CN115272853B - Industrial abandoned land identification method and product based on artificial intelligence technology and big data - Google Patents
Industrial abandoned land identification method and product based on artificial intelligence technology and big data Download PDFInfo
- Publication number
- CN115272853B CN115272853B CN202210892261.3A CN202210892261A CN115272853B CN 115272853 B CN115272853 B CN 115272853B CN 202210892261 A CN202210892261 A CN 202210892261A CN 115272853 B CN115272853 B CN 115272853B
- Authority
- CN
- China
- Prior art keywords
- industrial waste
- land
- waste land
- surface temperature
- potential
- 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
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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural 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
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W90/00—Enabling technologies or technologies with a potential or indirect contribution to greenhouse gas [GHG] emissions mitigation
Abstract
The embodiment of the invention provides an industrial abandoned land identification method and product based on an artificial intelligence technology and big data. According to the industrial waste land identification method provided by the embodiment of the invention, the pre-trained industrial waste land identification model is adopted, the potential industrial waste land is segmented from the image to be identified, and the multi-source data (at least comprising the people stream thermal data and the ground surface temperature data) is utilized to screen the idle industrial waste land from the potential industrial waste land. In the embodiment of the invention, a mature and efficient industrial waste land automatic identification technical method is established by means of a mature artificial intelligent image identification technology, open-source urban big data information and remote sensing data to form an industrial waste land identification method with space and time universality and strong operability, and a series of important strategic theoretical researches such as homeland space planning, urban updating and the like can be practically helped, and scientific and efficient promotion of ground practice can be realized.
Description
Technical Field
The embodiment of the invention relates to the technical field of information processing, in particular to an industrial waste land identification method and product based on an artificial intelligence technology and big data.
Background
The effective and reasonable reuse of the industrial waste land is an important grip for urban development and industrial transformation in the background of the post-industrial age, and the construction of a basic database of the industrial waste land is one of key technical links. The construction of the basic database of the industrial waste land firstly needs to identify the industrial waste land, and the technical bottleneck of the industrial waste land identification is that the land type is complex, the characteristic difference of the surface remote sensing image is large, and whether the identification target is abandoned or not is difficult to be determined by simply relying on the remote sensing image. In order to obtain a more accurate industrial abandoned land identification result, a multi-time phase image comparison is often required by a manual visual interpretation method, and the multi-time phase image comparison is confirmed by field investigation. Therefore, in the related technology, industrial abandoned land identification can be performed only in a limited and specific city or region level, and most of the dependent key data information is not open source data, so that the acquisition difficulty is high, the application range of the method is too limited, and the method does not have mobility in space and time levels.
The industrial waste land identification methods proposed in the related art are mainly classified into the following two types: an identification method based on high-resolution remote sensing images and an identification method based on multi-source spatial and non-spatial information.
The industrial waste identification method based on the high-resolution remote sensing image is mostly based on the spectrum information of the remote sensing image, and is used for carrying out industrial waste identification through remote sensing interpretation technologies such as supervision classification, support vector machines and the like, and has large fluctuation of accuracy and large difference of identification accuracy of different types of industrial waste. In addition, although the industrial abandoned land identification method based on the high-resolution remote sensing image has low labor cost and time input cost, the identification result is greatly influenced by the image data, and the acquisition of the open-source high-resolution remote sensing image is difficult.
The identification result of the industrial abandoned land identification method based on the multi-source space and non-space information is accurate, but a great deal of manpower and time are needed, and most of key information is not public information, and the method is provided by government and is difficult to popularize under the current national conditions of China.
It follows that a new method for identifying industrial waste is currently needed.
Disclosure of Invention
Embodiments of the present invention provide an industrial waste identification method, apparatus, electronic device, computer readable storage medium and computer program product based on artificial intelligence technology and big data, so as to solve at least some of the problems in the related art.
The first aspect of the embodiment of the invention provides an industrial abandoned land identification method based on an artificial intelligence technology and big data, which comprises the following steps:
inputting an image to be identified into a pre-trained industrial waste land identification model to obtain a potential industrial waste land contained in the image to be identified;
screening the potential industrial waste land from which the artificial flow heating power is lower than a first preset threshold value and the surface temperature is lower than a second preset threshold value, and determining the potential industrial waste land as the industrial waste land;
wherein the pre-trained industrial waste identification model pre-learns image features of a sample potential industrial waste comprising: industrial sites and industrial waste sites that are still running.
Optionally, the industrial waste identification model is trained according to the following steps:
obtaining a sample of a potential industrial waste;
training a preset semantic segmentation model by utilizing a sample potential industrial waste.
Optionally, screening a portion of the potential industrial waste having a low heat of people stream and a low surface temperature from the boundary of the potential industrial waste comprises:
acquiring the heat quantity of the people stream of the potential industrial waste land in a first preset time period;
inputting the people stream heat number to a GIS platform, and drawing a people stream heat distribution diagram;
acquiring a surface temperature value of the potential industrial waste land in a second preset time period;
inputting the surface temperature value into a GIS platform, and drawing a surface temperature distribution map;
and screening partial potential industrial waste land with the people stream heat lower than a first preset threshold and the ground surface temperature lower than a second preset threshold according to the people stream heat distribution diagram, the ground surface temperature distribution diagram and the potential industrial waste land.
Optionally, screening a part of the potential industrial waste land with the people stream heat lower than the first preset threshold and the ground surface temperature lower than the second preset threshold according to the people stream heat distribution map, the ground surface temperature distribution map and the potential industrial waste land comprises:
screening out a first potential industrial waste land with the people stream heat lower than a first preset threshold according to the people stream heat distribution diagram and the potential industrial waste land;
and screening out partial potential industrial waste land with the surface temperature lower than a second preset threshold according to the surface temperature distribution diagram and the first potential industrial waste land.
Optionally, the image to be identified is an open-source low-resolution remote sensing image of the target area.
Optionally, before screening the portion of the potential industrial waste from which the artificial flow heating power is below the first preset threshold and the surface temperature is below the second preset threshold, the method further comprises:
dividing the industrial waste land by using traffic road data to subdivide the boundary of the industrial waste land;
screening a portion of the potential industrial waste from which the artificial flow heating power is below a first preset threshold and the surface temperature is below a second preset threshold, comprising:
and screening the part of the potential industrial waste land with the artificial flow heat lower than the first preset threshold and the surface temperature lower than the second preset threshold from the segmented potential industrial waste land.
A second aspect of the embodiments of the present invention provides an industrial waste land identification device based on an artificial intelligence technique and big data, the device comprising:
the identification module is used for inputting an image to be identified into a pre-trained industrial waste land identification model to obtain a potential industrial waste land contained in the image to be identified;
a screening module, configured to screen a part of the potential industrial waste land from the potential industrial waste land, where the heating power of the artificial flow is lower than a first preset threshold value and the surface temperature is lower than a second preset threshold value, to determine the part of the potential industrial waste land as the industrial waste land;
wherein the pre-trained industrial waste identification model pre-learns image features of a sample potential industrial waste comprising: industrial sites and industrial waste sites that are still running.
Optionally, the industrial waste identification model is trained according to the following steps:
obtaining a sample of a potential industrial waste;
training a preset semantic segmentation model by utilizing a sample potential industrial waste.
Optionally, the screening module includes:
the first acquisition submodule is used for acquiring the heat quantity of the people flow of the potential industrial waste land in a first preset time period;
the first drawing submodule is used for inputting the people flow heat number into the GIS platform and drawing a people flow heat distribution diagram;
the second acquisition submodule is used for acquiring the surface temperature value of the potential industrial waste land in a second preset time period;
the second drawing submodule is used for inputting the ground surface temperature value into the GIS platform and drawing a ground surface temperature distribution map;
and the screening sub-module is used for screening partial potential industrial waste land with the people stream heat lower than a first preset threshold value and the ground surface temperature lower than a second preset threshold value according to the people stream heat distribution diagram, the ground surface temperature distribution diagram and the potential industrial waste land.
Optionally, the screening submodule is specifically configured to:
screening out a first potential industrial waste land with the people stream heat lower than a first preset threshold according to the people stream heat distribution diagram and the potential industrial waste land;
and screening out partial potential industrial waste land with the surface temperature lower than a second preset threshold according to the surface temperature distribution diagram and the first potential industrial waste land.
Optionally, the image to be identified is an open-source low-resolution remote sensing image of the target area.
Optionally, the apparatus further comprises:
the dividing module is used for dividing the industrial waste land by utilizing traffic road data so as to subdivide the boundary of the industrial waste land;
the screening module is specifically configured to: and screening the part of the potential industrial waste land with the artificial flow heat lower than the first preset threshold and the surface temperature lower than the second preset threshold from the segmented potential industrial waste land.
A third aspect of an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the industrial waste identification method based on artificial intelligence technology and big data according to the first aspect of the present invention when the program is executed.
A fourth aspect of the embodiments of the present invention provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method for identifying industrial waste land based on artificial intelligence technology and big data according to the first aspect of the present invention.
A fifth aspect of an embodiment of the invention provides a computer program product comprising computer programs/instructions which when executed by a processor implement the steps in the method for identifying industrial waste based on artificial intelligence technology and big data according to the first aspect of the invention.
According to the industrial waste land identification method based on the artificial intelligence technology and the big data, a pre-trained industrial waste land identification model is adopted, potential industrial waste lands are segmented from images to be identified, and multi-source data (at least comprising people flow thermal data and ground surface temperature data) are utilized to screen idle industrial waste lands from the potential industrial waste lands, so that the idle industrial waste lands are used as identification results.
In the embodiment of the invention, a mature and efficient industrial waste land automatic identification technical method is established by means of a mature artificial intelligent image identification technology, open-source urban big data information and remote sensing data to form an industrial waste land identification method with space and time universality and strong operability, and a series of important strategic theoretical researches such as homeland space planning, urban updating and the like can be practically helped, and scientific and efficient promotion of ground practice can be realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method of industrial waste identification in accordance with an embodiment of the present invention;
FIG. 2 is a flow chart of another industrial waste identification method of an embodiment of the present invention;
FIG. 3 is a schematic diagram of potential industrial waste prediction results obtained by a specific embodiment of an industrial waste identification method according to an embodiment of the present invention;
FIG. 4 is a schematic illustration of a heat distribution of a human flow obtained by a specific embodiment of an industrial waste land identification method according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a surface temperature distribution obtained by a specific example of an industrial waste land identification method according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of industrial waste identification results obtained by a specific embodiment of an industrial waste identification method according to an embodiment of the present invention;
fig. 7 is a block diagram of another industrial waste identification device according to an embodiment of the present invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Land utilization identification and monitoring technology based on artificial intelligence method and remote sensing image are all developing rapidly in training model construction and research application scene. The remote sensing image scene classification technology is gradually opened in the process of continuous maturity and is combined with multi-source big data, so that the remote sensing image scene classification technology can be widely used in various subjects.
Based on this, the embodiment of the present invention proposes an inventive concept: based on the semantic segmentation model, a potential industrial waste is identified from the image to be identified by utilizing a pre-trained industrial waste identification model. However, since the image features of the still-running industrial land and the industrial waste land are not greatly different, it is actually difficult for the industrial waste land identification model to accurately distinguish between them, and therefore, the embodiments of the present invention further propose: and screening the potential industrial waste land identified by the industrial waste land identification model based on the people stream heat and the ground surface temperature to determine the industrial waste land.
Referring to fig. 1, a flowchart of an industrial waste land identification method based on an artificial intelligence technology and big data according to an embodiment of the present invention is shown, and specifically, the industrial waste land identification method provided by the embodiment of the present invention may include the following steps:
s101, inputting an image to be identified into a pre-trained industrial waste land identification model to obtain a potential industrial waste land contained in the image to be identified.
Wherein the pre-trained industrial waste identification model pre-learns image features of a sample potential industrial waste comprising: industrial sites and industrial waste sites that are still running.
In the embodiment of the invention, the image to be identified is an open-source low-resolution remote sensing image of the target area. In the embodiment of the invention, after a target area to be identified (such as a certain province, a certain city or a certain area) is determined, an open-source low-resolution remote sensing image of the area can be obtained and used as an image to be identified, and a pre-trained industrial abandoned land identification model is input for identification.
Considering that the industrial land has obvious difference in surface characteristics from other urban construction lands (such as single surface characteristics of industrial factory buildings, group distribution characteristics and obvious changes of industrial production facilities on the surface), the industrial land and other urban construction lands can be further distinguished according to the image characteristics. But it is difficult to distinguish between an industrial site that is still running and an industrial site that has been left unused, based on image characteristics. Thus, in the embodiment of the invention, the pre-trained industrial waste identification model does not distinguish between the two potential industrial waste areas, but rather the industrial waste areas are screened from the potential industrial waste areas by means of a subsequent screening step.
The difficulty and the high cost of acquiring the high-resolution remote sensing image are considered, and the data volume for processing the high-resolution remote sensing influence is large, so that the processing time is high. The embodiment of the invention provides an industrial abandoned land identification method based on an open source low-resolution remote sensing image. Based on the above, in the embodiment of the invention, in the process of model training, the used sample potential industrial waste is also an open source low-resolution remote sensing image.
And S102, screening the part of the potential industrial waste land with the artificial flow heat lower than a first preset threshold value and the surface temperature lower than a second preset threshold value from the potential industrial waste land, and determining the part as the industrial waste land.
Considering that an industrial waste that has been left unused must be left unattended or have only a small number of people going to and staying, and that an industrial waste that is not being produced must produce significantly less heat radiation than an industrial waste that is still being produced. Thus, in the embodiment of the present invention, it is proposed that: the industrial abandoned land has two remarkable characteristics of low heat of people stream and low surface temperature. Considering that the situation that the industrial production only needs a small amount of personnel to manage on site possibly occurs due to the high degree of automation of the industrial production, the invention creates the condition of judging the industrial waste land to be defined as: the heat value of the traffic flow in the land is significantly lower and the highest value of the surface temperature in the land is significantly lower than the average surface temperature of the known running industrial area.
Thus, the embodiment of the invention provides: and selecting urban mass flow thermal data and ground surface temperature inversion data as the basis, and screening the idle industrial waste land from the potential industrial waste land as the identification result.
In an alternative embodiment, the step S102 includes the substeps of:
s1021, acquiring the heat quantity of the people stream of the potential industrial waste land in a first preset time period.
S1022, inputting the people stream heat number to a GIS platform, and drawing a people stream heat distribution diagram.
S1023, obtaining the surface temperature value of the potential industrial waste land in a second preset time period.
S1024, inputting the surface temperature value into a GIS platform, and drawing a surface temperature distribution map.
S1025, screening partial potential industrial waste land with the people stream heat lower than a first preset threshold and the ground surface temperature lower than a second preset threshold according to the people stream heat distribution diagram, the ground surface temperature distribution diagram and the potential industrial waste land.
According to the embodiment of the invention, a Geographic Information System (GIS) platform can be utilized to obtain the induced abortion thermal power distribution diagram and the surface temperature distribution diagram by acquiring the induced abortion thermal power data and the surface temperature data.
In the embodiment of the invention, the first preset time period can be a working time period of a working day so as to determine the people stream thermal data of the working time period, thereby more accurately screening out the still-running industrial land.
In an embodiment of the present invention, the second preset time period may be an operation time period of a general factory to determine a surface temperature of the potential industrial waste in the operation time period, so as to accurately reflect whether the potential industrial waste is still operating.
In the embodiment of the present invention, the first preset time period and the second preset time period may be the same preset time period or different preset time periods.
In the embodiment of the invention, after the induced abortion thermal distribution diagram, the ground surface temperature distribution diagram and the potential industrial waste land are obtained, the idle industrial waste land can be determined according to the first preset threshold value and the second preset threshold value.
In the embodiment of the invention, the first preset threshold and the second preset threshold can be determined by a technician according to the actual situation of the target area. For example: the first preset threshold may be determined based on an average flow of people in the target area and the second preset threshold may be determined based on an average surface temperature of a common industrial area being operated in the target area.
Specifically, S1025 includes:
screening out a first potential industrial waste land with the people stream heat lower than a first preset threshold according to the people stream heat distribution diagram and the potential industrial waste land; and screening out partial potential industrial waste land with the surface temperature lower than a second preset threshold according to the surface temperature distribution diagram and the first potential industrial waste land.
In the embodiment of the invention, first-round screening can be performed by using the artificial flow thermodynamic data, and further second-round screening can be performed by using the surface temperature data, so that the still-running industrial land is scientifically and efficiently stripped from the identification result of the potential industrial waste land, and the identification result of the industrial waste land is obtained.
In an alternative embodiment, the industrial waste identification model is trained according to the following steps:
s1, obtaining a sample potential industrial waste.
In the embodiment of the invention, the sample potential industrial waste is an open-source low-resolution remote sensing image.
Specifically, in the embodiment of the invention, a Sentinel-2A open source remote sensing image with a pixel resolution of 10 meters is applied to a geographic information system platform to visually identify 536 potential industrial waste lands of yellow stones, tangshan and new 3 cities, so that sample potential industrial waste lands including still-running industrial lands and industrial waste lands are obtained and serve as data sources of a training set.
In the embodiment of the invention, the model training is performed by using the open-source low-resolution remote sensing image data, so that the data acquisition cost can be reduced, and the model training efficiency and the model processing efficiency can be improved. Therefore, the industrial waste land identification method provided by the embodiment of the invention is low in application cost and high in operability.
In the embodiment of the invention, a grid layer of the 3-band pseudo-color remote sensing image formed by fusing RGB wave bands and a grid layer of a sample potential industrial waste land are assigned corresponding assignment in a GIS platform, a potential industrial waste land part is assigned a 1 value, a non-industrial attribute land part is assigned a 0 value, and the potential industrial waste land part is divided into training samples of 256 pixels by 256 pixels. In order to ensure the training effect of the subsequent model, the training sample set in the embodiment of the invention meets the following requirements: (1) Including a large number of potential industrial waste and non-industrial property sites with different image features to promote robustness of the model; (2) A sufficient number of training samples are included to ensure that model training does not suffer from overfitting problems. Finally, a training sample set consisting of 1140 256 pixel by 256 pixel assigned tif image samples containing the potential industrial site is obtained.
S2, training a preset semantic segmentation model by utilizing a sample potential industrial abandoned place.
In the embodiment of the invention, the preset semantic segmentation model can distinguish the region of the target category from the image to be processed.
In the embodiment of the invention, all samples of the training sample set are distributed according to 20% -80% to obtain a test set and a training set. Specifically, in the embodiment of the present invention, the preset semantic segmentation model may be an HRNet model. Training of the HRNet model can be understood as two processes of forward propagation, which is responsible for computing the image on the input network model and outputting the predicted value, and backward propagation, which is responsible for optimizing and adjusting the weights and biased gradients of the layers.
In the embodiment of the invention, the input of the model is label samples with the size of 256 multiplied by 256 of 3 channels, 2 groups of training samples x and corresponding labels y are randomly extracted in each batch (batch size) in a training set, each layer in the HRNet model is sequentially entered for operation until the training set traverses one time and belongs to one training round (epoch), the class probability distribution diagram of each pixel of the image is obtained through the activation of a Sigmoid function, the predicted value of the network is expressed as y_pred, and the whole process is the forward propagation of the model. The forward propagation obtains an image probability distribution diagram, and the difference (loss) between the predicted value y_pred of the model and the true value y of the data set is calculated by using a loss function, wherein the loss is more towards 0, so that the better the model effect is, and the worse the model effect is. The back propagation process can be understood simply as the process of minimizing the loss function, typically using a gradient descent algorithm to optimize tens of thousands of parameters in the network to find the best combination of parameters for the model. In the embodiment of the invention, an Adam optimization algorithm is adopted, through 50 training iterations, when the gradient is reduced to global optimum, the loss function is not reduced any more, the network realizes the optimum parameter combination, and the model training is completed, so that the industrial abandoned land identification model is obtained.
In the embodiment of the invention, in the training process of the model, the industrial land still running and the idle industrial waste land are not distinguished, but are unified as potential industrial waste land parts to be assigned 1 values, and non-industrial attribute land parts to be assigned 0 values. The obtained industrial waste identification model is trained, the potential industrial waste obtained through identification comprises the industrial waste still in operation and the industrial waste which is idle, the identification errors of the industrial waste still in operation and the industrial waste which is idle are reduced by avoiding the dependence on image characteristics.
Referring to fig. 2, a flowchart of an industrial waste land identification method based on an artificial intelligence technology and big data according to an embodiment of the present invention is shown, and specifically, the industrial waste land identification method provided by the embodiment of the present invention may include the following steps:
s201, inputting an image to be identified into a pre-trained industrial waste land identification model to obtain a potential industrial waste land contained in the image to be identified.
This step is similar to the step S101 described above, and will not be described again here.
S202, dividing the industrial waste land by using traffic road data to subdivide the boundary of the industrial waste land.
In the embodiment of the invention, due to the problem of remote sensing image precision, the prediction and identification result cannot reasonably divide a large-area potential industrial waste area through an urban road. It can be understood that the land divided by the main traffic road of the city is rarely the same factory or enterprise, so in order to improve the recognition accuracy, the situation that a large land is an industrial waste land due to the heat of people flow or abnormal value of the surface temperature in a certain place in the subsequent screening process is avoided.
And S203, selecting a part of the potential industrial waste land with the artificial flow heat lower than a first preset threshold value and the surface temperature lower than a second preset threshold value from the segmented potential industrial waste land, and determining the part as the industrial waste land.
This step is similar to step S102 described above, and will not be described again here.
For easy understanding, an industrial waste land identification method based on artificial intelligence technology and big data provided in the embodiment of the present invention is further explained below by means of a specific embodiment, and it can be understood that this embodiment is only an example:
the following will describe in detail a central urban area of Tangshan city as an example of the target recognition area. And determining a Feng nan region, an open flat region, a Feng run region, a Lu nan region and part of 5 administrative regions of the Lu North region as identification target regions according to the land layout plan of the central urban region of Tang mountain city and the administrative region range vector data of Tang mountain city. Meanwhile, in order to ensure the accuracy of the results obtained after the screening of the people stream thermal data and the ground surface temperature data, the data date used in the embodiment is 12 months of working day data of Tangshan city. The working days ensure that the people stream thermal data of the target area cannot be subjected to holiday images, so that the situation that the people stream thermal value is low in the industrial land still running and is confused with the industrial abandoned land is caused; the winter time is selected to ensure that the surface temperature of the still-running industrial land is significantly higher than that of the industrial waste land which does not generate heat radiation due to heat radiation generated by industrial production. Therefore, the remote sensing image and the people stream heating power and the ground surface temperature data on the 12 th month and 10 th year of 2020 are selected as the basic data of the image and the screening of the potential industrial abandoned land AI identification.
First, a pre-trained industrial waste identification model is used to predict a remote sensing image of a target area, and a prediction result of a potential industrial waste is obtained, as shown in fig. 3, wherein a white part represents the potential industrial waste.
And then, crawling the heat numbers of the people stream in the working time period of 10:00 am working days in the case range of 2020 by a crawler in real time at the background of the hundred-degree map website, performing vectorization calculation after splicing, and importing the vector data into a GIS platform. The precision of the crawled people stream thermal data is 100 meters multiplied by 100 meters, and after vectorization calculation, the POP value of each vector point in the image layer is the people stream thermal average value in the square range of 1 hectare with the vector point as the center. If the people flow thermal value is less than 100 people in the 1 hectare range with the vector point as the center, the POP value is recorded as 0, and the vectorization calculation result is not counted, so that the people flow thermal distribution diagram is obtained, as shown in fig. 4. The statistics shows that the thermal maximum value of the people flow in the central urban area of Tangshan city at the time point is 2277 people/hectare, the minimum value counted into the statistics is 247 people/hectare, and the average value is 888.01 people/hectare.
Then, performing ground surface temperature inversion on Landsat8 images of the prediction range in the period by adopting an atmosphere correction method to obtain a ground surface temperature data range of-7.3-19.2 ℃, wherein the ground surface temperature data range is shown in FIG. 5, and a schematic diagram of ground surface temperature distribution in Tangshan city is shown. Wherein lighter color of each region indicates lower temperature. After the industrial land which is determined by combining the thermal data of the people stream and is still running is calculated, if the land surface temperature exceeds 7 ℃, a large amount of heat radiation formed by self-heating exists in the area, so that the highest value of the land surface temperature of the industrial abandoned land at the time point is inevitably lower than 7 ℃.
Finally, inputting the identification result of the potential abandoned land in the central urban area 2020 of Tangshan city, 12 months and 10 days, the people stream thermal data and the ground surface temperature data into a GIS platform, and screening out the land blocks with the people stream thermal value smaller than 100 people/hectare and the highest ground surface temperature value smaller than 7 ℃ in the area to obtain the final identification result of the industrial abandoned land, as shown in fig. 6, wherein the dark gray part represents the industrial abandoned land as the obtained result. Through statistics, total area of 431 industrial abandoned lands is about 24.24 square kilometers, and other geographic information data such as the area of each abandoned land, longitude and latitude of a central point and the like can be obtained from a GIS map layer attribute table.
In the embodiment of the invention, the Sentinel-2A open source remote sensing image with the resolution of 10 m pixels is used as a data source to perform model training and industrial waste land identification, and the identification result of the model is further screened according to multi-source data (at least comprising people stream data, ground surface temperature data and road data) so as to make more accurate and scientific judgment on whether the potential industrial waste land obtained by the model is truly abandoned or not, thereby ensuring that the identification result is more accurate.
By adopting the industrial abandoned land identification method provided by the embodiment of the invention, the identification efficiency can be greatly improved and the manpower and material resources can be saved under the condition that the accuracy of the identification result is ensured to be reliable. Meanwhile, the open source resolution data is used as a data basis, the data acquisition path is simple, the technical use threshold is low, and the application scene can be effectively enlarged.
Based on the same inventive concept, the embodiment of the invention provides an industrial waste land identification device based on an artificial intelligence technology and big data, and referring to fig. 7, fig. 7 is a schematic diagram of the industrial waste land identification device provided by the embodiment of the invention. As shown in fig. 7, the apparatus includes:
the identifying module 701 is configured to input an image to be identified into a pre-trained industrial waste land identifying model, so as to obtain a potential industrial waste land contained in the image to be identified;
a screening module 702, configured to screen a part of the potential industrial waste land from the potential industrial waste land, where the heating power of the artificial flow is lower than a first preset threshold value and the surface temperature is lower than a second preset threshold value, and determine the part as the industrial waste land;
wherein the pre-trained industrial waste identification model pre-learns image features of a sample potential industrial waste comprising: industrial sites and industrial waste sites that are still running.
Optionally, the industrial waste identification model is trained according to the following steps:
obtaining a sample of a potential industrial waste;
training a preset semantic segmentation model by utilizing a sample potential industrial waste.
Optionally, the screening module 702 includes:
the first acquisition submodule is used for acquiring the heat quantity of the people flow of the potential industrial waste land in a first preset time period;
the first drawing submodule is used for inputting the people flow heat number into the GIS platform and drawing a people flow heat distribution diagram;
the second acquisition submodule is used for acquiring the surface temperature value of the potential industrial waste land in a second preset time period;
the second drawing submodule is used for inputting the ground surface temperature value into the GIS platform and drawing a ground surface temperature distribution map;
and the screening sub-module is used for screening partial potential industrial waste land with the people stream heat lower than a first preset threshold value and the ground surface temperature lower than a second preset threshold value according to the people stream heat distribution diagram, the ground surface temperature distribution diagram and the potential industrial waste land.
Optionally, the screening submodule is specifically configured to:
screening out a first potential industrial waste land with the people stream heat lower than a first preset threshold according to the people stream heat distribution diagram and the potential industrial waste land;
and screening out partial potential industrial waste land with the surface temperature lower than a second preset threshold according to the surface temperature distribution diagram and the first potential industrial waste land.
Optionally, the image to be identified is an open source low resolution remote sensing image.
Optionally, the apparatus further comprises:
and the segmentation module is used for segmenting the industrial waste land by using the traffic road data so as to subdivide the boundary of the industrial waste land.
Based on the same inventive concept, an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the steps in the industrial waste identification method according to any of the embodiments above when executing the program.
Based on the same inventive concept, an embodiment of the present invention provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps in the industrial waste identification method based on artificial intelligence technology and big data as described in any of the above embodiments.
Based on the same inventive concept, embodiments of the present invention provide a computer program product comprising a computer program/instruction which, when executed by a processor, implements the steps of the industrial waste identification method based on artificial intelligence technology and big data described in any of the above embodiments.
For the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.
It will be apparent to those skilled in the art that embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The above description of the invention provides an industrial waste identification method, device, electronic equipment, computer readable storage medium and computer program product based on artificial intelligence technology and big data, and specific examples are applied to illustrate the principle and implementation of the invention, and the description of the above examples is only used to help understand the method and core idea of the invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.
Claims (8)
1. An industrial waste land identification method based on artificial intelligence technology and big data, which is characterized by comprising the following steps:
inputting an image to be identified into a pre-trained industrial waste land identification model to obtain a potential industrial waste land contained in the image to be identified;
screening the potential industrial waste land from which the artificial flow heating power is lower than a first preset threshold value and the surface temperature is lower than a second preset threshold value, and determining the potential industrial waste land as the industrial waste land;
wherein the pre-trained industrial waste identification model pre-learns image features of a sample potential industrial waste comprising: industrial sites and industrial waste sites that are still running;
screening a portion of the potential industrial waste having a low heat of human flow and a low surface temperature from the boundary of the potential industrial waste, comprising:
acquiring the heat quantity of the people stream of the potential industrial waste land in a first preset time period;
inputting the people stream heat number to a GIS platform, and drawing a people stream heat distribution diagram;
acquiring a surface temperature value of the potential industrial waste land in a second preset time period;
inputting the surface temperature value into a GIS platform, and drawing a surface temperature distribution map;
and screening partial potential industrial waste land with the people stream heat lower than a first preset threshold and the ground surface temperature lower than a second preset threshold according to the people stream heat distribution diagram, the ground surface temperature distribution diagram and the potential industrial waste land.
2. The industrial waste identification method based on artificial intelligence technology and big data according to claim 1, wherein the industrial waste identification model is trained according to the following steps:
obtaining a sample of a potential industrial waste;
training a preset semantic segmentation model by utilizing a sample potential industrial waste.
3. The method for identifying industrial waste land based on artificial intelligence technology and big data according to claim 1, wherein screening the part of the potential industrial waste land with the people stream heat lower than a first preset threshold and the ground surface temperature lower than a second preset threshold according to the people stream heat distribution map, the ground surface temperature distribution map and the potential industrial waste land comprises:
screening out a first potential industrial waste land with the people stream heat lower than a first preset threshold according to the people stream heat distribution diagram and the potential industrial waste land;
and screening out partial potential industrial waste land with the surface temperature lower than a second preset threshold according to the surface temperature distribution diagram and the first potential industrial waste land.
4. The method for identifying industrial waste land based on artificial intelligence technology and big data according to any one of claims 1-3, wherein the image to be identified is an open source low resolution remote sensing image of a target area.
5. A method of identifying industrial waste based on artificial intelligence technology and big data according to any of claims 1-3, wherein prior to screening the portion of the potential industrial waste from which the artificial flow heat is below a first preset threshold and the surface temperature is below a second preset threshold, the method further comprises:
dividing the industrial waste land by using traffic road data to subdivide the boundary of the industrial waste land;
screening a portion of the potential industrial waste from which the artificial flow heating power is below a first preset threshold and the surface temperature is below a second preset threshold, comprising:
and screening the part of the potential industrial waste land with the artificial flow heat lower than the first preset threshold and the surface temperature lower than the second preset threshold from the segmented potential industrial waste land.
6. An industrial waste land identification device based on artificial intelligence technology and big data, the device comprising:
the identification module is used for inputting an image to be identified into a pre-trained industrial waste land identification model to obtain a potential industrial waste land contained in the image to be identified;
a screening module, configured to screen a part of the potential industrial waste land from the potential industrial waste land, where the heating power of the artificial flow is lower than a first preset threshold value and the surface temperature is lower than a second preset threshold value, to determine the part of the potential industrial waste land as the industrial waste land;
wherein the pre-trained industrial waste identification model pre-learns image features of a sample potential industrial waste comprising: industrial sites and industrial waste sites that are still running;
the screening module comprises:
the first acquisition submodule is used for acquiring the heat quantity of the people flow of the potential industrial waste land in a first preset time period;
the first drawing submodule is used for inputting the people flow heat number into the GIS platform and drawing a people flow heat distribution diagram;
the second acquisition submodule is used for acquiring the surface temperature value of the potential industrial waste land in a second preset time period;
the second drawing submodule is used for inputting the ground surface temperature value into the GIS platform and drawing a ground surface temperature distribution map;
and the screening sub-module is used for screening partial potential industrial waste land with the people stream heat lower than a first preset threshold value and the ground surface temperature lower than a second preset threshold value according to the people stream heat distribution diagram, the ground surface temperature distribution diagram and the potential industrial waste land.
7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the artificial intelligence technology and big data based industrial waste identification method of any of claims 1-5 when the program is executed.
8. A computer readable storage medium having stored thereon a computer program, characterized in that the program when executed by a processor implements the steps of the artificial intelligence technology and big data based industrial waste identification method according to any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210892261.3A CN115272853B (en) | 2022-07-27 | 2022-07-27 | Industrial abandoned land identification method and product based on artificial intelligence technology and big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210892261.3A CN115272853B (en) | 2022-07-27 | 2022-07-27 | Industrial abandoned land identification method and product based on artificial intelligence technology and big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115272853A CN115272853A (en) | 2022-11-01 |
| CN115272853B true CN115272853B (en) | 2023-06-27 |
Family
ID=83770121
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210892261.3A Active CN115272853B (en) | 2022-07-27 | 2022-07-27 | Industrial abandoned land identification method and product based on artificial intelligence technology and big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115272853B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117408438A (en) * | 2023-12-14 | 2024-01-16 | 中国测绘科学研究院 | Industrial park low-utility land extraction method and system based on multi-source big data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109101955A (en) * | 2018-09-12 | 2018-12-28 | 北京英视睿达科技有限公司 | Industrial heat anomaly area recognizing method based on Multi-sensor satellite remote sensing |
| CN113516059A (en) * | 2021-06-23 | 2021-10-19 | 南京华高生态环境遥感技术研究院有限公司 | Solid waste identification method and device, electronic device and storage medium |
| CN114021348A (en) * | 2021-11-05 | 2022-02-08 | 中国矿业大学(北京) | Mining area vegetation carbon sink remote sensing inversion method of fine land utilization type |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8675917B2 (en) * | 2011-10-31 | 2014-03-18 | International Business Machines Corporation | Abandoned object recognition using pedestrian detection |
| CN108038116A (en) * | 2017-10-24 | 2018-05-15 | 安徽四创电子股份有限公司 | A kind of density of stream of people monitoring method based on GIS |
| CN110263675B (en) * | 2019-06-03 | 2024-02-20 | 武汉联一合立技术有限公司 | Garbage target identification system and method of community security robot |
| CN111629335A (en) * | 2020-05-29 | 2020-09-04 | 四川亨通网智科技有限公司 | Method and system for realizing real-time people flow thermodynamic diagram of scenic spot based on big data |
| CN112561971A (en) * | 2020-12-16 | 2021-03-26 | 珠海格力电器股份有限公司 | People flow statistical method, device, equipment and storage medium |
| CN113780057A (en) * | 2021-07-23 | 2021-12-10 | 贵州图智信息技术有限公司 | Idle land identification method and device |
| CN114241332A (en) * | 2021-12-17 | 2022-03-25 | 深圳博沃智慧科技有限公司 | Deep learning-based solid waste field identification method and device and storage medium |
-
2022
- 2022-07-27 CN CN202210892261.3A patent/CN115272853B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109101955A (en) * | 2018-09-12 | 2018-12-28 | 北京英视睿达科技有限公司 | Industrial heat anomaly area recognizing method based on Multi-sensor satellite remote sensing |
| CN113516059A (en) * | 2021-06-23 | 2021-10-19 | 南京华高生态环境遥感技术研究院有限公司 | Solid waste identification method and device, electronic device and storage medium |
| CN114021348A (en) * | 2021-11-05 | 2022-02-08 | 中国矿业大学(北京) | Mining area vegetation carbon sink remote sensing inversion method of fine land utilization type |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115272853A (en) | 2022-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liping et al. | Monitoring and predicting land use and land cover changes using remote sensing and GIS techniques—A case study of a hilly area, Jiangle, China | |
| Yatoo et al. | Monitoring land use changes and its future prospects using cellular automata simulation and artificial neural network for Ahmedabad city, India | |
| Alsharif et al. | Urban sprawl analysis of Tripoli Metropolitan city (Libya) using remote sensing data and multivariate logistic regression model | |
| Hegazy et al. | Monitoring urban growth and land use change detection with GIS and remote sensing techniques in Daqahlia governorate Egypt | |
| Ke et al. | A partitioned and asynchronous cellular automata model for urban growth simulation | |
| CN116385689B (en) | Visual information management method, system and medium for site pollution data | |
| CN112132264A (en) | Regional exhaust emission prediction method and system based on space-time residual perception network | |
| CN110727717B (en) | Monitoring method, device, equipment and storage medium for gridding atmospheric pollution intensity | |
| Zheng et al. | An eiot system designed for ecological and environmental management of the xianghe segment of china’s grand canal | |
| CN115272853B (en) | Industrial abandoned land identification method and product based on artificial intelligence technology and big data | |
| Chettry et al. | Delineating urban growth boundary using remote sensing, ANN-MLP and CA model: a case study of Thiruvananthapuram urban agglomeration, India | |
| Yang et al. | Feature selection in conditional random fields for map matching of GPS trajectories | |
| Salman Aal-shamkhi et al. | Extraction and modeling of urban sprawl development in Karbala City using VHR satellite imagery | |
| CN116703004B (en) | Water system river basin intelligent patrol method and device based on pre-training model | |
| CN113780175B (en) | Remote sensing identification method for typhoon and storm landslide in high vegetation coverage area | |
| Albertengo et al. | Short term urban traffic forecasting using deep learning | |
| Gao et al. | Determining the weights of influencing factors of construction lands with a neural network algorithm: a case study based on Ya’an City | |
| CN115169466A (en) | Method and device for drawing image of land, electronic equipment and computer readable medium | |
| Azizi et al. | Prediction of groundwater level using GMDH artificial neural network based on climate change scenarios | |
| CN113919533A (en) | Air quality tracing and forecasting method based on artificial intelligence | |
| Tamma et al. | Past influences present: mammalian species from different biogeographic pools sort environmentally in the Indian subcontinent | |
| CN112215096B (en) | Remote sensing image town extraction method and device based on scene and pixel information | |
| EP4209951A1 (en) | A method, a computer program product and a device for dynamic spatial anonymization of vehicle data in a cloud environment | |
| CN114821339A (en) | Method for identifying state of building rubbish digestion field by using high-resolution remote sensing image | |
| Ferreira et al. | SLkit: An R package for property extraction and analysis of multiple Sensing Layers |
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 |