CN112184718B - Method and device for automatically extracting high-resolution remote sensing images of urban buildings - Google Patents

Method and device for automatically extracting high-resolution remote sensing images of urban buildings Download PDF

Info

Publication number
CN112184718B
CN112184718B CN202010850697.7A CN202010850697A CN112184718B CN 112184718 B CN112184718 B CN 112184718B CN 202010850697 A CN202010850697 A CN 202010850697A CN 112184718 B CN112184718 B CN 112184718B
Authority
CN
China
Prior art keywords
morphological
attribute
sequence
differential
feature
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
CN202010850697.7A
Other languages
Chinese (zh)
Other versions
CN112184718A (en
Inventor
李治
傅俏燕
郑琎琎
李晓进
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.)
China Center for Resource Satellite Data and Applications CRESDA
Original Assignee
China Center for Resource Satellite Data and Applications CRESDA
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 China Center for Resource Satellite Data and Applications CRESDA filed Critical China Center for Resource Satellite Data and Applications CRESDA
Priority to CN202010850697.7A priority Critical patent/CN112184718B/en
Publication of CN112184718A publication Critical patent/CN112184718A/en
Application granted granted Critical
Publication of CN112184718B publication Critical patent/CN112184718B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06T7/155—Segmentation; Edge detection involving morphological operators
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20221—Image fusion; Image merging
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Geometry (AREA)
  • Image Processing (AREA)

Abstract

The application discloses a method and a device for automatically extracting high-resolution remote sensing images of urban buildings, wherein the method comprises the following steps: respectively calculating differential morphological structure sequence features and differential morphological attribute sequence features of the remote sensing images according to the original high-resolution remote sensing images; constructing a feature optimization model according to the differential morphological structure sequence features and the differential morphological attribute sequence features, and optimizing the differential morphological structure sequence features and the differential morphological attribute sequence features according to the feature optimization model to obtain optimized morphological structure optimization features and morphology attribute optimization features; and automatically dividing the city building information according to the optimized morphological structure optimized characteristics and the optimized morphological attribute optimized characteristics of the preset multisource prior information to obtain initial information of a plurality of city buildings, and fusing the initial information to obtain final information of the city buildings. The application solves the technical problem of poor quality of the urban building information extracted in the prior art.

Description

Method and device for automatically extracting high-resolution remote sensing images of urban buildings
Technical Field
The application relates to the technical field of urban building high-resolution remote sensing image extraction, in particular to a method and a device for automatically extracting urban building high-resolution remote sensing images.
Background
Urban building information is used as an urban core index element, plays an important role in urban basic information, urban ecological environment research and the like, and is an important link for urban management along with the acceleration of urban development progress and the need of urban intelligent management. With the rapid development of domestic high-resolution remote sensing satellite industry, the remote sensing image can provide large-scale, fine and high-frequency monitoring information for cities, thereby becoming a main data source for urban building monitoring.
At present, the method for extracting the urban building high-resolution remote sensing image mainly comprises two methods, namely extracting the urban building high-resolution remote sensing image according to the high-resolution remote sensing morphological building index of the urban area, namely introducing the morphological structure sequence characteristics of multiple scales and multiple angles, effectively expressing the attribute characteristics of different types of urban buildings from individuals, and further extracting the urban building high-resolution remote sensing image according to the attribute characteristics of the different types of buildings; and extracting the urban building high-resolution remote sensing image according to the high-resolution remote sensing morphological attribute sequence characteristics, namely effectively expressing the regional characteristics of the urban building on the whole according to the attribute rules of the regional buildings, and extracting the urban building high-resolution remote sensing image according to the regional characteristics of the urban building. However, whether the urban building high-resolution remote sensing image is extracted based on the morphological structure sequence features of multiple scales and multiple angles or the urban building high-resolution remote sensing image is extracted based on the high-resolution remote sensing morphological attribute sequence features, the urban building high-resolution remote sensing image is extracted based on the morphological feature method of a single mode, and the building attribute is difficult to fully express by the morphological feature method of the single mode due to the complexity of the building in the urban scene, so that the quality of the extracted urban building information is affected.
Disclosure of Invention
The application solves the technical problems that: aiming at the problem of poor quality of urban building information extracted in the prior art, the embodiment of the application provides a method and a device for automatically extracting urban building high-resolution remote sensing images.
In a first aspect, an embodiment of the present application provides a method for automatically extracting high-resolution remote sensing images of urban buildings, where the method includes:
respectively calculating differential morphological structure sequence features and differential morphological attribute sequence features of the remote sensing images according to the original high-resolution remote sensing images;
constructing a feature optimization model according to the differential morphological structure sequence feature and the differential morphological attribute sequence feature, and optimizing the differential morphological structure sequence feature and the differential morphological attribute sequence feature according to the feature optimization model to obtain an optimized morphological structure optimization feature and an optimized morphological attribute optimization feature;
And automatically dividing the city building information according to the optimized morphological structure optimized characteristics and the optimized morphological attribute optimized characteristics according to the preset multi-source prior information to obtain initial information of a plurality of city buildings, and fusing the initial information to obtain final information of the city buildings.
Optionally, calculating the differential morphological structure sequence feature and the differential morphological attribute sequence feature of the remote sensing image according to the original high-resolution remote sensing image respectively includes:
Respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original high-resolution remote sensing image according to a preset structural element to obtain a morphological open reconstruction sequence and a morphological closed reconstruction sequence, and calculating according to the morphological open reconstruction sequence and the morphological closed reconstruction sequence to obtain the differential morphological structure sequence characteristics;
And respectively carrying out attribute open operation and attribute close operation on the original high-resolution remote sensing image according to a preset scale set and a preset area attribute to obtain differential morphology attribute open sequence features and differential morphology attribute close sequence features taking the area of the image spot as a reference, and calculating to obtain the differential morphology attribute sequence features according to the differential morphology attribute open sequence features and the differential morphology attribute close sequence features.
Optionally, calculating the differential morphological structure sequence feature according to the morphological open reconstruction sequence and the morphological closed reconstruction sequence includes:
the fractal morphology structure sequence characteristics are obtained through calculation according to the following formula:
DMP(x)＝(ΔγR(f)∪ΔφR(f))(x)
Wherein DMP (x) represents the differential morphological structural sequence feature; Δ γR (f) represents the differential morphological structure open reconstruction sequence characteristics of the original high-resolution remote sensing image; Δ φR (f) represents the differential morphological structure closed reconstruction sequence characteristic of the original high-resolution remote sensing image; Representing a differential characteristic image result obtained by carrying out morphological open reconstruction on the original image by using a structural element lambda when the bit number i of the sequence is; /(I) Performing morphological closed reconstruction on the original image by using a structural element lambda when the bit number i of the sequence is represented to obtain a differential characteristic image result; /(I)And/>Respectively representing the characteristic image results of respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda when the bit number i of the sequence is; ii γR (f) and ii φR (f) represent a morphological open reconstruction sequence and a morphological closed reconstruction sequence, respectively; /(I)And/>When the bit number i of the sequence is expressed, respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda; f represents input high-resolution remote sensing images; x represents the pixel value of the input high-resolution remote sensing image; λ represents a morphological structure element; i represents the number of bits of the sequence; i represents the value in the number of sequence bits I.
Optionally, calculating the differential morphology attribute sequence feature according to the differential morphology attribute open sequence feature and the differential morphology attribute closed sequence feature includes:
the differential morphological attribute sequence features are calculated by the following formula:
wherein DAP (x) represents the differential morphological attribute sequence feature; Representing the differential morphological attribute open-sequence features; /(I) Representing the differential morphological attribute closed sequence feature; /(I)And/>The method comprises the steps of representing that morphological attribute open operation and morphological attribute close operation are carried out on the original high-resolution remote sensing image in a preset scale set by area attributes; a represents a preset area attribute; μ represents a set of scales of preset morphological properties; th k (f) represents a binary image of the image f with the gray level k as a threshold; k represents the gray level of the image f; Γ m () and Φ n () represent the feature map of m and n in taking Th k (f) binary image, respectively; /(I)And/>The scale set μ is represented by an area attribute a attribute on operation and an attribute off operation, respectively.
Optionally, the optimizing the differential morphological structure sequence feature and the differential morphological attribute sequence feature according to the feature optimizing model to obtain a preferred morphological structure optimizing feature and a preferred morphological attribute optimizing feature, which includes:
and optimizing the differential morphological structure sequence characteristic and the differential morphological attribute sequence characteristic through the following formula to obtain an optimized morphological structure optimized characteristic and an optimized morphological attribute optimized characteristic:
Wherein DMP opt (x) represents the preferred morphological structure preference feature; DAP opt (x) represents the preferred morphological attribute preference feature; A maximum value representing the differential morphological structure open reconstruction feature; A maximum value representing a characteristic of the closed reconstruction sequence of the differential morphological structure; /(I) A maximum value representing an open-sequence feature of the differential morphology attribute; /(I)And the maximum value of the differential morphological property closed sequence characteristic.
Optionally, according to preset multi-source prior information, automatically dividing the city building information according to the preferable morphological structure preferable characteristics and the preferable morphological attribute preferable characteristics to obtain initial information of a plurality of city buildings, including:
Determining first mask data corresponding to the preferred morphology structure preferred feature and second mask data corresponding to the preferred morphology structure preferred feature according to the multi-source prior information and the preferred morphology structure preferred feature and morphology attribute preferred feature;
Threshold data is obtained through calculation according to the multi-source prior information, the first mask data and the second mask data, and initial information is obtained through calculation according to the optimized morphological structure optimized feature and the morphological attribute optimized feature, the first mask data, the second mask data and the threshold data.
Optionally, determining the first mask data corresponding to the preferred morphology structure preferred feature and the second mask data corresponding to the preferred morphology property preferred feature according to the multi-source prior information and the preferred morphology structure preferred feature and morphology property preferred feature includes:
the first mask data and the second mask data are obtained through calculation according to the following formula:
Wherein Mask (DMP opt (x)) represents the first Mask data; mask (DAP opt (x)) represents second Mask data; globeLand30, 30 water represents a water body type attribute value of a GlobeLand information product in the multi-source prior information; globeLand30 to bare represents the bare land type attribute value of GlobeLand information product in the multi-source prior information; the OSM represents the road type attribute value of Openstreetmap information products in the multi-source prior information; hansen represents forest type attribute values of Hansen information products in the multisource prior information;
calculating threshold data according to the multi-source prior information, the first mask data and the second mask data, wherein the threshold data comprises:
the threshold data is calculated by the following formula:
Wherein T 1、T2、T3 and T 4 each represent the threshold data; area t () represents the Area at which the preferred feature takes the pixel value t; t 1、t2、t3 and t 4 denote pixel values; area GUF (BU) represents the building Area value of the GUF information product in the multisource prior information; area GHSL (BU) represents the building Area value of the GHSL information product in the multi-source a priori information.
Optionally, calculating the initial information according to the preferred morphological structure preferred feature and the preferred morphological attribute preferred feature, the first mask data, the second mask data and the threshold data includes:
the initial information of a plurality of city buildings is obtained by the following formula:
Wherein, And/>Representing the initial information.
Optionally, fusing the initial information to obtain final information of the city building, including:
the final information of the city building is calculated by the following formula:
Wherein O BU represents the final information of the city building; k represents a preset threshold.
In a second aspect, an embodiment of the present application provides an apparatus for automatically extracting high-resolution remote sensing images of urban buildings, where the apparatus includes:
the computing unit is used for respectively computing differential morphological structure sequence features and differential morphological attribute sequence features of the remote sensing image according to the original high-resolution remote sensing image;
The optimizing unit is used for constructing a feature optimizing model according to the differential morphological structure sequence feature and the differential morphological attribute sequence feature, and optimizing the differential morphological structure sequence feature and the differential morphological attribute sequence feature according to the feature optimizing model to obtain an optimized morphological structure optimizing feature and an optimized morphological attribute optimizing feature;
And the fusion unit is used for automatically segmenting the city building information according to the optimized morphological structure optimized characteristics and the optimized morphological attribute optimized characteristics according to the preset multi-source prior information to obtain initial information of a plurality of city buildings, and fusing the initial information to obtain final information of the city buildings.
Optionally, the computing unit is specifically configured to:
Respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original high-resolution remote sensing image according to a preset structural element to obtain a morphological open reconstruction sequence and a morphological closed reconstruction sequence, and calculating according to the morphological open reconstruction sequence and the morphological closed reconstruction sequence to obtain the differential morphological structure sequence characteristics;
And respectively carrying out attribute open operation and attribute close operation on the original high-resolution remote sensing image according to a preset scale set and a preset area attribute to obtain differential morphology attribute open sequence features and differential morphology attribute close sequence features taking the area of the image spot as a reference, and calculating to obtain the differential morphology attribute sequence features according to the differential morphology attribute open sequence features and the differential morphology attribute close sequence features.
Optionally, the computing unit is specifically configured to:
the fractal morphology structure sequence characteristics are obtained through calculation according to the following formula:
DMP(x)＝(ΔγR(f)∪ΔφR(f))(x)
Wherein DMP (x) represents the differential morphological structural sequence feature; Δ γR (f) represents the differential morphological structure open reconstruction sequence characteristics of the original high-resolution remote sensing image; Δ φR (f) represents the differential morphological structure closed reconstruction sequence characteristic of the original high-resolution remote sensing image; Representing a differential characteristic image result obtained by carrying out morphological open reconstruction on the original image by using a structural element lambda when the bit number i of the sequence is; /(I) Performing morphological closed reconstruction on the original image by using a structural element lambda when the bit number i of the sequence is represented to obtain a differential characteristic image result; /(I)And/>Respectively representing the characteristic image results of respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda when the bit number i of the sequence is; pi γR (f) and pi φR (f) represent a morphological open reconstruction sequence and a morphological closed reconstruction sequence, respectively; /(I)And/>When the bit number i of the sequence is expressed, respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda; f represents input high-resolution remote sensing images; x represents the pixel value of the input high-resolution remote sensing image; λ represents a morphological structure element; i represents the number of bits of the sequence; i represents the value in the number of sequence bits I.
Optionally, the computing unit is specifically configured to:
the differential morphological attribute sequence features are calculated by the following formula:
wherein DAP (x) represents the differential morphological attribute sequence feature; Representing the differential morphological attribute open-sequence features; /(I) Representing the differential morphological attribute closed sequence feature; /(I)And/>The method comprises the steps of representing that morphological attribute open operation and morphological attribute close operation are carried out on the original high-resolution remote sensing image in a preset scale set by area attributes; a represents a preset area attribute; μ represents a set of scales of preset morphological properties; th k (f) represents a binary image of the image f with the gray level k as a threshold; k represents the gray level of the image f; Γ m () and Φ n () represent the feature map of m and n in taking Th k (f) binary image, respectively; /(I)And/>The scale set μ is represented by an area attribute a attribute on operation and an attribute off operation, respectively.
Optionally, the preferred unit is specifically configured to:
and optimizing the differential morphological structure sequence characteristic and the differential morphological attribute sequence characteristic through the following formula to obtain an optimized morphological structure optimized characteristic and an optimized morphological attribute optimized characteristic:
Wherein DMP opt (x) represents the preferred morphological structure preference feature; DAP opt (x) represents the preferred morphological attribute preference feature; A maximum value representing the differential morphological structure open reconstruction feature; A maximum value representing a characteristic of the closed reconstruction sequence of the differential morphological structure; /(I) A maximum value representing an open-sequence feature of the differential morphology attribute; /(I)And the maximum value of the differential morphological property closed sequence characteristic.
Optionally, the fusion unit is specifically configured to:
Determining first mask data corresponding to the preferred morphology structure preferred feature and second mask data corresponding to the preferred morphology structure preferred feature according to the multi-source prior information and the preferred morphology structure preferred feature and morphology attribute preferred feature;
Threshold data is obtained through calculation according to the multi-source prior information, the first mask data and the second mask data, and initial information is obtained through calculation according to the optimized morphological structure optimized feature and the morphological attribute optimized feature, the first mask data, the second mask data and the threshold data.
Optionally, the fusion unit is specifically configured to:
the first mask data and the second mask data are obtained through calculation according to the following formula:
Wherein Mask (DMP opt (x)) represents the first Mask data; mask (DAP opt (x)) represents second Mask data; globeLand30, 30 water represents a water body type attribute value of a GlobeLand information product in the multi-source prior information; globeLand30 to bare represents the bare land type attribute value of GlobeLand information product in the multi-source prior information; the OSM represents the road type attribute value of Openstreetmap information products in the multi-source prior information; hansen represents forest type attribute values of Hansen information products in the multisource prior information;
the threshold data is calculated by the following formula:
Wherein T 1、T2、T3 and T 4 each represent the threshold data; area t () represents the Area at which the preferred feature takes the pixel value t; t 1、t2、t3 and t 4 denote pixel values; area GUF (BU) represents the building Area value of the GUF information product in the multisource prior information; area GHSL (BU) represents the building Area value of the GHSL information product in the multi-source a priori information.
Optionally, the fusion unit is specifically configured to:
the initial information of a plurality of city buildings is obtained by the following formula:
Wherein, And/>Representing the initial information.
Optionally, the fusion unit is specifically configured to:
the final information of the city building is calculated by the following formula:
Wherein O BU represents the final information of the city building; k represents a preset threshold.
Compared with the prior art, the scheme provided by the embodiment of the application has the following advantages:
1. In the scheme provided by the embodiment of the application, the urban building characteristics of high-resolution remote sensing are expressed by utilizing a mode of combining the morphological structure sequence characteristics and the morphological attribute sequence characteristics, so that the comprehensive expression capacity of various urban building forms is enhanced, the problem that the building attributes are difficult to fully express by a single-mode morphological characteristic method is avoided, and the quality of the extracted urban building information is further improved.
2. In the scheme provided by the embodiment of the application, the urban building information is automatically segmented according to the optimized morphological structure optimized characteristics and the optimized morphological attribute optimized characteristics after optimized multi-source prior information to obtain the initial information of a plurality of urban buildings, and the initial information is fused to obtain the final information of the urban buildings, so that the degree of automatic extraction of the urban building information is improved.
3. In the scheme provided by the embodiment of the application, the morphological structure sequence features and the morphological attribute sequence features are subjected to feature optimization through the feature optimization model, so that the problem of reduced extraction precision caused by redundancy of feature information is avoided, and the extraction precision of urban building high-resolution remote sensing images is further improved.
4. According to the scheme provided by the embodiment of the application, the urban building information is automatically segmented according to the optimized morphological structure optimized characteristic and the optimized morphological attribute optimized characteristic by introducing the multi-source prior information, so that the problems of errors and low efficiency caused by segmentation through a manual threshold are solved, and the stability and the high efficiency of the urban building segmentation result are further ensured.
5. In the scheme provided by the embodiment of the application, the final information of the final urban building is obtained by fusing the extracted initial information, so that the robustness and generalization capability of the result are improved, and the automatic extraction of the building in the urban scene is finally realized.
Drawings
Fig. 1 is a schematic flow chart of a method for automatically extracting high-resolution remote sensing images of urban buildings according to an embodiment of the application;
FIG. 2 is a schematic diagram of an original high-resolution remote sensing image according to an embodiment of the present application;
FIG. 3 is a diagram of a differential morphological structure sequence according to an embodiment of the present application;
FIG. 4 is a sequence of feature diagrams of differential morphology according to an embodiment of the present application;
FIG. 5a is a diagram showing a preferred morphological structure sequence feature according to an embodiment of the present application;
FIG. 5b is a diagram of a preferred morphological attribute sequence feature according to an embodiment of the present application;
FIG. 6a is a schematic diagram of initial information of a city building according to an embodiment of the present application;
FIG. 6b is a schematic diagram of initial information of a city building according to an embodiment of the present application;
FIG. 6c is a schematic diagram of initial information of a city building according to an embodiment of the present application;
FIG. 6d is a schematic diagram of initial information of a city building according to an embodiment of the present application;
FIG. 7 is a schematic diagram of final information of a city building according to an embodiment of the present application;
Fig. 8 is a schematic structural diagram of an apparatus for automatically extracting high-resolution remote sensing images of urban buildings according to an embodiment of the present application.
Detailed Description
In the solutions provided by the embodiments of the present application, the described embodiments are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The method for automatically extracting the high-resolution remote sensing image of the urban building provided by the embodiment of the application is further described in detail below by combining with the attached drawings of the specification, and the specific implementation mode of the method can comprise the following steps (the method flow is shown as a figure 1):
And step 101, respectively calculating differential morphological structure sequence features and differential morphological attribute sequence features of the remote sensing image according to the original high-resolution remote sensing image.
In the scheme provided by the embodiment of the application, referring to fig. 2, a schematic diagram of an original high-resolution remote sensing image provided by the embodiment of the application is shown, the original high-resolution remote sensing image includes, but is not limited to, a high-resolution first-order high-spatial resolution remote sensing image of 2016, 4 months and 9 days, wherein multispectral and full-color resolutions of the high-resolution first-order high-spatial resolution remote sensing image are 8 meters and 2 meters respectively, four wave bands are included, namely a blue wave band (0.45-0.52 μm), a green wave band (0.52-0.59 μm), a red wave band (0.63-0.69 μm), a near infrared wave band (0.77-0.89 μm), and a radiation quantization level is 16 bits.
Specifically, there are various ways to calculate the differential morphological structure sequence feature and the differential morphological attribute sequence feature of the remote sensing image according to the original high-resolution remote sensing image, and a preferred way is described below as an example.
In one possible implementation manner, the calculating the differential morphological structure sequence feature and the differential morphological attribute sequence feature of the remote sensing image according to the original high-resolution remote sensing image includes: respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original high-resolution remote sensing image according to a preset structural element to obtain a morphological open reconstruction sequence and a morphological closed reconstruction sequence, and calculating according to the morphological open reconstruction sequence and the morphological closed reconstruction sequence to obtain the differential morphological structure sequence characteristics; and respectively carrying out attribute open operation and attribute close operation on the original high-resolution remote sensing image according to a preset scale set and a preset area attribute to obtain differential morphology attribute open sequence features and differential morphology attribute close sequence features taking the area of the image spot as a reference, and calculating to obtain the differential morphology attribute sequence features according to the differential morphology attribute open sequence features and the differential morphology attribute close sequence features.
Further, in one possible implementation manner, the calculating the differential morphological structure sequence feature according to the morphological open reconstruction sequence and the morphological closed reconstruction sequence includes:
the fractal morphology structure sequence characteristics are obtained through calculation according to the following formula:
DMP(x)＝(ΔγR(f)∪ΔφR(f))(x)
Wherein DMP (x) represents the differential morphological structural sequence feature; Δ γR (f) represents the differential morphological structure open reconstruction sequence characteristics of the original high-resolution remote sensing image; Δ φR (f) represents the differential morphological structure closed reconstruction sequence characteristic of the original high-resolution remote sensing image; Representing a differential characteristic image result obtained by carrying out morphological open reconstruction on the original image by using a structural element lambda when the bit number i of the sequence is; /(I) Performing morphological closed reconstruction on the original image by using a structural element lambda when the bit number i of the sequence is represented to obtain a differential characteristic image result; /(I)And/>Respectively representing the characteristic image results of respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda when the bit number i of the sequence is; ii γR (f) and ii φR (f) represent a morphological open reconstruction sequence and a morphological closed reconstruction sequence, respectively; /(I)And/>When the bit number i of the sequence is expressed, respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda; f represents input high-resolution remote sensing images; x represents the pixel value of the input high-resolution remote sensing image; λ represents a morphological structure element; i represents the number of bits of the sequence; i represents the value in the number of sequence bits I.
Further, in one possible implementation manner, the differential morphology attribute sequence feature is obtained by calculating according to the differential morphology attribute open sequence feature and the differential morphology attribute closed sequence feature, and the method includes:
the differential morphological attribute sequence features are calculated by the following formula:
wherein DAP (x) represents the differential morphological attribute sequence feature; Representing the differential morphological attribute open-sequence features; /(I) Representing the differential morphological attribute closed sequence feature; /(I)And/>The method comprises the steps of representing that morphological attribute open operation and morphological attribute close operation are carried out on the original high-resolution remote sensing image in a preset scale set by area attributes; a represents a preset area attribute; μ represents a set of scales of preset morphological properties; th k (f) represents a binary image of the image f with the gray level k as a threshold; k represents the gray level of the image f; Γ m () and Φ n () represent the feature map of m and n in taking Th k (f) binary image, respectively; /(I)And/>The scale set μ is represented by an area attribute a attribute on operation and an attribute off operation, respectively.
Specifically, referring to fig. 3, a differential morphological structure sequence feature map provided by an embodiment of the present application is shown, where the differential morphological structure sequence feature map includes a structure open reconstruction sequence and a structure closed reconstruction sequence; referring to fig. 4, a differential morphological attribute sequence feature map provided by an embodiment of the present application is shown, where the differential morphological attribute sequence feature map includes an attribute open reconstruction sequence and an attribute closed reconstruction sequence.
Step 102, a feature optimization model is constructed according to the differential morphological structure sequence feature and the differential morphological attribute sequence feature, and the differential morphological structure sequence feature and the differential morphological attribute sequence feature are optimized according to the feature optimization model to obtain an optimized morphological structure optimization feature and an optimized morphological attribute optimization feature.
Specifically, in the scheme provided by the embodiment of the application, after the structural sequence feature and the attribute sequence feature are calculated, the differential morphological structural sequence feature and the differential morphological attribute sequence feature construct a feature optimization model, wherein the feature optimization model can be a feature significant level model or other feature selection models, and is not limited herein.
Further, after the optimization model, the differential morphological structure sequence feature and the differential morphological attribute sequence feature are optimized according to the feature optimization model to obtain an optimized morphological structure optimization feature and an optimized morphological attribute optimization feature, and specifically, the preferred morphological structure optimization feature and the optimized morphological attribute optimization feature are obtained by optimizing the structure sequence feature and the attribute sequence feature in various ways, and a preferred way will be described below.
In one possible implementation manner, the optimizing the differential morphological structure sequence feature and the differential morphological attribute sequence feature according to the feature optimization model to obtain a preferred morphological structure preferred feature and a preferred morphological attribute preferred feature includes:
and optimizing the differential morphological structure sequence characteristic and the differential morphological attribute sequence characteristic through the following formula to obtain an optimized morphological structure optimized characteristic and an optimized morphological attribute optimized characteristic:
Wherein DMP opt (x) represents the preferred morphological structure preference feature; DAP opt (x) represents the preferred morphological attribute preference feature; A maximum value representing the differential morphological structure open reconstruction feature; A maximum value representing a characteristic of the closed reconstruction sequence of the differential morphological structure; /(I) A maximum value representing an open-sequence feature of the differential morphology attribute; /(I)And the maximum value of the differential morphological property closed sequence characteristic.
In particular, in the scheme provided by the embodiment of the application,AndCan be calculated by the following formula:
In order to facilitate the visual understanding of the above preferred morphological structure preferred features and morphological property preferred features, preferred morphological structure preferred feature schematics and morphological property preferred feature schematics are given below in the form of drawings. Specifically, referring to fig. 5a and fig. 5b, fig. 5a shows a preferred characteristic diagram of a preferred morphology structure provided by an embodiment of the present application, and fig. 5b shows a preferred characteristic diagram of a preferred morphology attribute provided by an embodiment of the present application.
In the scheme provided by the embodiment of the application, the morphological structure sequence features and the morphological attribute sequence features are subjected to feature optimization through the feature optimization model, so that the problem of reduced extraction precision caused by redundancy of feature information is avoided, and the extraction precision of urban building information is further improved.
And 103, automatically dividing the city building information according to the optimized morphological structure optimized characteristics and the optimized morphological attribute optimized characteristics according to the preset multi-source prior information to obtain initial information of a plurality of city buildings, and fusing the initial information to obtain final information of the city buildings.
Specifically, in the scheme provided by the embodiment of the application, the multi-source prior information includes, but is not limited to: water body type information in the GlobeLand information product in 2010, bare land type information in the GlobeLand information product in 2010, forest type information in the Hansen information product in 2013, road type information in the Openstreetmap information product in 2015, building area information in the GUF information product in 2011 and in the GHSL information product in 2014.
In one possible implementation manner, the automatic segmentation of the city building information is performed according to the preferred morphological structure preferred feature and the preferred morphological attribute feature according to the preset multi-source prior information, so as to obtain initial information of a plurality of city buildings, including: determining first mask data corresponding to the preferred morphology structure preferred feature and second mask data corresponding to the preferred morphology structure preferred feature according to the multi-source prior information and the preferred morphology structure preferred feature and morphology attribute preferred feature; threshold data is obtained through calculation according to the multi-source prior information, the first mask data and the second mask data, and initial information is obtained through calculation according to the optimized morphological structure optimized feature and the morphological attribute optimized feature, the first mask data, the second mask data and the threshold data.
Further, in one possible implementation, determining, according to the multi-source prior information and the preferred morphology structure preferred feature and morphology attribute preferred feature, first mask data corresponding to the preferred morphology structure preferred feature and second mask data corresponding to the preferred morphology attribute preferred feature includes:
the first mask data and the second mask data are obtained through calculation according to the following formula:
Wherein Mask (DMP opt (x)) represents the first Mask data; mask (DAP opt (x)) represents second Mask data; globeLand30, 30 water represents a water body type attribute value of a GlobeLand information product in the multi-source prior information; globeLand30 to bare represents the bare land type attribute value of GlobeLand information product in the multi-source prior information; the OSM represents the road type attribute value of Openstreetmap information products in the multi-source prior information; hansen represents forest type attribute values of Hansen information products in the multisource prior information;
calculating threshold data according to the multi-source prior information, the first mask data and the second mask data, wherein the threshold data comprises:
the threshold data is calculated by the following formula:
/>
Wherein T 1、T2、T3 and T 4 each represent the threshold data; area t () represents the Area at which the preferred feature takes the pixel value t; t 1、t2、t3 and t 4 denote pixel values; area GUF (BU) represents the building Area value of the GUF information product in the multisource prior information; area GHSL (BU) represents the building Area value of the GHSL information product in the multi-source a priori information.
According to the scheme provided by the embodiment of the application, the urban building information is automatically segmented according to the optimized morphological structure optimized characteristic and the optimized morphological attribute optimized characteristic by introducing the multi-source prior information, so that the problems of errors and low efficiency caused by segmentation through a manual threshold are solved, and the stability and the high efficiency of the urban building segmentation result are further ensured.
Further, in one possible implementation manner, calculating initial information of a plurality of city buildings according to the preferred morphological structure preferred feature and the morphological attribute preferred feature, the first mask data, the second mask data and the threshold data includes:
the initial information of a plurality of city buildings is obtained by the following formula:
Wherein, And/>Representing initial information of the city building.
To facilitate visual understanding of the initial informationAnd/>The following is given in the form of a figureAnd/>Is a remote sensing image of (a). See, in particular, fig. 6a, 6b, 6c and 6d.
Further, in order to improve robustness and generalization capability of urban building information extraction, in a possible implementation manner, the fusing the initial information to obtain final information of the urban building includes:
the final information of the city building is calculated by the following formula:
Wherein O BU represents the final information of the city building; k represents a preset threshold.
In order to facilitate visual understanding of the final information of the above-mentioned city building, the final information of the city building is given below in the form of a drawing. In particular, see fig. 7.
In the scheme provided by the embodiment of the application, differential morphological structure sequence features and differential morphological attribute sequence features of a remote sensing image are calculated respectively according to an original high-resolution remote sensing image, a feature optimization model is built according to the differential morphological structure sequence features and the differential morphological attribute sequence features, the differential morphological structure sequence features and the differential morphological attribute sequence features are optimized according to the feature optimization model to obtain optimized morphological structure optimization features and morphology attribute optimization features, then urban building information is automatically segmented according to preset multi-source priori information to obtain initial information of a plurality of urban buildings, and the initial information is fused to obtain final information of the urban buildings. Therefore, in the scheme provided by the embodiment of the application, the urban building characteristics of high-resolution remote sensing are expressed by utilizing a mode of combining morphological structure sequence characteristics and morphological attribute sequence characteristics, the comprehensive expression capacity of various urban building forms is enhanced, the problem that the building attributes are difficult to fully express by a single-mode morphological characteristic method is avoided, and the quality of the extracted urban building information is further improved.
Based on the same inventive concept as the method shown in fig. 1, the embodiment of the application provides a device for automatically extracting high-resolution remote sensing images of urban buildings, see fig. 8, which comprises:
a calculating unit 801, configured to calculate a differential morphological structure sequence feature and a differential morphological attribute sequence feature of the remote sensing image according to the original high-resolution remote sensing image respectively;
A optimizing unit 802, configured to construct a feature optimizing model according to the differential morphological structure sequence feature and the differential morphological attribute sequence feature, and optimize the differential morphological structure sequence feature and the differential morphological attribute sequence feature according to the feature optimizing model to obtain an optimized morphological structure optimizing feature and an optimized morphological attribute optimizing feature;
And the fusion unit 803 is configured to automatically segment the city building information according to the optimized morphology structure optimized feature and morphology attribute optimized feature according to the preset multisource prior information, obtain initial information of a plurality of city buildings, and fuse the initial information to obtain final information of the city buildings.
Optionally, the computing unit 801 is specifically configured to:
Respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original high-resolution remote sensing image according to a preset structural element to obtain a morphological open reconstruction sequence and a morphological closed reconstruction sequence, and calculating according to the morphological open reconstruction sequence and the morphological closed reconstruction sequence to obtain the differential morphological structure sequence characteristics;
And respectively carrying out attribute open operation and attribute close operation on the original high-resolution remote sensing image according to a preset scale set and a preset area attribute to obtain differential morphology attribute open sequence features and differential morphology attribute close sequence features taking the area of the image spot as a reference, and calculating to obtain the differential morphology attribute sequence features according to the differential morphology attribute open sequence features and the differential morphology attribute close sequence features.
Optionally, the computing unit 801 is specifically configured to:
the fractal morphology structure sequence characteristics are obtained through calculation according to the following formula:
DMP(x)＝(ΔγR(f)∪ΔφR(f))(x)
Wherein DMP (x) represents the differential morphological structural sequence feature; Δ γR (f) represents the differential morphological structure open reconstruction sequence characteristics of the original high-resolution remote sensing image; Δ φR (f) represents the differential morphological structure closed reconstruction sequence characteristic of the original high-resolution remote sensing image; Representing a differential characteristic image result obtained by carrying out morphological open reconstruction on the original image by using a structural element lambda when the bit number i of the sequence is; /(I) Performing morphological closed reconstruction on the original image by using a structural element lambda when the bit number i of the sequence is represented to obtain a differential characteristic image result; /(I)And/>Respectively representing the characteristic image results of respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda when the bit number i of the sequence is; pi γR (f) and pi φR (f) represent a morphological open reconstruction sequence and a morphological closed reconstruction sequence, respectively; /(I)And/>When the bit number i of the sequence is expressed, respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda; f represents input high-resolution remote sensing images; x represents the pixel value of the input high-resolution remote sensing image; λ represents a morphological structure element; i represents the number of bits of the sequence; i represents the value in the number of sequence bits I.
Optionally, the computing unit 801 is specifically configured to:
the differential morphological attribute sequence features are calculated by the following formula:
wherein DAP (x) represents the differential morphological attribute sequence feature; Representing the differential morphological attribute open-sequence features; /(I) Representing the differential morphological attribute closed sequence feature; /(I)And/>The method comprises the steps of representing that morphological attribute open operation and morphological attribute close operation are carried out on the original high-resolution remote sensing image in a preset scale set by area attributes; a represents a preset area attribute; μ represents a set of scales of preset morphological properties; th k (f) represents a binary image of the image f with the gray level k as a threshold; k represents the gray level of the image f; Γ m () and Φ n () represent the feature map of m and n in taking Th k (f) binary image, respectively; /(I)And/>The scale set μ is represented by an area attribute a attribute on operation and an attribute off operation, respectively.
Optionally, the preferred unit 802 is specifically configured to:
Wherein DMP opt (x) represents the preferred morphological structure preference feature; DAP opt (x) represents the preferred morphological attribute preference feature; A maximum value representing the differential morphological structure open reconstruction feature; A maximum value representing a characteristic of the closed reconstruction sequence of the differential morphological structure; /(I) A maximum value representing an open-sequence feature of the differential morphology attribute; /(I)And the maximum value of the differential morphological property closed sequence characteristic.
Optionally, the fusion unit 803 is specifically configured to:
Determining first mask data corresponding to the preferred morphology structure preferred feature and second mask data corresponding to the preferred morphology structure preferred feature according to the multi-source prior information and the preferred morphology structure preferred feature and morphology attribute preferred feature;
Threshold data is obtained through calculation according to the multi-source prior information, the first mask data and the second mask data, and initial information is obtained through calculation according to the optimized morphological structure optimized feature and the morphological attribute optimized feature, the first mask data, the second mask data and the threshold data.
Optionally, the fusion unit 803 is specifically configured to:
the first mask data and the second mask data are obtained through calculation according to the following formula:
Wherein Mask (DMP opt (x)) represents the first Mask data; mask (DAP opt (x)) represents second Mask data; globeLand30, 30 water represents a water body type attribute value of a GlobeLand information product in the multi-source prior information; globeLand30 to bare represents the bare land type attribute value of GlobeLand information product in the multi-source prior information; the OSM represents the road type attribute value of Openstreetmap information products in the multi-source prior information; hansen represents forest type attribute values of Hansen information products in the multisource prior information;
the threshold data is calculated by the following formula:
/>
Wherein T 1、T2、T3 and T 4 each represent the threshold data; area t () represents the Area at which the preferred feature takes the pixel value t; t 1、t2、t3 and t 4 denote pixel values; area GUF (BU) represents the building Area value of the GUF information product in the multisource prior information; area GHSL (BU) represents the building Area value of the GHSL information product in the multi-source a priori information.
Optionally, the fusion unit 803 is specifically configured to:
the initial information of a plurality of city buildings is obtained by the following formula:
Wherein, And/>Representing the initial information.
Optionally, the fusion unit 803 is specifically configured to:
the final information of the city building is calculated by the following formula:
Wherein O BU represents the final information of the city building; k represents a preset threshold.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, magnetic disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.

Claims (5)

1. The method for automatically extracting the high-resolution remote sensing image of the urban building is characterized by comprising the following steps of:
Respectively calculating differential morphological structure sequence features and differential morphological attribute sequence features of the remote sensing images according to the original high-resolution remote sensing images; comprising the following steps: respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original high-resolution remote sensing image according to a preset structural element to obtain a morphological open reconstruction sequence and a morphological closed reconstruction sequence, and calculating according to the morphological open reconstruction sequence and the morphological closed reconstruction sequence to obtain the differential morphological structure sequence characteristics; performing attribute open operation and attribute close operation on the original high-resolution remote sensing image according to a preset scale set and a preset area attribute to obtain differential morphology attribute open sequence features and differential morphology attribute close sequence features taking the area of a pattern spot as a reference, and calculating to obtain the differential morphology attribute sequence features according to the differential morphology attribute open sequence features and the differential morphology attribute close sequence features;
Constructing a feature optimization model according to the differential morphological structure sequence feature and the differential morphological attribute sequence feature, and optimizing the differential morphological structure sequence feature and the differential morphological attribute sequence feature according to the feature optimization model to obtain an optimized morphological structure optimization feature and an optimized morphological attribute optimization feature; the feature optimization model is as follows:
Wherein DMP opt (x) represents the preferred morphological structure preference feature; DAP opt (x) represents the preferred morphological attribute preference feature; a maximum value representing a reconstruction characteristic of the differential morphological structure; /(I) A maximum value representing a characteristic of the closed reconstruction sequence of the differential morphological structure; /(I)Maximum value representing open sequence characteristics of differential morphology attribute; /(I)Maximum value representing the closed sequence characteristic of the differential morphology attribute;
Automatically dividing the city building information according to the optimized morphological structure optimized characteristics and the optimized morphological attribute optimized characteristics according to preset multi-source prior information to obtain initial information of a plurality of city buildings, and fusing the initial information to obtain final information of the city buildings;
The method for automatically dividing the city building information according to the optimized morphological structure optimized feature and the optimized morphological attribute optimized feature according to the preset multi-source prior information comprises the following steps:
Determining first mask data corresponding to the preferred morphology structure preferred feature and second mask data corresponding to the preferred morphology structure preferred feature according to the multi-source prior information and the preferred morphology structure preferred feature and morphology attribute preferred feature;
the first mask data and the second mask data are obtained through calculation according to the following formula:
Wherein Mask (DMP opt (x)) represents the first Mask data; mask (DAP opt (x)) represents second Mask data; globeLand30, 30 water represents a water body type attribute value of a GlobeLand information product in the multi-source prior information; globeLand30 to bare represents the bare land type attribute value of GlobeLand information product in the multi-source prior information; the OSM represents the road type attribute value of Openstreetmap information products in the multi-source prior information; hansen represents forest type attribute values of Hansen information products in the multisource prior information;
Calculating to obtain threshold value data according to the multi-source prior information, the first mask data and the second mask data, and calculating to obtain initial information of the urban building according to the optimized morphological structure optimized characteristic and the optimized morphological attribute optimized characteristic, the first mask data, the second mask data and the threshold value data;
the threshold data is calculated by the following formula:
Wherein T 1、T2、T3 and T 4 each represent the threshold data; area t () represents the Area at which the preferred feature takes the pixel value t; t 1、t2、t3 and t 4 denote pixel values; area GUF (BU) represents the building Area value of the GUF information product in the multisource prior information; area GHSL (BU) represents the building Area value of GHSL information product in the multisource prior information
The initial information of a plurality of city buildings is obtained by the following formula:
Wherein, And/>Representing initial information of the city building.
2. The method of claim 1, wherein computing the differential morphological structural sequence features from the morphological open reconstruction sequence and the morphological closed reconstruction sequence comprises:
The differential morphological structure sequence features are calculated by the following formula:
DMP(x)＝(ΔγR(f)∪ΔφR(f))(x)
Wherein DMP (x) represents the differential morphological structural sequence feature; Δ γR (f) represents the differential morphological structure open reconstruction sequence characteristics of the original high-resolution remote sensing image; Δ φR (f) represents the differential morphological structure closed reconstruction sequence characteristic of the original high-resolution remote sensing image; Representing a differential characteristic image result obtained by carrying out morphological open reconstruction on the original high-resolution remote sensing image by using a structural element lambda when the bit number i of the sequence is set; /(I) Performing morphological closed reconstruction on the original image by using a structural element lambda when the bit number i of the sequence is represented to obtain a differential characteristic image result; /(I)And/>Respectively representing the characteristic image results of respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda when the bit number i of the sequence is; ii γR (f) and ii φR (f) represent a morphological open reconstruction sequence and a morphological closed reconstruction sequence, respectively; /(I)And/>When the bit number i of the sequence is expressed, respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original image by using a structural body element lambda; f represents input high-resolution remote sensing images; x represents the pixel value of the input high-resolution remote sensing image; λ represents a morphological structure element; i represents the number of bits of the sequence; i represents the value in the number of sequence bits I.
3. The method of claim 2, wherein computing the differential morphology attribute sequence feature from the differential morphology attribute open sequence feature and the differential morphology attribute closed sequence feature comprises:
the differential morphological attribute sequence features are calculated by the following formula:
wherein DAP (x) represents the differential morphological attribute sequence feature; Representing the differential morphological attribute open-sequence features; /(I) Representing the differential morphological attribute closed sequence feature; /(I)And/>The method comprises the steps of representing that morphological attribute open operation and morphological attribute close operation are carried out on the original high-resolution remote sensing image in a preset scale set by area attributes; a represents a preset area attribute; μ represents a set of scales of preset morphological properties; th k (f) represents a binary image of the image f with the gray level k as a threshold; k represents the gray level of the image f; Γ m () and Φ n () represent the feature map of m and n in taking Th k (f) binary image, respectively; /(I)And/>The scale set μ is represented by an area attribute a attribute on operation and an attribute off operation, respectively.
4. The method of claim 1, wherein fusing the initial information of the city building to obtain the final information of the city building comprises:
the final information of the city building is calculated by the following formula:
Wherein O BU represents the final information of the city building; k represents a preset threshold.
5. The utility model provides a device that city building high-resolution remote sensing image draws automatically which characterized in that includes:
The computing unit is used for respectively computing differential morphological structure sequence features and differential morphological attribute sequence features of the remote sensing image according to the original high-resolution remote sensing image; comprising the following steps: respectively carrying out morphological open reconstruction and morphological closed reconstruction on the original high-resolution remote sensing image according to a preset structural element to obtain a morphological open reconstruction sequence and a morphological closed reconstruction sequence, and calculating according to the morphological open reconstruction sequence and the morphological closed reconstruction sequence to obtain the differential morphological structure sequence characteristics; performing attribute open operation and attribute close operation on the original high-resolution remote sensing image according to a preset scale set and a preset area attribute to obtain differential morphology attribute open sequence features and differential morphology attribute close sequence features taking the area of a pattern spot as a reference, and calculating to obtain the differential morphology attribute sequence features according to the differential morphology attribute open sequence features and the differential morphology attribute close sequence features;
The optimizing unit is used for constructing a feature optimizing model according to the differential morphological structure sequence feature and the differential morphological attribute sequence feature, and optimizing the differential morphological structure sequence feature and the differential morphological attribute sequence feature according to the feature optimizing model to obtain an optimized morphological structure optimizing feature and an optimized morphological attribute optimizing feature; the feature optimization model is as follows:
Wherein DMP opt (x) represents the preferred morphological structure preference feature; DAP opt (x) represents the preferred morphological attribute preference feature; a maximum value representing a reconstruction characteristic of the differential morphological structure; /(I) A maximum value representing a characteristic of the closed reconstruction sequence of the differential morphological structure; /(I)Maximum value representing open sequence characteristics of differential morphology attribute; /(I)Maximum value representing the closed sequence characteristic of the differential morphology attribute;
the fusion unit is used for automatically segmenting the city building information according to the optimized morphological structure optimized characteristics and the optimized morphological attribute optimized characteristics after the optimized multi-source prior information to obtain initial information of a plurality of city buildings, and fusing the initial information to obtain final information of the city buildings;
The method for automatically dividing the city building information according to the optimized morphological structure optimized feature and the optimized morphological attribute optimized feature according to the preset multi-source prior information comprises the following steps:
Determining first mask data corresponding to the preferred morphology structure preferred feature and second mask data corresponding to the preferred morphology structure preferred feature according to the multi-source prior information and the preferred morphology structure preferred feature and morphology attribute preferred feature;
the first mask data and the second mask data are obtained through calculation according to the following formula:
Wherein Mask (DMP opt (x)) represents the first Mask data; mask (DAP opt (x)) represents second Mask data; globeLand30, 30 water represents a water body type attribute value of a GlobeLand information product in the multi-source prior information; globeLand30 to bare represents the bare land type attribute value of GlobeLand information product in the multi-source prior information; the OSM represents the road type attribute value of Openstreetmap information products in the multi-source prior information; hansen represents forest type attribute values of Hansen information products in the multisource prior information;
Calculating to obtain threshold value data according to the multi-source prior information, the first mask data and the second mask data, and calculating to obtain initial information of the urban building according to the optimized morphological structure optimized characteristic and the optimized morphological attribute optimized characteristic, the first mask data, the second mask data and the threshold value data;
the threshold data is calculated by the following formula:
Wherein T 1、T2、T3 and T 4 each represent the threshold data; area t () represents the Area at which the preferred feature takes the pixel value t; t 1、t2、t3 and t 4 denote pixel values; area GUF (BU) represents the building Area value of the GUF information product in the multisource prior information; area GHSL (BU) represents the building Area value of GHSL information product in the multisource prior information
The initial information of a plurality of city buildings is obtained by the following formula:
Wherein, And/>Representing initial information of the city building.
CN202010850697.7A 2020-08-21 2020-08-21 Method and device for automatically extracting high-resolution remote sensing images of urban buildings Active CN112184718B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010850697.7A CN112184718B (en) 2020-08-21 2020-08-21 Method and device for automatically extracting high-resolution remote sensing images of urban buildings

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010850697.7A CN112184718B (en) 2020-08-21 2020-08-21 Method and device for automatically extracting high-resolution remote sensing images of urban buildings

Publications (2)

Publication Number Publication Date
CN112184718A CN112184718A (en) 2021-01-05
CN112184718B true CN112184718B (en) 2024-05-21

Family

ID=73924250

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010850697.7A Active CN112184718B (en) 2020-08-21 2020-08-21 Method and device for automatically extracting high-resolution remote sensing images of urban buildings

Country Status (1)

Country Link
CN (1) CN112184718B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103839267A (en) * 2014-02-27 2014-06-04 西安科技大学 Building extracting method based on morphological building indexes
KR20190080009A (en) * 2017-12-28 2019-07-08 한국국토정보공사 Automatic drawing method using lane information
CN110569751A (en) * 2019-08-23 2019-12-13 南京信息工程大学 High-resolution remote sensing image building extraction method
CN111339948A (en) * 2020-02-25 2020-06-26 武汉大学 Automatic identification method for newly-added buildings of high-resolution remote sensing images
CN111539965A (en) * 2020-04-10 2020-08-14 中国资源卫星应用中心 Urban surface water domestic high-resolution remote sensing self-adaptive extraction method cooperating with multi-source information

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8712142B2 (en) * 2011-03-14 2014-04-29 The University Of Warwick Method and apparatus for analysis of histopathology images and its application to cancer diagnosis and grading

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103839267A (en) * 2014-02-27 2014-06-04 西安科技大学 Building extracting method based on morphological building indexes
KR20190080009A (en) * 2017-12-28 2019-07-08 한국국토정보공사 Automatic drawing method using lane information
CN110569751A (en) * 2019-08-23 2019-12-13 南京信息工程大学 High-resolution remote sensing image building extraction method
CN111339948A (en) * 2020-02-25 2020-06-26 武汉大学 Automatic identification method for newly-added buildings of high-resolution remote sensing images
CN111539965A (en) * 2020-04-10 2020-08-14 中国资源卫星应用中心 Urban surface water domestic high-resolution remote sensing self-adaptive extraction method cooperating with multi-source information

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
An Automatic Morphological Attribute Building Extraction Approach for Satellite High Spatial Resolution Imagery;Weixuan Ma et al;《remote sensing》;第1-24页 *
Morphological Profiles Based on Differently Shaped Structuring Elements for Classification of Images With Very High Spatial Resolution;Zhi Yong Lv et al;《IEEE》;第4644-4653页 *
一种面向变化检测的建筑物指数;王慧贤 等;《测绘通报》;第47-50页 *
基于形态学序列和多源先验信息的城市建筑物 高分遥感提取;李治 等;《遥感学报》;第998-1006页 *
结合NASFs滤波策略与形态学属性剖面的SAR影像建筑物检测;胡世明 等;《地理与地理信息科学》;第34卷(第6期);第28-32页 *

Also Published As

Publication number Publication date
CN112184718A (en) 2021-01-05

Similar Documents

Publication Publication Date Title
Wu et al. Generative Adversarial Networks in the built environment: A comprehensive review of the application of GANs across data types and scales
CN112084923B (en) Remote sensing image semantic segmentation method, storage medium and computing device
CN111553963B (en) Meta-grid generation method and device based on geographic information
CN109635748B (en) Method for extracting road characteristics in high-resolution image
CN113343775B (en) Deep learning generalization method for remote sensing image ground surface coverage classification
CN109993064A (en) A kind of method of connection path in extraction picture between road network node
Stanislawski et al. Generalisation operators
CN113822832A (en) Natural resource multi-source vector data fusion method
CN113724279A (en) System, method, equipment and storage medium for automatically dividing traffic cells into road networks
Li et al. Exploring multiple crowdsourced data to learn deep convolutional neural networks for road extraction
Tran et al. Pp-linknet: Improving semantic segmentation of high resolution satellite imagery with multi-stage training
Dai et al. Multiscale residual convolution neural network and sector descriptor-based road detection method
CN112927237A (en) Honeycomb lung focus segmentation method based on improved SCB-Unet network
CN115937461A (en) Multi-source fusion model construction and texture generation method, device, medium and equipment
CN113139511B (en) Water and soil conservation production and construction project disturbance pattern spot extraction method, medium and equipment
CN101887522A (en) Method for extracting remote sensing image road based on multipoint simultaneous and rapid advance
CN114373073A (en) Method and system for road scene semantic segmentation
CN112184718B (en) Method and device for automatically extracting high-resolution remote sensing images of urban buildings
CN117727046A (en) Novel mountain torrent front-end instrument and meter reading automatic identification method and system
CN115984603A (en) Fine classification method and system for urban green land based on GF-2 and open map data
CN115773744A (en) Model training and road network processing method, device, equipment, medium and product
Mao et al. City object detection from airborne Lidar data with OpenStreetMap‐tagged superpixels
Qiao et al. Revolutionizing building damage detection: A novel weakly supervised approach using high-resolution remote sensing images
Qin et al. UrbanEvolver: function-aware urban layout regeneration
Simpson et al. An open-source automatic survey of green roofs in London using segmentation of aerial imagery

Legal Events

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