CN111311596A - Remote sensing image change detection method based on improved LBP (local binary pattern) characteristics - Google Patents

Remote sensing image change detection method based on improved LBP (local binary pattern) characteristics Download PDF

Info

Publication number
CN111311596A
CN111311596A CN202010209215.XA CN202010209215A CN111311596A CN 111311596 A CN111311596 A CN 111311596A CN 202010209215 A CN202010209215 A CN 202010209215A CN 111311596 A CN111311596 A CN 111311596A
Authority
CN
China
Prior art keywords
images
remote sensing
lbp
change detection
segmentation
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
Application number
CN202010209215.XA
Other languages
Chinese (zh)
Inventor
朱邦彦
毛羽丰
张琪
储征伟
刘文伍
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.)
Nanjing Surveying And Mapping Research Institute Co ltd
Original Assignee
Nanjing Surveying And Mapping Research Institute Co ltd
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 Nanjing Surveying And Mapping Research Institute Co ltd filed Critical Nanjing Surveying And Mapping Research Institute Co ltd
Priority to CN202010209215.XA priority Critical patent/CN111311596A/en
Publication of CN111311596A publication Critical patent/CN111311596A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
    • 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
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G06V10/467—Encoded features or binary features, e.g. local binary patterns [LBP]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a remote sensing image change detection method based on improved LBP characteristics, which comprises the following steps: preprocessing the optical remote sensing images at different time intervals to finish the registration between the images; respectively carrying out multi-scale segmentation on the optical remote sensing images in two different time periods; merging the segmentation results of the optical remote sensing images in the two time periods; extracting four local binary pattern characteristic maps of the remote sensing images in two periods; calculating a joint statistical histogram of LBP characteristic graphs of each object in the two images; and comparing the similarity of the joint statistical histograms of the objects matched with the positions of the two images by using the correlation coefficient, setting a threshold value for classification, and extracting a change region. Compared with the traditional method, the method has the advantages of high accuracy, low requirement on matching precision, stability in linear brightness change, fault tolerance on nonlinear illumination change, low calculation complexity, no need of training and learning and the like, improves the reliability of change detection, and reduces the calculation complexity.

Description

Remote sensing image change detection method based on improved LBP (local binary pattern) characteristics
Technical Field
The invention relates to the field of shadow measurement and remote sensing, in particular to a remote sensing image change detection method based on improved LBP characteristics, which is used for detecting optical remote sensing image change areas in different periods.
Background
The change detection of the remote sensing image is one of important application fields of the remote sensing technology, and mainly quantitatively analyzes and determines the characteristics and the process of the surface change from remote sensing data in different periods. Generally, feature extraction and classification are performed on optical images, however, image feature extraction techniques are often affected by imaging condition factors such as different angles, illumination changes, scale changes, complex backgrounds and the like, so that it is difficult to obtain image features with high discrimination and robustness, especially differences of spectral and spatial features of a same ground object in high-resolution images at different time periods caused by illumination changes are difficult to eliminate only through image preprocessing, and finally, the accuracy of change detection is reduced.
Local Binary Pattern (LBP) is a theoretically simple, computationally efficient non-parametric Local texture feature descriptor. The basic principle is as follows: and carrying out binary coding according to the comparison between the central pixel point and the neighborhood pixel point, and converting the binary coding into a decimal value so as to obtain an LBP value for replacing the central pixel point. The algorithm has the characteristics of strong classification capability, low calculation complexity, insensitivity of image gray scale change, capability of combining the overall characteristics of the image and the like, is not easily influenced by factors such as posture, partial shielding, changed illumination and the like, and can obtain image characteristics with high discrimination and high robustness even under the complicated illumination condition.
Therefore, the remote sensing image change detection method based on the improved LBP characteristics is provided, an improved local binary pattern LBP calculation method and an object-oriented idea are used for processing the optical remote sensing image, the robustness to factors such as illumination and scale change is good, and the precision of remote sensing image change detection can be effectively improved.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides an improved local binary pattern LBP calculation method, and an object-oriented method is used for detecting the change of remote sensing images.
The object of the invention is achieved in that:
a remote sensing image change detection method based on improved LBP characteristics comprises the following steps:
firstly, preprocessing optical remote sensing images in two time periods to finish the registration between the images;
secondly, respectively carrying out multi-scale segmentation on the optical remote sensing images in two time periods;
thirdly, registering the segmentation results of the remote sensing images in two time periods, and then segmenting each image to finally enable the segmented objects of the two images to be completely overlapped;
fourthly, extracting improved Local Binary Pattern (LBP) feature maps of the images in different periods;
fifthly, segmenting each LBP characteristic graph according to the segmentation result of the third step to obtain the LBP characteristic graph of each object;
sixthly, calculating a joint statistical histogram of the LBP characteristic graph of each object in the images at different time periods;
and seventhly, comparing the similarity of the statistical histograms of the objects matched with the positions of the two images by using a correlation coefficient method, setting a threshold value for classification, and extracting a change region.
The working principle of the method comprises the following steps: the principle of image segmentation; local binary pattern principle; computer vision principles. The working process is as follows:
① preprocessing two images to be subjected to change detection, including geometric correction, radiation correction, image registration, etc., to obtain two images with pixels in strict one-to-one correspondence with geographical positions;
②, respectively segmenting the remote sensing images of the two pre-processed and registered periods by utilizing a multi-scale segmentation method, namely a fractal network evolution method, fractional net evolution approach and FNEA;
③ matching the segmentation results of the two time intervals to obtain the same segmentation object of the two images;
Figure 501858DEST_PATH_IMAGE001
calculating four Local Binary Pattern (LBP) feature maps of two period images, including two types of radial and tangential directions;
Figure 622261DEST_PATH_IMAGE002
the LBP characteristics are mapped according to
Figure 401998DEST_PATH_IMAGE003
Segmenting the segmentation result of the step to obtain an LBP characteristic map of each object, and counting the combined histogram of the four local binary pattern characteristics of all the objects in the images of two time periods;
Figure 639950DEST_PATH_IMAGE004
calculating the similarity of local binary pattern feature joint histograms of object regions corresponding to the geographic positions in the two time interval images by using a correlation coefficient method;
Figure 719902DEST_PATH_IMAGE005
setting a threshold value, regarding the area with the similarity lower than the threshold value as a change area, and outputting an extraction result.
Improved LBP feature in the fourth step: the Y direction of the image coordinates is uniformly adopted as the main direction, namely the direction is taken as the LBP initial coding direction, and coding is carried out clockwise along 8 neighborhoods.
In the fourth step, the LBP features include four types, and the calculation formulas are respectively:
Figure 542364DEST_PATH_IMAGE007
in the above formula, the first and second carbon atoms are,g c is the gray value of a pixel centered at a certain point,g p （p=0,1, …, 7) is the gray value of the surrounding 8 field sample points pixels.
In the sixth step, the threshold value of the correlation coefficient is set to 0.9, and a change region is considered when the threshold value is lower than the threshold value.
Has the positive and beneficial effects that: compared with the traditional method in the prior art, the method has the advantages of high accuracy, low requirement on matching precision, improvement on LBP (local binary pattern), adoption of the joint participation of radial and tangential pixel neighborhoods for calculation, improvement on the reliability of regional characteristics, stability in linear brightness change, fault tolerance on nonlinear illumination change, low calculation complexity, no need of training and learning and the like, improvement on the reliability of change detection, and reduction in calculation complexity.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following drawings and specific examples:
as shown in fig. 1, a method for detecting changes in remote sensing images based on improved LBP features includes the following steps:
firstly, preprocessing optical remote sensing images in two time periods to finish the registration between the images;
secondly, respectively carrying out multi-scale segmentation on the optical remote sensing images in two time periods;
thirdly, registering the segmentation results of the remote sensing images in two time periods, and then segmenting each image to finally enable the segmented objects of the two images to be completely overlapped;
fourthly, extracting improved Local Binary Pattern (LBP) feature maps of the images in different periods;
fifthly, segmenting each LBP characteristic graph according to the segmentation result of the third step to obtain the LBP characteristic graph of each object;
sixthly, calculating a joint statistical histogram of the LBP characteristic graph of each object in the images at different time periods;
and seventhly, comparing the similarity of the statistical histograms of the objects matched with the positions of the two images by using a correlation coefficient method, setting a threshold value for classification, and extracting a change region.
The working principle of the method comprises the following steps: the principle of image segmentation; local binary pattern principle; computer vision principles. The working process is as follows:
① preprocessing two images to be subjected to change detection, including geometric correction, radiation correction, image registration, etc., to obtain two images with pixels in strict one-to-one correspondence with geographical positions;
②, segmenting the preprocessed and registered remote sensing images in two periods by respectively utilizing a multi-scale segmentation method (namely a Fractal Net Evolution (FNEA));
③ matching the segmentation results of the two time intervals to obtain the same segmentation object of the two images;
Figure 12660DEST_PATH_IMAGE001
calculating four Local Binary Pattern (LBP) feature maps of two period images, including two types of radial and tangential directions;
Figure 805035DEST_PATH_IMAGE002
the LBP characteristics are mapped according to
Figure 473914DEST_PATH_IMAGE003
Segmenting the segmentation result of the step to obtain an LBP characteristic map of each object, and counting the combined histogram of the four local binary pattern characteristics of all the objects in the images of two time periods;
Figure 732857DEST_PATH_IMAGE004
calculating two time intervals by using correlation coefficient methodThe similarity of local binary pattern features of an object region corresponding to a geographical position in the image and the histogram is combined;
Figure 956028DEST_PATH_IMAGE005
setting a threshold value, regarding the area with the similarity lower than the threshold value as a change area, and outputting an extraction result.
Improved LBP feature in the fourth step: the Y direction of the image coordinates is uniformly adopted as the main direction, namely the direction is taken as the LBP initial coding direction, and coding is carried out clockwise along 8 neighborhoods.
In the fourth step, the LBP features include four types, and the calculation formulas are respectively:
Figure 37247DEST_PATH_IMAGE008
in the above formula, the first and second carbon atoms are,g c is the gray value of a pixel centered at a certain point,g p （p=0,1, …, 7) is the gray value of the surrounding 8 field sample points pixels.
In the sixth step, the threshold value of the correlation coefficient is set to 0.9, and a change region is considered when the threshold value is lower than the threshold value.
Example 1
The invention is further clarified by taking the high-score second image change detection of a certain area as an application example:
as shown in fig. 1, there are two beijing No. two satellite remote sensing images of a certain area in different time periods, and the change area needs to be extracted, the process is as follows:
(1) and image preprocessing, including radiometric calibration, atmospheric correction, geometric correction, image registration and cutting, to obtain two orthoimages with pixels in one-to-one correspondence and in the same geographic position.
(2) The preprocessed remote sensing images of two time intervals corresponding to the pixel geographic positions one by one are respectively segmented by a multi-scale segmentation method (namely a fractal network evolution method, FNEA), so that each image is divided into a plurality of small areas, the area boundaries are generally matched with the ground object boundaries, and the images are segmented into a plurality of objects.
(3) And overlapping the boundaries of the objects respectively segmented from the remote sensing images in two time periods, and then segmenting each image to finally enable the objects in the two images to be in one-to-one correspondence and completely coincide.
(4) Because the resolution of the high-resolution second panchromatic band is the highest, four Local Binary Pattern (LBP) feature maps of the remote sensing images of two periods are calculated by using the panchromatic band according to the following formula:
Figure 763895DEST_PATH_IMAGE010
in the above formula, the first and second carbon atoms are,g cis the gray value of a pixel centered at a certain point,g p（p=0,1, …, 7) is the gray value of the surrounding 8 field sample points pixels.
(5) And (3) dividing all the feature maps by using the division result obtained in the step (3) to obtain LBP feature maps of all the objects, normalizing the gray level histograms of the four LBP feature maps and connecting the normalized gray level histograms together to obtain the LBP feature vector of each object in different time periods.
(6) And (3) comparing the similarity of LBP characteristic vectors (statistical histograms) of the objects at the same position in different time periods by using a correlation coefficient method:
Figure 193739DEST_PATH_IMAGE011
wherein:
Figure 700944DEST_PATH_IMAGE012
，H 1andH 2the LBP feature vectors (histogram vectors) of the two comparison objects respectively,Nis the total number of gray levels in the histogram,I∈[0,N]，J∈[0,N]。
the similarity value range is 0 to 1 ifH 1=H 2I.e. the histograms of both graphs are identical, the value is 1, and the image is considered to be completely unchanged.
(7) And setting a similarity threshold according to actual conditions, setting the similarity threshold as 0.9 by default, regarding the similarity as a change area below the threshold, marking the change area by using a vector, and finally outputting the vector range of the change area.
Compared with the traditional method in the prior art, the method has the advantages of high accuracy, low requirement on matching precision, improvement on LBP (local binary pattern), adoption of the joint participation of radial and tangential pixel neighborhoods for calculation, improvement on the reliability of regional characteristics, stability in linear brightness change, fault tolerance on nonlinear illumination change, low calculation complexity, no need of training and learning and the like, improvement on the reliability of change detection, and reduction in calculation complexity.

Claims (5)

1. The utility model provides a special-shaped precast pile zip fastener formula supporting construction which characterized in that: the remote sensing image change detection method based on the improved LBP characteristics is characterized by comprising the following steps of:
firstly, preprocessing optical remote sensing images in two time periods to finish the registration between the images;
secondly, respectively carrying out multi-scale segmentation on the optical remote sensing images in two time periods;
thirdly, registering the remote sensing image segmentation results of the two time periods, and then segmenting each image, so that the segmented objects of the two images can be completely overlapped;
fourthly, extracting improved local LBP characteristic graphs of the images in different time periods;
fifthly, segmenting each LBP characteristic graph according to the segmentation result of the third step to obtain the LBP characteristic graph of each object;
sixthly, calculating a joint statistical histogram of the LBP characteristic graph of each object in the images at different time periods;
and seventhly, comparing the similarity of the joint statistical histograms of the objects matched with the positions of the two images by using the correlation coefficient, setting a threshold value for classification, and extracting a change region.
2. The remote sensing image change detection method based on the improved LBP characteristics as claimed in claim 1, wherein: the working principle comprises the following steps: the principle of image segmentation; local binary pattern principle; the computer vision principle works as follows:
① preprocessing two images to be subjected to change detection, including geometric correction, radiation correction, image registration, etc., to obtain two images with pixels in strict one-to-one correspondence with geographical positions;
②, segmenting the preprocessed and registered remote sensing images in two periods by respectively utilizing a multi-scale segmentation method (namely a Fractal Net Evolution (FNEA));
③ matching the segmentation results of the two time intervals to obtain the same segmentation object of the two images;
Figure 444459DEST_PATH_IMAGE001
calculating four Local Binary Pattern (LBP) feature maps of two period images, including two types of radial and tangential directions;
Figure 702265DEST_PATH_IMAGE002
the LBP characteristics are mapped according to
Figure 132109DEST_PATH_IMAGE003
Segmenting the segmentation result of the step to obtain an LBP characteristic map of each object, and counting the combined histogram of the four local binary pattern characteristics of all the objects in the images of two time periods;
Figure 717942DEST_PATH_IMAGE004
calculating local binary pattern characteristics of object regions corresponding to geographic positions in two time interval images by using correlation coefficient methodCharacterizing the similarity of the joint histogram;
Figure 727487DEST_PATH_IMAGE005
setting a threshold value, regarding the area with the similarity lower than the threshold value as a change area, and outputting an extraction result.
3. The remote sensing image change detection method based on the improved LBP characteristics as claimed in claim 1, wherein: the improved LBP characteristic in the fourth step is as follows: the Y direction of the image coordinates is uniformly adopted as the main direction, namely the direction is taken as the LBP initial coding direction, and coding is carried out clockwise along 8 neighborhoods.
4. The remote sensing image change detection method based on the improved LBP characteristics as claimed in claim 1, wherein: the LBP characteristics in the fourth step include four types, and the calculation formulas are respectively as follows:
Figure 370958DEST_PATH_IMAGE006
in the above formula, the first and second carbon atoms are,g c is the gray value of a pixel centered at a certain point,g p （p=0,1, …, 7) is the gray value of the surrounding 8 field sample points pixels.
5. The remote sensing image change detection method based on the improved LBP characteristics as claimed in claim 1, wherein: in the sixth step, the threshold value of the correlation coefficient is set to 0.9, and a change region is considered when the threshold value is lower than the threshold value.
CN202010209215.XA 2020-03-23 2020-03-23 Remote sensing image change detection method based on improved LBP (local binary pattern) characteristics Pending CN111311596A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010209215.XA CN111311596A (en) 2020-03-23 2020-03-23 Remote sensing image change detection method based on improved LBP (local binary pattern) characteristics

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010209215.XA CN111311596A (en) 2020-03-23 2020-03-23 Remote sensing image change detection method based on improved LBP (local binary pattern) characteristics

Publications (1)

Publication Number Publication Date
CN111311596A true CN111311596A (en) 2020-06-19

Family

ID=71160739

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010209215.XA Pending CN111311596A (en) 2020-03-23 2020-03-23 Remote sensing image change detection method based on improved LBP (local binary pattern) characteristics

Country Status (1)

Country Link
CN (1) CN111311596A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112633070A (en) * 2020-11-27 2021-04-09 中国科学院空天信息创新研究院 High-resolution remote sensing image building extraction method and system
CN113298755A (en) * 2021-04-13 2021-08-24 生态环境部卫星环境应用中心 Method and device for rapidly detecting ecological environment change patch based on time sequence image
CN115410096A (en) * 2022-11-03 2022-11-29 成都国星宇航科技股份有限公司 Satellite remote sensing image multi-scale fusion change detection method, medium and electronic device

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100098342A1 (en) * 2008-10-16 2010-04-22 Curators Of The University Of Missouri Detecting geographic-area change using high-resolution, remotely sensed imagery
CN101937079A (en) * 2010-06-29 2011-01-05 中国农业大学 Remote sensing image variation detection method based on region similarity
CN105956557A (en) * 2016-05-04 2016-09-21 长江水利委员会长江科学院 Object-oriented timing sequence remote sensing image cloud coverage area automatic detection method
CN108805915A (en) * 2018-04-19 2018-11-13 南京市测绘勘察研究院股份有限公司 A kind of close-range image provincial characteristics matching process of anti-visual angle change
CN109523516A (en) * 2018-10-19 2019-03-26 中国科学院遥感与数字地球研究所 A kind of object level land cover pattern change detecting method based on double constraints condition
CN109657610A (en) * 2018-12-18 2019-04-19 北京航天泰坦科技股份有限公司 A kind of land use change survey detection method of high-resolution multi-source Remote Sensing Images

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100098342A1 (en) * 2008-10-16 2010-04-22 Curators Of The University Of Missouri Detecting geographic-area change using high-resolution, remotely sensed imagery
CN101937079A (en) * 2010-06-29 2011-01-05 中国农业大学 Remote sensing image variation detection method based on region similarity
CN105956557A (en) * 2016-05-04 2016-09-21 长江水利委员会长江科学院 Object-oriented timing sequence remote sensing image cloud coverage area automatic detection method
CN108805915A (en) * 2018-04-19 2018-11-13 南京市测绘勘察研究院股份有限公司 A kind of close-range image provincial characteristics matching process of anti-visual angle change
CN109523516A (en) * 2018-10-19 2019-03-26 中国科学院遥感与数字地球研究所 A kind of object level land cover pattern change detecting method based on double constraints condition
CN109657610A (en) * 2018-12-18 2019-04-19 北京航天泰坦科技股份有限公司 A kind of land use change survey detection method of high-resolution multi-source Remote Sensing Images

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
DAIFENG PENG 等: "Object-based change detection from satellite imagery by segmentation optimization and multi-features fusion", 《INTERNATIONAL JOURNAL OF REMOTE SENSING》 *
姜慧伟: "利用多特征融合的面向对象的高分辨率遥感影像变化检测", 《陕西理工学院学报(自然科学版)》 *
张忠芳: "面向对象的高分辨率SAR影像变化检测方法研究", 《中国优秀博硕士学位论文全文数据库（硕士）基础科学辑》 *
李亮等: "融合光谱及纹理异质度的遥感影像变化检测", 《测绘通报》 *
马红等: "一种多特征自适应组合的遥感影像变化检测方法", 《首都师范大学学报(自然科学版)》 *
黄杰等: "高分辨率彩色遥感影像变化检测方法研究", 《测绘通报》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112633070A (en) * 2020-11-27 2021-04-09 中国科学院空天信息创新研究院 High-resolution remote sensing image building extraction method and system
CN113298755A (en) * 2021-04-13 2021-08-24 生态环境部卫星环境应用中心 Method and device for rapidly detecting ecological environment change patch based on time sequence image
CN115410096A (en) * 2022-11-03 2022-11-29 成都国星宇航科技股份有限公司 Satellite remote sensing image multi-scale fusion change detection method, medium and electronic device

Similar Documents

Publication Publication Date Title
CN109409292B (en) Heterogeneous image matching method based on refined feature optimization extraction
Gonçalves et al. HAIRIS: A method for automatic image registration through histogram-based image segmentation
CN111311596A (en) Remote sensing image change detection method based on improved LBP (local binary pattern) characteristics
CN110097093A (en) A kind of heterologous accurate matching of image method
CN112766184B (en) Remote sensing target detection method based on multi-level feature selection convolutional neural network
CN109461132B (en) SAR image automatic registration method based on feature point geometric topological relation
CN113256653B (en) Heterogeneous high-resolution remote sensing image registration method for high-rise ground object
CN112215085A (en) Power transmission corridor foreign matter detection method and system based on twin network
CN112633185A (en) Image processing method and device
CN110569861A (en) Image matching positioning method based on point feature and contour feature fusion
CN114022459A (en) Multi-temporal satellite image-based super-pixel change detection method and system
CN114494378A (en) Multi-temporal remote sensing image automatic registration method based on improved SIFT algorithm
CN114972453B (en) Improved SAR image region registration method based on LSD and template matching
Xing et al. An adaptive change threshold selection method based on land cover posterior probability and spatial neighborhood information
CN111523342A (en) Two-dimensional code detection and correction method in complex scene
Huang et al. SAR and optical images registration using shape context
CN109544608B (en) Unmanned aerial vehicle image acquisition characteristic registration method
Li et al. A novel framework for urban change detection using VHR satellite images
Changjie et al. Algorithm of remote sensing image matching based on corner-point
CN113822361A (en) SAR image similarity measurement method and system based on Hamming distance
CN103337080A (en) Registration technology of infrared image and visible image based on Hausdorff distance in gradient direction
Sui et al. Automatic image registration based on shape features and multi-scale image segmentation
CN107832793A (en) The sorting technique and system of a kind of high spectrum image
CN111178175A (en) Automatic building information extraction method and system based on high-view satellite image
CN116385477A (en) Tower image registration method based on image segmentation

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: 20200619