CN117095364B - Urban environment early warning method and system based on AI intelligent garbage recognition - Google Patents

Urban environment early warning method and system based on AI intelligent garbage recognition Download PDF

Info

Publication number
CN117095364B
CN117095364B CN202311363846.7A CN202311363846A CN117095364B CN 117095364 B CN117095364 B CN 117095364B CN 202311363846 A CN202311363846 A CN 202311363846A CN 117095364 B CN117095364 B CN 117095364B
Authority
CN
China
Prior art keywords
garbage
monitoring area
urban
acquiring
monitoring
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202311363846.7A
Other languages
Chinese (zh)
Other versions
CN117095364A (en
Inventor
楚敏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Runjiang Technology Co ltd
Original Assignee
Shenzhen Runjiang Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Runjiang Technology Co ltd filed Critical Shenzhen Runjiang Technology Co ltd
Priority to CN202311363846.7A priority Critical patent/CN117095364B/en
Publication of CN117095364A publication Critical patent/CN117095364A/en
Application granted granted Critical
Publication of CN117095364B publication Critical patent/CN117095364B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/50Context or environment of the image
    • G06V20/52Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/0464Convolutional networks [CNN, ConvNet]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • G06N3/084Backpropagation, e.g. using gradient descent
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0631Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06316Sequencing of tasks or work
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10Services
    • G06Q50/26Government or public services
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/05Geographic models
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/26Segmentation 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
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/30Noise filtering
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/44Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/70Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/70Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Software Systems (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Economics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Tourism & Hospitality (AREA)
  • Strategic Management (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Development Economics (AREA)
  • Medical Informatics (AREA)
  • Educational Administration (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Databases & Information Systems (AREA)
  • Biophysics (AREA)
  • Marketing (AREA)
  • General Engineering & Computer Science (AREA)
  • Molecular Biology (AREA)
  • Data Mining & Analysis (AREA)
  • General Business, Economics & Management (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Geometry (AREA)
  • Game Theory and Decision Science (AREA)
  • Primary Health Care (AREA)
  • Remote Sensing (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Computer Graphics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses an urban environment early warning method and system based on AI intelligent garbage recognition, comprising the following steps: acquiring a garbage monitoring area based on an urban monitoring system, and determining garbage types in the garbage monitoring area by using an AI intelligent garbage identification model; acquiring space-time characteristics and garbage characteristics of a garbage monitoring area, and realizing associated early warning of the garbage monitoring area; and finally, carrying out vehicle scheduling on the urban cleaning vehicles by acquiring the garbage cleaning sequence and analyzing the distribution condition, the load condition and the urban traffic condition of the urban cleaning vehicles. The method can intelligently identify and classify the garbage on the urban road, clear and treat the garbage, beautify urban environment, improve the air quality and life quality of the city, and is beneficial to beautiful construction of the city.

Description

Urban environment early warning method and system based on AI intelligent garbage recognition
Technical Field
The invention relates to the field of garbage identification, in particular to an urban environment early warning method and system based on AI intelligent garbage identification.
Background
In urban roads, due to factors such as daily activities of people or work production, garbage such as food garbage, industrial garbage and the like can be generated on the urban roads, so that urban appearance environment appearance of cities is seriously influenced, inconvenience is brought to various aspects of life of people, and even human health is influenced. Because the types of garbage are different, the garbage treatment modes are different, the garbage types are required to be identified, the mode of manually identifying the garbage is relatively backward, the efficiency is low, the identification efficiency of the garbage types can be improved by using the AI intelligent garbage identification, and the garbage clearing and transporting sequence is correspondingly generated and the clearing and transporting vehicles are planned based on the garbage type identification result.
Disclosure of Invention
The invention overcomes the defects of the prior art and provides an urban environment early warning method and system based on AI intelligent garbage identification.
In order to achieve the above purpose, the invention adopts the following technical scheme:
the invention provides a city appearance environment early warning method based on AI intelligent garbage recognition, which comprises the following steps:
based on the urban monitoring system, real-time monitoring is carried out on the urban road, and a garbage monitoring area is determined based on the real-time monitoring result;
constructing an AI intelligent garbage identification model, and determining garbage types in a garbage monitoring area by using the AI intelligent garbage identification model;
acquiring space-time characteristics of the garbage monitoring area, acquiring the garbage characteristics of the garbage monitoring area based on the space-time characteristics, and carrying out associated early warning on the garbage monitoring area according to the garbage characteristics;
and acquiring a garbage clearing sequence, analyzing the distribution condition, the load condition and the urban traffic condition of the urban clearing vehicles, and carrying out vehicle scheduling on the urban clearing vehicles.
Further, in a preferred embodiment of the present invention, the urban road is monitored in real time based on the urban monitoring system, and the garbage monitoring area is determined based on the real-time monitoring result, which specifically includes:
Acquiring a real-time monitoring image of an urban road based on an urban monitoring system;
performing image graying treatment and image noise reduction treatment on the real-time monitoring image of the urban road to obtain a preprocessed urban road image;
based on an edge feature segmentation method, extracting image features of the preprocessed urban road image to obtain road surface parameters of the urban road;
based on historical data acquisition, constructing an urban three-dimensional model, and importing road surface parameters of the urban road into the urban three-dimensional model to update model parameters to obtain the urban road three-dimensional model;
obtaining model parameters of the urban road three-dimensional model, guiding the model parameters into a convolutional neural network model for prediction to obtain garbage position prediction data, and guiding the garbage position prediction data into the urban road three-dimensional model to obtain the coordinate position of garbage on the urban road;
and (3) presetting a standard garbage packing density and a standard garbage packing area, analyzing the coordinate position of the garbage on the urban road, and defining a region with the garbage packing density and the garbage packing area larger than the standard garbage packing density and the standard garbage packing area as a garbage monitoring region.
Further, in a preferred embodiment of the present invention, the construction of the AI-intelligent garbage identification model, and the determination of the garbage type in the garbage monitoring area using the AI-intelligent garbage identification model, specifically includes:
Based on big data retrieval, acquiring all types of junk information, and dividing the junk information into a training set and a testing set, wherein the training set contains sample data of the junk information of all types;
introducing a decision tree model, acquiring a characteristic value and a segmentation value of sample data, determining a segmentation point of the sample data in the decision tree model based on the characteristic value and the segmentation value, dividing the sample data at the segmentation point by the decision tree model, re-acquiring the segmentation point after classification, continuing to divide, analyzing the sample data after each division, stopping the corresponding division operation of the sample data when only one garbage information exists in the sample data, and outputting the latest segmentation point which is defined as a leaf node;
introducing a singular decomposition algorithm, performing singular decomposition on samples in leaf nodes to generate a singular matrix, introducing a cosine measurement algorithm to perform cosine measurement value calculation on the singular matrix, and defining leaf nodes with cosine measurement values larger than a preset value as outliers;
performing iterative splitting treatment on the outliers by using a genetic algorithm, ending the iterative splitting treatment when the iteration times reach a preset value and the outliers are not present, outputting all leaf nodes to obtain a trained decision tree model, and constructing an AI intelligent garbage recognition model based on the trained decision tree model;
And importing model parameters of the urban road three-dimensional model of the garbage monitoring area into an AI intelligent garbage identification model to identify garbage types, so as to obtain garbage type information in the garbage monitoring area.
Further, in a preferred embodiment of the present invention, the acquiring the space-time characteristic of the garbage monitoring area, acquiring the garbage characteristic of the garbage monitoring area based on the space-time characteristic, and performing the associated early warning on the garbage monitoring area according to the garbage characteristic, specifically includes:
acquiring the garbage filling density and the garbage filling area of a garbage monitoring area based on the coordinate position of the garbage on the urban road;
combining the garbage filling density and the garbage filling area of the garbage monitoring area to generate a training set and a testing set, introducing the training set into a convolutional layer of a TCN model to perform causal convolutional processing to obtain data features of different time scales, adding the data in the training set with the data features of different time scales through residual connection, introducing the data features into a pooling layer to perform maximum pooling processing, and outputting pooling values of different time scales;
importing the pooling values of different time scales into a full-connection layer of a TCN model for reverse training to obtain a trained TCN model, and generating time characteristics of a garbage monitoring area based on the trained TCN model;
Combining the garbage filling density, the garbage filling area and the surrounding environment parameters of the garbage monitoring area, acquiring the spatial characteristics of the garbage monitoring area based on an RNN model, and combining the temporal characteristics of the garbage monitoring area and the spatial characteristics of the garbage monitoring area to generate the space-time characteristics of the garbage monitoring area, wherein the space-time characteristics of the garbage monitoring area represent the quantity and position change condition of garbage in the garbage monitoring area in different time periods;
based on the space-time characteristics of the garbage monitoring areas, the garbage characteristics of the garbage monitoring areas are obtained, and association early warning is carried out on different garbage monitoring areas based on the garbage characteristics.
Further, in a preferred embodiment of the present invention, the space-time characteristics based on the garbage monitoring area obtain garbage characteristics of the garbage monitoring area, and perform associated early warning on different garbage monitoring areas based on the garbage characteristics, specifically:
acquiring surrounding environment parameters of the garbage monitoring area, and acquiring garbage characteristics of the garbage monitoring area by combining space-time characteristics and garbage type information of the garbage monitoring area;
based on the urban road three-dimensional model, acquiring relative distances among all the garbage monitoring areas, clustering the garbage monitoring areas with the relative distances within a preset value, defining the garbage monitoring areas as a type of garbage monitoring areas, combining garbage characteristics of the type of garbage monitoring areas, acquiring association values among the garbage characteristics of the type of garbage monitoring areas by using a gray association method, and dividing the type of garbage monitoring areas into two types of garbage monitoring areas if the association values are larger than the preset value;
Monitoring the garbage characteristics of the garbage monitoring area in real time, and if the garbage characteristics abnormal condition exists in the garbage monitoring area, generating a garbage characteristics abnormal signal in the garbage monitoring area based on the urban road three-dimensional model;
when the second class of garbage monitoring areas have garbage characteristic abnormal signals, defining the garbage monitoring areas without the garbage characteristic abnormal signals as garbage monitoring areas to be detected, automatically detecting the signal intensity of the garbage characteristic abnormal signals in the garbage monitoring areas to be detected, and presetting a first signal intensity threshold and a second signal intensity threshold;
when the signal intensity of the garbage characteristic abnormal signal is between a first signal intensity threshold value and a second signal intensity threshold value, generating an early warning signal in a garbage monitoring area to be detected;
and when the signal intensity of the garbage characteristic abnormal signal is larger than the second signal intensity threshold value, generating the garbage characteristic abnormal signal in the garbage monitoring area to be detected.
Further, in a preferred embodiment of the present invention, the garbage collection sequence is obtained, and the distribution condition, the load condition and the urban traffic condition of the urban collection vehicle are analyzed, so as to schedule the urban collection vehicle, specifically:
Marking a garbage monitoring area with a garbage characteristic abnormal signal, acquiring the urban environment importance degree of the marking area, the hazard degree of garbage types in the marking area and the people flow density degree in the marking area based on big data retrieval, and acquiring a garbage clearing sequence through convolutional neural network model prediction;
acquiring a distribution map of the city clearance vehicles, acquiring real-time positions of different city clearance vehicles, and acquiring load conditions of different city clearance vehicles through thermodynamic diagrams of the city clearance vehicles;
acquiring the real-time position of the city clearance vehicle and the relative distance between each marking area, screening and obtaining the city clearance vehicle with the relative distance within the preset distance, and simultaneously carrying out combination analysis on the load condition of the city clearance vehicle and the garbage characteristics in each marking area, and further screening the city clearance vehicle based on the analysis result to obtain the proper city clearance vehicle in each marking area;
and dispatching city clearing vehicles in the marked area based on city traffic conditions and garbage clearing sequence, and clearing garbage in the marked area by using the city clearing vehicles.
The invention also provides a city appearance environment early warning system based on the AI intelligent garbage identification, which comprises a memory and a processor, wherein the memory stores a city appearance environment early warning method based on the AI intelligent garbage identification, and when the city appearance environment early warning method based on the AI intelligent garbage identification is executed by the processor, the following steps are realized:
Based on the urban monitoring system, real-time monitoring is carried out on the urban road, and a garbage monitoring area is determined based on the real-time monitoring result;
constructing an AI intelligent garbage identification model, and determining garbage types in a garbage monitoring area by using the AI intelligent garbage identification model;
acquiring space-time characteristics of the garbage monitoring area, acquiring the garbage characteristics of the garbage monitoring area based on the space-time characteristics, and carrying out associated early warning on the garbage monitoring area according to the garbage characteristics;
and acquiring a garbage clearing sequence, analyzing the distribution condition, the load condition and the urban traffic condition of the urban clearing vehicles, and carrying out vehicle scheduling on the urban clearing vehicles.
The invention solves the technical defects in the background technology, and has the following beneficial effects: acquiring a garbage monitoring area based on an urban monitoring system, and determining garbage types in the garbage monitoring area by using an AI intelligent garbage identification model; acquiring space-time characteristics and garbage characteristics of a garbage monitoring area, and realizing associated early warning of the garbage monitoring area; and finally, carrying out vehicle scheduling on the urban cleaning vehicles by acquiring the garbage cleaning sequence and analyzing the distribution condition, the load condition and the urban traffic condition of the urban cleaning vehicles. The method can intelligently identify and classify the garbage on the urban road, clear and treat the garbage, beautify urban environment, improve the air quality and life quality of the city, and is beneficial to beautiful construction of the city.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other embodiments of the drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 shows a flow chart of a urban environment early warning method based on AI intelligent garbage recognition;
FIG. 2 is a flow chart of a method for analyzing the space-time characteristics of a garbage monitoring area and performing associated pre-warning on the garbage monitoring area;
fig. 3 shows a view of a urban environment early warning system based on AI intelligent garbage recognition.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
Fig. 1 shows a flow chart of a city appearance environment early warning method based on AI intelligent garbage recognition, which comprises the following steps:
s102: based on the urban monitoring system, real-time monitoring is carried out on the urban road, and a garbage monitoring area is determined based on the real-time monitoring result;
s104: constructing an AI intelligent garbage identification model, and determining garbage types in a garbage monitoring area by using the AI intelligent garbage identification model;
s106: acquiring space-time characteristics of the garbage monitoring area, acquiring the garbage characteristics of the garbage monitoring area based on the space-time characteristics, and carrying out associated early warning on the garbage monitoring area according to the garbage characteristics;
s108: and acquiring a garbage clearing sequence, analyzing the distribution condition, the load condition and the urban traffic condition of the urban clearing vehicles, and carrying out vehicle scheduling on the urban clearing vehicles.
Further, in a preferred embodiment of the present invention, the urban road is monitored in real time based on the urban monitoring system, and the garbage monitoring area is determined based on the real-time monitoring result, which specifically includes:
acquiring a real-time monitoring image of an urban road based on an urban monitoring system;
performing image graying treatment and image noise reduction treatment on the real-time monitoring image of the urban road to obtain a preprocessed urban road image;
Based on an edge feature segmentation method, extracting image features of the preprocessed urban road image to obtain road surface parameters of the urban road;
based on historical data acquisition, constructing an urban three-dimensional model, and importing road surface parameters of the urban road into the urban three-dimensional model to update model parameters to obtain the urban road three-dimensional model;
obtaining model parameters of the urban road three-dimensional model, guiding the model parameters into a convolutional neural network model for prediction to obtain garbage position prediction data, and guiding the garbage position prediction data into the urban road three-dimensional model to obtain the coordinate position of garbage on the urban road;
and (3) presetting a standard garbage packing density and a standard garbage packing area, analyzing the coordinate position of the garbage on the urban road, and defining a region with the garbage packing density and the garbage packing area larger than the standard garbage packing density and the standard garbage packing area as a garbage monitoring region.
It should be noted that, due to activities of people or other reasons, garbage can be generated in urban roads, and the garbage can affect urban appearance, which is bad for people's travel, health and other aspects. The urban monitoring system can acquire real-time monitoring video of the urban road, and road surface parameters of the urban road can be acquired according to the real-time monitoring video of the urban road, wherein the road surface parameters of the urban road comprise road surface garbage conditions. The road surface condition of the city can be intuitively known by constructing the three-dimensional model. And acquiring the coordinate position of the garbage, acquiring the garbage filling density and the garbage filling area according to the coordinate position, and further analyzing to obtain a garbage monitoring area. The method can acquire real-time images of urban roads through the urban monitoring system, so that a garbage monitoring area is acquired.
Further, in a preferred embodiment of the present invention, the construction of the AI-intelligent garbage identification model, and the determination of the garbage type in the garbage monitoring area using the AI-intelligent garbage identification model, specifically includes:
based on big data retrieval, acquiring all types of junk information, and dividing the junk information into a training set and a testing set, wherein the training set contains sample data of the junk information of all types;
introducing a decision tree model, acquiring a characteristic value and a segmentation value of sample data, determining a segmentation point of the sample data in the decision tree model based on the characteristic value and the segmentation value, dividing the sample data at the segmentation point by the decision tree model, re-acquiring the segmentation point after classification, continuing to divide, analyzing the sample data after each division, stopping the corresponding division operation of the sample data when only one garbage information exists in the sample data, and outputting the latest segmentation point which is defined as a leaf node;
introducing a singular decomposition algorithm, performing singular decomposition on samples in leaf nodes to generate a singular matrix, introducing a cosine measurement algorithm to perform cosine measurement value calculation on the singular matrix, and defining leaf nodes with cosine measurement values larger than a preset value as outliers;
Performing iterative splitting treatment on the outliers by using a genetic algorithm, ending the iterative splitting treatment when the iteration times reach a preset value and the outliers are not present, outputting all leaf nodes to obtain a trained decision tree model, and constructing an AI intelligent garbage recognition model based on the trained decision tree model;
and importing model parameters of the urban road three-dimensional model of the garbage monitoring area into an AI intelligent garbage identification model to identify garbage types, so as to obtain garbage type information in the garbage monitoring area.
It should be noted that the garbage types in the garbage monitoring area may be different, and the cleaning modes, cleaning difficulties, and the like of different garbage are different, so that the garbage types in the garbage monitoring area need to be determined. And importing all types of junk information into a decision tree model for training, wherein the decision tree model can finish classification operation by carrying out feature division on sample data, and defines the latest segmentation point after stopping division as a leaf node. In the dividing process, the situation that a local optimal solution possibly exists, the singular decomposition algorithm can be used for reducing the operation complexity of a decision tree, generating a singular matrix, introducing the cosine metric algorithm can calculate the superposition degree of sample data, and obtaining a cosine metric value, wherein leaf nodes with the cosine metric value larger than a preset value are defined as outliers. The genetic algorithm can perform iterative split optimization on the outlier, so that the coincidence degree of the sample data of the outlier is reduced. And outputting the optimized leaf nodes, constructing an AI intelligent garbage identification model, and identifying garbage types in the garbage monitoring area. The invention can construct an AI intelligent garbage identification model through a decision tree algorithm, thereby acquiring garbage type information in a garbage monitoring area.
Further, in a preferred embodiment of the present invention, the garbage collection sequence is obtained, and the distribution condition, the load condition and the urban traffic condition of the urban collection vehicle are analyzed, so as to schedule the urban collection vehicle, specifically:
marking a garbage monitoring area with a garbage characteristic abnormal signal, acquiring the urban environment importance degree of the marking area, the hazard degree of garbage types in the marking area and the people flow density degree in the marking area based on big data retrieval, and acquiring a garbage clearing sequence through convolutional neural network model prediction;
acquiring a distribution map of the city clearance vehicles, acquiring real-time positions of different city clearance vehicles, and acquiring load conditions of different city clearance vehicles through thermodynamic diagrams of the city clearance vehicles;
acquiring the real-time position of the city clearance vehicle and the relative distance between each marking area, screening and obtaining the city clearance vehicle with the relative distance within the preset distance, and simultaneously carrying out combination analysis on the load condition of the city clearance vehicle and the garbage characteristics in each marking area, and further screening the city clearance vehicle based on the analysis result to obtain the proper city clearance vehicle in each marking area;
And dispatching city clearing vehicles in the marked area based on city traffic conditions and garbage clearing sequence, and clearing garbage in the marked area by using the city clearing vehicles.
The garbage monitoring area with the garbage characteristic abnormal signals is an area needing to be cleared by preferential garbage treatment, and if a plurality of garbage monitoring areas containing the garbage characteristic abnormal signals exist, the area is marked first, and the urban environment importance degree, the garbage type hazard degree and the people stream density degree of the marked area are searched. The marked places exist in one city, the urban environment of the marked places has higher importance degree, and garbage clearing treatment is required to be carried out preferentially; some garbage has larger hazard degree and needs to be treated preferentially; the people flow in some garbage monitoring areas is large, and garbage collection is inconvenient. According to the above elements, the garbage collection order can be obtained by the prediction model. There are a plurality of clearing vehicles in the city, and the function of the clearing vehicles is to clear and transport garbage. The positions of the urban cleaning vehicles are different in different time periods, and the load condition of the urban cleaning vehicles can be acquired according to the thermodynamic diagram. Some garbage monitoring areas have more garbage, the load of the needed city clearance vehicles is smaller, and the city clearance vehicles with closer distances are preferentially used for garbage clearance treatment. Therefore, each marked area is provided with the corresponding most suitable urban cleaning vehicle for garbage cleaning treatment, and finally, the urban cleaning vehicles are subjected to vehicle scheduling treatment according to urban traffic conditions and garbage cleaning sequences. The method can analyze the real-time position and the load condition of the clearing vehicles in the city, and realize the vehicle dispatching of the clearing vehicles in the city.
Fig. 2 shows a flow chart of a method for analyzing the space-time characteristics of the garbage monitoring area and performing associated early warning on the garbage monitoring area, which comprises the following steps:
s202: acquiring space-time characteristics of the garbage monitoring area based on the garbage filling density and the garbage filling area of the garbage monitoring area;
s204: acquiring a second class of garbage monitoring areas based on the relative distance between the garbage monitoring areas;
s206: and carrying out association early warning on garbage in the second class garbage monitoring area based on the garbage characteristics.
Further, in a preferred embodiment of the present invention, the garbage filling density and the garbage filling area based on the garbage monitoring area obtain the space-time characteristics of the garbage monitoring area, specifically:
acquiring the garbage filling density and the garbage filling area of a garbage monitoring area based on the coordinate position of the garbage on the urban road;
combining the garbage filling density and the garbage filling area of the garbage monitoring area to generate a training set and a testing set, introducing the training set into a convolutional layer of a TCN model to perform causal convolutional processing to obtain data features of different time scales, adding the data in the training set with the data features of different time scales through residual connection, introducing the data features into a pooling layer to perform maximum pooling processing, and outputting pooling values of different time scales;
Importing the pooling values of different time scales into a full-connection layer of a TCN model for reverse training to obtain a trained TCN model, and generating time characteristics of a garbage monitoring area based on the trained TCN model;
combining the garbage filling density, the garbage filling area and the surrounding environment parameters of the garbage monitoring area, acquiring the spatial characteristics of the garbage monitoring area based on the RNN model, and combining the temporal characteristics of the garbage monitoring area and the spatial characteristics of the garbage monitoring area to generate the space-time characteristics of the garbage monitoring area, wherein the space-time characteristics of the garbage monitoring area represent the change condition of the quantity and the position of garbage in the garbage monitoring area in different time periods.
It should be noted that, there are space-time features in the garbage monitoring area, and the space-time features include temporal features and spatial features. The time characteristics and the space characteristics of the garbage monitoring area can be obtained by using the TCN model and the RNN model respectively. The time characteristic means the change condition of the quantity of garbage in the garbage monitoring area in different time periods, and the space characteristic means the change condition of the position of the garbage in the garbage monitoring area. Acquiring the space-time characteristics of the garbage monitoring area is helpful for providing conditions for the subsequent acquisition of the garbage characteristics of the garbage monitoring area and the associated early warning of the garbage monitoring area.
Further, in a preferred embodiment of the present invention, the obtaining the second type of garbage monitoring area based on the relative distance between the garbage monitoring areas specifically includes:
acquiring surrounding environment parameters of the garbage monitoring area, and acquiring garbage characteristics of the garbage monitoring area by combining space-time characteristics and garbage type information of the garbage monitoring area;
based on the urban road three-dimensional model, acquiring the relative distance between all the garbage monitoring areas, clustering the garbage monitoring areas with the relative distance within a preset value, defining the garbage monitoring areas as a type of garbage monitoring areas, combining the garbage characteristics of the type of garbage monitoring areas, acquiring the correlation value between the garbage characteristics of the type of garbage monitoring areas by using a gray correlation method, and dividing the type of garbage monitoring areas into two types of garbage monitoring areas if the correlation value is larger than the preset value.
It should be noted that, there may be large activities around the garbage monitoring area, such as food streets, and garbage is easy to be generated, so that it is necessary to obtain environmental parameters around the garbage monitoring area, and combine the space-time characteristics of the garbage monitoring area and the garbage type information to obtain the garbage characteristics of the garbage monitoring area. The trash characteristics represent the number, location, type and impact of trash in the trash monitoring area. The garbage in the garbage monitoring area may be related to each other, for example, the garbage in the first garbage monitoring area may flow to the second garbage monitoring area along with the factors of people flow, traffic flow, etc. And acquiring the relative distance between the garbage monitoring areas, wherein the garbage monitoring area with the relative distance within a preset value is regarded as a large monitoring area, and is defined as a garbage monitoring area. The garbage monitoring areas in the garbage monitoring areas are not necessarily associated, so that a gray association method is required to be used for acquiring association values of garbage features in the garbage monitoring areas, and the garbage monitoring areas with association values larger than a preset value are defined as the garbage monitoring areas. All the garbage monitoring areas in the second-class garbage monitoring area can be mutually influenced.
Further, in a preferred embodiment of the present invention, the association early warning of garbage in the second class garbage monitoring area based on the garbage characteristics specifically includes:
monitoring the garbage characteristics of the garbage monitoring area in real time, and if the garbage characteristics abnormal condition exists in the garbage monitoring area, generating a garbage characteristics abnormal signal in the garbage monitoring area based on the urban road three-dimensional model;
when the second class of garbage monitoring areas have garbage characteristic abnormal signals, defining the garbage monitoring areas without the garbage characteristic abnormal signals as garbage monitoring areas to be detected, automatically detecting the signal intensity of the garbage characteristic abnormal signals in the garbage monitoring areas to be detected, and presetting a first signal intensity threshold and a second signal intensity threshold;
when the signal intensity of the garbage characteristic abnormal signal is between a first signal intensity threshold value and a second signal intensity threshold value, generating an early warning signal in a garbage monitoring area to be detected;
and when the signal intensity of the garbage characteristic abnormal signal is larger than the second signal intensity threshold value, generating the garbage characteristic abnormal signal in the garbage monitoring area to be detected.
It should be noted that the number and distribution position of the garbage in the garbage monitoring area may change, and the abnormal condition of the garbage features includes that the number of the garbage is increased, the position where the garbage is located is not suitable, and the like, so that the garbage features of the garbage monitoring area are monitored in real time in the three-dimensional model of the urban road, and a garbage feature abnormal signal is generated. Because each rubbish monitoring area in the second class rubbish monitoring area can influence each other, when rubbish characteristic abnormal signals exist in the second class rubbish monitoring area, other rubbish monitoring areas need to be subjected to early warning treatment, and the purpose is to predict rubbish characteristic changes of the rubbish monitoring areas in advance and to make corresponding measures in advance. When the characteristic abnormal signal of the garbage is stronger, the garbage abnormal condition in the garbage monitoring area is proved to be more serious. And correspondingly generating a garbage characteristic abnormal signal of the early warning signal in the garbage monitoring area to be detected according to the magnitude of the signal intensity threshold value so as to perform associated early warning processing of the garbage characteristic monitoring area. The method can realize the associated early warning operation of the second class garbage monitoring area through the garbage characteristic abnormal signal.
In addition, the urban environment early warning method based on AI intelligent garbage recognition further comprises the following steps:
dividing all types of garbage information into recyclable garbage and unrecoverable garbage based on big data, and further training the AI intelligent garbage recognition model to obtain an optimized AI intelligent garbage recognition model;
classifying the garbage in the garbage monitoring area into recoverable garbage and non-recoverable garbage based on the optimized AI intelligent garbage identification model, updating the garbage characteristics of the garbage monitoring area, and scheduling and updating the urban garbage cleaning vehicles;
in the process of cleaning garbage in a garbage monitoring area by an urban cleaning vehicle, acquiring air quality parameters of the garbage monitoring area, guiding the air quality parameters of the garbage monitoring area into a big data network for searching, defining the garbage monitoring area as an air pollution area if the air quality parameters are unqualified, and acquiring monitoring images in the air pollution area in real time;
and extracting character feature data of the monitoring image in the air pollution area, analyzing the character feature data, and carrying out warning driving processing on characters through broadcasting on urban cleaning vehicles in the air pollution area when characters with non-shielding faces are detected.
The garbage is divided into recyclable matters and non-recyclable matters, and the recyclable matters and the non-recyclable matters are required to be classified and cleaned, so that the garbage can be recycled. The AI intelligent garbage recognition model is further trained, so that the AI intelligent garbage recognition model can recognize recoverable objects and non-recoverable objects, and therefore garbage characteristics of a garbage monitoring area are updated, and urban garbage collection vehicles are scheduled and updated.
In addition, it should be noted that the garbage may pollute the air, and part of the garbage, such as industrial garbage, chemical garbage, etc., may pollute the air and then pollute the human health, so in the garbage cleaning process, air quality parameters of the garbage monitoring area are obtained, where the air quality parameters include element components in the air, volatile concentration of compounds, etc. If the air quality parameters are unqualified, elemental components and the like in the air can cause harm to the environment and human body. Therefore, when the air quality parameters are unqualified, people in the garbage monitoring area need to be monitored in real time, the people with the face without the obstructer mean people without wearing a mask or people with the face fully exposed in the air, and the people with the face without the obstructer are easily affected by air pollution, so that if the people with the face without the obstructer are detected, warning and driving treatment is needed to be carried out on the people.
In addition, the urban environment early warning method based on AI intelligent garbage recognition further comprises the following steps:
the method comprises the steps of monitoring garbage monitoring points for a long time, obtaining the occurrence frequency of garbage characteristic abnormal signals, and defining the garbage monitoring points with the occurrence frequency higher than a preset value as high-frequency garbage monitoring points;
acquiring a character garbage throwing action data set in a big data network, wherein the character garbage throwing action data set stores all preparation actions of the character before throwing garbage, and the character garbage throwing action data set is imported into an AI intelligent garbage recognition model for model training to obtain an updated AI intelligent garbage recognition model;
based on the updated AI intelligent garbage identification model, acquiring the relative distance between the character and each garbage discarding position in the high-frequency garbage monitoring points, and defining the relative distance as a discarding distance;
if the discarding distance is greater than the preset distance, the actions of the characters are analyzed in real time, and if the actions of the characters are identified as the preparation actions for discarding garbage, the updated AI intelligent garbage identification model is connected with the broadcast nearest to the target characters through the Internet of things, and a warning is sent to the target characters;
the garbage generation frequency of people business circles, perennial activities, places where the garbage is held in cities is high, and the garbage generation frequency is defined as a high-frequency garbage monitoring point. Based on the figure throwing garbage action data set, further model training is carried out on the AI intelligent garbage recognition model, so that the AI intelligent garbage recognition model can monitor the action intention of the figure and judge whether the figure throws garbage or not. Acquiring the discarding distance of the person, and throwing garbage in the point where the discarding distance is smaller than the preset distance to prove that the garbage discarding point of the person is a garbage discarding position such as a garbage can; when the AI intelligent garbage recognition model judges that the character throws garbage in the point with the throwing distance larger than the preset distance according to the action of the character, the AI intelligent garbage recognition model proves that the character has possibility of throwing garbage randomly, and the character is reminded not to throw garbage randomly by connecting the broadcast nearest to the task.
As shown in fig. 3, the second aspect of the present invention further provides an urban environment early warning system based on AI intelligent garbage recognition, where the urban environment early warning system based on AI intelligent garbage recognition includes a memory 31 and a processor 32, where the memory 31 stores an urban environment early warning method based on AI intelligent garbage recognition, and when the urban environment early warning method based on AI intelligent garbage recognition is executed by the processor 32, the following steps are implemented:
based on the urban monitoring system, real-time monitoring is carried out on the urban road, and a garbage monitoring area is determined based on the real-time monitoring result;
constructing an AI intelligent garbage identification model, and determining garbage types in a garbage monitoring area by using the AI intelligent garbage identification model;
acquiring space-time characteristics of the garbage monitoring area, acquiring the garbage characteristics of the garbage monitoring area based on the space-time characteristics, and carrying out associated early warning on the garbage monitoring area according to the garbage characteristics;
and acquiring a garbage clearing sequence, analyzing the distribution condition, the load condition and the urban traffic condition of the urban clearing vehicles, and carrying out vehicle scheduling on the urban clearing vehicles.
The foregoing is merely illustrative embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think about variations or substitutions within the technical scope of the present invention, and the invention should be covered. Therefore, the protection scope of the invention is subject to the protection scope of the claims.

Claims (6)

1. The urban environment early warning method based on AI intelligent garbage recognition is characterized by comprising the following steps:
based on the urban monitoring system, real-time monitoring is carried out on the urban road, and a garbage monitoring area is determined based on the real-time monitoring result;
constructing an AI intelligent garbage identification model, and determining garbage types in a garbage monitoring area by using the AI intelligent garbage identification model;
acquiring space-time characteristics of the garbage monitoring area, acquiring the garbage characteristics of the garbage monitoring area based on the space-time characteristics, and carrying out associated early warning on the garbage monitoring area according to the garbage characteristics;
acquiring a garbage clearing sequence, analyzing the distribution condition, the load condition and the urban traffic condition of the urban clearing vehicles, and carrying out vehicle scheduling on the urban clearing vehicles;
the method comprises the steps of acquiring space-time characteristics of a garbage monitoring area, acquiring the garbage characteristics of the garbage monitoring area based on the space-time characteristics, and carrying out association early warning on the garbage monitoring area according to the garbage characteristics, wherein the method specifically comprises the following steps:
acquiring the garbage filling density and the garbage filling area of a garbage monitoring area based on the coordinate position of the garbage on the urban road;
combining the garbage filling density and the garbage filling area of the garbage monitoring area to generate a training set and a testing set, introducing the training set into a convolutional layer of a TCN model to perform causal convolutional processing to obtain data features of different time scales, adding the data in the training set with the data features of different time scales through residual connection, introducing the data features into a pooling layer to perform maximum pooling processing, and outputting pooling values of different time scales;
Importing the pooling values of different time scales into a full-connection layer of a TCN model for reverse training to obtain a trained TCN model, and generating time characteristics of a garbage monitoring area based on the trained TCN model;
combining the garbage filling density, the garbage filling area and the surrounding environment parameters of the garbage monitoring area, acquiring the spatial characteristics of the garbage monitoring area based on an RNN model, and combining the temporal characteristics of the garbage monitoring area and the spatial characteristics of the garbage monitoring area to generate the space-time characteristics of the garbage monitoring area, wherein the space-time characteristics of the garbage monitoring area represent the quantity and position change condition of garbage in the garbage monitoring area in different time periods;
based on the space-time characteristics of the garbage monitoring areas, the garbage characteristics of the garbage monitoring areas are obtained, and association early warning is carried out on different garbage monitoring areas based on the garbage characteristics.
2. The urban environment early warning method based on AI intelligent garbage recognition according to claim 1, wherein the urban road is monitored in real time based on the urban monitoring system, and the garbage monitoring area is determined based on the real-time monitoring result, specifically:
acquiring a real-time monitoring image of an urban road based on an urban monitoring system;
Performing image graying treatment and image noise reduction treatment on the real-time monitoring image of the urban road to obtain a preprocessed urban road image;
based on an edge feature segmentation method, extracting image features of the preprocessed urban road image to obtain road surface parameters of the urban road;
based on historical data acquisition, constructing an urban three-dimensional model, and importing road surface parameters of the urban road into the urban three-dimensional model to update model parameters to obtain the urban road three-dimensional model;
obtaining model parameters of the urban road three-dimensional model, guiding the model parameters into a convolutional neural network model for prediction to obtain garbage position prediction data, and guiding the garbage position prediction data into the urban road three-dimensional model to obtain the coordinate position of garbage on the urban road;
and (3) presetting a standard garbage packing density and a standard garbage packing area, analyzing the coordinate position of the garbage on the urban road, and defining a region with the garbage packing density and the garbage packing area larger than the standard garbage packing density and the standard garbage packing area as a garbage monitoring region.
3. The urban environment early warning method based on AI intelligent garbage recognition according to claim 1, wherein the constructing an AI intelligent garbage recognition model and determining garbage types in a garbage monitoring area by using the AI intelligent garbage recognition model is specifically as follows:
Based on big data retrieval, acquiring all types of junk information, and dividing the junk information into a training set and a testing set, wherein the training set contains sample data of the junk information of all types;
introducing a decision tree model, acquiring a characteristic value and a segmentation value of sample data, determining a segmentation point of the sample data in the decision tree model based on the characteristic value and the segmentation value, dividing the sample data at the segmentation point by the decision tree model, re-acquiring the segmentation point after classification, continuing to divide, analyzing the sample data after each division, stopping the corresponding division operation of the sample data when only one garbage information exists in the sample data, and outputting the latest segmentation point which is defined as a leaf node;
introducing a singular decomposition algorithm, performing singular decomposition on samples in leaf nodes to generate a singular matrix, introducing a cosine measurement algorithm to perform cosine measurement value calculation on the singular matrix, and defining leaf nodes with cosine measurement values larger than a preset value as outliers;
performing iterative splitting treatment on the outliers by using a genetic algorithm, ending the iterative splitting treatment when the iteration times reach a preset value and the outliers are not present, outputting all leaf nodes to obtain a trained decision tree model, and constructing an AI intelligent garbage recognition model based on the trained decision tree model;
And importing model parameters of the urban road three-dimensional model of the garbage monitoring area into an AI intelligent garbage identification model to identify garbage types, so as to obtain garbage type information in the garbage monitoring area.
4. The urban environment early warning method based on AI intelligent garbage identification according to claim 1, wherein the space-time characteristics based on the garbage monitoring area obtain the garbage characteristics of the garbage monitoring area, and perform associated early warning on different garbage monitoring areas based on the garbage characteristics, specifically comprises:
acquiring surrounding environment parameters of the garbage monitoring area, and acquiring garbage characteristics of the garbage monitoring area by combining space-time characteristics and garbage type information of the garbage monitoring area;
based on the urban road three-dimensional model, acquiring relative distances among all the garbage monitoring areas, clustering the garbage monitoring areas with the relative distances within a preset value, defining the garbage monitoring areas as a type of garbage monitoring areas, combining garbage characteristics of the type of garbage monitoring areas, acquiring association values among the garbage characteristics of the type of garbage monitoring areas by using a gray association method, and dividing the type of garbage monitoring areas into two types of garbage monitoring areas if the association values are larger than the preset value;
Monitoring the garbage characteristics of the garbage monitoring area in real time, and if the garbage characteristics abnormal condition exists in the garbage monitoring area, generating a garbage characteristics abnormal signal in the garbage monitoring area based on the urban road three-dimensional model;
when the second class of garbage monitoring areas have garbage characteristic abnormal signals, defining the garbage monitoring areas without the garbage characteristic abnormal signals as garbage monitoring areas to be detected, automatically detecting the garbage monitoring areas to be detected, acquiring the signal intensity of the garbage characteristic abnormal signals, and presetting a first signal intensity threshold and a second signal intensity threshold;
when the signal intensity of the garbage characteristic abnormal signal is between a first signal intensity threshold value and a second signal intensity threshold value, generating an early warning signal in a garbage monitoring area to be detected;
and when the signal intensity of the garbage characteristic abnormal signal is larger than a second signal intensity threshold value, generating an associated early warning in the garbage monitoring area to be detected.
5. The urban environment early warning method based on AI intelligent garbage identification of claim 1, wherein the steps of obtaining garbage clearance sequence, analyzing distribution condition, load condition and urban traffic condition of urban clearance vehicles, and carrying out vehicle scheduling on the urban clearance vehicles are as follows:
Marking a garbage monitoring area with a garbage characteristic abnormal signal, acquiring the urban environment importance degree of the marking area, the hazard degree of garbage types in the marking area and the people flow density degree in the marking area based on big data retrieval, and acquiring a garbage clearing sequence through convolutional neural network model prediction;
acquiring a distribution map of the city clearance vehicles, acquiring real-time positions of different city clearance vehicles, and acquiring load conditions of different city clearance vehicles through thermodynamic diagrams of the city clearance vehicles;
acquiring the real-time position of the city clearance vehicle and the relative distance between each marking area, screening and obtaining the city clearance vehicle with the relative distance within the preset distance, and simultaneously carrying out combination analysis on the load condition of the city clearance vehicle and the garbage characteristics in each marking area, and further screening the city clearance vehicle based on the analysis result to obtain the proper city clearance vehicle in each marking area;
and dispatching city clearing vehicles in the marked area based on city traffic conditions and garbage clearing sequence, and clearing garbage in the marked area by using the city clearing vehicles.
6. The urban environment early warning system based on the AI intelligent garbage identification is characterized by comprising a memory and a processor, wherein the memory stores an urban environment early warning method based on the AI intelligent garbage identification, and when the urban environment early warning method based on the AI intelligent garbage identification is executed by the processor, the following steps are realized:
Based on the urban monitoring system, real-time monitoring is carried out on the urban road, and a garbage monitoring area is determined based on the real-time monitoring result;
constructing an AI intelligent garbage identification model, and determining garbage types in a garbage monitoring area by using the AI intelligent garbage identification model;
acquiring space-time characteristics of the garbage monitoring area, acquiring the garbage characteristics of the garbage monitoring area based on the space-time characteristics, and carrying out associated early warning on the garbage monitoring area according to the garbage characteristics;
acquiring a garbage clearing sequence, analyzing the distribution condition, the load condition and the urban traffic condition of the urban clearing vehicles, and carrying out vehicle scheduling on the urban clearing vehicles;
the method comprises the steps of acquiring space-time characteristics of a garbage monitoring area, acquiring the garbage characteristics of the garbage monitoring area based on the space-time characteristics, and carrying out association early warning on the garbage monitoring area according to the garbage characteristics, wherein the method specifically comprises the following steps:
acquiring the garbage filling density and the garbage filling area of a garbage monitoring area based on the coordinate position of the garbage on the urban road;
combining the garbage filling density and the garbage filling area of the garbage monitoring area to generate a training set and a testing set, introducing the training set into a convolutional layer of a TCN model to perform causal convolutional processing to obtain data features of different time scales, adding the data in the training set with the data features of different time scales through residual connection, introducing the data features into a pooling layer to perform maximum pooling processing, and outputting pooling values of different time scales;
Importing the pooling values of different time scales into a full-connection layer of a TCN model for reverse training to obtain a trained TCN model, and generating time characteristics of a garbage monitoring area based on the trained TCN model;
combining the garbage filling density, the garbage filling area and the surrounding environment parameters of the garbage monitoring area, acquiring the spatial characteristics of the garbage monitoring area based on an RNN model, and combining the temporal characteristics of the garbage monitoring area and the spatial characteristics of the garbage monitoring area to generate the space-time characteristics of the garbage monitoring area, wherein the space-time characteristics of the garbage monitoring area represent the quantity and position change condition of garbage in the garbage monitoring area in different time periods;
based on the space-time characteristics of the garbage monitoring areas, the garbage characteristics of the garbage monitoring areas are obtained, and association early warning is carried out on different garbage monitoring areas based on the garbage characteristics.
CN202311363846.7A 2023-10-20 2023-10-20 Urban environment early warning method and system based on AI intelligent garbage recognition Active CN117095364B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311363846.7A CN117095364B (en) 2023-10-20 2023-10-20 Urban environment early warning method and system based on AI intelligent garbage recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311363846.7A CN117095364B (en) 2023-10-20 2023-10-20 Urban environment early warning method and system based on AI intelligent garbage recognition

Publications (2)

Publication Number Publication Date
CN117095364A CN117095364A (en) 2023-11-21
CN117095364B true CN117095364B (en) 2024-02-13

Family

ID=88775711

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311363846.7A Active CN117095364B (en) 2023-10-20 2023-10-20 Urban environment early warning method and system based on AI intelligent garbage recognition

Country Status (1)

Country Link
CN (1) CN117095364B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111598476A (en) * 2020-05-22 2020-08-28 济源职业技术学院 Smart city environmental sanitation resource scheduling system based on sparse self-coding and SVM
CN113344262A (en) * 2021-05-28 2021-09-03 淮阴工学院 Intelligent clearing system and method based on urban garbage classification
CN113762624A (en) * 2021-09-09 2021-12-07 安徽自然美环境科技有限公司 Garbage clearing and transporting vehicle route optimization method and urban garbage clearing and transporting ecological system
CN114283378A (en) * 2021-12-22 2022-04-05 京东方科技集团股份有限公司 Monitoring method, monitoring device, storage medium and electronic equipment
CN115169466A (en) * 2022-07-15 2022-10-11 京东城市(北京)数字科技有限公司 Method and device for drawing image of land, electronic equipment and computer readable medium
CN115860403A (en) * 2022-12-15 2023-03-28 成都秦川物联网科技股份有限公司 Smart city garbage disposal equipment management method and Internet of things system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111598476A (en) * 2020-05-22 2020-08-28 济源职业技术学院 Smart city environmental sanitation resource scheduling system based on sparse self-coding and SVM
CN113344262A (en) * 2021-05-28 2021-09-03 淮阴工学院 Intelligent clearing system and method based on urban garbage classification
CN113762624A (en) * 2021-09-09 2021-12-07 安徽自然美环境科技有限公司 Garbage clearing and transporting vehicle route optimization method and urban garbage clearing and transporting ecological system
CN114283378A (en) * 2021-12-22 2022-04-05 京东方科技集团股份有限公司 Monitoring method, monitoring device, storage medium and electronic equipment
CN115169466A (en) * 2022-07-15 2022-10-11 京东城市(北京)数字科技有限公司 Method and device for drawing image of land, electronic equipment and computer readable medium
CN115860403A (en) * 2022-12-15 2023-03-28 成都秦川物联网科技股份有限公司 Smart city garbage disposal equipment management method and Internet of things system

Also Published As

Publication number Publication date
CN117095364A (en) 2023-11-21

Similar Documents

Publication Publication Date Title
CN111462488B (en) Intersection safety risk assessment method based on deep convolutional neural network and intersection behavior characteristic model
CN109145954B (en) Network taxi appointment travel safety evaluation method and system based on multi-source time-space data
CN109410588B (en) Traffic accident evolution analysis method based on traffic big data
CN111144446B (en) Driver identity recognition method and system based on space-time grid
CN116168356B (en) Vehicle damage judging method based on computer vision
CN113538072A (en) Intelligent travel chain identification method and device for freight vehicle and electronic equipment
Lorintiu et al. Transportation mode recognition based on smartphone embedded sensors for carbon footprint estimation
CN106447137A (en) Traffic passenger flow forecasting method based on information fusion and Markov model
CN117554574B (en) Miniature air quality automatic monitor based on internet of things
CN115620471A (en) Image identification security system based on big data screening
CN111931986A (en) Garbage clearing and transporting vehicle route optimization method and urban garbage clearing and transporting ecological system
CN116206451A (en) Intelligent traffic flow data analysis method
CN118334604B (en) Accident detection and data set construction method and equipment based on multi-mode large model
CN117095364B (en) Urban environment early warning method and system based on AI intelligent garbage recognition
CN112188478B (en) Resident population data acquisition method based on big data analysis
CN114091581A (en) Vehicle operation behavior type identification method based on sparse track
CN112651992A (en) Trajectory tracking method and system
CN112801181A (en) Urban signaling traffic flow user classification and prediction method, storage medium and system
CN109614926B (en) Distributed optical fiber sensing signal mode identification method and system based on prediction model
CN107194505A (en) A kind of method and system that bus travel amount is predicted based on city big data
Puntumapon et al. Classification of cellular phone mobility using naive Bayes model
CN109447306A (en) Metro accidents delay time at stop prediction technique based on maximum likelihood regression tree
Tejima et al. Mm-aqi: A novel framework to understand the associations between urban traffic, visual pollution, and air pollution
CN112861621A (en) Subway passenger behavior identification information system, method, medium and equipment
CN116935625B (en) Intelligent detection method for illegal behaviors of illegal inter-provincial passenger transport operation vehicles

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