CN113313006A - Urban illegal construction supervision method and system based on unmanned aerial vehicle and storage medium - Google Patents
Urban illegal construction supervision method and system based on unmanned aerial vehicle and storage medium Download PDFInfo
- Publication number
- CN113313006A CN113313006A CN202110572819.5A CN202110572819A CN113313006A CN 113313006 A CN113313006 A CN 113313006A CN 202110572819 A CN202110572819 A CN 202110572819A CN 113313006 A CN113313006 A CN 113313006A
- Authority
- CN
- China
- Prior art keywords
- building
- supervision
- image
- supervised
- area
- 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
- 238000000034 method Methods 0.000 title claims abstract description 72
- 238000010276 construction Methods 0.000 title claims abstract description 24
- 238000003860 storage Methods 0.000 title claims abstract description 11
- 239000013598 vector Substances 0.000 claims abstract description 89
- 238000012545 processing Methods 0.000 claims abstract description 43
- 238000000605 extraction Methods 0.000 claims abstract description 29
- 239000010410 layer Substances 0.000 claims description 50
- 230000011218 segmentation Effects 0.000 claims description 27
- 238000012937 correction Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 239000011229 interlayer Substances 0.000 claims description 3
- 238000012806 monitoring device Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 11
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 13
- 238000012544 monitoring process Methods 0.000 description 12
- 238000005516 engineering process Methods 0.000 description 6
- 238000012800 visualization Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 102220621241 Proline-rich membrane anchor 1_S32A_mutation Human genes 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/176—Urban or other man-made structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Image Processing (AREA)
- Alarm Systems (AREA)
Abstract
The invention relates to a city illegal building supervision method, a system and a storage medium based on an unmanned aerial vehicle, which comprises the steps of utilizing the unmanned aerial vehicle to carry out aerial photography on an area to be supervised periodically to obtain a aerial photo set of the area to be supervised; processing the aerial photo set to obtain a slice image set of an area to be supervised; the aerial photo set comprises original aerial images of an area to be supervised at a plurality of supervision moments respectively; performing feature extraction on the slice image set to obtain and display a vector feature layer of the region to be supervised at each supervision time; and judging whether illegal buildings exist in the area to be supervised according to the vector feature map layers at all supervision moments, and sending illegal building prompt information and displaying when the illegal buildings exist in the area to be supervised. The invention automatically judges whether the illegal construction exists based on the intelligent identification of the unmanned aerial vehicle and the image, generates and displays the illegal construction prompt message in time, has short discovery period, high efficiency, low labor cost and good supervision effect, and is beneficial to the orderly planning of cities.
Description
Technical Field
The invention relates to the technical field of city supervision, in particular to a city illegal building supervision method and system based on an unmanned aerial vehicle and a storage medium.
Background
Along with the continuous development of economic society of China, the urbanization process is accelerated, urban buildings are increased continuously, the number and the scale of illegal buildings are increased continuously, the phenomenon destroys urban planning and urban landscape, influences urban image and resident life, and is one of the core problems of urban management. The illegal buildings are found in time, and the illegal buildings are monitored, so that the system has important significance for city development.
At present, the monitoring aspect of illegal buildings is weak, an automatic monitoring means is lacked, discovery and field confirmation are mainly reported by people, but uncontrollable factors of the means are more, the discovery period is long, the workload of workers is large, the monitoring effect is poor, and the illegal buildings are still forbidden frequently. The purpose of monitoring illegal construction is achieved by processing and identifying the remote sensing monitoring image, but the method still has the defects that the image processing and identifying effect is poor and the identifying result cannot be notified and displayed in time, so that the higher and higher monitoring requirements of people on urban illegal construction cannot be met.
Disclosure of Invention
The invention aims to solve the technical problems of the prior art and provides a city illegal construction supervision method, a system and a storage medium based on an unmanned aerial vehicle.
The technical scheme for solving the technical problems is as follows:
an unmanned aerial vehicle-based city illegal construction supervision method comprises the following steps:
step 1: the method comprises the steps that an unmanned aerial vehicle is used for carrying out aerial photography on an area to be supervised periodically to obtain an aerial photo set of the area to be supervised; processing the aerial photo set to obtain a slice image set of the area to be supervised; the aerial photo set comprises original aerial images of the area to be supervised at a plurality of supervision moments respectively;
step 2: performing feature extraction on the slice image set to obtain and display a vector feature layer of the region to be supervised at each supervision time;
and step 3: and judging whether illegal buildings exist in the area to be supervised according to the vector feature map layers at all supervision moments, and sending illegal building prompt information and displaying when illegal buildings exist in the area to be supervised.
According to another aspect of the invention, the invention also provides an unmanned aerial vehicle-based city illegal building supervision system, which is applied to the unmanned aerial vehicle-based city illegal building supervision method, and comprises an unmanned aerial vehicle and a supervision device, wherein the unmanned aerial vehicle and the supervision device are in communication connection through a wireless network;
the monitoring device comprises an aerial photography module, a processing module, an extraction module, a judgment module and a display module;
the aerial photography module is used for utilizing the unmanned aerial vehicle to carry out aerial photography on the area to be supervised periodically to obtain an aerial photo set of the area to be supervised; the aerial photo set comprises original aerial images of the area to be supervised at a plurality of supervision moments respectively;
the processing module is used for processing the aerial photo set to obtain a slice image set of the area to be supervised;
the extraction module is used for extracting the features of the slice image set to obtain a vector feature map layer of the region to be supervised at each supervision time;
the judging module is used for judging whether illegal buildings exist in the area to be supervised according to the vector feature map layers at all supervision moments, and sending illegal building prompt information when the illegal buildings exist in the area to be supervised;
the display module is used for displaying the violation prompt information and the vector feature map layer of the area to be supervised at each supervision moment.
According to another aspect of the present invention, there is provided a city violation supervision system based on unmanned aerial vehicles, comprising a processor, a memory and a computer program stored in the memory and executable on the processor, wherein the computer program realizes the steps of the city violation supervision method based on unmanned aerial vehicles in the present invention when running.
In accordance with another aspect of the present invention, there is provided a computer storage medium comprising: at least one instruction which, when executed, implements the steps in the drone-based city violation supervision method of the present invention.
The city illegal building supervision method, system and storage medium based on the unmanned aerial vehicle have the advantages that: the method has the advantages that the original aerial images of the area to be supervised at different supervision moments can be obtained by utilizing the regular aerial photography of the unmanned aerial vehicle, the difference of each supervision moment can be conveniently compared through the intelligent identification of subsequent images, whether illegal buildings exist in the area to be supervised is further conveniently judged, the intelligent degree is high, and the realization difficulty is low; the intelligent recognition of the image comprises a processing process, a feature extraction process and a judgment process of the image, wherein a navigation sheet set is converted into a slice image set in the processing process of the image, so that the subsequent feature extraction is more refined, the obtained vector feature image layer is more consistent with a real scene, and the intelligent recognition effect is further improved; in the feature extraction process, the vector feature map layers at all supervision moments are obtained and displayed, so that the building model in the area to be supervised can be displayed better, the subsequent judgment process is facilitated, and the accuracy of illegal supervision is improved; finally, comparing, analyzing and judging according to the vector characteristic map layers at all supervision moments, realizing efficient intelligent supervision of illegal building phenomena of the area to be supervised, and sending illegal building prompt information and displaying when illegal buildings exist in the area to achieve the purpose of timely informing related personnel of timely intervening; the vector feature map layers at all supervision moments and the final illegal construction prompt information are displayed in the intelligent identification process, so that the whole supervision process can be conveniently visualized and displayed in a diversified manner, and the intelligent supervision effect of urban illegal construction is improved;
according to the city illegal construction supervision method, system and medium based on the unmanned aerial vehicle, intelligent identification is carried out based on the unmanned aerial vehicle and the image, whether illegal construction exists or not is automatically judged according to the result of the intelligent identification, illegal construction prompt messages are generated and displayed in time, the purposes of timely notification and display are achieved, the discovery period is short, the efficiency is high, the labor cost is low, the supervision effect is good, and orderly planning of cities is facilitated.
Drawings
Fig. 1 is a schematic flow chart of a city illegal building supervision method based on an unmanned aerial vehicle according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a process of obtaining a slice image set according to an embodiment of the present invention;
fig. 3 is a schematic flow chart illustrating a process of obtaining and displaying a vector feature layer at each supervision time in the first embodiment of the present invention;
FIG. 4 is a schematic flow chart illustrating a process of determining whether an illegal building exists in an area to be supervised according to a first embodiment of the present invention;
fig. 5 is a schematic structural diagram of a city illegal building supervision system based on an unmanned aerial vehicle in the second embodiment of the present invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
The present invention will be described with reference to the accompanying drawings.
In an embodiment, as shown in fig. 1, a city illegal building supervision method based on an unmanned aerial vehicle includes the following steps:
s1: the method comprises the steps that an unmanned aerial vehicle is used for carrying out aerial photography on an area to be supervised periodically to obtain an aerial photo set of the area to be supervised; processing the aerial photo set to obtain a slice image set of the area to be supervised; the aerial photo set comprises original aerial images of the area to be supervised at a plurality of supervision moments respectively;
step 2: performing feature extraction on the slice image set to obtain and display a vector feature layer of the region to be supervised at each supervision time;
and step 3: and judging whether illegal buildings exist in the area to be supervised according to the vector feature map layers at all supervision moments, and sending illegal building prompt information and displaying when illegal buildings exist in the area to be supervised.
The city illegal construction supervision method based on the unmanned aerial vehicle is based on intelligent identification of the unmanned aerial vehicle and images, whether illegal construction exists or not is automatically judged according to the intelligent identification result, illegal construction prompt messages are generated and displayed in time, the purposes of timely notification and display are achieved, the discovery period is short, the efficiency is high, the labor cost is low, and the supervision effect is good.
Specifically, the area to be supervised is one of cities to be supervised, the city to be supervised can be divided into a plurality of areas to be supervised according to actual conditions, each area to be supervised is provided with an unmanned aerial vehicle for regular aerial photography, each area to be supervised is also provided with a supervision device, aerial photo sets obtained by regular aerial photography of each unmanned aerial vehicle are transmitted to the corresponding supervision devices, each supervision device is intelligently supervised according to methods from S1 to S3, urban illegal building supervision of the areas is realized, and a regional leader can master urban planning conditions of the respective areas conveniently; a remote decision center is configured for the city to be supervised, and the results and data obtained by the supervision devices are uploaded to the remote decision center, so that the city leader can master the planning condition of the whole city conveniently; in the supervision device, a display screen can be arranged according to the existing visualization technology and GIS technology, so that the visualization display and diversified display of city supervision are realized; in the remote decision center, a large screen can be set according to the existing visualization technology and GIS technology, so that the data sharing of urban illegal building supervision by leaders at all levels is facilitated, and the remote decision capability of urban planning is improved conveniently.
Specifically, when the unmanned aerial vehicle takes a photo by plane regularly to obtain a photo set (i.e. original aerial images at each supervision time), the original aerial images can be sequenced and numbered in sequence according to the supervision time before processing, which is beneficial to improving the efficiency of the subsequent feature extraction process and judgment process.
Preferably, as shown in fig. 2, in S1, the step of obtaining the slice image set includes:
s11: respectively carrying out geometric correction preprocessing on the original aerial image at each supervision moment to obtain a processed aerial image set of the area to be supervised;
s12: respectively carrying out standardized segmentation processing on the processed aerial images of the processed aerial image set at each supervision moment according to a preset segmentation standard to obtain slice image subsets corresponding to the processed aerial images at each supervision moment;
s13: and obtaining the slice image set according to all the slice image subsets.
The method has the advantages that the geometric correction pretreatment is carried out on the original aerial image, so that the influence of factors such as a shooting angle, a height and an illumination angle on the image when the unmanned aerial vehicle carries out aerial photography can be effectively overcome, the picture quality of the original aerial image is improved, and the slicing effect of subsequent slicing treatment is improved; according to the preset segmentation standard, the image set after preprocessing (namely the aviation image set) is subjected to standardized segmentation processing, so that the fine processing and analysis of the image are in a unified standard, and the accuracy of subsequent feature extraction is improved.
Preferably, the specific step of S11 includes:
s111: respectively carrying out internal orientation processing and external orientation processing on the original aerial image at each supervision moment in sequence to obtain an orientation parameter file corresponding to each original aerial image;
s112: performing orthorectification on the corresponding original aerial image according to the directional parameter file of each original aerial image by adopting a digital elevation model correction method to obtain an orthoimage corresponding to each original aerial image one to one;
s113: acquiring the shadow area of each orthoimage, and taking the orthoimage corresponding to the minimum shadow area in all the shadow areas as a reference image;
s114: respectively carrying out center correction on each orthoimage except the reference image by using the reference image to obtain a corrected image corresponding to each orthoimage except the reference image one by one;
s115: respectively taking the reference image and each correction image as processing aerial images at corresponding supervision moments, and obtaining a processing aerial image set according to all the processing aerial images;
the preset segmentation standard in the step S12 is specifically a five-layer fifteen-level segmentation standard, and the intra-layer segmentation ratio is 5:2.5:1, and the inter-layer segmentation ratio is 10: 1.
The unmanned aerial vehicle shoots images at different monitoring moments during aerial shooting, and the images have larger difference due to camera factors, ground control factors and the like during each shooting, so that the difference caused by the camera factors (such as the type, the focal length and other parameters of the camera) can be effectively overcome by carrying out internal orientation processing on the original aerial images at each monitoring moment in S111, and the difference caused by the ground control factors (such as ground controller information and the like) can be effectively overcome by carrying out external orientation processing; in step S112, according to the obtained orientation parameter file corresponding to each original aerial image, based on a digital elevation model correction method, real ortho-correction can be achieved, an ortho-image that can feed back the real situation of the building can be obtained, and then a more accurate stereo image of the building on the ground and related geometric and physical information (such as shadow area) can be obtained according to the more real ortho-image; because the smaller the shadow area of the ortho image, the smaller the influence of the angle, the light brightness and the like shot by the unmanned aerial vehicle on the feature extraction and analysis of the image, after the shadow areas of the ortho images are obtained in the step S113, the ortho image with the minimum shadow area is taken as a reference image, and the center correction is performed on other ortho images, so that the image features of the building can be obviously reflected by all corrected images as far as possible, and the subsequent accurate extraction of the vector feature image layer is facilitated.
In S12, standardized segmentation processing is performed according to a five-layer fifteen-level segmentation standard, the intra-layer segmentation proportion is 5:2.5:1, and the inter-layer segmentation proportion is 10:1, so that each slice image in each slice image subset is 1000 x 1000 pixels in size, the image resolution on each segmentation level can be guaranteed to have consistency, and the subsequent feature extraction process is further supported better.
Preferably, as shown in fig. 3, the specific step of S2 includes:
s21: building feature extraction models are constructed based on a deep learning method;
s22: selecting a slice image subset at any supervision moment, inputting the selected slice image subset serving as a test set into the building feature extraction model, and extracting all feature point description vectors in the selected slice image subset;
s23: selecting one feature point description vector from all feature point description vectors of the selected slice image subset, and respectively obtaining Euclidean distances between the selected feature point description vector and each of the rest feature point description vectors; according to all Euclidean distances, searching out two nearest characteristic point description vectors corresponding to the selected characteristic point description vector from the rest characteristic point description vectors by adopting a preferential k-d tree searching method;
s24: traversing each feature point description vector in the selected slice image subset to obtain two nearest feature point description vectors corresponding to each feature point description vector in the selected slice image subset;
s25: image splicing is carried out according to all feature point description vectors and all nearest feature point description vectors corresponding to all feature point description vectors by adopting an image weighted average method, and a vector feature image layer corresponding to the selected slice image subset is obtained;
s26: and traversing the slice image subsets at each supervision time to obtain and display the vector feature map layers of the region to be supervised at each supervision time.
Firstly, a model capable of extracting the building characteristics (namely a building characteristic extraction model) is constructed by utilizing an artificial intelligence technology, and then a slice image subset at any supervision moment is input into the model, so that the intelligent extraction of the characteristics of each slice image can be realized; after all feature point description vectors in each slice image subset are extracted, in order to subsequently analyze the overall features of the building images in the images, the sliced images also need to be spliced, and the image processing method with the segmentation and splicing can ensure that the building features are obtained as much as possible by segmentation first and then the building is subjected to overall analysis by splicing, so that the analysis and judgment of the building are improved to the greatest extent, and the accuracy of illegal building supervision is improved; in S23 and S24, two nearest feature point description vectors corresponding to the selected feature point description vectors are obtained based on Euclidean distance and a preferential k-d tree searching method, so that the feature points can be guaranteed to be far away and evenly distributed in an image region as much as possible, the accuracy and reliability of image splicing in S25 are guaranteed, image splicing is carried out in S25 by adopting an image weighted average method, better splicing among multiple images can be guaranteed, and a reliable vector feature map layer capable of accurately describing building features at each supervision time is obtained.
Specifically, in S21, a large number of model sample preparation data sets are obtained by using a big data technique, a training set is extracted from the data sets, and the training set is trained by using a convolutional neural network to obtain a building feature extraction model, wherein the specific operation method is the prior art; in S22, extracting a feature point description vector by using the SIFT operator and a building feature extraction model; in S23, S24, and S25, the specific methods of calculating the euclidean distance, the preferential k-d tree search method, and the image weighted average method are also the prior art, and the specific details are not repeated.
Preferably, in S1, the method further includes the following steps after obtaining the slice image set:
and publishing the slice image set into a map service by adopting a WFS/WMTS service publishing method, and pushing the map service.
By publishing and pushing the slice image set to a map service, the sharing of metadata information and slice data of the image can be realized, and the sharing of data in each system and each department can be realized. The specific operation steps of the WFS/WMTS service publishing method are prior art, and the details are not described herein.
Preferably, as shown in fig. 4, the specific step of S3 includes:
s31: selecting a vector feature layer at any supervision time, and obtaining all building images and basic parameters of each building image in the area to be supervised at the selected supervision time according to the selected vector feature layer;
s32: building a corresponding building object set at the supervision time according to all the building images at the selected supervision time and the basic parameters of all the building images;
s33: traversing each vector feature map layer to obtain a building object set at each supervision moment;
s34: selecting a building object set at any two supervision moments, wherein the building object sets are M1And M2In building object set M1Optionally a building image A1From the building image A1And a set of building objects M2Determining the building image A according to the preset determination formula based on the basic parameters of all the building images1Relative to a building object set M2The building type of (a);
s35: traversing building object set M1In the building image, a building object set M is determined one by one1Each building image in (a) relative to a set of building objects M2And a building object set M is obtained according to all building types1Set M of objects with building2A set of building types in between;
s36: when the building object set M1Set M of objects with building2When at least one building type in the building type set between the buildings is indicated as illegal building, executing S37; when the building object set M1Set M of objects with building2When the building type in the building type set does not exist and indicates that the building type is illegal, returning to S34, and reselecting the building pairs at two supervision momentsThe image sets are traversed until the building object sets of the area to be supervised at every two supervision moments;
s37: and sending out and displaying the illegal construction prompt information.
Specifically, the basic parameters include the minimum circumscribed rectangle area of the building image;
determining a selected set of building objects M1Middle building image A1With respect to the selected building object set M2The preset judgment formula of the building type is specifically as follows:
wherein A is2For selected building object sets M2Of any one of the images of the building,
and
respectively building image A1Minimum circumscribed rectangle area and building image a2The area of the minimum circumscribed rectangle of (a),
and
respectively building image A1Corresponding supervision instants and building images a2At the corresponding supervision time, δ and K are both preset threshold values, and s.t. represents a constraint condition.
For the vector feature map layer at any supervision time, all the building images and corresponding basic parameters can be obtained from the vector feature map layer, each building image and corresponding basic parameters are used as a building object, all the building objects at the corresponding supervision time are obtained, further, a building object set at the supervision time is built, and the building object set at the supervision time is built through the methods of S31 and S32A building object set is obtained, and the building images and the basic image parameters are associated one by one, so that the building type set is conveniently analyzed and judged subsequently, and further illegal building judgment is realized; in S34 and S35, based on the basic parameters of the building image and the preset judgment formula, the building object set M can be accurately judged1Each building image in the set of building objects M2The building type set between the building object sets at the two supervision times is obtained, and the building type set embodies image comparison results at the two supervision times; in S36, when at least one building type in the building type set between the building object sets at two monitoring times is an illegal building, it indicates that the illegal building occurs in the area to be monitored by comparing the two monitoring times, and it is necessary to send an illegal building prompt message to inform relevant staff of intervention; when the building type set between the building object sets at the two supervision moments does not have a building type which is illegal, the situation that the building type set is illegal in the two supervision moments through the region to be supervised is proved, and whether the illegal building phenomenon occurs or not is judged again through the comparison of the other two supervision moments; in the process of secondary judgment, if the illegal building phenomenon occurs, sending illegal building prompt information and a time notice, if the illegal building phenomenon does not occur, continuously repeating the judgment process until the building object sets at every two supervision moments are traversed; when the traversal termination condition is reached, if the illegal building phenomenon does not occur, the judgment result that the illegal building does not exist in the area to be supervised can be represented by not taking any action, and the judgment result that the illegal building exists in the area to be supervised can be represented by sending a message that the illegal building phenomenon does not occur; the judging method has the advantages of high accuracy, low labor cost and good intelligent supervision effect.
With respect to the preset decision formula,
building object set M representing selected two surveillance moments1Is later than the building object set M2At the supervision moment, i.e. building object set M2Obtained for a preceding air photo set, and a building object set M1Obtained for the following air photo set; under the constraint conditions, when the building object set M2In which there is a building image a1And there is no image A of any building1Building image A is obtained by overlapping building images with equivalent areas and positions1Relative to a building object set M2Is an illegal building.
Specifically, in S2, for the display of the vector feature map layer at each monitoring time, the three-dimensional display of the multi-source data may be performed based on the GIS technology, including the superposition, rendering, and labeling of the image, and a search box is further set on the display interface to implement the space search and measurement. On this three-dimensional show interface, can realize a key contrast function, can carry out a key contrast to the vector feature picture layer of different supervision moments, realize difference visualization between them, to difference visualization can provide roll curtain contrast and quick contrast two kinds of modes. The rolling curtain comparison means that images and differences at different supervision moments are displayed in a same window in an overlapping mode; the quick comparison means that different differences are extracted through a left window and a right window, so that the purpose of quick display is achieved. In the embodiment, for the display of the illegal construction prompting information in S3, the alarm information may be displayed on the display interface, and the extracted difference is highlighted in red or the like, so as to quickly locate the illegal construction phenomenon; an independent alarm can be arranged, and alarm sound is sent out through the alarm to prompt related personnel.
In the second embodiment, as shown in fig. 5, a city illegal building supervision system based on an unmanned aerial vehicle is applied to the city illegal building supervision method based on the unmanned aerial vehicle in the first embodiment, and comprises the unmanned aerial vehicle and a supervision device, wherein the unmanned aerial vehicle and the supervision device are in communication connection through a wireless network;
the monitoring device comprises an aerial photography module, a processing module, an extraction module, a judgment module and a display module;
the aerial photography module is used for utilizing the unmanned aerial vehicle to carry out aerial photography on the area to be supervised periodically to obtain an aerial photo set of the area to be supervised; the aerial photo set comprises original aerial images of the area to be supervised at a plurality of supervision moments respectively;
the processing module is used for processing the aerial photo set to obtain a slice image set of the area to be supervised;
the extraction module is used for extracting the features of the slice image set to obtain a vector feature map layer of the region to be supervised at each supervision time;
the judging module is used for judging whether illegal buildings exist in the area to be supervised according to the vector feature map layers at all supervision moments, and sending illegal building prompt information when the illegal buildings exist in the area to be supervised;
the display module is used for displaying the violation prompt information and the vector feature map layer of the area to be supervised at each supervision moment.
The city illegal building supervisory system based on unmanned aerial vehicle in this embodiment, based on the intelligent recognition of unmanned aerial vehicle and image, carry out automatic judgement according to intelligent recognition's result and whether have and violate the building, in time generate illegal building prompt message and show, reach the purpose of in time notice and show, discovery cycle is short, efficient, the human cost is low, and the supervision effect is good, is favorable to the orderly planning in city.
Preferably, the processing module is specifically configured to:
respectively carrying out geometric correction preprocessing on the original aerial image at each supervision moment to obtain a processed aerial image set of the area to be supervised;
respectively carrying out standardized segmentation processing on the processed aerial images of the processed aerial image set at each supervision moment according to a preset segmentation standard to obtain slice image subsets corresponding to the processed aerial images at each supervision moment;
and obtaining the slice image set according to all the slice image subsets.
Preferably, the specific step of obtaining the processed aerial image set by the processing module includes:
respectively carrying out internal orientation processing and external orientation processing on the original aerial image at each supervision moment in sequence to obtain an orientation parameter file corresponding to each original aerial image;
performing orthorectification on the corresponding original aerial image according to the directional parameter file of each original aerial image by adopting a digital elevation model correction method to obtain an orthoimage corresponding to each original aerial image one to one;
acquiring the shadow area of each orthoimage, and taking the orthoimage corresponding to the minimum shadow area in all the shadow areas as a reference image;
respectively carrying out center correction on each orthoimage except the reference image by using the reference image to obtain a corrected image corresponding to each orthoimage except the reference image one by one;
respectively taking the reference image and each correction image as processing aerial images at corresponding supervision moments, and obtaining a processing aerial image set according to all the processing aerial images;
the preset segmentation standard is a five-layer fifteen-level segmentation standard, the segmentation ratio in each layer is 5:2.5:1, and the segmentation ratio between layers is 10: 1.
Preferably, the extraction module is specifically configured to:
building feature extraction models are constructed based on a deep learning method;
selecting a slice image subset at any supervision moment, inputting the selected slice image subset serving as a test set into the building feature extraction model, and extracting all feature point description vectors in the selected slice image subset;
selecting one feature point description vector from all feature point description vectors of the selected slice image subset, and respectively obtaining Euclidean distances between the selected feature point description vector and each of the rest feature point description vectors; according to all Euclidean distances, searching out two nearest characteristic point description vectors corresponding to the selected characteristic point description vector from the rest characteristic point description vectors by adopting a preferential k-d tree searching method;
traversing each feature point description vector in the selected slice image subset to obtain two nearest feature point description vectors corresponding to each feature point description vector in the selected slice image subset;
image splicing is carried out according to all feature point description vectors and all nearest feature point description vectors corresponding to all feature point description vectors by adopting an image weighted average method, and a vector feature image layer corresponding to the selected slice image subset is obtained;
and traversing the slice image subsets at each supervision time to obtain the vector feature map layers of the areas to be supervised at each supervision time.
Preferably, the map service distribution system further comprises a distribution module, and the distribution module is configured to adopt a WFS/WMTS service distribution method to distribute the slice image set into a map service and push the map service.
Preferably, the determining module is specifically configured to:
selecting a vector feature layer at any supervision time, and obtaining all building images and basic parameters of each building image in the area to be supervised at the selected supervision time according to the selected vector feature layer;
building a corresponding building object set at the supervision time according to all the building images at the selected supervision time and the basic parameters of all the building images;
traversing each vector feature map layer to obtain a building object set at each supervision moment;
selecting a building object set at any two supervision moments, wherein the building object sets are M1And M2In building object set M1Optionally a building image A1From the building image A1And a set of building objects M2Determining the building image A according to the preset determination formula based on the basic parameters of all the building images1Relative to a building object set M2The building type of (a);
traversing building object set M1Each building image in the building image is determined one by oneBuilding object set M1Each building image in (a) relative to a set of building objects M2And a building object set M is obtained according to all building types1Set M of objects with building2A set of building types in between;
when the building object set M1Set M of objects with building2When at least one building type in the building type set between the buildings indicates that the building is an illegal building, sending out the illegal building prompt message; when the building object set M1Set M of objects with building2And when the building type in the building type set between the building type sets does not exist and indicates that the building type is illegal, reselecting the building object sets at the two supervision moments to obtain the building type set between the two reselected building objects until the building object sets of the area to be supervised at every two supervision moments are traversed.
Details of the embodiment are not described in detail in the first embodiment and the specific descriptions in fig. 1 to 4, which are not repeated herein.
Third embodiment, based on the first embodiment and the second embodiment, the present embodiment further discloses a city waterlogging monitoring and early warning system based on the mobile internet, which includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor, and the computer program implements the specific steps of S1 to S3 when running.
Through the computer program stored on the memory, the intelligent identification based on the unmanned aerial vehicle and the image is operated on the processor, whether illegal construction exists is automatically judged according to the result of the intelligent identification, and illegal construction prompt messages are generated and displayed in time, so that the aims of timely notification and display are fulfilled.
The present embodiment also provides a computer storage medium having at least one instruction stored thereon, where the instruction when executed implements the specific steps of S1-S3.
Through carrying out the computer storage medium who contains at least one instruction, based on the intelligent recognition of unmanned aerial vehicle and image, carry out automatic judgement according to intelligent recognition's result and whether have the violation of construction, in time generate the suggestion message of violating construction and show, reach the purpose of in time notice and show, discovery cycle is short, efficient, the human cost is low, and the supervision effect is good, is favorable to the orderly planning in city.
Details of the embodiment are not described in detail in the first embodiment and the specific descriptions in fig. 1 to 4, which are not repeated herein.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A city illegal building supervision method based on an unmanned aerial vehicle is characterized by comprising the following steps:
step 1: the method comprises the steps that an unmanned aerial vehicle is used for carrying out aerial photography on an area to be supervised periodically to obtain an aerial photo set of the area to be supervised; processing the aerial photo set to obtain a slice image set of the area to be supervised; the aerial photo set comprises original aerial images of the area to be supervised at a plurality of supervision moments respectively;
step 2: performing feature extraction on the slice image set to obtain and display a vector feature layer of the region to be supervised at each supervision time;
and step 3: and judging whether illegal buildings exist in the area to be supervised according to the vector feature map layers at all supervision moments, and sending illegal building prompt information and displaying when illegal buildings exist in the area to be supervised.
2. The city violation supervision method based on unmanned aerial vehicle as claimed in claim 1, wherein in step 1, the specific step of obtaining the slice image set comprises:
step 11: respectively carrying out geometric correction preprocessing on the original aerial image at each supervision moment to obtain a processed aerial image set of the area to be supervised;
step 12: respectively carrying out standardized segmentation processing on the processed aerial images of the processed aerial image set at each supervision moment according to a preset segmentation standard to obtain slice image subsets corresponding to the processed aerial images at each supervision moment;
step 13: and obtaining the slice image set according to all the slice image subsets.
3. The city violation supervision method based on unmanned aerial vehicle as claimed in claim 2, wherein the specific steps of step 11 comprise:
step 111: respectively carrying out internal orientation processing and external orientation processing on the original aerial image at each supervision moment in sequence to obtain an orientation parameter file corresponding to each original aerial image;
step 112: performing orthorectification on the corresponding original aerial image according to the directional parameter file of each original aerial image by adopting a digital elevation model correction method to obtain an orthoimage corresponding to each original aerial image one to one;
step 113: acquiring the shadow area of each orthoimage, and taking the orthoimage corresponding to the minimum shadow area in all the shadow areas as a reference image;
step 114: respectively carrying out center correction on each orthoimage except the reference image by using the reference image to obtain a corrected image corresponding to each orthoimage except the reference image one by one;
step 115: respectively taking the reference image and each correction image as processing aerial images at corresponding supervision moments, and obtaining a processing aerial image set according to all the processing aerial images;
the preset segmentation standard in the step 12 is a five-layer fifteen-level segmentation standard, the intra-layer segmentation ratio is 5:2.5:1, and the inter-layer segmentation ratio is 10: 1.
4. The city violation supervision method based on unmanned aerial vehicle as claimed in claim 2, wherein the specific steps of step 2 comprise:
step 21: building feature extraction models are constructed based on a deep learning method;
step 22: selecting a slice image subset at any supervision moment, inputting the selected slice image subset serving as a test set into the building feature extraction model, and extracting all feature point description vectors in the selected slice image subset;
step 23: selecting one feature point description vector from all feature point description vectors of the selected slice image subset, and respectively obtaining Euclidean distances between the selected feature point description vector and each of the rest feature point description vectors; according to all Euclidean distances, searching out two nearest characteristic point description vectors corresponding to the selected characteristic point description vector from the rest characteristic point description vectors by adopting a preferential k-d tree searching method;
step 24: traversing each feature point description vector in the selected slice image subset to obtain two nearest feature point description vectors corresponding to each feature point description vector in the selected slice image subset;
step 25: image splicing is carried out according to all feature point description vectors and all nearest feature point description vectors corresponding to all feature point description vectors by adopting an image weighted average method, and a vector feature image layer corresponding to the selected slice image subset is obtained;
step 26: and traversing the slice image subsets at each supervision time to obtain and display the vector feature map layers of the region to be supervised at each supervision time.
5. The city violation supervision method based on unmanned aerial vehicles according to any of claims 1-4, wherein in step 1, after obtaining the slice image set, further comprising the following steps:
and publishing the slice image set into a map service by adopting a WFS/WMTS service publishing method, and pushing the map service.
6. The city violation supervision method based on unmanned aerial vehicles according to any of claims 1-4, wherein the specific steps of step 3 comprise:
step 31: selecting a vector feature layer at any supervision time, and obtaining all building images and basic parameters of each building image in the area to be supervised at the selected supervision time according to the selected vector feature layer;
step 32: building a corresponding building object set at the supervision time according to all the building images at the selected supervision time and the basic parameters of all the building images;
step 33: traversing each vector feature map layer to obtain a building object set at each supervision moment;
step 34: selecting a building object set at any two supervision moments, wherein the building object sets are M1And M2In building object set M1Optionally a building image A1From the building image A1And a set of building objects M2Determining the building image A according to the preset determination formula based on the basic parameters of all the building images1Relative to a building object set M2The building type of (a);
step 35: traversing building object set M1In the building image, a building object set M is determined one by one1Each building image in (a) relative to a set of building objects M2And a building object set M is obtained according to all building types1Set M of objects with building2A set of building types in between;
step 36: when the building object set M1Set M of objects with building2When at least one building type in the building type set between indicates as an illegal building, executing step 37; when the building object set M1Set M of objects with building2When the building type in the building type set does not exist and indicates that the building type is illegal, returning to the step 34, and reselecting the building object sets at the two supervision moments until the area to be supervised is in the area to be supervisedBuilding object sets at every two supervision moments are traversed;
step 37: and sending out and displaying the illegal construction prompt information.
7. The unmanned-aerial-vehicle-based city violation supervision method of claim 6, wherein the basic parameters comprise a minimum bounding rectangle area of the building image;
determining a selected set of building objects M1Middle building image A1With respect to the selected building object set M2The preset judgment formula of the building type is specifically as follows:
wherein A is2For selected building object sets M2Of any one of the images of the building,
and
respectively building image A1Minimum circumscribed rectangle area and building image a2The area of the minimum circumscribed rectangle of (a),
and
respectively building image A1Corresponding supervision instants and building images a2At the corresponding supervision time, δ and K are both preset threshold values, and s.t. represents a constraint condition.
8. An unmanned aerial vehicle-based city illegal building supervision system is applied to the unmanned aerial vehicle-based city illegal building supervision method of any one of claims 1 to 7, and comprises an unmanned aerial vehicle and a supervision device, wherein the unmanned aerial vehicle and the supervision device are in communication connection through a wireless network;
the monitoring device comprises an aerial photography module, a processing module, an extraction module, a judgment module and a display module;
the aerial photography module is used for utilizing the unmanned aerial vehicle to carry out aerial photography on the area to be supervised periodically to obtain an aerial photo set of the area to be supervised; the aerial photo set comprises original aerial images of the area to be supervised at a plurality of supervision moments respectively;
the processing module is used for processing the aerial photo set to obtain a slice image set of the area to be supervised;
the extraction module is used for extracting the features of the slice image set to obtain a vector feature map layer of the region to be supervised at each supervision time;
the judging module is used for judging whether illegal buildings exist in the area to be supervised according to the vector feature map layers at all supervision moments, and sending illegal building prompt information when the illegal buildings exist in the area to be supervised;
the display module is used for displaying the violation prompt information and the vector feature map layer of the area to be supervised at each supervision moment.
9. A city violation supervision system based on unmanned aerial vehicles, comprising a processor, a memory and a computer program stored in the memory and executable on the processor, the computer program when executed implementing the method steps of any of claims 1 to 7.
10. A computer storage medium, the computer storage medium comprising: at least one instruction which, when executed, implements the method steps of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110572819.5A CN113313006B (en) | 2021-05-25 | 2021-05-25 | Urban illegal building supervision method, system and storage medium based on unmanned aerial vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110572819.5A CN113313006B (en) | 2021-05-25 | 2021-05-25 | Urban illegal building supervision method, system and storage medium based on unmanned aerial vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113313006A true CN113313006A (en) | 2021-08-27 |
| CN113313006B CN113313006B (en) | 2023-08-22 |
Family
ID=77374714
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110572819.5A Active CN113313006B (en) | 2021-05-25 | 2021-05-25 | Urban illegal building supervision method, system and storage medium based on unmanned aerial vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113313006B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113723373A (en) * | 2021-11-02 | 2021-11-30 | 深圳市勘察研究院有限公司 | Unmanned aerial vehicle panoramic image-based illegal construction detection method |
| CN115424155A (en) * | 2022-11-04 | 2022-12-02 | 浙江大华技术股份有限公司 | Illegal construction detection method, illegal construction detection device and computer storage medium |
| CN115439672A (en) * | 2022-11-04 | 2022-12-06 | 浙江大华技术股份有限公司 | Image matching method, illicit detection method, terminal device, and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108765234A (en) * | 2018-05-07 | 2018-11-06 | 重庆睿宇测绘有限责任公司 | Doubtful illegal building information management system |
| CN110243354A (en) * | 2019-07-04 | 2019-09-17 | 桂林理工大学 | A kind of city illegal building object dynamic monitoring method and system |
| CN111401345A (en) * | 2020-06-04 | 2020-07-10 | 常州市新翼空间信息科技有限公司 | DSM automatic comparison system based on aerial photography measurement |
| CN111539296A (en) * | 2020-04-17 | 2020-08-14 | 河海大学常州校区 | Method and system for identifying illegal building based on remote sensing image change detection |
| WO2021003824A1 (en) * | 2019-07-11 | 2021-01-14 | 平安科技（深圳）有限公司 | Image recognition-based illegal building identification method and device |
-
2021
- 2021-05-25 CN CN202110572819.5A patent/CN113313006B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108765234A (en) * | 2018-05-07 | 2018-11-06 | 重庆睿宇测绘有限责任公司 | Doubtful illegal building information management system |
| CN110243354A (en) * | 2019-07-04 | 2019-09-17 | 桂林理工大学 | A kind of city illegal building object dynamic monitoring method and system |
| WO2021003824A1 (en) * | 2019-07-11 | 2021-01-14 | 平安科技（深圳）有限公司 | Image recognition-based illegal building identification method and device |
| CN111539296A (en) * | 2020-04-17 | 2020-08-14 | 河海大学常州校区 | Method and system for identifying illegal building based on remote sensing image change detection |
| CN111401345A (en) * | 2020-06-04 | 2020-07-10 | 常州市新翼空间信息科技有限公司 | DSM automatic comparison system based on aerial photography measurement |
Non-Patent Citations (4)
| Title |
|---|
| LIANG ZHEHENG 等: "The application of illegal building detection from VHR UAV remote sensing images based on convolutional neural network", 《BULLETIN OF SURVEYING AND MAPPING》, vol. 0, no. 4, pages 111 - 115 * |
| N. KHALILI MOGHADAM 等: "Automatic Urban Illegal Building Detection Using Multi-Temporal Satellite Images and Geospatial Information Systems", 《INT.ARCH.PHOTOGRAMM.REMOTE SENS.SPATIAL INF.SCI.》, pages 387 - 393 * |
| 赵建才;蔡靖宇;许苏苏;: "基于无人机多期影像对比的违章建筑监测***", 《现代测绘》, vol. 43, no. 04, pages 62 - 64 * |
| 钟波;: "应用无人机对于违法建筑物的影像对比判别", 《城市地理》, no. 22, pages 150 - 151 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113723373A (en) * | 2021-11-02 | 2021-11-30 | 深圳市勘察研究院有限公司 | Unmanned aerial vehicle panoramic image-based illegal construction detection method |
| CN115424155A (en) * | 2022-11-04 | 2022-12-02 | 浙江大华技术股份有限公司 | Illegal construction detection method, illegal construction detection device and computer storage medium |
| CN115439672A (en) * | 2022-11-04 | 2022-12-06 | 浙江大华技术股份有限公司 | Image matching method, illicit detection method, terminal device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113313006B (en) | 2023-08-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113313006B (en) | Urban illegal building supervision method, system and storage medium based on unmanned aerial vehicle | |
| CN104303193B (en) | Target classification based on cluster | |
| CN111818313B (en) | Vehicle real-time tracking method and device based on monitoring video | |
| JP2021531462A (en) | Intelligent navigation methods and systems based on topology maps | |
| CN109935080B (en) | Monitoring system and method for real-time calculation of traffic flow on traffic line | |
| CN111458721B (en) | Exposed garbage identification and positioning method, device and system | |
| CN113822247B (en) | Method and system for identifying illegal building based on aerial image | |
| CN103700261A (en) | Video-based road traffic flow feature parameter monitoring and traffic comprehensive information service system | |
| CN108256554A (en) | Vehicle reverse stopping judgment method, server and system based on deep learning | |
| CN112287827A (en) | Complex environment pedestrian mask wearing detection method and system based on intelligent lamp pole | |
| CN112668375B (en) | Tourist distribution analysis system and method in scenic spot | |
| CN113284144B (en) | Tunnel detection method and device based on unmanned aerial vehicle | |
| CN113887412A (en) | Detection method, detection terminal, monitoring system and storage medium for pollution emission | |
| US20210104059A1 (en) | Method for size estimation by image recognition of specific target using given scale | |
| WO2020211593A1 (en) | Digital reconstruction method, apparatus, and system for traffic road | |
| CN114035606A (en) | Pole tower inspection system, pole tower inspection method, control device and storage medium | |
| CN113240906A (en) | Vehicle guiding method and system based on real-time monitoring of road congestion in logistics park | |
| CN116993681A (en) | Substation inspection defect detection method and system | |
| CN111723656A (en) | Smoke detection method and device based on YOLO v3 and self-optimization | |
| CN114926791A (en) | Method and device for detecting abnormal lane change of vehicles at intersection, storage medium and electronic equipment | |
| CN114494845A (en) | Artificial intelligence hidden danger troubleshooting system and method for construction project site | |
| CN117768610A (en) | High-speed railway perimeter intrusion risk monitoring method and system based on multi-target recognition | |
| CN112418081A (en) | Method and system for air-ground joint rapid investigation of traffic accidents | |
| CN113569801B (en) | Distribution construction site live equipment and live area identification method and device thereof | |
| CN113361968B (en) | Power grid infrastructure worker safety risk assessment method based on artificial intelligence and big data |
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 |