CN112967286A - Method and device for detecting newly added construction land - Google Patents
Method and device for detecting newly added construction land Download PDFInfo
- Publication number
- CN112967286A CN112967286A CN202110543709.6A CN202110543709A CN112967286A CN 112967286 A CN112967286 A CN 112967286A CN 202110543709 A CN202110543709 A CN 202110543709A CN 112967286 A CN112967286 A CN 112967286A
- Authority
- CN
- China
- Prior art keywords
- land
- construction land
- detection result
- detection
- layer
- 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.)
- Granted
Links
- 238000010276 construction Methods 0.000 title claims abstract description 236
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000001514 detection method Methods 0.000 claims abstract description 272
- 239000013598 vector Substances 0.000 claims abstract description 49
- 230000004927 fusion Effects 0.000 claims abstract description 39
- 238000012545 processing Methods 0.000 claims abstract description 29
- 238000007499 fusion processing Methods 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims description 26
- 230000008569 process Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 description 18
- 230000011218 segmentation Effects 0.000 description 18
- 238000013135 deep learning Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 11
- 238000002372 labelling Methods 0.000 description 8
- 230000008901 benefit Effects 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- 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
- 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
- G06T2207/20081—Training; Learning
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
A detection method and a device for newly added construction land relate to the technical field of remote sensing, and the detection method for the newly added construction land comprises the following steps: firstly, acquiring a satellite high-resolution image of a target detection area; then, performing prediction processing on the satellite high-resolution image through a newly-built construction land prediction model built in advance to obtain a construction land heat map of the target detection area; then, detecting and processing newly added construction land for a land layer and an object layer of the satellite high-resolution image according to preset ground surface coverage vector data to obtain a land layer detection result and an object layer detection result; further, performing decision-level fusion processing on the detection result of the land parcel layer and the detection result of the object layer to obtain a fusion detection result; and finally, determining a target new construction land detection result of the target detection area according to the fusion detection result, so that the new construction land can be quickly and accurately detected, and the method is small in error and good in applicability.
Description
Technical Field
The application relates to the technical field of remote sensing, in particular to a method and a device for detecting newly added construction land.
Background
The newly added construction land is used as an important component of urban construction and development, and the rapid discovery and accurate extraction of the newly added construction land can assist the urban natural resource management department in the development trend and the development speed of the urban, so that dynamic monitoring and comprehensive analysis are realized. The existing newly added construction land detection method generally identifies a change region by utilizing front and rear two-phase images and combining a front time phase land utilization/coverage classification map through multi-change detection (MAD) and K-mean value clustering, and then supervises and classifies the rear time phase images of the change region to extract the newly added construction land. However, in practice, it is found that the existing method for detecting the newly added construction land is easily affected by interference factors such as atmospheric conditions, solar altitude, soil humidity, vegetation season changes, matching errors and the like by detecting changes of images, and the images in the two stages before and after are difficult to acquire and have poor applicability. Therefore, the existing detection method for the newly added construction land has the defects of large error, low accuracy and poor applicability.
Disclosure of Invention
The embodiment of the application aims to provide a method and a device for detecting newly added construction land, which can be used for quickly and accurately detecting the newly added construction land, and are small in error and good in applicability.
The first aspect of the embodiment of the present application provides a method for detecting a newly added construction land, including:
acquiring a satellite high-resolution image of a target detection area;
predicting the satellite high-resolution image through a newly-built construction land prediction model to obtain a construction land heat map of the target detection area;
carrying out detection processing on newly added construction land of a land layer and an object layer on the satellite high-resolution image according to preset ground surface coverage vector data to obtain a land layer detection result and an object layer detection result;
performing decision-level fusion processing on the detection result of the land layer and the detection result of the object layer to obtain a fusion detection result;
and determining a detection result of the target newly added construction land of the target detection area according to the fusion detection result.
In the implementation process, firstly, a satellite high-resolution image of a target detection area is obtained; then, performing prediction processing on the satellite high-resolution image through a newly-built construction land prediction model built in advance to obtain a construction land heat map of the target detection area; then, detecting and processing newly added construction land for a land layer and an object layer of the satellite high-resolution image according to preset ground surface coverage vector data to obtain a land layer detection result and an object layer detection result; further, performing decision-level fusion processing on the detection result of the land parcel layer and the detection result of the object layer to obtain a fusion detection result; and finally, determining a target new construction land detection result of the target detection area according to the fusion detection result, so that the new construction land can be quickly and accurately detected, and the method is small in error and good in applicability.
Further, before the obtaining the satellite high score image of the target detection area, the method further includes:
acquiring training samples, wherein the training samples comprise building marking samples of a sample area and road marking samples of the sample area;
constructing an original detection model;
and respectively training the detection model through the building marking sample and the road marking sample to obtain a trained newly-added construction land prediction model.
In the implementation process, the building marking sample and the road marking sample are easy to obtain, the generalization capability of the model is favorably improved, and the dependence on the sample is reduced.
Further, the step of performing prediction processing on the satellite high-resolution image through a newly-added construction land prediction model which is constructed in advance to obtain a construction land heat map of the target detection area includes:
predicting the satellite high-resolution image through a newly-built construction land prediction model to obtain a road heat map of the target detection area and a house heat map of the target detection area;
and splicing and combining the road heat map and the house heat map to obtain a construction land heat map.
In the implementation process, the construction land heat map is obtained through the newly added construction land prediction model, and the improvement of the detection precision of the newly added construction land is facilitated.
Further, the detecting and processing of the newly added construction land for the land layer and the object layer of the satellite high-resolution image according to the preset earth surface coverage vector data to obtain a land layer detection result and an object layer detection result includes:
segmenting the satellite high-resolution image according to the preset earth surface covering vector data to obtain block layer data and object layer data;
carrying out feature extraction processing on the data of the land parcel layer to obtain the features of the land parcel layer;
performing parameter statistics on the characteristics of the land parcel layer to obtain a newly added construction land index system;
performing knowledge reasoning on the data of the land parcel layer according to the newly added construction land index system to obtain a land parcel layer detection result;
and carrying out construction land target detection on the object layer data according to the construction land heat map and the land parcel layer detection result to obtain an object layer detection result.
In the implementation process, the detection result of the land parcel layer provides a starting point and possibility of the newly added construction land, and the detection result of the object layer provides a terminal point and evidence, so that the overflow of changes caused by geometric registration errors of vectors and images can be eliminated.
Further, the determining a detection result of the target new construction land for the target detection area according to the fusion detection result includes:
denoising the fusion detection result to obtain a denoising detection result;
and generating a vector format target newly-added construction land detection result according to the denoising detection result.
In the implementation process, the denoising processing is performed on the fusion detection result, so that the precision of the target newly-added construction land detection result can be improved, and the error is reduced.
The second aspect of the embodiment of the present application provides a newly-increased construction land detection device, newly-increased construction land detection device includes:
the image acquisition unit is used for acquiring a satellite high-resolution image of a target detection area;
the model prediction unit is used for predicting the satellite high-resolution image through a newly-added construction land prediction model which is constructed in advance to obtain a construction land heat map of the target detection area;
the detection unit is used for detecting and processing newly added construction land for a land layer and an object layer of the satellite high-resolution image according to preset ground surface coverage vector data to obtain a land layer detection result and an object layer detection result;
the fusion unit is used for performing decision-level fusion processing on the land parcel layer detection result and the object layer detection result to obtain a fusion detection result;
and the determining unit is used for determining the detection result of the target newly-added construction land of the target detection area according to the fusion detection result.
In the implementation process, the image acquisition unit firstly acquires a satellite high-resolution image of a target detection area; the model prediction unit carries out prediction processing on the satellite high-resolution image through a newly-built construction land prediction model built in advance to obtain a construction land heat map of the target detection area; then, the detection unit detects and processes newly added construction land for a land layer and an object layer of the satellite high-resolution image according to preset ground surface coverage vector data to obtain a land layer detection result and an object layer detection result; further, the fusion unit performs decision-level fusion processing on the detection result of the land parcel layer and the detection result of the object layer to obtain a fusion detection result; and finally, the determining unit determines the detection result of the target new construction land in the target detection area according to the fusion detection result, so that the new construction land can be quickly and accurately detected, the error is small, and the applicability is good.
Further, the newly-added construction land detection device further includes:
the system comprises a sample acquisition unit, a detection unit and a control unit, wherein the sample acquisition unit is used for acquiring training samples before acquiring a satellite high-resolution image of a target detection area, and the training samples comprise building marking samples of a sample area and road marking samples of the sample area;
the construction unit is used for constructing an original detection model;
and the training unit is used for respectively training the detection model through the building marking sample and the road marking sample to obtain a trained newly-added construction land prediction model.
In the implementation process, the building marking sample and the road marking sample are easy to obtain, the generalization capability of the model is favorably improved, and the dependence on the sample is reduced.
Further, the model prediction unit includes:
the forecasting sub-unit is used for forecasting the satellite high-resolution images through a newly-built construction land forecasting model built in advance to obtain a road heat map of the target detection area and a house heat map of the target detection area;
and the splicing and merging subunit is used for splicing and merging the road heat map and the house heat map to obtain a construction geothermal map.
In the implementation process, the construction land heat map is obtained through the newly added construction land prediction model, and the improvement of the detection precision of the newly added construction land is facilitated.
A third aspect of the embodiments of the present application provides an electronic device, including a memory and a processor, where the memory is used to store a computer program, and the processor runs the computer program to make the electronic device execute the method for detecting a newly added construction site according to any one of the first aspect of the embodiments of the present application.
A fourth aspect of the embodiments of the present application provides a computer-readable storage medium, which stores computer program instructions, and when the computer program instructions are read and executed by a processor, the method for detecting a newly added construction land according to any one of the first aspect of the embodiments of the present application is executed.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
Fig. 1 is a schematic flow chart of a method for detecting a newly added construction land according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a method for detecting a newly added construction land according to a second embodiment of the present application;
fig. 3 is a schematic structural diagram of a newly added construction land detection device provided in the third embodiment of the present application;
fig. 4 is a schematic structural diagram of a newly added construction land detection device provided in the fourth embodiment of the present application;
fig. 5 is a schematic view of an image slice of a sample region according to a second embodiment of the present application;
fig. 6 is a schematic view of a road marking sample of a sample region according to a second embodiment of the present application;
FIG. 7 is a schematic view of a building annotation sample for a sample area provided in example two of the present application;
fig. 8 is a schematic grayscale image of an image slice of a satellite high-resolution image of a target detection area according to a second embodiment of the present application;
fig. 9 is a partial schematic view of a road heat map of a target detection area according to a second embodiment of the present application;
fig. 10 is a partial schematic view of a house heat map of a target detection area according to a second embodiment of the present application;
fig. 11 is a partial schematic view of a construction geothermal map of a target detection area according to a second embodiment of the present application;
fig. 12 is a schematic view of an automatic detection process for a newly added construction land according to a second embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Example 1
Referring to fig. 1, fig. 1 is a schematic flow chart of a method for detecting a newly added construction land according to an embodiment of the present application. The method for detecting the newly added construction land comprises the following steps:
s101, obtaining a satellite high-resolution image of a target detection area.
In this embodiment of the present application, the satellite high-resolution image of the target detection area may be a three-band or four-band high-resolution remote sensing image, and the like, which is not limited in this embodiment of the present application.
In the embodiment of the present application, the execution subject of the method may be a computing device such as a computer and a server, and is not limited in this embodiment.
In this embodiment, an execution subject of the method may also be an intelligent device such as a smart phone and a tablet computer, which is not limited in this embodiment.
S102, conducting prediction processing on the satellite high-resolution image through a newly-built construction land prediction model which is built in advance, and obtaining a construction land heat map of the target detection area.
In the embodiment of the application, the newly added construction land prediction model may be specifically a D-LinkNet semantic segmentation model, and the embodiment of the application is not limited.
As an optional implementation manner, the D-LinkNet semantic segmentation model adopts a Pytorch 1.7 deep learning framework under Windows, and uses Pycharm software to perform program training optimization, wherein the learning rate is set to 0.0002, the number of road iterations is 100, the number of building iterations is 150, the batch processing size is set to 4 for one GPU, and the two GPUs are 8 in total. The configuration of the training platform is as follows: a CPU: AMD Virtual CPU v53.70GHz; GPU: NVIDIA GeForce RTX 309024 GB.
In the embodiment of the present application, D-LinkNet uses LinkNet with precoders as its backbone network and has an additional hole convolution layer (scaled convolution) in the central part. Linknet is a semantic segmentation neural network, has the advantages of jump connection, a residual block and an encoder-decoder system structure, and has higher operation speed; the cavity convolution layer added on the basis of the D-LinkNet can increase the receptive field and reserve the multi-scale spatial characteristics of the remote sensing image on the premise of not reducing the image resolution. And respectively training a D-LinkNet deep learning semantic segmentation model by using house and road samples.
As an alternative implementation manner, before step S102, the building marking sample and the road marking sample in the sample area may be used as samples, and the original detection model is trained separately to obtain a trained new construction land prediction model.
In the embodiment of the application, the satellite high-resolution images are processed through the newly added construction land prediction model to obtain the building heat map and the road heat map of the target detection area, and then the building heat map and the road heat map are combined to obtain the construction land heat map of the target detection area.
After step S102, the method further includes the following steps:
s103, detecting and processing newly added construction land for a land layer and an object layer of the satellite high-resolution image according to preset ground surface coverage vector data to obtain a land layer detection result and an object layer detection result.
In the embodiment of the present application, the surface coverage vector data is specifically pre-registered, and may specifically be a national survey vector or a geographic national situation and land coverage vector, and the like, which is not limited in this embodiment of the present application.
In the embodiment of the application, in actual use, geometric registration of the earth surface coverage vector data and the satellite high-resolution image is only needed, requirements on sensor and time phase selection and preprocessing of a data source are low, and the method is suitable for large-range and high-precision business detection projects.
In the embodiment of the application, the satellite high-resolution image is a high-resolution remote sensing image, specifically can be a meter-level high-resolution satellite image, and can be a three-band or four-band satellite image.
And S104, performing decision-level fusion processing on the land parcel layer detection result and the object layer detection result to obtain a fusion detection result.
And S105, determining a target newly-added construction land detection result of the target detection area according to the fusion detection result.
In the embodiment of the application, the detection result of the target newly-added construction land comprises a target layer newly-added land and a plot layer newly-added land, 6 types of newly-added construction lands comprise a T1 cultivated land-T2 construction land, a T1 grassland-T2 construction land, a T1 forest land-T2 construction land, a T1 garden-T2 construction land, a T1 water body-T2 construction land and a T1 unused land-T2 construction land 6 types.
Therefore, the method for detecting the newly added construction land described in the embodiment can be used for quickly and accurately detecting the newly added construction land, and is small in error and good in applicability.
Example 2
Please refer to fig. 2, and fig. 2 is a schematic flow chart of a method for detecting a newly added construction land according to an embodiment of the present application. As shown in fig. 2, the method for detecting a newly added construction land includes:
s201, acquiring a satellite high-resolution image of a target detection area.
S202, performing prediction processing on the satellite high-resolution images through a newly-built construction land prediction model built in advance to obtain a road heat map of a target detection area and a house heat map of the target detection area.
As an optional implementation manner, before acquiring the satellite high-resolution image of the target detection area, the method may further include the following steps:
acquiring training samples, wherein the training samples comprise building marking samples of a sample area and road marking samples of the sample area;
constructing an original detection model;
and respectively training the detection model through the building labeling sample and the road labeling sample to obtain a trained newly-added construction land prediction model.
In the above embodiment, the building marking sample and the road marking sample in the sample area are used as samples, and the original detection model is trained respectively to obtain the trained new construction land prediction model.
Referring to fig. 5, fig. 6 and fig. 7 together, fig. 5 is a schematic image slice diagram of a sample region according to an embodiment of the present disclosure, fig. 6 is a schematic road marking sample diagram of a sample region according to an embodiment of the present disclosure, and fig. 7 is a schematic building marking sample diagram of a sample region according to an embodiment of the present disclosure.
In the embodiment, the building marking sample and the road marking sample are easier to obtain than a newly added construction land sample, the model generalization capability is strong, and the dependence on the sample is low.
In the above embodiment, the original detection model may specifically be a D-LinkNet semantic segmentation model, and the newly trained construction land prediction model may be obtained by training the D-LinkNet semantic segmentation model with the building object labeling sample and the road labeling sample, respectively.
In the above embodiment, the D-LinkNet semantic segmentation model adopts a Pytorch 1.7 deep learning framework under Windows, and performs program training optimization using Pycharm software, where the learning rate is set to 0.0002, the number of road iterations is 100, the number of building iterations is 150, the batch processing size is set to 4 for one GPU, and the number of two GPUs is 8 in total. The configuration of the training platform is as follows: a CPU: AMD Virtual CPU v53.70GHz; GPU: NVIDIA GeForce RTX 309024 GB.
In the above embodiment, the D-LinkNet uses LinkNet with a precoder as its backbone network, and has an additional hole convolution layer (partitioned convolution) in the central portion. Linknet is a semantic segmentation neural network, has the advantages of jump connection, a residual block and an encoder-decoder system structure, and has higher operation speed; the cavity convolution layer added on the basis of the D-LinkNet can increase the receptive field and reserve the multi-scale spatial characteristics of the remote sensing image on the premise of not reducing the image resolution. And respectively training a D-LinkNet deep learning semantic segmentation model by using house and road samples.
Referring to fig. 8, 9 and 10 together, fig. 8 is a schematic grayscale image of an image slice of a satellite true color image according to an embodiment of the present disclosure, fig. 9 is a partial schematic view of a road heat map according to an embodiment of the present disclosure, and fig. 10 is a partial schematic view of a house heat map according to an embodiment of the present disclosure.
After step S202, the following steps are also included:
and S203, splicing and combining the road heat map and the house heat map to obtain a construction land heat map.
Referring to fig. 11, fig. 11 is a partial schematic view of a construction geothermal map according to an embodiment of the present application.
In the embodiment of the application, by implementing the steps S202 to S203, the construction land heat map of the target detection area can be obtained by performing prediction processing on the satellite high-resolution images through a newly-built construction land prediction model constructed in advance.
And S204, segmenting the satellite high-resolution image according to preset earth surface covering vector data to obtain block layer data and object layer data.
In the embodiment of the application, the earth surface coverage vector data comprises vector data of the area to be detected.
In the embodiment of the application, the earth surface covering vector data is used as an initial condition and a control boundary, and the satellite high-resolution image is subjected to two different segmentations to obtain the data of the land block layer and the data of the object layer.
In the embodiment of the application, during the first segmentation, the satellite high-resolution image is not considered, and knowledge-based segmentation is completely carried out according to the earth surface covering vector data to obtain the earth block data.
In the embodiment of the present application, in the second segmentation, on the basis of the data of the land parcel level, the data is finely segmented in consideration of the spectrum, shape, texture, and the like of the satellite high-resolution image, so as to obtain the object level data with high homogeneity.
In the embodiment of the application, one land block in the land block data is composed of one to a plurality of objects, and the difference between the objects represents the change and the differentiation of the land block. An object is composed of a plurality of homogeneous pixels. Conversely, one pixel belongs to a unique object and one object belongs to a unique parcel. The method is characterized in that pixels are used as leaves, objects and plots are used as branches, scenes are used as roots, detection units with different pixels, objects, plots and scenes are combined by inverted tree structures to form a hierarchical structure, time and space fields are provided for automatic detection of newly added construction land, and fusion of feature levels and decision levels of different detection units is facilitated.
And S205, performing feature extraction processing on the data of the land parcel layer to obtain the features of the land parcel layer.
And S206, carrying out parameter statistics on the characteristics of the land parcel layer to obtain a newly added construction land index system.
According to the embodiment of the application, according to the comprehensive performance of rough texture, discrete gray scale and high brightness of the construction land in the image space and the feature space, the feature extraction and parameter statistics of the land parcel layer are carried out according to the statistical principle in a classified mode, and a self-adaptive index system for the newly added construction land of the land parcel layer is constructed.
In the embodiment of the present application, the common feature of the parcel level may be extracted from two angles of the pixel level and the object level, the statistical features of the pixel level such as the blue band standard deviation (stddv B), the red band standard deviation (stddv R) and the brightness (Intensity), and the statistical features of the object level such as the sub-object red band standard deviation (stddv SO R), and the like, which are not limited in this embodiment.
In the embodiment of the present application, the parameter statistics of the features are performed in a classified manner, and the mean (m) and the standard deviation (σ) of each feature are respectively counted2). One standard deviation (m + σ) from the mean distance2) And (4) intercepting the threshold value and storing the threshold value into an index variable Px (x =1, 2, 3, …, n) to form an adaptive index system for detecting the newly added construction land of the land parcel layer (namely, the index system for the newly added construction land).
In the embodiment of the application, in the image feature space, the difference between individuals of the same land feature value is small (aggregation distribution), and once the land utilization manner in the land is partially or completely changed, the corresponding features of the land will have a significant deviation (discrete distribution) compared with the statistical features of the whole land. When non-construction lands such as cultivated lands, grasslands, gardens, woodlands, water bodies, unused lands, and the like become the areas for increasing construction, the spectrum discretization among pixels in the plot is firstly shown, and the spectrum difference among the objects constituting the plot is increased. In addition, the non-construction land is changed to a construction land, and generally, the humidity is reduced and the brightness is also significantly increased.
In the embodiment of the application, based on the comprehensive performance of the construction land in the image space and the feature space, based on the statistical principle, the feature extraction is carried out by the land classification, the parameters such as the mean value, the standard deviation and the like are counted, and a newly added construction land index system is established.
After step S206, the following steps are also included:
and S207, carrying out knowledge reasoning on the data of the land parcel layer according to the newly added land index system for construction to obtain a land parcel layer detection result.
As an optional implementation mode, a newly added construction land index system is used for carrying out knowledge reasoning and locking the potential change plots. The specific method is that the system adopts If … then statement to construct a knowledge algorithm, such as If StdDev B > P0 or StdDev R > P1 or Stdev SO R > P2 or Intensity > P3 then potentially changing plot in the farmland plot range. Meanwhile, the interference of 'pseudo change' in a geothermal map of the construction land in the next step is eliminated, and particularly the situation that the highlighted ridge in the non-construction land is wrongly divided into the construction land is avoided.
In the above embodiment, the potentially changing plots can be classified into the plot layers by the above method and locked by knowledge inference, so as to obtain the plot layer detection result.
In the above embodiment, in actual use, when the local block vector boundary overflows to a surrounding road, a "pseudo-change" is caused, which can be excluded by a knowledge algorithm in which the construction site is distributed near the block boundary and has a long and narrow shape.
After step S207, the following steps are also included:
and S208, carrying out construction land target detection on the object layer data according to the construction land heat map and the land parcel layer detection result to obtain an object layer detection result.
As an alternative embodiment, the potential change plots in the plot layer detection result are used as control boundaries, and the object that becomes the construction land is detected within the control boundaries, so as to obtain the object layer detection result. Specifically, in the range of the potential change plots, the construction land heat map is taken as a characteristic, and the objects with the construction land heat occupation ratio larger than a preset threshold (for example, 20%) in the detection sub-objects are taken as newly-added construction land objects, so as to obtain the object layer detection result.
In the embodiment of the application, by implementing the step S208, the pixel-level deep learning result can be fused into the object aiming at the defect of the newly added construction land prediction model, and the object boundary is used for replacing the deep learning semantic segmentation boundary, so that the cavity and the isolated point are eliminated, and the boundary is more regular.
In the embodiment of the present application, by implementing the steps S204 to S208, the newly added construction land detection processing of the land layer and the object layer can be performed on the satellite high-resolution image according to the preset ground surface coverage vector data, so as to obtain a land layer detection result and an object layer detection result.
S209, performing decision-level fusion processing on the land parcel layer detection result and the object layer detection result to obtain a fusion detection result.
In the embodiment of the application, the potential change plots in the plot layer detection result provide a starting point and possibility of a newly added construction land, the newly added construction land object in the object layer detection result provides a terminal point and evidence, and change overflow caused by geometric registration errors of vectors and images is eliminated according to the shape of the object layer construction land and the distribution position in the plot layer.
In the embodiment of the application, the detection result of the land parcel layer and the detection result of the object layer are subjected to decision-level fusion through relation characteristics and Boolean operation according to related industrial standards, and the final newly-added construction land for the land parcel layer and the newly-added construction land for the object layer are obtained.
In the embodiment of the application, the fusion detection result comprises the land for the newly added construction of the parcel level and the newly added construction land for the object level.
In the embodiment of the application, the fusion detection result may specifically include 6 major categories of T1 cultivated land-T2 construction land, T1 grassland-T2 construction land, T1 forest land-T2 construction land, T1 garden land-T2 construction land, T1 water body-T2 construction land, and T1 unused land-T2 construction land.
After step S209, the following steps are also included:
s210, denoising the fusion detection result to obtain a denoising detection result.
In the embodiment of the application, the land types of the newly-added construction land used for the land parcel layer and the newly-added construction land used for the object layer are merged, the denoising processing is performed on the small land parcel or the object, and the like, so that the upper map area of the land parcel reaches the preset area threshold, and specifically, the preset area threshold can be the minimum upper map area of the construction land of 100m2Minimum upper drawing area of 400m for agricultural land2Minimum unused area 600m2This embodiment is not limited at all.
And S211, generating a vector format target newly-added construction land detection result according to the denoising detection result.
In the embodiment of the application, the denoising detection result, the index and the change type are output in a vector format.
In the embodiment of the present application, the denoised embodiment of the present application is output in a vector format together with the variation type and the diagnostic index. The newly added construction land of the land parcel layer can be directly hung on the earth surface coverage vector data to dynamically update the vector database. The newly added construction land of the object layer provides a specific change position reference for updating the vector data.
In the embodiment of the application, by implementing the steps S210 to S211, the detection result of the target new construction land of the target detection area can be determined according to the fusion detection result.
Referring to fig. 12, fig. 12 is a schematic view of an automatic detection process of a new construction land according to an embodiment of the present application, for example, taking a target detection area B as an example, the detection of the new construction land for B may include the following steps:
the first step is as follows: and acquiring a satellite high-resolution image of the B place.
The second step is that: and (3) predicting the satellite high-resolution image of the B place by using a newly-added construction land prediction model trained in advance to obtain a road heat map and a house heat map (an example is shown in the following figure 6), and splicing and combining the road heat map and the house heat map to obtain a construction land heat map of the B place.
The third step: and acquiring preset earth surface coverage vector data (namely a T1 vector), and dividing the satellite high-resolution image (namely the T2 high-resolution image) twice by taking the T1 vector as an initial condition and a control boundary. The high-resolution image of T2 is not considered for the first time, and knowledge-based segmentation is completely carried out according to the vector boundary of T1 to obtain the data of the plot layer; in the second segmentation, the object layer data with relatively high internal homogeneity is obtained by performing data-based sub-segmentation in consideration of the spectral, shape, and texture characteristics of the T2 video image, on the basis of the block layer data. The method is characterized in that pixels are used as leaves, objects and plots are used as branches, scenes are used as roots, different detection units of the pixels, the objects, the plots and the scenes are organized by using an inverted tree structure to form a time and space field for detecting newly added construction land, and feature level and decision level fusion of different detection units is facilitated.
The fourth step: and respectively extracting the characteristics of the land parcel layers of the land parcel data from cultivated land, grassland, forest land, garden land, water body and unused land, and carrying out parameter statistics such as the mean value, standard deviation and the like of each characteristic to form a newly added construction land index system.
The fifth step: and (3) judging the land parcel layer data according to the newly added construction land index system by land classification based on knowledge reasoning, locking the potential newly added construction land parcels to obtain a land parcel layer detection result, and aiming at eliminating pseudo-change interference of some high-brightness ridges in the construction land heat map.
And a sixth step: and judging that the construction geothermal map accounts for more than 20% of the object layer data as the construction geothermal object by taking the construction geothermal map as a characteristic, and further obtaining an object layer detection result. Aiming at the defect of fuzzy boundary of the deep learning result, the deep learning heat map is fused into the object boundary, isolated points and holes in the deep learning result are eliminated, and the boundary is more regular.
The seventh step: and according to the distribution position and shape of the object layer detection result in the block layer detection result, eliminating the change overflow caused by vector and image registration error, and performing decision-level fusion processing on the block layer detection result and the object layer detection result to obtain a fusion detection result.
Eighth step: and (3) carrying out noise reduction treatment on the small-area object in the fusion detection result to obtain a denoising detection result, and outputting the denoising detection result together with the index and the change type (figure 7) in a vector format to obtain a target newly-added construction land detection result. The block layer new construction land vector can be hung to the T1 vector to perform vector dynamic update. The newly added construction land of the object layer can provide specific change position reference for interior and exterior business inspection.
In the embodiment of the application, knowledge reasoning and deep learning are combined, the cognition and perception capability of human is simulated, and the detection efficiency is improved. The T1 vector is used as an initial condition and a control boundary, and the newly added construction land detection is carried out in a classified mode, so that the complexity of the problem is simplified, the uncertainty of the remote sensing image is reduced, the detection precision, the reliability and the automation level are improved, all the 6 types of detection results of the block layer and the object layer of the newly added construction land can be obtained, and the requirements of natural resource management services are met.
In the embodiment of the application, the method provides a change detection framework of multi-source data integration, knowledge guidance and overall analysis. Based on the T1 vector and the T2 image, the traditional knowledge reasoning and the emerging deep learning semantic segmentation are combined, different detection units of pixels, objects, plots and scenes are comprehensively judged, and the newly added construction land and the change types of the construction land of the plot layer and the object layer are obtained.
In the embodiment of the application, the method can combine traditional knowledge reasoning with emerging deep learning, simulate the cognitive and perception abilities of human beings, improve the detection efficiency, automatically detect the newly added construction land and judge the change type of the newly added construction land based on the T2 image and the T1 vector, and provide technical support for homeland space planning and natural resource management.
In the embodiment of the application, the method can use the T1 vector as an initial condition and a control boundary, reduce the uncertainty of the remote sensing image and the complexity of the change detection problem, improve the detection precision, reliability and automation level, obtain the detection results of all 6 types of target newly-added construction land, and better meet the requirements of natural resource management services.
Therefore, the method for detecting the newly added construction land described in the embodiment can be used for quickly and accurately detecting the newly added construction land, and is small in error and good in applicability.
Example 3
Please refer to fig. 3, and fig. 3 is a schematic structural diagram of a new construction land detection device according to an embodiment of the present application. As shown in fig. 3, the new construction land detection apparatus includes:
an image obtaining unit 310, configured to obtain a satellite high-resolution image of a target detection area;
the model prediction unit 320 is used for performing prediction processing on the satellite high-resolution images through a newly-added construction land prediction model which is constructed in advance to obtain a construction land heat map of the target detection area;
the detection unit 330 is used for detecting and processing newly added construction land for a land layer and an object layer of the satellite high-resolution image according to preset ground surface coverage vector data to obtain a land layer detection result and an object layer detection result;
the fusion unit 340 is configured to perform decision-level fusion processing on the detection result of the parcel layer and the detection result of the object layer to obtain a fusion detection result;
and a determining unit 350, configured to determine a detection result of the target new construction land of the target detection area according to the fusion detection result.
In the embodiment of the present application, for explanation of the new construction land detection device, reference may be made to the description in embodiment 1 or embodiment 2, and details are not repeated in this embodiment.
Therefore, the newly added construction land detection device described in the embodiment can quickly and accurately detect the newly added construction land, and has the advantages of small error and good applicability.
Example 4
Referring to fig. 4, fig. 4 is a schematic structural diagram of a new construction land detection device according to an embodiment of the present application. The new construction land detection device shown in fig. 4 is optimized by the new construction land detection device shown in fig. 3. As shown in fig. 4, the new construction land detection apparatus further includes:
the sample acquiring unit 360 is configured to acquire a training sample before acquiring the satellite high-resolution image of the target detection area, where the training sample includes a building labeling sample of the sample area and a road labeling sample of the sample area.
A construction unit 370 for constructing the original detection model.
And the training unit 380 is used for respectively training the detection model through the building labeling sample and the road labeling sample to obtain a trained newly-added construction land prediction model.
As an alternative embodiment, the model prediction unit 320 includes:
and the predicting subunit 321 is configured to perform prediction processing on the satellite high-resolution image through a newly-built construction land prediction model, so as to obtain a road heat map of the target detection area and a house heat map of the target detection area.
And the splicing and merging subunit 322 is configured to splice and merge the road heat map and the house heat map to obtain a construction geothermal map.
As an alternative embodiment, the detection unit 330 includes:
the segmentation subunit 331 is configured to segment the satellite high-resolution image according to preset earth surface coverage vector data to obtain a block layer data and an object layer data;
an extraction subunit 332, configured to perform feature extraction processing on the data of the geological formation to obtain characteristics of the geological formation;
a statistics subunit 333, configured to perform parameter statistics on the feature of the land parcel layer to obtain a newly added construction land index system;
the reasoning subunit 334 is configured to perform knowledge reasoning on the data of the parcel level according to the newly added construction land index system to obtain a parcel level detection result;
and the detection subunit 335 is configured to perform construction land target detection on the object layer data according to the construction land heat map and the parcel layer detection result, so as to obtain an object layer detection result.
As an alternative embodiment, the fusion unit 340 includes:
and the denoising subunit 341 is configured to perform denoising processing on the fusion detection result to obtain a denoising detection result.
The generating subunit 342 is configured to generate a detection result of the newly added target construction land in the vector format according to the denoising detection result.
In the embodiment of the present application, for explanation of the new construction land detection device, reference may be made to the description in embodiment 1 or embodiment 2, and details are not repeated in this embodiment.
Therefore, the newly added construction land detection device described in the embodiment can quickly and accurately detect the newly added construction land, and has the advantages of small error and good applicability.
The embodiment of the application provides electronic equipment, which comprises a memory and a processor, wherein the memory is used for storing a computer program, and the processor runs the computer program to enable the electronic equipment to execute the method for detecting the newly added construction land in any one of embodiment 1 or embodiment 2 of the application.
An embodiment of the present application provides a computer-readable storage medium, which stores computer program instructions, and when the computer program instructions are read and executed by a processor, the method for detecting a newly added construction site in any one of embodiments 1 and 2 of the present application is executed.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. A method for detecting newly added construction land is characterized by comprising the following steps:
acquiring a satellite high-resolution image of a target detection area;
predicting the satellite high-resolution image through a newly-built construction land prediction model to obtain a construction land heat map of the target detection area;
carrying out detection processing on newly added construction land of a land layer and an object layer on the satellite high-resolution image according to preset ground surface coverage vector data to obtain a land layer detection result and an object layer detection result;
performing decision-level fusion processing on the detection result of the land layer and the detection result of the object layer to obtain a fusion detection result;
and determining a detection result of the target newly added construction land of the target detection area according to the fusion detection result.
2. The method for detecting a newly added construction land according to claim 1, wherein before the step of obtaining the satellite high-resolution image of the target detection area, the method further comprises:
acquiring training samples, wherein the training samples comprise building marking samples of a sample area and road marking samples of the sample area;
constructing an original detection model;
and respectively training the detection model through the building marking sample and the road marking sample to obtain a trained newly-added construction land prediction model.
3. The method for detecting newly added construction land according to claim 1, wherein the obtaining of the construction land heat map of the target detection area by performing prediction processing on the satellite high-resolution image through a newly added construction land prediction model constructed in advance comprises:
predicting the satellite high-resolution image through a newly-built construction land prediction model to obtain a road heat map of the target detection area and a house heat map of the target detection area;
and splicing and combining the road heat map and the house heat map to obtain a construction land heat map.
4. The method for detecting a newly added construction land according to claim 1, wherein the step of performing a detection process of the newly added construction land on a land layer and an object layer on the satellite high-resolution image according to preset earth surface coverage vector data to obtain a land layer detection result and an object layer detection result comprises:
segmenting the satellite high-resolution image according to the preset earth surface covering vector data to obtain block layer data and object layer data;
carrying out feature extraction processing on the data of the land parcel layer to obtain the features of the land parcel layer;
performing parameter statistics on the characteristics of the land parcel layer to obtain a newly added construction land index system;
performing knowledge reasoning on the data of the land parcel layer according to the newly added construction land index system to obtain a land parcel layer detection result;
and carrying out construction land target detection on the object layer data according to the construction land heat map and the land parcel layer detection result to obtain an object layer detection result.
5. The method for detecting a newly added construction land according to claim 1, wherein the determining the detection result of the target newly added construction land for the target detection area based on the fusion detection result comprises:
denoising the fusion detection result to obtain a denoising detection result;
and generating a vector format target newly-added construction land detection result according to the denoising detection result.
6. The utility model provides a newly-increased construction land detection device which characterized in that, newly-increased construction land detection device includes:
the image acquisition unit is used for acquiring a satellite high-resolution image of a target detection area;
the model prediction unit is used for predicting the satellite high-resolution image through a newly-added construction land prediction model which is constructed in advance to obtain a construction land heat map of the target detection area;
the detection unit is used for detecting and processing newly added construction land for a land layer and an object layer of the satellite high-resolution image according to preset ground surface coverage vector data to obtain a land layer detection result and an object layer detection result;
the fusion unit is used for performing decision-level fusion processing on the land parcel layer detection result and the object layer detection result to obtain a fusion detection result;
and the determining unit is used for determining the detection result of the target newly-added construction land of the target detection area according to the fusion detection result.
7. The newly added construction land detection device according to claim 6, further comprising:
the system comprises a sample acquisition unit, a detection unit and a control unit, wherein the sample acquisition unit is used for acquiring training samples before acquiring a satellite high-resolution image of a target detection area, and the training samples comprise building marking samples of a sample area and road marking samples of the sample area;
the construction unit is used for constructing an original detection model;
and the training unit is used for respectively training the detection model through the building marking sample and the road marking sample to obtain a trained newly-added construction land prediction model.
8. The newly added construction land detection apparatus according to claim 6, wherein the model prediction unit includes:
the forecasting sub-unit is used for forecasting the satellite high-resolution images through a newly-built construction land forecasting model built in advance to obtain a road heat map of the target detection area and a house heat map of the target detection area;
and the splicing and merging subunit is used for splicing and merging the road heat map and the house heat map to obtain a construction geothermal map.
9. An electronic device, characterized in that the electronic device comprises a memory for storing a computer program and a processor for operating the computer program to make the electronic device execute the new construction land detection method according to any one of claims 1 to 5.
10. A readable storage medium, in which computer program instructions are stored, which, when read and executed by a processor, perform the newly added construction land detection method according to any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110543709.6A CN112967286B (en) | 2021-05-19 | 2021-05-19 | Method and device for detecting newly added construction land |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110543709.6A CN112967286B (en) | 2021-05-19 | 2021-05-19 | Method and device for detecting newly added construction land |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112967286A true CN112967286A (en) | 2021-06-15 |
CN112967286B CN112967286B (en) | 2021-09-17 |
Family
ID=76275628
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110543709.6A Active CN112967286B (en) | 2021-05-19 | 2021-05-19 | Method and device for detecting newly added construction land |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112967286B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113780057A (en) * | 2021-07-23 | 2021-12-10 | 贵州图智信息技术有限公司 | Idle land identification method and device |
CN113807301A (en) * | 2021-09-26 | 2021-12-17 | 武汉汉达瑞科技有限公司 | Automatic extraction method and automatic extraction system for newly-added construction land |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102855487A (en) * | 2012-08-27 | 2013-01-02 | 南京大学 | Method for automatically extracting newly added construction land change image spot of high-resolution remote sensing image |
CN103279954A (en) * | 2013-05-21 | 2013-09-04 | 武汉中测晟图遥感技术有限公司 | Remote-sensing image change detecting method based on land utilization database |
CN106384081A (en) * | 2016-08-30 | 2017-02-08 | 水利部水土保持监测中心 | Slope farmland extracting method and system based on high-resolution remote sensing image |
CN108776772A (en) * | 2018-05-02 | 2018-11-09 | 北京佳格天地科技有限公司 | Across the time building variation detection modeling method of one kind and detection device, method and storage medium |
CN109377480A (en) * | 2018-09-27 | 2019-02-22 | 中国电子科技集团公司第五十四研究所 | Arable land use change detection method based on deep learning |
US20190213414A1 (en) * | 2018-01-11 | 2019-07-11 | Intelinair, Inc | Row Detection System |
WO2020151528A1 (en) * | 2019-01-25 | 2020-07-30 | 东南大学 | Urban land automatic identification system integrating industrial big data and building forms |
CN111598048A (en) * | 2020-05-31 | 2020-08-28 | 中国科学院地理科学与资源研究所 | Urban village-in-village identification method integrating high-resolution remote sensing image and street view image |
CN112101325A (en) * | 2020-11-18 | 2020-12-18 | 航天宏图信息技术股份有限公司 | Method and device for detecting farmland change, electronic equipment and storage medium |
CN112183416A (en) * | 2020-09-30 | 2021-01-05 | 北京吉威数源信息技术有限公司 | Automatic extraction method of newly added construction land based on deep learning method |
CN112270291A (en) * | 2020-11-11 | 2021-01-26 | 中山大学 | Automatic detection method for illegal construction land development based on multi-source optical remote sensing image |
-
2021
- 2021-05-19 CN CN202110543709.6A patent/CN112967286B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102855487A (en) * | 2012-08-27 | 2013-01-02 | 南京大学 | Method for automatically extracting newly added construction land change image spot of high-resolution remote sensing image |
CN103279954A (en) * | 2013-05-21 | 2013-09-04 | 武汉中测晟图遥感技术有限公司 | Remote-sensing image change detecting method based on land utilization database |
CN106384081A (en) * | 2016-08-30 | 2017-02-08 | 水利部水土保持监测中心 | Slope farmland extracting method and system based on high-resolution remote sensing image |
US20190213414A1 (en) * | 2018-01-11 | 2019-07-11 | Intelinair, Inc | Row Detection System |
CN108776772A (en) * | 2018-05-02 | 2018-11-09 | 北京佳格天地科技有限公司 | Across the time building variation detection modeling method of one kind and detection device, method and storage medium |
CN109377480A (en) * | 2018-09-27 | 2019-02-22 | 中国电子科技集团公司第五十四研究所 | Arable land use change detection method based on deep learning |
WO2020151528A1 (en) * | 2019-01-25 | 2020-07-30 | 东南大学 | Urban land automatic identification system integrating industrial big data and building forms |
CN111598048A (en) * | 2020-05-31 | 2020-08-28 | 中国科学院地理科学与资源研究所 | Urban village-in-village identification method integrating high-resolution remote sensing image and street view image |
CN112183416A (en) * | 2020-09-30 | 2021-01-05 | 北京吉威数源信息技术有限公司 | Automatic extraction method of newly added construction land based on deep learning method |
CN112270291A (en) * | 2020-11-11 | 2021-01-26 | 中山大学 | Automatic detection method for illegal construction land development based on multi-source optical remote sensing image |
CN112101325A (en) * | 2020-11-18 | 2020-12-18 | 航天宏图信息技术股份有限公司 | Method and device for detecting farmland change, electronic equipment and storage medium |
Non-Patent Citations (4)
Title |
---|
GUOYIN C.等: "Detection of land cover and thermal environment change in Beijing from TM images", 《2013 21ST INTERNATIONAL CONFERENCE ON GEOINFORMATICS》 * |
张泽瑞 等: "基于深度学习与多源遥感数据的新增建设用地自动检测", 《中山大学学报(自然科学版)》 * |
范树印 等: "《土地整治遥感监测技术方法与实践》", 30 April 2016, 北京:地质出版社 * |
马文康: "基于深度神经网络的遥感图像道路提取", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113780057A (en) * | 2021-07-23 | 2021-12-10 | 贵州图智信息技术有限公司 | Idle land identification method and device |
CN113807301A (en) * | 2021-09-26 | 2021-12-17 | 武汉汉达瑞科技有限公司 | Automatic extraction method and automatic extraction system for newly-added construction land |
CN113807301B (en) * | 2021-09-26 | 2024-06-07 | 武汉汉达瑞科技有限公司 | Automatic extraction method and automatic extraction system for newly-added construction land |
Also Published As
Publication number | Publication date |
---|---|
CN112967286B (en) | 2021-09-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhang et al. | A feature difference convolutional neural network-based change detection method | |
Lv et al. | Novel land cover change detection method based on K-means clustering and adaptive majority voting using bitemporal remote sensing images | |
WO2021258758A1 (en) | Coastline change identification method based on multiple factors | |
CN111986099B (en) | Tillage monitoring method and system based on convolutional neural network with residual error correction fused | |
US11521380B2 (en) | Shadow and cloud masking for remote sensing images in agriculture applications using a multilayer perceptron | |
Yu et al. | Context-based hierarchical unequal merging for SAR image segmentation | |
CN111028255B (en) | Farmland area pre-screening method and device based on priori information and deep learning | |
CN112967286B (en) | Method and device for detecting newly added construction land | |
CN112950780B (en) | Intelligent network map generation method and system based on remote sensing image | |
Tehrani et al. | Multi-regional landslide detection using combined unsupervised and supervised machine learning | |
Zaabar et al. | Application of convolutional neural networks with object-based image analysis for land cover and land use mapping in coastal areas: A case study in Ain Témouchent, Algeria | |
Chen et al. | Object-based multi-modal convolution neural networks for building extraction using panchromatic and multispectral imagery | |
Karydas | Optimization of multi-scale segmentation of satellite imagery using fractal geometry | |
CN116343048A (en) | Accurate land block boundary extraction method and system for plain crop type complex region | |
Zhou et al. | Individual tree crown segmentation based on aerial image using superpixel and topological features | |
CN114596489A (en) | High-precision multisource remote sensing city built-up area extraction method for human habitation index | |
CN115546671A (en) | Unmanned aerial vehicle change detection method and system based on multitask learning | |
CN115019163A (en) | City factor identification method based on multi-source big data | |
Chandra et al. | Human cognition based framework for detecting roads from remote sensing images | |
Ran et al. | The status and influencing factors of surface water dynamics on the Qinghai-Tibet Plateau during 2000–2020 | |
CN112465821A (en) | Multi-scale pest image detection method based on boundary key point perception | |
Zhao et al. | Improving object-oriented land use/cover classification from high resolution imagery by spectral similarity-based post-classification | |
CN117075138A (en) | Remote sensing measurement and calculation method, system and medium for canopy height of 30-meter forest in area | |
CN114431124B (en) | Intelligent spraying control method for agricultural planting | |
Patel et al. | Road Network Extraction Methods from Remote Sensing Images: A Review Paper. |
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 |