CN109934110B - Method for identifying illegal buildings near river channel - Google Patents

Method for identifying illegal buildings near river channel Download PDF

Info

Publication number
CN109934110B
CN109934110B CN201910107890.9A CN201910107890A CN109934110B CN 109934110 B CN109934110 B CN 109934110B CN 201910107890 A CN201910107890 A CN 201910107890A CN 109934110 B CN109934110 B CN 109934110B
Authority
CN
China
Prior art keywords
house
outline
area
image
house outline
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
CN201910107890.9A
Other languages
Chinese (zh)
Other versions
CN109934110A (en
Inventor
潘屹峰
杨骥
李勇
刘文祥
李国华
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.)
Guangzhou Imapcloud Intelligent Technology Co ltd
Original Assignee
Guangzhou Imapcloud Intelligent Technology 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 Guangzhou Imapcloud Intelligent Technology Co ltd filed Critical Guangzhou Imapcloud Intelligent Technology Co ltd
Priority to CN201910107890.9A priority Critical patent/CN109934110B/en
Publication of CN109934110A publication Critical patent/CN109934110A/en
Application granted granted Critical
Publication of CN109934110B publication Critical patent/CN109934110B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Image Analysis (AREA)

Abstract

The invention relates to a method for identifying buildings illegally built near a river channel, which comprises the following steps: s1: splicing a plurality of river orthoimages shot by an unmanned aerial vehicle into a panoramic image; s2: carrying out house characteristic training on an initial semantic segmentation algorithm through the panoramic image to obtain a house outline recognition model; s3: extracting house outline pixels in the panoramic image through the house outline recognition model, and vectorizing the house outline pixels; s4: and carrying out house violation identification on the vectorized house outline pixels. The method for identifying the buildings illegally built near the riverway replaces the traditional manual work of carrying out the outline marking and vectorization work of the buildings, and improves the efficiency of extracting the outlines of the buildings.

Description

Method for identifying illegal buildings near river channel
Technical Field
The invention relates to the field of building illegal building identification, in particular to a method for identifying illegal buildings near a river channel.
Background
In order to ensure smooth flood discharge of the river channel, it is not allowed to set houses in a certain distance near the river channel according to relevant regulations, so that the buildings near the river channel need to be checked, wherein identification and area extraction of buildings such as houses are very important steps. At present, a river violation house identification and area extraction method generally comprises the steps of carrying out downward shooting on a river at a specific distance along a river center through an unmanned aerial vehicle, fusing downward shot images into a complete river ortho-image, guiding the ortho-image into professional software, carrying out contour identification on houses near the river through manual work, simultaneously vectorizing into an SHP file, finally carrying out overlap area extraction on vectorized houses through establishing a river buffer area, and identifying the violation houses. Because the river to be photographed is very long and the buildings on both sides of the river are very dense, it takes time and labor to manually perform the outline marking and vectorization work of the house. Statistics shows that the process of manually carrying out outline marking and vectorization of the house accounts for about 40% of the time and 80% of human resources of the whole house area extraction and marking violation process. Therefore, the efficiency of the traditional process of manually identifying the outline of the house and vectorizing is too low to meet the project construction period requirement.
Disclosure of Invention
Based on this, the invention aims to provide a method for identifying buildings illegally built near a river channel, which has the advantages of improving the working efficiency of contour marking and vectorization of the buildings.
A method for identifying illegal buildings near a river channel comprises the following steps:
s1: splicing a plurality of river orthoimages shot by an unmanned aerial vehicle into a panoramic image;
s2: carrying out house characteristic training on an initial semantic segmentation algorithm through the panoramic image to obtain a house outline recognition model;
s3: extracting house outline pixels in the panoramic image through the house outline recognition model, and vectorizing the house outline pixels;
s4: and carrying out house violation identification on the vectorized house outline pixels.
The method for identifying the buildings illegally built near the riverway replaces the traditional manual work of carrying out the outline marking and vectorization work of the buildings, and improves the efficiency of extracting the outlines of the buildings.
Further, the step S2 includes the following steps:
s201: constructing a backbone feature extractor to extract house features and obtain a feature information graph;
s202: performing convolution operation on the characteristic information graph to reduce the number of channels of the characteristic information graph and obtain a first characteristic graph;
s203: constructing a spatial pyramid pool to pool the first feature map, and performing bilinear interpolation deconvolution to obtain a second feature map;
s204: performing feature fusion on the first feature map and the second feature map to obtain a feature map set, and performing convolution and bilinear interpolation deconvolution operation on the feature map set to obtain a prediction image;
s205: according to a preset marked image, performing cross entropy calculation on the predicted image and the marked image to obtain a loss amount;
s206: and performing iterative optimization on the house outline recognition model by using the loss amount to obtain an optimized house outline recognition model.
Further, in the step S2, after the step S206, the method further includes the following steps: s207: and packaging the optimized house outline recognition model into a house outline extraction tool.
Further, the step S3 includes the following steps: and extracting the house outline from the house outline recognition model and outputting an SHP file from the vectorized characteristic diagram.
Further, in step S3, the predicted image is morphologically processed before the house outline is extracted.
Further, the morphological treatments include dilation, erosion, open and close operations.
Further, the step S4 includes the following steps:
s401, calculating the area of the house outline in the prediction image;
s402, constructing buffer areas on two sides of the river channel in the prediction image and calculating the area of the buffer areas;
s403, performing overlapping calculation on the area of the buffer area and the area of the house outline to obtain the area of an intersection area of the house outline and the buffer area;
s404: and identifying the intersection area of the house outline and the buffer area as an illegal house area.
Further, the step S401 specifically includes the following steps: presetting an area threshold value, extracting a house outline point set, calculating an outline area, comparing the outline area with the preset area threshold value, and removing a region with the outline area smaller than the preset area threshold value; presetting a rectangle degree threshold value, performing rectangle degree calculation on each house outline, and removing the outlines smaller than the preset rectangle degree threshold value; and uniformly diluting the house outline point set, and vectorizing the diluted house outline point set.
The present invention also provides a computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, realizes the steps of the method for identifying illicit premises in the vicinity of a river channel as described in any one of the above.
The invention also provides a computer device, which comprises a storage, a processor and a computer program stored in the storage and executable by the processor, wherein the processor executes the computer program to realize the steps of the method for identifying the illegal buildings near the river channel.
For a better understanding and practice, the invention is described in detail below with reference to the accompanying drawings.
Drawings
FIG. 1 is a flow chart of the method for identifying illegal buildings near a river channel according to the present invention;
FIG. 2 is a flow chart of step 2 of the present invention;
FIG. 3 is a schematic diagram of the house outline extraction and vectorization according to the present invention;
fig. 4 is a functional block diagram of the house outline extraction tool according to the present invention.
Detailed Description
In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on those shown in the drawings, and are used only for convenience in describing the present invention and for simplicity in description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention. In the description of the present invention, "a plurality" means two or more unless otherwise specified.
Referring to fig. 1-3, fig. 1 is a flow chart of the method for identifying illegal buildings near a river according to the present invention; FIG. 2 is a flow chart of step 2 of the present invention; fig. 3 is a schematic diagram of the house contour extraction and vectorization according to the present invention.
The equipment used in the method for identifying the buildings illegally built near the riverway comprises an unmanned aerial vehicle, a camera carried on the unmanned aerial vehicle and image processing equipment arranged on a ground workstation; the image processing equipment is internally provided with Arcgis software, a tensoflow learning framework and an Open CV image processing library.
In the embodiment, the unmanned aerial vehicle is controlled to fly along the center of the river; and the camera performs overhead shooting on the river to obtain a plurality of river ortho-images, and transmits the ortho-images to professional software of the image processing equipment for house illegal building identification.
The method for identifying the buildings illegally built near the riverway comprises the following steps:
s1: splicing a plurality of river orthoimages shot by an unmanned aerial vehicle into a panoramic image;
s2: carrying out house characteristic training on an initial semantic segmentation algorithm through the panoramic image to obtain a house outline recognition model;
s3: extracting house outline pixels in the panoramic image through the house outline recognition model, and vectorizing the house outline pixels;
s4: and carrying out house violation identification on the vectorized house outline pixels.
In this embodiment, the initial semantic segmentation algorithm adopts a deep bv3+ algorithm, and in other embodiments, the initial semantic segmentation algorithm may also adopt other algorithms, such as mask-RCNN, FCN, U-Net, and the like.
In this embodiment, step S2 specifically includes the following steps:
s201: constructing a backbone feature extractor to extract house features and obtain a feature information graph;
s202: performing convolution operation on the characteristic information graph to reduce the number of channels of the characteristic information graph and obtain a first characteristic graph;
s203: constructing a spatial pyramid pool to pool the first feature map, and performing bilinear interpolation deconvolution to obtain a second feature map;
s204: performing feature fusion on the first feature map and the second feature map to obtain a feature map set, and performing convolution and bilinear interpolation deconvolution operation on the feature map set to obtain a prediction image;
s205: according to a preset marked image, performing cross entropy calculation on the predicted image and the marked image to obtain a loss amount;
s206: performing iterative optimization on the house outline identification model by using the loss amount to obtain an optimized house outline identification model;
s207: and packaging the optimized house outline recognition model into a house outline extraction tool.
The house features extracted by the backbone feature extractor comprise: contour, shape, color, texture, and the like.
And a space pyramid pool is constructed to pool the feature map, so that the problem that the size of an input image is required to be fixed is solved, and the aspect ratio and the size of the input image can be arbitrary.
The step S3 further includes the steps of: and extracting the house outline from the house outline recognition model and outputting an SHP file from the vectorized characteristic diagram.
In step S3, before the house outline is extracted from the predicted image, morphological processing is performed by Open CV to distinguish the connected house outlines, so that the house outlines are closer to the actual outlines.
And after morphological processing, extracting the house outline to obtain a house outline pixel point set, and vectorizing the house outline pixels.
In addition, the user can select whether to vector to the SHP file according to requirements, and when the vector does not need to be the SHP file, the house outline image can be directly extracted and stored as the JPG file.
In particular, the morphological treatments include dilation, erosion, opening and closing operations.
1. Expansion: the method is an operation of solving a local maximum value, and judges whether an original image in a template area is set to be 1 or 0 at all by using the maximum value, so that for a binary image, a template with a specific structure is used for traversing each position on the image, and if one point of the position of the specific structure of the template is 1, the position of the specific structure on the image is 1.
2. And (3) corrosion: the role of corrosion in mathematical morphology operations is to eliminate boundary points of objects. In digital image processing, for a certain structural element, points smaller than the structural element can be eliminated by erosion operation. Meanwhile, if a target region contains fine connective parts, the region can be segmented by the Vaccinium uliginosum treatment.
Wherein the relationship between erosion and expansion is: if the target image is A and the structural element is S, the corrosion operation is performed on A by using A-S, and the expansion budget is performed on A by using A + S.
The application of erosion and dilation is the extraction of the boundary: the boundary extraction is realized by carrying out corrosion or expansion processing on the target image and comparing the difference between the result image and the original image; extracting the inner boundary to obtain a contraction of the original image by utilizing the corrosion treatment of the image, and then carrying out XOR operation on the contraction result and the target image; and extracting the outer boundary to perform expansion processing on the target image, and then performing XOR operation on the expansion result and the original target image.
3. Open and close operations:
opening operation: a ≈ S ═ a-S) + S (i.e., erosion and then dilation on the target image)
And (3) closed operation: a ● S ═ A + S-S (i.e., the target image was dilated and then eroded)
The image opening operation is often used for denoising a target image, and meanwhile, the image opening operation can selectively reserve the part which meets the geometric property of the structural primary colors in the target image and filter out the damaged part relative to the structural elements.
The closed operation of the image is often used for connecting separated areas of a target image and filling small gaps in the image, the closed operation of the image can enable the filling result of the image to have a geometric characteristic of one point by properly selecting structural elements, the closed operation of the image can enable the image to be clearer and more coherent sometimes, and meanwhile, the thickening of lines in the original image can be avoided.
In this embodiment, the step S4 specifically includes the following steps:
s401, calculating the area of the house outline in the prediction image;
s402, constructing buffer areas on two sides of the river channel in the prediction image and calculating the area of the buffer areas;
s403, performing overlapping calculation on the area of the buffer area and the area of the house outline to obtain the area of an intersection area of the house outline and the buffer area;
s404: and identifying the intersection area of the house outline and the buffer area as an illegal house area.
The step S401 specifically includes the following steps: presetting an area threshold value, extracting a house outline point set, calculating an outline area, comparing the outline area with the preset area threshold value, and removing a region with the outline area smaller than the preset area threshold value; presetting a rectangle degree threshold value, performing rectangle degree calculation on each house outline, and removing the outlines smaller than the preset rectangle degree threshold value; and uniformly diluting the house outline point set, and vectorizing the diluted house outline point set.
Specifically, a minimum outline area threshold value is preset; simultaneously calculating the area of each house outline, comparing the area with a minimum outline area threshold value, and if the area is smaller than the minimum outline area threshold value, removing the area to obtain a house outline area A contourr; importing the orthographic image into Arcgis software, constructing buffer areas with specific distances on two banks of a river channel, calculating the area A buffer of the buffer areas, and performing overlapping calculation on the area A buffer of the river channel buffer area and the house outline area A contourer: a IoU ═ Abuffer ≈ Acontour; and obtaining a cross-over area A Io U, namely obtaining an intersection area of the house outline and the buffer area, setting the intersection area of the house outline and the buffer area as an illegal house area, and identifying. In this embodiment, the buffer area is a six-meter buffer area or a ten-meter buffer area extending from the river channel to both banks, and buffer areas at other distances.
In addition, graphics displayed in computers can generally be divided into two broad categories: vector graphics and bitmap. Vector diagrams describe graphics using straight lines and curved lines, the elements of which are points, lines, rectangles, polygons, circles and arcs, etc., which are calculated by mathematical formulas. For example, a vector graphic of a picture is actually an outline of an outer frame formed by line segments, and the color of the outer frame and the color enclosed by the outer frame determine the color displayed by the flower. Vector graphics files are generally small in size, since vector graphics can be obtained by formulaic calculations. Vector images have many advantages: firstly, a vector image is composed of simple geometric primitives, the representation is compact, and the occupied storage space is small; secondly, the vector image is easy to edit, and only parameter information of corresponding geometric primitives needs to be modified when the vector image is edited, such as rotation, stretching, translation and other operations; thirdly, the object represented by the vector is easy to enlarge or compress without reducing the display quality of the object in a computer, the vector image can be enlarged or compressed with the characteristics of sharp corners and the like, and the display quality is not affected by the blurring. Therefore, in this embodiment, vectorization is performed after the feature of the house outline of the ortho-image is extracted, which is an important step.
Referring to fig. 4, fig. 4 is a functional block diagram of the house outline extraction tool according to the present invention. The functions of the house outline extraction and vectorization tool comprise image management, image operation, image display, house outline extraction, parameter setting and file export. Specifically, the image management may include functional operations such as single-picture import, multi-picture import, and predictive picture saving; the image operation may include image scaling, up-down image selection, image rotation, and other functional operations; the image display can comprise functional operations of loading a whole image, loading a partial prediction image and the like; the house outline extraction can comprise the functional operations of deep learning module calling, super-large image segmentation, image splicing and the like; the parameter setting can comprise the function operations of model parameter setting, cutting image parameter setting and the like; the file export can comprise functional operations of profile vectorization file export, prediction report generation and the like.
The method for identifying the buildings illegally built near the river channel comprises the steps of firstly collecting river ortho-images through an unmanned aerial vehicle, and then fusing and splicing a plurality of collected river ortho-images into a complete panoramic image through Arcgis software; then importing the spliced panoramic image into house outline extraction software, extracting the house outline according to the user requirement and vectorizing the outline into an shp file; and finally, importing the shp file and the orthoimage into arcgis software, establishing a buffer area for calculating the overlapping area, and identifying the illegal house. The method can greatly improve the working efficiency of house outline extraction and can save the time of house outline extraction and vectorization; meanwhile, the labor is saved, and only one person needs to operate the software; and the extraction precision of the deep learning algorithm in the whole image can reach 93.53%.
The present invention also provides a computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, realizes the steps of the method for identifying illegal buildings near a river channel according to any one of the above.
The present invention may take the form of a computer program product embodied on one or more storage media including, but not limited to, disk storage, CD-ROM, optical storage, and the like, in which program code is embodied. Computer readable storage media, which include both non-transitory and non-transitory, removable and non-removable media, may implement any method or technology for storage of information. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of the storage medium of the computer include, but are not limited to: phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic tape storage or other magnetic storage devices, or any other non-transmission medium, may be used to store information that may be accessed by a computing device.
The invention also provides computer equipment, which comprises a storage, a processor and a computer program stored in the storage and executable by the processor, wherein the processor executes the computer program to realize the steps of the method for identifying the illegal buildings near the river channel.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention.

Claims (8)

1. A method for identifying buildings illegally built near a river channel is characterized by comprising the following steps: the method comprises the following steps:
s1: splicing a plurality of river orthoimages shot by an unmanned aerial vehicle into a panoramic image;
s2: house characteristic training is carried out on the initial semantic segmentation algorithm through the panoramic image to obtain a house outline recognition model, and the house outline recognition method comprises the following steps:
s201: constructing a backbone feature extractor to extract house features and obtain a feature information graph;
s202: performing convolution operation on the characteristic information graph to reduce the number of channels of the characteristic information graph and obtain a first characteristic graph;
s203: constructing a spatial pyramid pool to pool the first feature map, and performing bilinear interpolation deconvolution to obtain a second feature map;
s204: performing feature fusion on the first feature map and the second feature map to obtain a feature map set, and performing convolution and bilinear interpolation deconvolution operation on the feature map set to obtain a prediction image;
s205: according to a preset marked image, performing cross entropy calculation on the predicted image and the marked image to obtain a loss amount;
s206: performing iterative optimization on the house outline identification model by using the loss amount to obtain an optimized house outline identification model;
s3: extracting house outline pixels in the panoramic image through the house outline recognition model, and vectorizing the house outline pixels;
s4: the method for carrying out house default construction identification on the vectorized house outline pixels comprises the following steps:
s401, calculating the area of the house outline in the prediction image;
s402, constructing buffer areas on two sides of the river channel in the prediction image and calculating the area of the buffer areas;
s403, performing overlapping calculation on the area of the buffer area and the area of the house outline to obtain the area of an intersection area of the house outline and the buffer area;
s404, identifying the intersection area of the house outline and the buffer area as a building violation area.
2. The method for identifying buildings illegal to build near a river according to claim 1, characterized in that: in step S2, after step S206, the method further includes the following steps: s207: and packaging the optimized house outline recognition model into a house outline extraction tool.
3. The method for identifying buildings illegal to build near a river according to claim 1, characterized in that: the step S3 further includes the steps of: and extracting the house outline from the house outline recognition model and outputting an SHP file from the vectorized characteristic diagram.
4. The method for identifying buildings in the vicinity of river channels as claimed in claim 2, wherein: in step S3, the predicted image is further morphologically processed before the house outline is extracted.
5. The method for identifying the buildings illegal in the vicinity of the river according to claim 4, wherein the method comprises the following steps: the morphological treatments include dilation, erosion, open and close operations.
6. The method for identifying buildings illegal to build near a river according to claim 1, characterized in that: the step S401 specifically includes the following steps: presetting an area threshold value, extracting a house outline point set, calculating an outline area, comparing the outline area with the preset area threshold value, and removing a region with the outline area smaller than the preset area threshold value; presetting a rectangle degree threshold value, performing rectangle degree calculation on each house outline, and removing the outlines smaller than the preset rectangle degree threshold value; and uniformly diluting the house outline point set, and vectorizing the diluted house outline point set.
7. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for identifying illicit premises in the vicinity of a river according to any one of claims 1 to 6.
8. Computer arrangement, comprising a memory, a processor and a computer program stored in said memory and executable by said processor, said processor implementing the steps of the method for identification of illicit premises in the vicinity of a river according to any one of claims 1 to 6 when executing said computer program.
CN201910107890.9A 2019-02-02 2019-02-02 Method for identifying illegal buildings near river channel Active CN109934110B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910107890.9A CN109934110B (en) 2019-02-02 2019-02-02 Method for identifying illegal buildings near river channel

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910107890.9A CN109934110B (en) 2019-02-02 2019-02-02 Method for identifying illegal buildings near river channel

Publications (2)

Publication Number Publication Date
CN109934110A CN109934110A (en) 2019-06-25
CN109934110B true CN109934110B (en) 2021-01-12

Family

ID=66985575

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910107890.9A Active CN109934110B (en) 2019-02-02 2019-02-02 Method for identifying illegal buildings near river channel

Country Status (1)

Country Link
CN (1) CN109934110B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111104879B (en) * 2019-12-09 2020-11-27 贝壳找房(北京)科技有限公司 Method and device for identifying house functions, readable storage medium and electronic equipment
CN111062361B (en) * 2019-12-27 2023-06-20 中水北方勘测设计研究有限责任公司 River and lake shoreline sand production monitoring and analyzing method and device
CN111382695A (en) * 2020-03-06 2020-07-07 北京百度网讯科技有限公司 Method and apparatus for detecting boundary points of object
CN111382696A (en) * 2020-03-06 2020-07-07 北京百度网讯科技有限公司 Method and apparatus for detecting boundary points of object
CN111523391B (en) * 2020-03-26 2021-11-05 上海刻羽信息科技有限公司 Building identification method, system, electronic device and readable storage medium
CN113536836A (en) * 2020-04-15 2021-10-22 宁波弘泰水利信息科技有限公司 Method for monitoring river and lake water area encroachment based on unmanned aerial vehicle remote sensing technology
CN111652892A (en) * 2020-05-02 2020-09-11 王磊 Remote sensing image building vector extraction and optimization method based on deep learning
CN112651338B (en) * 2020-12-26 2022-02-15 广东电网有限责任公司电力科学研究院 Method and device for distinguishing hidden danger of illegal construction of power transmission line
CN113724117A (en) * 2020-12-28 2021-11-30 京东城市(北京)数字科技有限公司 Model training method and device for house abnormal use recognition
CN114549569B (en) * 2022-03-01 2024-04-09 中国电建集团中南勘测设计研究院有限公司 House contour extraction method based on two-dimensional orthographic image

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2418606A2 (en) * 2010-08-13 2012-02-15 Pantech Co., Ltd. Apparatus and method for recognizing objects using filter information
CN102930540A (en) * 2012-10-26 2013-02-13 中国地质大学(武汉) Method and system for detecting contour of urban building
CN105608691A (en) * 2015-12-17 2016-05-25 武汉大学 High-resolution SAR image individual building extraction method
CN105929845A (en) * 2016-05-18 2016-09-07 中国计量大学 Unmanned aerial vehicle network-based river channel cruise system and cruise method
CN106934156A (en) * 2017-03-13 2017-07-07 中国水利水电科学研究院 It is a kind of to build dangerous spot recognition methods along river course
CN108764039A (en) * 2018-04-24 2018-11-06 中国科学院遥感与数字地球研究所 Building extracting method, medium and the computing device of neural network, remote sensing image
CN109242291A (en) * 2018-08-28 2019-01-18 天津大学 River and lake basin water environment wisdom management method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2418606A2 (en) * 2010-08-13 2012-02-15 Pantech Co., Ltd. Apparatus and method for recognizing objects using filter information
CN102930540A (en) * 2012-10-26 2013-02-13 中国地质大学(武汉) Method and system for detecting contour of urban building
CN105608691A (en) * 2015-12-17 2016-05-25 武汉大学 High-resolution SAR image individual building extraction method
CN105929845A (en) * 2016-05-18 2016-09-07 中国计量大学 Unmanned aerial vehicle network-based river channel cruise system and cruise method
CN106934156A (en) * 2017-03-13 2017-07-07 中国水利水电科学研究院 It is a kind of to build dangerous spot recognition methods along river course
CN108764039A (en) * 2018-04-24 2018-11-06 中国科学院遥感与数字地球研究所 Building extracting method, medium and the computing device of neural network, remote sensing image
CN109242291A (en) * 2018-08-28 2019-01-18 天津大学 River and lake basin water environment wisdom management method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Research on Illegal Building Detection Based on the Change Discovery and Spatial Morphological Characteristics of Image;Yu WANG等;《2018 2nd International Conference on Applied Mathematics, Modeling and Simulation》;20180826;136-143 *
无人机河道巡检系统设计与实现;鲁佳隽;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》;20190115(第01期);C031-499 *

Also Published As

Publication number Publication date
CN109934110A (en) 2019-06-25

Similar Documents

Publication Publication Date Title
CN109934110B (en) Method for identifying illegal buildings near river channel
Cornillere et al. Blind image super-resolution with spatially variant degradations
JP6705912B2 (en) Method and apparatus for recognizing character areas in an image
CN109815865B (en) Water level identification method and system based on virtual water gauge
CN111583097A (en) Image processing method, image processing device, electronic equipment and computer readable storage medium
CN105528614B (en) A kind of recognition methods of the cartoon image space of a whole page and automatic recognition system
CN105678347A (en) Pedestrian detection method and device
CN107273608B (en) Oil reservoir geological profile vectorization method
CN112785710B (en) Rapid unitization method, system, memory and equipment for OSGB three-dimensional model building
US7127104B2 (en) Vectorized image segmentation via trixel agglomeration
US20190057532A1 (en) Realistic augmentation of images and videos with graphics
US8670615B2 (en) Refinement of segmentation markup
CN111275034B (en) Method, device, equipment and storage medium for extracting text region from image
Kiess et al. Seam carving with improved edge preservation
CN110544300B (en) Method for automatically generating three-dimensional model based on two-dimensional hand-drawn image characteristics
CN111127631A (en) Single image-based three-dimensional shape and texture reconstruction method, system and storage medium
CN111915628A (en) Single-stage instance segmentation method based on prediction target dense boundary points
US20130294707A1 (en) Geometric modelization of images and applications
CN113077477B (en) Image vectorization method and device and terminal equipment
US20150206344A1 (en) 3D Model Enhancement
CN113870196A (en) Image processing method, device, equipment and medium based on anchor point cutting graph
CN111260564A (en) Image processing method and device and computer storage medium
CN111179289B (en) Image segmentation method suitable for webpage length graph and width graph
CN116091709A (en) Three-dimensional reconstruction method and device for building, electronic equipment and storage medium
CN114511862A (en) Form identification method and device and electronic equipment

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