, the 5 multiplied by 5 window is adopted to carry out Gaussian filtering based on the town land data, the sparsely distributed rural residential points are removed, and the town land patches with large areas are reserved.
the DEM data adopts a TanDEM-X global digital elevation model. The TanDEM-X global digital elevation model is drawn by an SAR interferometer pair composed of TanDEM-X and TerrasAR-X satellites in the German space center, and the two satellites are separated by 120-500 meters. The digital elevation model comprises data products with three resolutions of 90 meters, 30 meters and 12 meters, wherein the elevation models with the resolutions of 90 meters and 30 meters are derived from the elevation model with the resolution of 12 meters, and the existing elevation model with the resolution of 90 meters can be freely obtained. DEM data includes the height of the terrain itself, the height of the building in a generally flat area, and the height of the building in the vicinity of the building in a generally flat open space, so that the height of the building is inverted by subtracting the minimum value in the window.
Step S02, extracting road information based on the noctilucent remote sensing data, and dividing the noctilucent remote sensing data into patch images of a block scale by combining town land data;
Extracting road information by adopting edge detection methods such as high-pass filtering, CANNY, SOBEL and the like based on noctilucent remote sensing data, and connecting road breakpoints by adopting a connection operator due to the fact that the road provided by an edge detection operator has breakpoints to obtain corrected road information; and cutting the night light data by using the town land data after the rural residential points are removed, and dividing the night light data into blocks with the size of the block by adopting the corrected road information. As shown in fig. 1, a minimum value map as shown in fig. 1b is generated in the original image 1a by finding the minimum value (20) in the window of 5 × 5, and then the minimum value image is subtracted from the original height image to obtain a building height image as shown in fig. 1 c.
Step S03, calculating the average luminous intensity, the internal difference of the luminous intensity and the proportion of the non-luminous intensity area of each plaque based on the plaque image of the block scale, and linearly fitting the three parameters to be used as the ghost index of the plaque;
The construction of the ghost index is considered from three aspects of average luminous intensity, internal difference of luminous intensity and proportion of a luminous intensity-free area, and a specific formula is as follows:
GNI＝(1-NTLave,mmn)x(1-NTLstd,mmn)×NTLn-lit，per (2)
Wherein GNI represents the Coprinus index, NTLave，mmnthe method comprises the steps of firstly, calculating the mean value of all pixel night light intensity on the urban land patch of the current block scale, and carrying out range standardization on the mean value; NTLstd，mmnthe method comprises the steps of firstly calculating the standard deviation of all pixel nighttime lamplight intensities on a town land patch of the current block scale, and carrying out range standardization on the standard deviation; NTLn-lit，perThe percentage of the pixels without the night light intensity on the urban land patches of the current block scale is represented; ALImax,DLImaxAnd PLAmaxRespectively representing the maximum value of the pixel night light intensity mean ALI, the maximum value of the pixel night light intensity standard deviation DLI of the pixel urban land patches in the research area and the maximum value of percentage PLA of the non-night light intensity pixels in the patches; wherein L isithe brightness value of a certain pixel in a patch for a certain street scale and town, i is the ith pixel in the patch, n is the total pixel number of the patch, and LBAnd the threshold value is the background pixel threshold value of the lamplight-free area in the noctilucent remote sensing image.
Since the areas are unevenly developed and background pixel thresholds of different cities are different, different background pixel thresholds are set according to each city. For each city, the light intensity difference of the original night light image is compressed through logarithm operation, and fig. 2a is the night light image after logarithm operation. A histogram of the image is obtained, as shown in fig. 2b, a significant threshold distribution between the background and the foreground object can be clearly seen from fig. 2b, and an OTSU algorithm is used to obtain an optimal threshold, so that all background pixels in the night light image are identified, as shown in fig. 2 c. And then, counting the proportion of the background pixels in the plaque to obtain the PLA.
The ghost index GNI ranges between 0 and 1, with closer to 1 being more likely to be ghost community blobs and closer to 0 being less likely to be ghost community blobs.
The following formula can also be used for constructing the Coprinus index:
GNI＝(1-NTLave，mmn)+(1-NTLstd，mmn)+NTLn-lit，per (6)
the ghost index GNI ranges between 0 and 3, with closer to 3 being more likely to be ghost community blobs and closer to 0 being less likely to be ghost community blobs, as shown in fig. 2 d.
Step S04, a potential ghost community blob is identified based on the ghost index setting a reasonable threshold.
Areas with faster economic development usually have better infrastructure construction, and urban areas with the same population density often have higher night light intensity, and the threshold setting of 'one-time cutting' will cause the 'ghost' identification of economically developed areas to be missing and the 'ghost' identification of economically laggard areas to be excessive. The method adopts an OTSU algorithm to search the optimal threshold of the ghost city indexes of different cities aiming at different cities, and the plaque larger than the threshold is regarded as a potential ghost community plaque.
All the plaques in one city are arranged in a descending order according to the calculated ghost index, and because the ghost community plaques occupy a smaller proportion and have a lower frequency relative to the plaques in the whole city, the ghost community plaques are usually positioned in front of the list and show a sharp descending trend, as shown in fig. 2 e; potential ghost communities can be identified by reasonable thresholding, as shown in figure 2 f.
and step S05, acquiring actual housing vacancy rate data of the street scale based on the investigation method, correcting the housing vacancy rate index by adopting the actual housing vacancy rate data, and calculating the housing vacancy rate of each patch.
The DEM data adopts a TanDEM-X global digital elevation model. The TanDEM-X global digital elevation model is drawn by an SAR interferometer pair composed of TanDEM-X and TerrasAR-X satellites in the German space center, and the two satellites are separated by 120-500 meters. The digital elevation model comprises data products with three resolutions of 90 meters, 30 meters and 12 meters, wherein the elevation models with the resolutions of 90 meters and 30 meters are derived from the elevation model with the resolution of 12 meters, and the existing elevation model with the resolution of 90 meters can be freely obtained. The DEM data comprises the height of the terrain, and the building height data needs to be inverted by the DEM data. Buildings are usually built in flat areas and have usually flat open spaces near the building, so the method of pixel minus minimum value in window is used to reverse the building height. And calculating the average building height of the block scale patches, and removing patches with the average building height smaller than the threshold value in the potential ghost communities identified in the step S04 to obtain a final ghost community distribution map.
A house vacancy rate inversion method based on multi-source remote sensing data is characterized by comprising the following steps:
S01, acquiring high-resolution luminous remote sensing data and town land data with the same coverage area, and respectively preprocessing the data;
Step S02, extracting road information based on the noctilucent remote sensing data, and dividing the noctilucent remote sensing data into patch images of a block scale by combining town land data;
Step S03, calculating the average noctilucence intensity, the internal difference of the noctilucence intensity and the proportion of the noctilucence intensity-free area of the block scale based on the block image of the block scale, and linearly fitting the three parameters to be used as a ghost index;
and step S04, introducing building height data, calculating the average building height of the block scale patch, and calculating the unit volume ghost index strength based on the ghost index average building height and the patch area.
wherein GNI represents the Coprinus index, GVR index strength per volume.
and step S05, acquiring actual housing vacancy rate data of the block scale based on a survey method, and acquiring a housing vacancy rate estimation model by adopting the actual housing vacancy rate data and the unit volume ghost index intensity data and adopting a linear fitting method.
The night light intensity on the coverage area of the residential area is formed by overlapping the light intensity of each house which is lighted, and theoretically, the night light intensity is linearly related to the housing vacancy rate. In practical situations, residential areas with the same housing area and housing vacancy rate have certain differences, and the differences are mainly caused by the difference of background light intensity. Areas with faster economic development usually have better infrastructure construction, and even if the housing vacancy rate of a certain residential area is 100%, the coverage area of partial residential areas can have weak night light intensity. In order to solve the problem, the invention aims to select a plurality of residential areas of the same city as samples through network real-time street view data and field sampling investigation, and acquires actual housing vacancy rate data NTL of n cell patches by adopting investigation data or other statistical dataiI belongs to (1 … … n), and acquiring the Coprinus comatus index strength GVR of the unit volumeii ∈ (1.... n). And linearly fitting the housing vacancy rate and the luminous intensity in unit volume by taking a city as a unit, solving coefficients a and b by adopting a least square method, and obtaining a linear equation NTL (absolute total variation) of housing vacancy rate estimation, namely aGVR + b. And in addition, the actual housing vacancy rate data of other cells can be adopted to verify the accuracy of the housing vacancy rate estimation model.
the above description is only a preferred embodiment of the present invention, and is not intended to limit the technical scope of the present invention, so that any minor modifications, equivalent changes and modifications made to the above embodiment according to the technical spirit of the present invention are within the technical scope of the present invention.
Claims (12)
1. A method for identifying a ghost community based on multi-source remote sensing data is characterized by comprising the following steps:
step S1: acquiring high-resolution luminous remote sensing data and town land data with the same coverage area, and respectively preprocessing the data;
Step S2: extracting road information based on the noctilucent remote sensing data, and dividing the noctilucent remote sensing data into patch images of a block scale by combining town land data;
step S3: calculating the average luminous intensity, the internal difference of the luminous intensity and the proportion of a non-luminous intensity area of the block scale plaque based on the block image of the block scale, and linearly fitting the three parameters to be used as a ghost index;
Step S4: potential ghost community patches are identified based on the ghost index setting a reasonable threshold.
2. The method of claim 1, further comprising the steps of:
step S5: introducing building height data, calculating the average building height of the block scale patches, and removing patches with the average building height smaller than a threshold value in the potential ghost community patches identified in the step S4 to obtain a final ghost community distribution map.
3. The method for identifying a ghost community according to claim 1, wherein the data preprocessing in step S1 is specifically:
step S11: the noctilucent remote sensing data preprocessing comprises the following steps: converting the DN value into one or more of radiance value, splicing, geometric correction and desaturation treatment of a noctilucent image;
step S12: the town land data preprocessing comprises the following steps: adopting a filtering window with proper size to carry out Gaussian filtering on the town land data to obtain the filtered town land data, wherein the step can remove sparsely distributed rural residential points and reserve the town land data with larger area;
Step S13: and carrying out spatial registration on the noctilucent remote sensing data and town land data.
4. the method for identifying the ghost community according to claim 3, wherein the step S2 is specifically as follows:
step S21: carrying out edge detection on the noctilucent remote sensing data by adopting an edge detection operator to extract road information, correcting the road information, and connecting road breakpoints to obtain a continuous road segment;
Step S22: and cutting the noctilucent remote sensing data based on the filtered town land data, and removing the continuous road line segment, thereby segmenting the noctilucent remote sensing data into patch images with the size of a block.
5. the method for identifying a ghost community according to claim 1, wherein the ghost index in step S3 is specifically:
GNI＝(1-NTLave，mmn)×(1-NTLstd，mmn)×NTLn-lit，per (2)
Or
GNI＝(1-NTLave，mmn)+(1-NTLstd，mmn)+NTLn-lit，per (6)
Wherein GNI represents the Coprinus index, NTLave，mmnthe method comprises the steps of firstly, calculating the mean value of all pixel night light intensity on the urban land patch of the current block scale, and carrying out range standardization on the mean value; NTLstd，mmnThe method comprises the steps of firstly calculating the standard deviation of all pixel nighttime lamplight intensities on a town land patch of the current block scale, and carrying out range standardization on the standard deviation; NTLn-lit，perThe percentage of the pixels without the night light intensity on the urban land patches of the current block scale is represented; ALImax，DLImaxAnd PLAmaxRespectively representing the maximum value of the pixel night light intensity mean ALI, the maximum value of the pixel night light intensity standard deviation DLI of the pixel urban land patches in the research area and the maximum value of percentage PLA of the non-night light intensity pixels in the patches; wherein L isiThe brightness value of a certain pixel in a patch for a certain street scale and town, i is the ith pixel in the patch, n is the total pixel number of the patch, and LBand the threshold value is the background pixel threshold value of the lamplight-free area in the noctilucent remote sensing image.
6. A house vacancy rate inversion method based on multi-source remote sensing data is characterized by comprising the following steps:
Step S1: acquiring high-resolution luminous remote sensing data and town land data with the same coverage area, and respectively preprocessing the data;
step S2: extracting road information based on the noctilucent remote sensing data, and dividing the noctilucent remote sensing data into patch images of a block scale by combining town land data;
Step S3: calculating the average luminous intensity, the internal difference of the luminous intensity and the proportion of a non-luminous intensity area of the block scale plaque based on the block image of the block scale, and linearly fitting the three parameters to be used as a ghost index;
step S4: building height data are introduced, the average building height of the block scale patches is calculated, and the unit volume ghost index strength is calculated based on the ghost index average building height and the patch area;
Step S5: the method comprises the steps of obtaining actual housing vacancy rate data of a block scale based on a survey method, obtaining a housing vacancy rate estimation model by adopting the actual housing vacancy rate data and unit volume ghost index intensity data and adopting a linear fitting method.
7. the utility model provides a ghost community recognition device based on multisource remote sensing data which characterized in that:
A preprocessing module: acquiring high-resolution luminous remote sensing data and town land data with the same coverage area, and respectively preprocessing the data;
A plaque segmentation module: extracting road information based on the noctilucent remote sensing data, and dividing the noctilucent remote sensing data into patch images of a block scale by combining town land data;
the Coprinus index construction module: calculating the average luminous intensity, the internal difference of the luminous intensity and the proportion of a non-luminous intensity area of the block scale plaque based on the block image of the block scale, and linearly fitting the three parameters to be used as a ghost index;
Ghost community draws module: potential ghost community patches are identified based on the ghost index setting a reasonable threshold.
8. The ghost community identification device of claim 7, further comprising the steps of:
An optimization module: building height data are introduced, the average building height of the block scale patches is calculated, patches with the average building height smaller than a threshold value in the potential ghost community patches identified by the ghost community extraction module are removed, and a final ghost community distribution diagram is obtained.
9. The ghost community recognition device of claim 7, wherein the preprocessing module is specifically:
The night light remote sensing data preprocessing module: converting the DN value into one or more of radiance value, splicing, geometric correction and desaturation treatment of a noctilucent image;
Town land data preprocessing module: the town land data preprocessing module can remove sparsely distributed rural residential points and reserve town land data with larger area;
A registration module: and carrying out spatial registration on the noctilucent remote sensing data and town land data.
10. The ghost community recognition device of claim 9, wherein the blob segmentation module is specifically: carrying out edge detection on the noctilucent remote sensing data by adopting an edge detection operator to extract road information, correcting the road information, and connecting road breakpoints to obtain a continuous road segment; and cutting the noctilucent remote sensing data based on the filtered town land data, and removing the continuous road line segment, thereby segmenting the noctilucent remote sensing data into patch images with the size of a block.
11. a ghost community recognition device according to claim 7, wherein the ghost index body formula in the ghost index constructing module is:
GNI＝(1-NTLave，mmn)×(1-NTLstd，mmn)×NTLn-lit，per (2)
Or
GNI＝(1-NTLave，mmn)+(1-NTLstd，mmn)+NTLn-lit，per (6)
Wherein GNI represents the Coprinus index, NTLave，mmnThe method comprises the steps of firstly, calculating the mean value of all pixel night light intensity on the urban land patch of the current block scale, and carrying out range standardization on the mean value; NTLstd，mmnthe method comprises the steps of firstly calculating the standard deviation of all pixel nighttime lamplight intensities on a town land patch of the current block scale, and carrying out range standardization on the standard deviation; NTLn-lit，perthe percentage of the pixels without the night light intensity on the urban land patches of the current block scale is represented; ALImax，DLImaxAnd PLAmaxrespectively representing the maximum value of the pixel night light intensity mean ALI, the maximum value of the pixel night light intensity standard deviation DLI of the pixel urban land patches in the research area and the maximum value of percentage PLA of the non-night light intensity pixels in the patches; wherein L isithe brightness value of a certain pixel in a patch for a certain street scale and town, i is the ith pixel in the patch, n is the total pixel number of the patch, and LBAnd the threshold value is the background pixel threshold value of the lamplight-free area in the noctilucent remote sensing image.
12. the utility model provides a house vacancy rate inversion device based on multisource remote sensing data which characterized in that:
A preprocessing module: acquiring high-resolution luminous remote sensing data and town land data with the same coverage area, and respectively preprocessing the data;
A plaque segmentation module: extracting road information based on the noctilucent remote sensing data, and dividing the noctilucent remote sensing data into patch images of a block scale by combining town land data;
the Coprinus index construction module: calculating the average luminous intensity, the internal difference of the luminous intensity and the proportion of a non-luminous intensity area of the block scale plaque based on the block image of the block scale, and linearly fitting the three parameters to be used as a ghost index;
An optimization module: building height data are introduced, the average building height of the block scale patches is calculated, and the unit volume ghost index strength is calculated based on the ghost index average building height and the patch area;
The housing vacancy rate model building module comprises: the method comprises the steps of obtaining actual housing vacancy rate data of a block scale based on a survey method, obtaining a housing vacancy rate estimation model by adopting the actual housing vacancy rate data and unit volume ghost index intensity data and adopting a linear fitting method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910840231.6A CN110570470A (en) | 2019-09-06 | 2019-09-06 | ghost community identification and housing vacancy rate estimation method based on multi-source remote sensing data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910840231.6A CN110570470A (en) | 2019-09-06 | 2019-09-06 | ghost community identification and housing vacancy rate estimation method based on multi-source remote sensing data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110570470A true CN110570470A (en) | 2019-12-13 |
Family
ID=68778072
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910840231.6A Pending CN110570470A (en) | 2019-09-06 | 2019-09-06 | ghost community identification and housing vacancy rate estimation method based on multi-source remote sensing data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110570470A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110765885A (en) * | 2019-09-29 | 2020-02-07 | 武汉大学 | City expansion detection method and device based on heterogeneous luminous remote sensing image |
| CN111414820A (en) * | 2020-03-11 | 2020-07-14 | 长光卫星技术有限公司 | Urban population density acquisition method based on noctilucent images |
| CN111553237A (en) * | 2020-04-23 | 2020-08-18 | 江西理工大学 | LJ1-01 night light data denoising method based on multi-state superposition Gamma distribution |
| CN112347926A (en) * | 2020-11-06 | 2021-02-09 | 天津市勘察设计院集团有限公司 | High-resolution image urban village detection method based on building form distribution |
| CN113610873A (en) * | 2021-10-09 | 2021-11-05 | 中国科学院地理科学与资源研究所 | Noctilucent remote sensing image data processing method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130332877A1 (en) * | 2012-06-07 | 2013-12-12 | Costar Group | Method and apparatus for managing real estate data |
| CN104809572A (en) * | 2015-04-27 | 2015-07-29 | 中国科学院城市环境研究所 | Method for inversing population density based on night lamplight data |
| CN106682830A (en) * | 2016-12-26 | 2017-05-17 | 中国科学院深圳先进技术研究院 | Urbanization ghost town index evaluating method and device |
| CN107239756A (en) * | 2017-05-23 | 2017-10-10 | 国家地理空间信息中心 | Density of population analysis system based on high score satellite remote sensing date combination type of ground objects |
| CN109784667A (en) * | 2018-12-21 | 2019-05-21 | 中国科学院遥感与数字地球研究所 | A kind of vacant monitoring method in house based on noctilucence remotely-sensed data |
-
2019
- 2019-09-06 CN CN201910840231.6A patent/CN110570470A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130332877A1 (en) * | 2012-06-07 | 2013-12-12 | Costar Group | Method and apparatus for managing real estate data |
| CN104809572A (en) * | 2015-04-27 | 2015-07-29 | 中国科学院城市环境研究所 | Method for inversing population density based on night lamplight data |
| CN106682830A (en) * | 2016-12-26 | 2017-05-17 | 中国科学院深圳先进技术研究院 | Urbanization ghost town index evaluating method and device |
| CN107239756A (en) * | 2017-05-23 | 2017-10-10 | 国家地理空间信息中心 | Density of population analysis system based on high score satellite remote sensing date combination type of ground objects |
| CN109784667A (en) * | 2018-12-21 | 2019-05-21 | 中国科学院遥感与数字地球研究所 | A kind of vacant monitoring method in house based on noctilucence remotely-sensed data |
Non-Patent Citations (11)
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110765885A (en) * | 2019-09-29 | 2020-02-07 | 武汉大学 | City expansion detection method and device based on heterogeneous luminous remote sensing image |
| CN111414820A (en) * | 2020-03-11 | 2020-07-14 | 长光卫星技术有限公司 | Urban population density acquisition method based on noctilucent images |
| CN111553237A (en) * | 2020-04-23 | 2020-08-18 | 江西理工大学 | LJ1-01 night light data denoising method based on multi-state superposition Gamma distribution |
| CN111553237B (en) * | 2020-04-23 | 2023-11-07 | 江西理工大学 | LJ1-01 night lamplight data denoising method based on polymorphic superposition Gamma distribution |
| CN112347926A (en) * | 2020-11-06 | 2021-02-09 | 天津市勘察设计院集团有限公司 | High-resolution image urban village detection method based on building form distribution |
| CN112347926B (en) * | 2020-11-06 | 2023-05-23 | 天津市勘察设计院集团有限公司 | High-resolution image city village detection method based on building morphology distribution |
| CN113610873A (en) * | 2021-10-09 | 2021-11-05 | 中国科学院地理科学与资源研究所 | Noctilucent remote sensing image data processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110570470A (en) | ghost community identification and housing vacancy rate estimation method based on multi-source remote sensing data | |
| Huang et al. | Toward accurate mapping of 30-m time-series global impervious surface area (GISA) | |
| CN109919875B (en) | High-time-frequency remote sensing image feature-assisted residential area extraction and classification method | |
| CN109598794B (en) | Construction method of three-dimensional GIS dynamic model | |
| Yang et al. | Semiautomated building facade footprint extraction from mobile LiDAR point clouds | |
| CN103984946A (en) | High resolution remote sensing map road extraction method based on K-means | |
| Zou et al. | Detecting individual abandoned houses from google street view: A hierarchical deep learning approach | |
| Ghorbanzadeh et al. | Dwelling extraction in refugee camps using cnn–first experiences and lessons learnt | |
| CN111597949B (en) | NPP-VIIRS night light data-based urban built-up area extraction method | |
| CN112365467B (en) | Foggy image visibility estimation method based on single image depth estimation | |
| CN102663385A (en) | Detection method for spot target on satellite | |
| Chunyang et al. | Sea fog detection using U-Net deep learning model based on MODIS data | |
| CN112232303B (en) | Grassland road information extraction method based on high-resolution remote sensing image | |
| CN113313107A (en) | Intelligent detection and identification method for multiple types of diseases on cable surface of cable-stayed bridge | |
| CN107886125A (en) | MODIS satellite remote sensing images mask methods based on local spectral factorization marking | |
| Yan et al. | Estimation of urban-scale photovoltaic potential: A deep learning-based approach for constructing three-dimensional building models from optical remote sensing imagery | |
| Bashit et al. | Analysis of built-up land spatial patterns using multitemporal satellite imagery in Pekalongan city | |
| Dong et al. | Pixel-level intelligent segmentation and measurement method for pavement multiple damages based on mobile deep learning | |
| CN117078925A (en) | Building rubbish annual output accurate calculation method based on RDSA-deep LabV3+ network | |
| CN111539356A (en) | Pavement gap accurate and efficient detection method based on linear array image | |
| CN103093241A (en) | Optical remote sensing image non-homogeneous cloud layer discriminating method based on homogenization processing | |
| Xia et al. | An approach for road material identification by dual-stage convolutional networks | |
| Guo et al. | Population estimation in Singapore based on remote sensing and open data | |
| Lei et al. | Impervious surface estimation by integrated use of Landsat and MODIS time series in Wuhan, China | |
| Fang et al. | Automatic cloud and snow detection for GF-1 and PRSS-1 remote sensing images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191213 |