CN111382645B - Method and system for identifying overdue building in electronic map - Google Patents
Method and system for identifying overdue building in electronic map Download PDFInfo
- Publication number
- CN111382645B CN111382645B CN201811646297.3A CN201811646297A CN111382645B CN 111382645 B CN111382645 B CN 111382645B CN 201811646297 A CN201811646297 A CN 201811646297A CN 111382645 B CN111382645 B CN 111382645B
- Authority
- CN
- China
- Prior art keywords
- building
- picture
- target area
- training
- map
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012549 training Methods 0.000 claims abstract description 73
- 238000003709 image segmentation Methods 0.000 claims abstract description 52
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 28
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 238000005520 cutting process Methods 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 10
- 230000011218 segmentation Effects 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 6
- 230000004931 aggregating effect Effects 0.000 claims description 5
- 238000004220 aggregation Methods 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000007689 inspection Methods 0.000 description 3
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000003550 marker Substances 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 241001270131 Agaricus moelleri Species 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
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
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A30/00—Adapting or protecting infrastructure or their operation
- Y02A30/60—Planning or developing urban green infrastructure
Abstract
The invention discloses a method and a system for identifying expired buildings in an electronic map, wherein the method for identifying the expired buildings in the electronic map comprises the following steps: acquiring a training data set and a label data set; preprocessing the training data set and the tag data set to obtain a training picture set and a tag picture set; training a deep convolutional neural network image segmentation model according to the training picture set and the label picture set and obtaining model parameters; inputting the target area picture set into a trained deep convolutional neural network image segmentation model, and carrying out building identification through trained model parameters to obtain a target area building probability image map; and processing the target area building probability image map to identify an expired building. The method disclosed by the invention improves the updating speed of the electronic navigation map, can control and modify the operation style and the operation quality of the electronic map, reduces the number of manually processed overdue buildings and reduces the labor cost.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to a method for processing an electronic map expired building based on deep convolutional neural network satellite image segmentation, in particular to a method and a system for identifying the expired building in the electronic map.
Background
The high-precision electronic map is one of basic data of the unmanned system, and the data facing the unmanned system must have the capability of updating in time. Because the existing consumer-level electronic map generally adopts a map operator to manually delete the outdated data, the update speed of the electronic map is between 3 months and 1 year, and the data requirement of an unmanned system on the electronic map cannot be met. Moreover, the conventional consumer electronic map has low efficiency in quality inspection at the later stage of map operation due to inconsistent proficiency, operation style and the like of operators.
On the other hand, the satellite pictures are updated faster in real time at present, but the elements on the satellite pictures are manually described in the prior art, and the manual operation method including construction cannot adapt to the rapidly updated satellite pictures.
Therefore, there is an urgent need to develop a new electronic map processing method by which the above-mentioned technical problems caused by the manual update and quality inspection of the electronic map by an operator are improved or solved, and to accommodate the rapidly updated satellite picture.
Disclosure of Invention
In view of the foregoing drawbacks and deficiencies of the prior art, an object of the present invention is to provide a method and system for identifying expired buildings in an electronic map. The method and the system for identifying the outdated building in the electronic map can solve the technical problems caused by manual updating and quality inspection of the electronic map at present, and adapt to the satellite picture which is updated rapidly through automatic judgment of the model.
To achieve the above object, according to one aspect of the present invention, there is provided a method for identifying an expired building in an electronic map, comprising:
acquiring a training data set and a label data set;
preprocessing the training data set and the tag data set to obtain a training picture set and a tag picture set;
training a deep convolutional neural network image segmentation model according to the training picture set and the label picture set and obtaining model parameters;
inputting the target area picture set into a trained deep convolutional neural network image segmentation model, performing building identification through trained model parameters, and taking an S-shaped growth curve function as an output layer to obtain a target area building probability image map;
and processing the target area building probability image map, screening probability values in the target area building probability image map through a preset value threshold, and identifying outdated buildings.
Further, the training data set is a satellite picture and/or the tag data set is city building data, and the building data at least comprises building contour pictures and building contour vector boundary data.
Further, the preprocessing the training data set and the tag data set to obtain a training picture set and a tag picture set includes:
unifying the projection coordinates of the satellite picture and the projection coordinates of the building outline vector boundary;
converting the building outline picture into a grid according to the resolution of the satellite picture;
cutting the boundary of the satellite picture and the boundary of the building outline picture according to a unified frame;
and cutting the satellite picture and the building outline picture according to the unified geographic grid respectively to obtain a training picture set and a label picture set.
Furthermore, the depth convolution neural network image segmentation model adopts a medical image segmentation model.
Further, screening probability values in the target area building probability image map through a preset value threshold, and deleting the expired building after identifying the expired building.
Further, building identification is performed through trained model parameters, and an S-shaped growth curve function is used as an output layer to obtain a target area building probability image map, which comprises the following steps:
dividing the target area picture set to obtain a plurality of sub-area picture sets;
aggregating the plurality of sub-region picture sets;
respectively processing the subarea image sets through a trained deep convolutional neural network image segmentation model to obtain a plurality of subarea building probability image maps;
and splicing the sub-region building probability image map to obtain a target region building probability image map.
In another aspect of the present invention, a system for identifying expired buildings in an electronic map is provided, including:
a data acquisition unit configured to acquire a training data set and a tag data set;
the image acquisition unit is configured to preprocess the training data set and the tag data set to obtain a training image set and a tag image set;
the model training unit is configured to train a depth convolutional neural network image segmentation model according to the training picture set and the label picture set and obtain model parameters;
the building probability image map acquisition unit is configured to input a target area image set into a trained deep convolutional neural network image segmentation model, perform building identification through trained model parameters, and take an S-shaped growth curve function as an output layer to obtain a target area building probability image map;
and the expired building processing unit is configured to process the target area building probability map, screen probability values in the target area building probability image map through a preset threshold value and identify expired buildings.
Further, the training data set is a satellite picture and/or the tag data set is city building data, and the building data at least comprises building contour pictures and building contour vector boundary data.
Further, the image acquisition unit includes:
the projection coordinate unification subunit is configured to unify the projection coordinates of the satellite picture and the projection coordinates of the building contour vector boundary;
the picture rasterization subunit is configured to convert the building outline picture into a grid according to the resolution of the satellite picture;
a picture boundary clipping subunit configured to clip a boundary of the satellite picture and a boundary of the building contour picture according to a unified frame;
and the picture grid cutting sub-unit is configured to cut the satellite picture and the building outline picture according to the uniform geographic grid respectively to obtain a training picture set and a label picture set.
Furthermore, the depth convolution neural network image segmentation model adopts a medical image segmentation model.
And the expired building processing unit is further configured to screen probability values in the target area building probability image graph through a preset value threshold, identify expired buildings and delete the expired buildings.
Further, the building probability image map obtaining unit includes:
a target area picture set segmentation subunit configured to segment the target area picture set to obtain a plurality of sub-area picture sets;
a sub-region picture set aggregation sub-unit configured to aggregate the plurality of sub-region picture sets;
the sub-region building probability image map acquisition subunit is configured to respectively process the sub-region image sets through a trained deep convolutional neural network image segmentation model to obtain a plurality of sub-region building probability image maps;
and the target area building probability image map acquisition subunit is configured to splice the sub-area building probability image maps to obtain the target area building probability image map.
In another aspect of the present invention, there is also provided an apparatus for identifying an expired building in an electronic map, the apparatus comprising:
one or more processors;
a data store for storing data and one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of the preceding claims.
In another aspect of the invention, there is also provided a computer readable storage medium storing a computer program which, when executed by a processor, performs a method as described in any of the above.
Compared with the prior art, the invention has the beneficial effects that:
1. compared with the existing manual processing mode, the method for identifying the outdated building in the electronic map improves the updating speed of the electronic navigation map, has higher speed and precision of identifying the building by the deep convolution neural network model, can control and modify the operation style and the operation quality of the electronic map, reduces the number of the outdated buildings processed manually and reduces the labor cost.
2. According to the system for identifying the outdated building in the electronic map, disclosed by the invention, the updating speed of the electronic navigation map is improved by arranging the picture acquisition unit, the model training unit, the building probability image acquisition unit, the outdated building processing unit and other units, and compared with the existing manual processing mode, the speed and the accuracy of identifying the building by the deep convolutional neural network model are higher, the operation style and the operation quality of modifying the electronic map can be controlled, the number of the outdated buildings processed manually is reduced, and the labor cost is reduced.
3. The device for identifying the expired building in the electronic map provided by the embodiment facilitates the popularization of the technology for identifying the expired building in the electronic map through the computer readable medium storing the computer program.
Drawings
Other features, objects and advantages of the present invention will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the accompanying drawings in which:
FIG. 1 is a flow chart of a method for identifying expired buildings in an electronic map according to an example of the invention;
FIG. 2 is a natural image of a method for identifying expired buildings in an electronic map according to an embodiment of the present invention;
FIG. 3 is a probability map obtained from a natural image through a trained deep convolutional neural network image segmentation model in an exemplary method for identifying expired buildings in an electronic map according to an embodiment of the present invention;
fig. 4 is a probability map of a method for identifying expired buildings in an electronic map according to a second embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be noted that, for convenience of description, only the portions related to the invention are shown in the drawings.
It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other. The invention will be described in detail below with reference to the drawings in connection with embodiments.
As shown in fig. 1, the present embodiment provides a method for identifying an expired building in an electronic map, which includes the following steps:
s1: the method comprises the steps of obtaining a training data set and a tag data set, wherein the training data set is a satellite picture, the tag data set is city building data, and the building data at least comprises building outline pictures and building outline vector boundary data.
S2: preprocessing a training data set and a label data set to obtain a training picture set and a label picture set which are used as training input of a deep convolutional neural network image segmentation model, wherein the training input comprises the following specific steps:
s2-1: unifying the projection coordinates of the satellite picture and the projection coordinates of the building outline vector boundary;
s2-2: converting the building outline picture into a grid according to the resolution of the satellite picture;
s2-3: cutting the boundary of the satellite picture and the boundary of the building outline picture according to a unified frame;
s2-4: and cutting the satellite picture and the building outline picture according to the unified geographic grid respectively to obtain a training picture set and a label picture set.
S3: and training a depth convolution neural network image segmentation model according to the training picture set and the label picture set, and obtaining model parameters, wherein the depth convolution neural network image segmentation model adopts a medical image segmentation model.
S4: inputting the target area picture set into a trained deep convolutional neural network image segmentation model, performing building identification through trained model parameters, and taking an S-shaped growth curve function (S i gmo i d function) as an output layer to obtain a target area building probability image map;
s4-1: dividing the target area picture set to obtain a plurality of sub-area picture sets;
s4-2: aggregating the plurality of sub-region picture sets;
s4-3: respectively processing the subarea image sets through a trained deep convolutional neural network image segmentation model to obtain a plurality of subarea building probability image maps;
s4-4: and splicing the sub-region building probability image map to obtain a target region building probability image map.
S5, processing the target area building probability image map, screening probability values in the target area building probability image map through a preset threshold, and deleting the outdated building.
In this embodiment, the acquired training data set is a satellite picture, and the tag data set is city building data, where the building data at least includes a building contour picture and building contour vector boundary data.
In this embodiment, preprocessing a training dataset and a tag dataset to obtain a training picture set and a tag picture set, including the steps of:
unifying the projection coordinates of the satellite picture and the projection coordinates of the building contour vector boundary;
converting the building outline picture into a grid according to the resolution of the satellite picture;
cutting the boundary of the satellite picture and the boundary of the building outline picture according to the unified frame;
and cutting the satellite picture and the building outline picture according to the unified geographic grid respectively to obtain a training picture set and a label picture set.
In the embodiment, a medical image segmentation model (Convolutional Networks for Biomedical Image Segmentation) is adopted as a depth convolution neural network image segmentation model, and the depth convolution neural network image (CNN) segmentation model has excellent automatic feature extraction capability, so that the accuracy of the CNN-based semantic segmentation model is 20% -30% higher than that of a traditional image segmentation method. The medical image segmentation model (Convolutional Networks for Biomedical Image Segmentation) is a CNN segmentation network architecture applied to cell image segmentation, and has the advantages that compared with other model architectures, the information can skip the middle full-connection layer from the low-feature layer and be directly connected to the high-feature layer, so that uncertainty of results caused by sampling is avoided, and finally, a high-quality pixel-level segmentation effect is achieved.
In this embodiment, the preset threshold is 50%, and the mark with the probability value lower than 50% in the target area building probability image map is identified as the expired building, and the expired building is deleted. The threshold value can theoretically be preset to 50%, i.e. within the framework of this building, at least half of the pixels represent the probability that it is not a building. However, in actual production, due to the need, erroneous deletion should be avoided as much as possible, so that a stricter threshold range can be preset, and the threshold range can be 20-50%.
In order to facilitate understanding of the present invention, the method for identifying expired buildings in the electronic map provided in this embodiment is further described below by using the principle of the method for identifying expired buildings in the electronic map of the present invention:
the first step: acquiring training data sets and tag data sets
Acquiring a training data set and a tag data set, wherein the training data set is used as public sub-meter satellite picture data provided by Google Earth as an alternative scheme, the number of pictures is formulated according to training requirements, thousands of satellite pictures are generally required, and the tag data set downloads vector boundary data of urban building blocks provided by an existing map development interface.
And a second step of: data preprocessing
The satellite picture acquired in the first step is aligned with a map building in space position through data preprocessing, and is cut into an input data set required by a model, and the method specifically comprises the following steps:
step 1: and unifying the projection coordinates of the satellite picture and the projection coordinates of the building outline vector boundary.
And 2, converting the vector building outline picture into a grid according to the resolution of the satellite picture.
And 3, cutting the satellite picture boundary and the building outline picture boundary according to the unified frame.
And 4, cutting the satellite picture and the building outline picture according to the uniform geographic grid to obtain a training picture set and a label picture set, and taking the training picture set and the label picture set as training input of the depth convolution neural network image segmentation model.
And a third step of: training a deep convolutional neural network image segmentation model and obtaining model parameters
The medical image segmentation model is applied to satellite image segmentation, parameters used when the medical image segmentation model is constructed cannot be directly applied due to different application scenes of the model in the image processing technology, parameters of the model need to be retrained by using the data set acquired in the first step, namely, a training picture set and a label picture set acquired in the second step are input into a deep convolutional neural network image segmentation model, and the training model acquires model parameters.
Fourth step: identifying buildings in satellite images
As shown in fig. 2 and fig. 3, after the satellite image of the target working area is segmented in the manner of preprocessing the satellite image in the first step, building identification is performed through trained model parameters, and an S-shaped growth curve function (Sigmoid function) is used as an output layer to obtain a building probability image of each area in the target working area. Fig. 2 is a natural image, and after the natural image shown in fig. 2 is put into a model and the model is segmented by a trained deep convolutional neural network image, a target area building probability image shown in fig. 3 is obtained. The probability image map identifies probability through a model and is influenced by the distinguishing degree of the features in the previous training sample. Inputting the target area picture set into a trained deep convolutional neural network image segmentation model, extracting corresponding features of the target area picture set by the model, comparing the features with previous training data, for example, the shadow places in 99% of the pictures of the training data are all buildings, and extracting the shadow features from a certain picture of the target area picture set, then judging that the region is a building with higher probability by the model, thereby giving a higher probability value, and obtaining a target area building probability image.
According to the method for identifying the outdated building in the electronic map, through inputting the training picture set and the label picture set into the deep convolutional neural network image segmentation model, the model can count some common characteristics of the building in the pictures, such as the difference of the roof material and the land, for example, a certain amount of shadows and the like along with the building. The advantage of deep learning through a deep convolutional neural network image segmentation model is that it can extract many such features, as many as some of which are not intuitively perceived. By these features, the model can map the pixel value 0-255 of the satellite picture to probability 0-1, and the correspondence between the pixel value and the probability is formed, for example, the probability value corresponding to 128 gray scales is 0.5.
Fifth step: screening and deleting the expired building
After the building is identified by the model, a high-resolution building probability image map of the whole target area is obtained. The building probability image map cannot be directly applied to processing the vector overdue building in the map, so that the building probability image map needs to be further processed, and the processing steps are as follows:
alternatively, a preset threshold value is 45%, the marker with the probability value lower than 45% in the target area building probability image map is identified as outdated buildings, and the outdated buildings are deleted.
Compared with the existing manual processing mode, the method for identifying the outdated building in the electronic map improves the updating speed of the electronic navigation map, has higher speed and precision in identifying the building by the deep convolutional neural network model, can control and modify the operation style and the operation quality of the electronic map, reduces the number of the outdated buildings processed manually, and reduces the labor cost.
The embodiment also provides a system for identifying expired buildings in the electronic map, which comprises:
a data acquisition unit configured to acquire a training data set and a tag data set; the training data set obtained in the embodiment is a satellite picture, and the tag data set is city building data, wherein the building data at least comprises building outline pictures and building outline vector boundary data.
The image acquisition unit is configured to preprocess the training data set and the tag data set to obtain a training image set and a tag image set; the method specifically comprises the following steps:
the projection coordinate unification subunit is configured to unify the projection coordinates of the satellite picture and the projection coordinates of the building outline vector boundary;
the picture rasterization subunit is configured to convert the building outline picture into a grid according to the resolution of the satellite picture;
the picture boundary cutting subunit is configured to cut the boundary of the satellite picture and the boundary of the building outline picture according to the unified frame;
and the picture grid cutting sub-unit is configured to cut the satellite picture and the building outline picture according to the uniform geographic grid respectively to obtain a training picture set and a label picture set.
The method adopts a medical image segmentation model (Convolutional Networks for Biomedical Image Segmentation, U-Net for short) as a depth convolution neural network image segmentation model.
The model training unit is configured to train the deep convolutional neural network image segmentation model according to the training picture set and the label picture set and obtain model parameters;
the building probability image map acquisition unit is configured to input a target area image set into a trained deep convolutional neural network image segmentation model, perform building identification through trained model parameters, and take an S-shaped growth curve function as an output layer to obtain a target area building probability image map;
and the expired building processing unit is configured to process the target area building probability map, screen probability values in the target area building probability image map through a preset threshold value, and delete expired buildings. The method specifically comprises the following steps:
a threshold setting subunit configured to preset a threshold;
the expired building identification subunit is configured to identify a marker with a probability value lower than a preset threshold value in the target area building probability image map as an expired building;
and an expired building deleting subunit configured to delete the expired building.
It should be appreciated that the steps in the method for identifying expired buildings in the electronic map described above correspond to the subunits described in the system for identifying expired buildings in the electronic map. Thus, the operations and features described above for the system and the units contained therein are equally applicable to the above-described methods and are not repeated here.
According to the system for identifying the outdated building in the electronic map, provided by the embodiment, the updating speed of the electronic navigation map is improved through arranging the picture acquisition unit, the model training unit, the building probability image acquisition unit, the outdated building processing unit and other units, and compared with the existing manual processing mode, the speed and the accuracy of identifying the building by the deep convolutional neural network model are higher, the operation style and the operation quality of modifying the electronic map can be controlled, the number of the outdated buildings processed manually is reduced, and the labor cost is reduced.
The embodiment also provides a device for identifying an expired building in an electronic map, the device comprising:
one or more processors;
a data store for storing data and one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of the preceding claims.
The present embodiment also provides a computer readable storage medium storing a computer program which when executed by a processor implements a method as claimed in any one of the preceding claims.
The device for identifying the expired building in the electronic map provided by the embodiment facilitates the popularization of the technology for identifying the expired building in the electronic map through the computer readable medium storing the computer program.
Embodiment two:
the same features as those of the first embodiment are not described in detail, and the different features of the first embodiment are as follows:
the method for identifying the expired building in the electronic map provided by the embodiment further comprises the following steps:
dividing a target area picture set to obtain a plurality of sub-area picture sets;
aggregating a plurality of sub-region picture sets;
respectively processing the subarea image sets through a trained deep convolutional neural network image segmentation model to obtain a plurality of subarea building probability image maps;
and splicing the sub-area building probability image map to obtain the target area building probability image map.
In order to facilitate understanding of the present invention, the method for identifying expired buildings in the electronic map provided in this embodiment is further described below by using the principle of the method for identifying expired buildings in the electronic map of the present invention:
1. picture splitting
If the target area is a city, the range of the single city may exceed the computable image size, so that the Map-Reduce method is used to divide the building probability image recognition result of the single city into a plurality of sub-areas for parallel processing, namely the Map-Reduce method is used to distribute the probability image recognition work into a plurality of parts of data which can be processed by a computer, and then the data are processed respectively, and finally the result is synthesized; for example, the scale of a ground city is about 50km-500km, the satellite picture for identification is an image with 0.5m resolution, if the whole process is that the identification is a picture with the possible 100000 x 100000 pixels, the software loading size is easily exceeded, the Map-Reduce method is adopted to divide the identification result of a single city into a plurality of sub-areas for parallel processing, and finally the result is synthesized.
2. Aggregate images
And cutting the building outline in the region according to the range of the subareas, and aggregating the identified picture slices according to geographic positions.
When the model is used for dividing, 200-1000 pixels can be theoretically used, but the geographic length corresponding to 200 pixels is 100 meters, and the model is likely to be smaller than a building, so that the extracted characteristics are incomplete, and the model is very slow to run due to 1000 pixels. Alternatively, the pictures are sliced into 400 x 400 pixels per picture (200 m x 200m geographic area).
3. Obtaining a sub-area building probability image map
As shown in fig. 4, the average value of the area statistics is performed on the identification images in the sub-areas, so as to obtain the probability that the building is identified as a building. Assume that 5 buildings are arranged on the original electronic map, and the original electronic map is marked as 1-5. Judging whether the 5 buildings are outdated buildings according to the identification result, counting the identification result corresponding to each building, and averaging the identification result to serve as the probability of the building or the building. That is, the identification result in the existing building area is used as the evidence for judging whether the building exists or not. The final result is that the probability value corresponding to the No. 1-4 building is possibly more than 0.8, and the probability value corresponding to the No. 5 building is possibly 0.1, so that the No. 5 building is identified as an overdue building directly through the probability value.
4. Spliced subarea building probability image map
After all the subareas are processed, vector building outlines in all the subareas are fused into a layer.
5. Handling expired building
The threshold value can theoretically be preset to 50%, i.e. within the framework of this building, at least half of the pixels represent the probability that it is not a building. However, in actual production, due to the need, erroneous deletion should be avoided as much as possible, so that a stricter threshold range can be preset, and the threshold range can be 20-50%. Alternatively, a preset threshold value is set to be 50%, the marks with probability values lower than 50% in the target area building probability image map are identified as outdated buildings, and the outdated buildings are deleted.
The embodiment also provides a system for identifying overdue buildings in the electronic map, which specifically comprises:
a target area picture set segmentation subunit configured to segment the target area picture set to obtain a plurality of sub-area picture sets;
a sub-region picture set aggregation sub-unit configured to aggregate the plurality of sub-region picture sets;
the sub-region building probability image map acquisition subunit is configured to respectively process the sub-region image sets through a trained deep convolutional neural network image segmentation model to obtain a plurality of sub-region building probability image maps;
and the target area building probability image map acquisition subunit is configured to splice the sub-area building probability image maps to obtain the target area building probability image map.
According to the method and the system for identifying the expired building in the electronic map, the target area pictures are segmented and aggregated, then the sub-area images are processed through the trained deep convolutional neural network image segmentation model respectively, finally all the sub-area building probability image pictures are spliced, the expired building in the sub-area building probability image pictures is deleted, the expired building in the electronic map can be processed for the target area with a city level or a larger area, and the popularization of the technology for identifying the expired building in the electronic map is facilitated.
The above description is only illustrative of the preferred embodiments of the present invention and of the principles of the technology employed. It will be appreciated by persons skilled in the art that the scope of the invention referred to in the present invention is not limited to the specific combinations of the technical features described above, but also covers other technical features formed by any combination of the technical features described above or their equivalents without departing from the inventive concept. Such as the above-mentioned features and the technical features disclosed in the present invention (but not limited to) having similar functions are replaced with each other.
Claims (4)
1. A method for identifying expired buildings in an electronic map, comprising:
acquiring a training data set and a tag data set, wherein the training data set is a satellite picture, the tag data set is city building data, and the building data at least comprises building contour pictures and building contour vector boundary data;
preprocessing the training data set and the tag data set to obtain a training picture set and a tag picture set;
training a deep convolutional neural network image segmentation model according to the training picture set and the label picture set and obtaining model parameters, wherein the deep convolutional neural network image segmentation model adopts a U-Net segmentation model;
inputting the target area picture set into the trained deep convolutional neural network image segmentation model, performing building identification through trained model parameters, and taking an S-shaped growth curve function as an output layer to obtain a target area building probability image map;
processing the target area building probability image map, screening probability values in the target area building probability image map through a preset threshold value, and identifying outdated buildings;
the preprocessing the training data set and the label data set to obtain a training picture set and a label picture set comprises the following steps:
unifying the projection coordinates of the satellite picture and the projection coordinates of the building outline vector boundary;
converting the building outline picture into a grid according to the resolution of the satellite picture;
cutting the boundary of the satellite picture and the boundary of the building outline picture according to a unified frame;
cutting the satellite picture and the building outline picture according to the unified geographic grid respectively to obtain the training picture set and the tag picture set;
the building identification is carried out through the trained model parameters, the S-shaped growth curve function is used as an output layer, and a target area building probability image map is obtained, and the building identification method comprises the following steps:
dividing the target area picture set to obtain a plurality of sub-area picture sets;
aggregating the plurality of sub-region picture sets;
respectively processing the subarea picture sets through the trained deep convolutional neural network image segmentation model to obtain a plurality of subarea building probability image maps;
and splicing the sub-region building probability image map to obtain the target region building probability image map.
2. The method for identifying an expired building in an electronic map according to claim 1, wherein the screening probability values in the target area building probability image map by a preset threshold value further comprises deleting the expired building after identifying the expired building.
3. A system for identifying expired buildings in an electronic map, comprising:
the system comprises a data acquisition unit, a data processing unit and a data processing unit, wherein the data acquisition unit is configured to acquire a training data set and a label data set, the training data set is a satellite picture, the label data set is city building data, and the building data at least comprises a building outline picture and building outline vector boundary data;
the image acquisition unit is configured to preprocess the training data set and the tag data set to obtain a training image set and a tag image set;
the model training unit is configured to train a depth convolution neural network image segmentation model according to the training picture set and the label picture set and obtain model parameters, and the depth convolution neural network image segmentation model adopts a U-Net segmentation model;
the building probability image map acquisition unit is configured to input the target area image set into the trained deep convolutional neural network image segmentation model, perform building identification through trained model parameters, and take an S-shaped growth curve function as an output layer to obtain a target area building probability image map;
the outdated building processing unit is configured to process the target area building probability image graph, screen probability values in the target area building probability image graph through a preset threshold value, and identify outdated buildings;
wherein, the picture acquisition unit includes:
the projection coordinate unification subunit is configured to unify the projection coordinates of the satellite picture and the projection coordinates of the building outline vector boundary;
a picture rasterization subunit configured to convert the building outline picture into a grid according to the resolution of the satellite picture;
a picture boundary clipping subunit configured to clip a boundary of the satellite picture and a boundary of the building contour picture according to a unified frame;
the picture grid cutting sub-unit is configured to cut the satellite picture and the building outline picture according to the unified geographic grid respectively to obtain the training picture set and the tag picture set;
wherein, the building probability image map acquisition unit includes:
a target area picture set segmentation subunit configured to segment the target area picture set to obtain a plurality of sub-area picture sets;
a sub-region picture set aggregation sub-unit configured to aggregate the plurality of sub-region picture sets;
the sub-region building probability image map obtaining subunit is configured to respectively process the sub-region picture sets through the trained deep convolutional neural network image segmentation model to obtain a plurality of sub-region building probability image maps;
and the target area building probability image map acquisition subunit is configured to splice the sub-area building probability image maps to acquire the target area building probability image map.
4. The system for identifying expired buildings in an electronic map of claim 3, wherein the expired building processing unit is further configured to filter probability values in the target area building probability image map by a preset threshold, identify expired buildings, and delete the expired buildings.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811646297.3A CN111382645B (en) | 2018-12-29 | 2018-12-29 | Method and system for identifying overdue building in electronic map |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811646297.3A CN111382645B (en) | 2018-12-29 | 2018-12-29 | Method and system for identifying overdue building in electronic map |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111382645A CN111382645A (en) | 2020-07-07 |
| CN111382645B true CN111382645B (en) | 2023-11-28 |
Family
ID=71218333
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811646297.3A Active CN111382645B (en) | 2018-12-29 | 2018-12-29 | Method and system for identifying overdue building in electronic map |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111382645B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115795070B (en) * | 2023-01-30 | 2023-05-02 | 智道网联科技(北京)有限公司 | Method, device, electronic equipment and storage medium for detecting map expiration guideboard |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107516096A (en) * | 2016-06-15 | 2017-12-26 | 阿里巴巴集团控股有限公司 | A kind of character identifying method and device |
| CN108229364A (en) * | 2017-12-28 | 2018-06-29 | 百度在线网络技术（北京）有限公司 | Contour of building generation method, device, computer equipment and storage medium |
| CN108346428A (en) * | 2017-09-13 | 2018-07-31 | 腾讯科技（深圳）有限公司 | Voice activity detection and its method for establishing model, device, equipment and storage medium |
| CN108376232A (en) * | 2018-01-04 | 2018-08-07 | 北京星衡科技有限公司 | A kind of method and apparatus of automatic interpretation for remote sensing image |
| CN108764226A (en) * | 2018-04-13 | 2018-11-06 | 顺丰科技有限公司 | Image text recognition methods, device, equipment and its storage medium |
| 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 |
| CN108898603A (en) * | 2018-05-29 | 2018-11-27 | 北京佳格天地科技有限公司 | Plot segmenting system and method on satellite image |
| CN109034073A (en) * | 2018-07-30 | 2018-12-18 | 深圳大学 | Predict method, system, equipment and the medium of building demolition waste yield |
-
2018
- 2018-12-29 CN CN201811646297.3A patent/CN111382645B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107516096A (en) * | 2016-06-15 | 2017-12-26 | 阿里巴巴集团控股有限公司 | A kind of character identifying method and device |
| CN108346428A (en) * | 2017-09-13 | 2018-07-31 | 腾讯科技（深圳）有限公司 | Voice activity detection and its method for establishing model, device, equipment and storage medium |
| CN108229364A (en) * | 2017-12-28 | 2018-06-29 | 百度在线网络技术（北京）有限公司 | Contour of building generation method, device, computer equipment and storage medium |
| CN108376232A (en) * | 2018-01-04 | 2018-08-07 | 北京星衡科技有限公司 | A kind of method and apparatus of automatic interpretation for remote sensing image |
| CN108764226A (en) * | 2018-04-13 | 2018-11-06 | 顺丰科技有限公司 | Image text recognition methods, device, equipment and its storage medium |
| 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 |
| CN108898603A (en) * | 2018-05-29 | 2018-11-27 | 北京佳格天地科技有限公司 | Plot segmenting system and method on satellite image |
| CN109034073A (en) * | 2018-07-30 | 2018-12-18 | 深圳大学 | Predict method, system, equipment and the medium of building demolition waste yield |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111382645A (en) | 2020-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111986099B (en) | Tillage monitoring method and system based on convolutional neural network with residual error correction fused | |
| CN111626947B (en) | Map vectorization sample enhancement method and system based on generation of countermeasure network | |
| CN110910343A (en) | Method and device for detecting pavement cracks and computer equipment | |
| CN111695488A (en) | Interest plane identification method, device, equipment and storage medium | |
| WO2021077947A1 (en) | Image processing method, apparatus and device, and storage medium | |
| CN110992384B (en) | Semi-automatic image data labeling method, electronic device and storage medium | |
| Rakhmadi et al. | Connected component labeling using components neighbors-scan labeling approach | |
| CN112883900B (en) | Method and device for bare-ground inversion of visible images of remote sensing images | |
| CN114241326B (en) | Progressive intelligent production method and system for ground feature elements of remote sensing images | |
| Grigillo et al. | Automated building extraction from IKONOS images in suburban areas | |
| CN111028261A (en) | High-precision semi-automatic image data annotation method, electronic device and storage medium | |
| CN111415364A (en) | Method, system and storage medium for converting image segmentation samples in computer vision | |
| CN110390724B (en) | SLAM method with instance segmentation | |
| CN111382645B (en) | Method and system for identifying overdue building in electronic map | |
| CN109657728B (en) | Sample production method and model training method | |
| CN115115954A (en) | Intelligent identification method for pine nematode disease area color-changing standing trees based on unmanned aerial vehicle remote sensing | |
| CN113158856B (en) | Processing method and device for extracting target area in remote sensing image | |
| CN110874170A (en) | Image area correction method, image segmentation method and device | |
| CN112632303A (en) | Distributed storage method, analysis method and device for image data | |
| CN112435274A (en) | Remote sensing image planar ground object extraction method based on object-oriented segmentation | |
| CN116597270A (en) | Road damage target detection method based on attention mechanism integrated learning network | |
| CN111144487A (en) | Method for establishing and updating remote sensing image sample library | |
| CN113282781A (en) | Image retrieval method and device | |
| CN113766147A (en) | Method for embedding image in video, and method and device for acquiring plane prediction model | |
| CN117593648B (en) | Remote sensing target building extraction method based on weak supervision learning |
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 |