[go: up one dir, main page]

CN117975220B - Air quality monitoring and management method and system - Google Patents

Air quality monitoring and management method and system Download PDF

Info

Publication number
CN117975220B
CN117975220B CN202410225414.8A CN202410225414A CN117975220B CN 117975220 B CN117975220 B CN 117975220B CN 202410225414 A CN202410225414 A CN 202410225414A CN 117975220 B CN117975220 B CN 117975220B
Authority
CN
China
Prior art keywords
features
feature
sewer
air quality
temperature
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
CN202410225414.8A
Other languages
Chinese (zh)
Other versions
CN117975220A (en
Inventor
胡国奇
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.)
Zhengzhou Zhenglong Environmental Protection Technology Co ltd
Original Assignee
Zhengzhou Zhenglong Environmental Protection 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 Zhengzhou Zhenglong Environmental Protection Technology Co ltd filed Critical Zhengzhou Zhenglong Environmental Protection Technology Co ltd
Priority to CN202410225414.8A priority Critical patent/CN117975220B/en
Publication of CN117975220A publication Critical patent/CN117975220A/en
Application granted granted Critical
Publication of CN117975220B publication Critical patent/CN117975220B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
    • G06V10/806—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of extracted features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION 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
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • 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
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/761—Proximity, similarity or dissimilarity measures
    • 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
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • 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
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/7715—Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods
    • 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
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
    • 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
    • Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
    • Y02B30/70—Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Databases & Information Systems (AREA)
  • Multimedia (AREA)
  • Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Primary Health Care (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses an air quality monitoring and managing method and system, wherein the method comprises the steps of collecting temperature and humidity data in a sewer through a sensor, and collecting environmental images of the sewer through a camera; extracting features of the environment image, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with maximum correlation from the feature quantities as target features; fusing the temperature and humidity data with target characteristics and inputting the fused temperature and humidity data into a trained prediction model to obtain a sewer air quality prediction result; generating an air purification instruction, and treating the sewer by using the air purification instruction. The invention can rapidly and accurately predict the air quality of the sewer, and has the advantages of low cost and high precision compared with a sensor for detecting and controlling various gases.

Description

Air quality monitoring and management method and system
Technical Field
The invention relates to the technical field of air monitoring, in particular to an air quality monitoring management method and system.
Background
Sewer is an important component of urban infrastructure, mainly used for collecting and transporting rainwater, domestic sewage and industrial wastewater. Because the internal environment of the sewer is special, a large amount of organic substances and inorganic substances exist, and various gases such as hydrogen sulfide, ammonia, methane and the like can be generated by the substances under the action of microorganisms, if the gases cannot be effectively controlled, the gases can not only cause injury to human bodies, but also cause safety accidents such as fire and explosion. Therefore, it is important to monitor the air quality of the sewer.
At present, the monitoring of the sewer is generally to directly detect the gas concentration in the sewer by using a sensor and the like, and immediately start an alarm once exceeding the standard, but the detection mode by the sensor is often error due to complex gas environment of the sewer and easy chemical reaction between the sewer and the sensor; secondly, one type of sensor can only detect one gas concentration, and if the whole air quality in the sewer is to be evaluated, a plurality of sensors are required to be distributed and controlled, so that the cost is high, and the later operation and maintenance difficulty is increased.
Disclosure of Invention
In order to solve at least one technical problem set forth above, the present invention provides an air quality monitoring and managing method and system.
In a first aspect, the present invention provides an air quality monitoring and management method, the method comprising:
acquiring temperature and humidity data in the sewer through a sensor, and acquiring an environment image of the sewer through a camera;
extracting features of the environment image, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features as target features;
fusing the temperature and humidity data with target characteristics, and inputting the fused temperature and humidity data into a trained prediction model to obtain a sewer air quality prediction result;
And generating an air purification instruction according to the air quality prediction result, and processing the sewer by using the air purification instruction.
Preferably, the fitting the feature of dimension reduction according to the gradient lifting tree algorithm includes:
Assuming that the feature dimensions have m in total and the category number is N, calculating the coefficient of the node's base under the decision tree:
In the method, in the process of the invention, The coefficient of the foundation for the node v under the kth decision tree; The proportion of the category l under the node v is calculated;
calculating the variation of the coefficient of the characteristic x z before and after the node v branches
In the method, in the process of the invention,AndRespectively the coefficient of the foundation of 2 new nodes after branching;
defining the node set of the feature x z under the decision tree k as Q, the importance of x z under the kth decision tree is:
wherein z' is a node in the node set Q;
Assuming that T trees are co-generated during training, the importance of feature x z in all trees is:
Wherein V z z' is a node in the node set Q; v z z' is the feature importance corresponding to each tree;
and (3) performing normalization operation to obtain a final importance score I GBz of the feature x z:
Where V w is the importance of the feature dimension w.
Preferably, the feature extraction of the environmental image includes:
Establishing a plurality of target detection frames, and respectively extracting the characteristics of the multi-channel color characteristics of a plurality of environmental images of the sewer by utilizing each target detection frame to generate a plurality of characteristic sub-graphs;
Calculating the intersection ratio between any two target detection frames to be used as the correlation between any two target detection frames; splicing the feature sub-graphs according to the correlation relationship to obtain a target feature graph;
and extracting color features, texture features, spectrum features and spatial distribution features in the target feature map.
Preferably, the method further comprises training the predictive model, comprising:
Acquiring a plurality of temperature and humidity data obtained by monitoring in a historical scene and an environment image corresponding to the temperature and humidity as training samples;
Extracting features of the environment images in the training samples, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features;
fusing the temperature and humidity data with the feature quantity with the maximum correlation to train the GRU network model with the self-attention mechanism;
And constructing a loss function by using root mean square error, screening out the feature quantity with secondary correlation from the fitted features if the value of the loss function is larger than a first threshold value, reconstructing fusion feature iteration, and training the GRU network model until the value of the loss function is smaller than the first threshold value, and generating a prediction model.
Preferably, the expression of the self-attention mechanism is:
where Q *,K*,V* is the find, key and value, Q *K*T is the find similarity to the key, and d k is the key dimension, respectively.
Preferably, the method further comprises:
Taking the difference value between the original input data and the GRU network output predicted value as a training set, training a model constructed based on an SVR algorithm, and obtaining an error prediction model;
and inputting the characteristics of the temperature and humidity data and the target characteristics after fusion into an error prediction model, and compensating a prediction result output by the prediction model by using the error prediction model to obtain a final air quality prediction result.
In a second aspect, the present invention also provides an air quality monitoring management system, the system comprising:
the acquisition unit is used for acquiring temperature and humidity data in the sewer through the sensor and acquiring an environment image of the sewer through the camera;
The feature extraction unit is used for extracting features of the environment image, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features to serve as target features;
the prediction unit is used for fusing the temperature and humidity data with target characteristics, inputting the temperature and humidity data into a trained prediction model, and obtaining a predicted result of the air quality of the sewer;
and the purification control unit is used for generating an air purification instruction according to the air quality prediction result and processing the sewer by using the air purification instruction.
Preferably, the feature extraction unit is further configured to:
Establishing a plurality of target detection frames, and respectively extracting the characteristics of the multi-channel color characteristics of a plurality of environmental images of the sewer by utilizing each target detection frame to generate a plurality of characteristic sub-graphs;
Calculating the intersection ratio between any two target detection frames to be used as the correlation between any two target detection frames; splicing the feature sub-graphs according to the correlation relationship to obtain a target feature graph;
and extracting color features, texture features, spectrum features and spatial distribution features in the target feature map.
In a third aspect, the present invention also provides an electronic device, including: a processor and a memory for storing computer program code comprising computer instructions which, when executed by the processor, cause the electronic device to perform a method as described in the first aspect and any one of its possible implementation manners.
In a fourth aspect, the present invention also provides a computer readable storage medium having stored therein a computer program comprising program instructions which, when executed by a processor of an electronic device, cause the processor to perform a method as in the first aspect and any one of the possible implementations thereof.
Compared with the prior art, the invention has the beneficial effects that:
1) The invention collects temperature and humidity data in the sewer through the sensor, and collects environmental images of the sewer through the camera; extracting features of the environment image, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features as target features; fusing the temperature and humidity data with target characteristics, and inputting the fused temperature and humidity data into a trained prediction model to obtain a sewer air quality prediction result; the air quality can be rapidly and accurately measured and predicted by combining the characteristic extraction of the environmental image with the temperature and humidity environment of the sewer, and compared with a sensor for detecting and controlling various gas concentrations, the sensor has the advantages of low cost and high precision.
2) When the feature fitting is carried out according to the gradient lifting tree algorithm, the method mainly calculates the coefficient of the foundation of different nodes under the decision tree, and determines the feature importance by calculating the variation of the coefficient of the foundation, so that the feature with high correlation is extracted, and the prediction result is more accurate.
3) According to the invention, the air purification instruction is generated according to the air quality prediction result to treat the sewer, and the air prediction result generated at the beginning is compensated by the error prediction model, so that the accuracy of the air quality prediction result is improved, the rationality of the purification instruction is further improved, and the safety of the air environment of the sewer is ensured.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
In order to more clearly describe the embodiments of the present invention or the technical solutions in the background art, the following description will describe the drawings that are required to be used in the embodiments of the present invention or the background art.
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the technical aspects of the disclosure.
FIG. 1 is a schematic flow chart of an air quality monitoring and managing method according to an embodiment of the present invention;
Fig. 2 is a flow chart illustrating the substeps of step S20 in fig. 1 according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an air quality monitoring and managing system according to an embodiment of the present invention;
Fig. 4 is a schematic hardware structure of an electronic device according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The terms first, second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.
The term "and/or" is herein merely an association relationship describing an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a exists alone, A and B exist together, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B, C, may mean including any one or more elements selected from the group consisting of A, B and C.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
Furthermore, in the following detailed description, numerous specific details are set forth in order to provide a better illustration of the invention. It will be understood by those skilled in the art that the present invention may be practiced without some of these specific details. In some instances, well known methods, procedures, components, and circuits have not been described in detail so as not to obscure the present invention.
At present, the air quality monitoring of the sewer is generally realized by arranging and controlling various sensors for detecting the concentration of the polluted gas, so that the cost is high, the precision cannot be ensured, and the later operation and maintenance difficulty is high. Therefore, the invention provides the air quality monitoring and managing method, which can quickly and accurately obtain the air quality prediction result according to the temperature and humidity in the environment and the environment image and has the advantages of low cost and high precision.
Referring to fig. 1, fig. 1 is a flow chart of an air quality monitoring and managing method according to an embodiment of the invention. As shown in fig. 1, an air quality monitoring and managing method includes the following steps:
S10, acquiring temperature and humidity data in the sewer through a sensor, and acquiring an environment image of the sewer through a camera.
There are many types of temperature and humidity sensors, such as DHT11, DHT22, SHT11, etc., which all have respective advantages and application scenarios. The proper sensor can be selected according to the actual condition of the sewer, such as the range of temperature and humidity, environmental conditions and other factors, and the sensor needs to be installed at the key position of the sewer during installation so as to accurately capture the temperature and humidity information.
Because the environment of the sewer is special, waterproof, dustproof and vibration-resistant cameras are required to be selected. The device is also required to be arranged at a key position of the sewer during installation so as to capture the environmental condition of the sewer in an omnibearing manner. When the images are acquired, photographing can be carried out at different angles according to preset intervals, so that environment images comprising sewage parts and air parts of the sewer are obtained, the camera can be rotated for 360 degrees to photograph videos, and then the video streams are subjected to framing to extract the images.
And S20, extracting features of the environment image, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features as target features.
Referring to fig. 2, in one embodiment, feature extraction of an ambient image includes the sub-steps of:
s201, establishing a plurality of target detection frames, and respectively extracting the characteristics of the multi-channel color characteristics of a plurality of environmental images of the sewer by utilizing each target detection frame to generate a plurality of characteristic sub-graphs;
S202, calculating the cross ratio between any two target detection frames to be used as the correlation between any two target detection frames; splicing the feature sub-graphs according to the correlation relationship to obtain a target feature graph;
s203, extracting color features, texture features, spectrum features and spatial distribution features in the target feature map.
The target detection frame is a rectangular frame for marking the position of the detection object in one picture or video frame. When a target detection frame is established, firstly acquiring an initial detection frame of a rectangle and a corresponding polygonal outline; and then calculating a first IOU value of the initial detection frame and the polygonal outline, when the first IOU value meets the requirement, calculating a second IOU value of the polygonal outline and the target actual outline, and when the second IOU value meets the requirement, screening out the corresponding initial detection frame as the target detection frame.
It can be understood that if the extraction of a target overall feature is adopted in the feature extraction, the overall extraction range is large, and granularity is not fine enough, so that the accuracy of finally identifying the feature is easy to be disturbed. Therefore, in this embodiment, feature extraction is performed on different target areas of the video image by using each target detection frame, so as to obtain a plurality of feature images, and then the feature images are spliced to obtain a feature extraction image of the target.
When feature mapping is spliced, the correlation between every two target detection frames needs to be determined, when the features are spliced, features with high correlation are spliced preferentially, then features with next high correlation are spliced, finally features without correlation can be discarded, feature interference is prevented, and finally an integral target feature mapping is obtained through splicing.
Finally, extracting color features, texture features, spectrum features and spatial distribution features in the target feature map:
color characteristics: color is one of the most intuitive features of an image, and different sewage may exhibit different colors, such as turbidity, contained chemicals, etc., which may affect the color appearance.
Texture features: texture features reflect the roughness, directionality, regularity, etc. of the image. For example, the sewage treatment image has texture features, and the texture features under different working conditions have the characteristics of local sequential repetition, non-random arrangement, approximately uniform texture areas and the like.
Spectral characteristics: spectral characteristics refer to the reflected or emitted intensity of an image at different wavelengths. For example, certain chemicals may absorb light of a particular wavelength, which may be reflected by spectral characteristics.
Spatial distribution characteristics: the spatial distribution features describe the spatial positional relationship of individual pixels or regions in the image. For example, certain contaminants may accumulate at particular spatial locations, which may be reflected by spatial distribution characteristics.
Thus, by the feature extraction through partitioning provided by the embodiment, finer image features can be extracted; by extracting the color, texture, spectrum and spatial distribution characteristics, the model can be used for deep learning of the environmental image characteristics, so that the air quality of the sewer can be predicted by combining temperature and humidity data. By considering the correlation of different target detection frames, invalid features can be effectively filtered during feature splicing, noise interference is reduced, and the quality of extracted features is improved.
Further, since the feature dimension is too high, in one embodiment, feature dimension reduction is performed on the extracted features by adopting a PCA algorithm, the feature dimension reduction is performed according to a gradient lifting tree algorithm, and the feature quantity with the largest correlation is screened out from the fitted features to serve as a target feature.
The PCA algorithm, principal component analysis, is a commonly used feature dimension reduction algorithm that maps raw data into a new low-dimensional space through linear transformation, while preserving the information of the data. The data after dimension reduction can be obtained by normalizing the data, calculating covariance matrix, then calculating eigenvalue and eigenvector, finally screening out principal components and projecting
In one embodiment, fitting the feature of dimension reduction according to a gradient-lifted tree algorithm includes:
Assuming that the feature dimensions have m in total and the category number is N, calculating the coefficient of the node's base under the decision tree:
In the method, in the process of the invention, The coefficient of the foundation for the node v under the kth decision tree; The proportion of the category l under the node v is calculated;
calculating the variation of the coefficient of the characteristic x z before and after the node v branches
In the method, in the process of the invention,AndRespectively the coefficient of the foundation of 2 new nodes after branching;
defining the node set of the feature x z under the decision tree k as Q, the importance of x z under the kth decision tree is:
wherein z' is a node in the node set Q;
Assuming that T trees are co-generated during training, the importance of feature x z in all trees is:
Wherein V z z' is a node in the node set Q; v z z' is the feature importance corresponding to each tree;
and (3) performing normalization operation to obtain a final importance score I GBz of the feature x z:
Where V w is the importance of the feature dimension w.
Therefore, the feature after dimension reduction is fitted through a gradient lifting tree algorithm (GBDT), the feature importance is determined by calculating the coefficient of the foundation of different nodes under the decision tree and calculating the variation of the coefficient of the foundation, and the feature quantity with the largest correlation is screened out from the coefficient of the foundation, so that the prediction result is more accurate.
S30, fusing the temperature and humidity data with target features, and inputting the fused temperature and humidity data into a trained prediction model to obtain a predicted result of the air quality of the sewer.
Because the temperature and humidity data are weather influencing factors influencing the air quality of the sewer, and the environmental images are extracted, and the environmental images are mainly analyzed for influencing factors of sewage and pollutants in the air, in the embodiment, the temperature and humidity data are combined with the extracted image features and are input into a trained prediction model, and then the air quality result can be obtained rapidly.
In one embodiment, preferably, the method further comprises training a predictive model, comprising:
1) Acquiring a plurality of temperature and humidity data obtained by monitoring in a historical scene and an environment image corresponding to the temperature and humidity as training samples;
2) Extracting features of the environment images in the training samples, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features;
3) And fusing the temperature and humidity data with the feature quantity with the maximum correlation to train the GRU network model with the self-attention mechanism.
The GRU network extracts information features in a time sequence forward manner. When an abnormal state occurs, the data will have a phenomenon of recovery trend, and if the trend is utilized reversely, the abnormal state will be captured more sensitively. The bi-directional gated loop cell network BiGRU is formed by combining a forward GRU and a backward GRU, and can extract the forward and backward characteristic information of the time sequence at the same time. BiGRU has two independent hidden layers, wherein the forward hidden layer learns history information, the reverse hidden layer learns future information, and finally the history and the future information are integrated as output results.
Wherein, the expression of the self-attention mechanism is:
where Q *,K*,V* is the find, key and value, Q *K*T is the find similarity to the key, and d k is the key dimension, respectively.
In this embodiment, the self-attention structure is composed of a search (query), a key (key) and a value (value), and the three are equal, which can link different information of the input sequence and is better at capturing important data. The mechanism is mainly divided into three steps, namely, the similarity between each query and each key is calculated firstly to obtain a corresponding value; then normalizing the weight by using a Softmax function to obtain a weight coefficient; and finally, weighting and summing the weight coefficient and the corresponding value to obtain a final attention value. By adding a self-attention mechanism, important characteristics affecting air quality can be focused more when the model is trained.
4) And constructing a loss function by using root mean square error, screening out the feature quantity with secondary correlation from the fitted features if the value of the loss function is larger than a first threshold value, reconstructing fusion feature iteration, and training the GRU network model until the value of the loss function is smaller than the first threshold value, and generating a prediction model.
Considering that the features with larger correlation may have an influence on the prediction accuracy of the training process of the prediction model, in one embodiment, the root mean square error is used to construct a loss function first, and it is assumed that the iteration number has already met the requirement, and the value of the loss function is still greater than the first threshold at this time, which indicates that the extracted feature quantity may be problematic. Therefore, when the value of the loss function is found to exceed the first threshold, the relevant secondary feature quantity is screened out from the fitted features, the fusion feature is reconstructed and iterated to train the GRU network model, after the preset times of training are assumed, the feature quantity with the importance degree arranged behind the secondary feature is still exceeded, or the target feature is screened out from the feature quantity with the largest relevance, secondary feature and the like according to the ranking of the importance degree, finally, the combined target feature is fused with the temperature and humidity data to train the GRU network until the value of the loss function is smaller than the first threshold, and the final prediction model is generated.
In one embodiment, the method further comprises:
1) Taking the difference value between the original input data and the GRU network output predicted value as a training set, training a model constructed based on an SVR algorithm, and obtaining an error prediction model;
2) And inputting the characteristics of the temperature and humidity data and the target characteristics after fusion into an error prediction model, and compensating a prediction result output by the prediction model by using the error prediction model to obtain a final air quality prediction result.
The SVR algorithm is based on SVM classification of a support vector machine, a kernel function and a loss function are introduced, input data is mapped to a high-dimensional feature space through a nonlinear mapping function, namely the kernel function, a best fit hyperplane is found, and the total deviation between all training samples and the plane is minimized. The regression function expression is as follows:
wherein ω is a weight coefficient, And b is an offset, which is a nonlinear mapping function.
The modeling is performed by introducing relaxation variables, using a minimized canonical deconcentration criterion, converting the original regression problem into an unconstrained quadratic programming problem.
In the embodiment, an error prediction model is obtained by training an SVR algorithm; and then, inputting the characteristics obtained by fusing the temperature and humidity data and the target characteristics into an error prediction model to obtain a predicted error value, compensating a prediction result of the prediction model based on GRU network training by utilizing the error value, and greatly improving the prediction precision of the prediction model through error correction so as to assist the generation of a purification instruction and ensure the air quality safety of a sewer.
S40, generating an air purification instruction according to the air quality prediction result, and processing the sewer by using the air purification instruction.
In summary, the air quality monitoring and managing method provided by the embodiment of the invention can rapidly and accurately measure and predict the air quality by combining the characteristic extraction of the environmental image with the temperature and humidity environment of the sewer, and has the advantages of low cost and high precision compared with a sensor for detecting and controlling various gas concentrations. In addition, the feature importance is determined by calculating the coefficient of the foundation of different nodes under the decision tree and calculating the variation of the coefficient of the foundation, and further, the feature with large correlation is extracted, so that the prediction result is more accurate. Finally, according to the embodiment of the invention, the air purification instruction is generated according to the air quality prediction result to process the sewer, and the air prediction result generated at the beginning is compensated through the error prediction model, so that the accuracy of the air quality prediction result is improved, the rationality of the purification instruction is further improved, and the safety of the air environment of the sewer is ensured.
Referring to fig. 3, in one embodiment of the present invention, there is also provided an air quality monitoring management system, the system comprising:
The acquisition unit 100 is used for acquiring temperature and humidity data in the sewer through a sensor and acquiring an environment image of the sewer through a camera;
The feature extraction unit 200 is configured to perform feature extraction on the environmental image, perform feature dimension reduction on the extracted features by using a PCA algorithm, fit the dimension-reduced features according to a gradient lifting tree algorithm, and screen out feature quantities with the largest correlation from the fitted features as target features;
the prediction unit 300 is used for fusing the temperature and humidity data with target characteristics, inputting the temperature and humidity data into a trained prediction model, and obtaining a predicted result of the air quality of the sewer;
and a purification control unit 400 for generating an air purification command according to the air quality prediction result, and treating the sewage by using the air purification command.
In a preferred embodiment, the feature extraction unit 200 is further configured to:
Establishing a plurality of target detection frames, and respectively extracting the characteristics of the multi-channel color characteristics of a plurality of environmental images of the sewer by utilizing each target detection frame to generate a plurality of characteristic sub-graphs;
Calculating the intersection ratio between any two target detection frames to be used as the correlation between any two target detection frames; splicing the feature sub-graphs according to the correlation relationship to obtain a target feature graph;
and extracting color features, texture features, spectrum features and spatial distribution features in the target feature map.
In some embodiments, the functions or modules included in the system provided by the present embodiment may be used to perform the methods described in the foregoing method embodiments, and specific implementations thereof may refer to descriptions of the foregoing method embodiments, which are not repeated herein for brevity.
The invention also provides an electronic device, comprising: a processor, a transmitting means, an input means, an output means and a memory for storing computer program code comprising computer instructions which, when executed by the processor, cause the electronic device to perform a method as any one of the possible implementations described above.
The invention also provides a computer readable storage medium having stored therein a computer program comprising program instructions which, when executed by a processor of an electronic device, cause the processor to perform a method as any one of the possible implementations described above.
Referring to fig. 4, fig. 4 is a schematic hardware structure of an electronic device according to an embodiment of the invention.
The electronic device 2 comprises a processor 21, a memory 22, input means 23, output means 24. The processor 21, memory 22, input device 23, and output device 24 are coupled by connectors including various interfaces, transmission lines or buses, etc., as are not limited by the present embodiments. It should be appreciated that in various embodiments of the invention, coupled is intended to mean interconnected by a particular means, including directly or indirectly through other devices, e.g., through various interfaces, transmission lines, buses, etc.
The processor 21 may be one or more graphics processors (graphics processing unit, GPUs), which in the case of a GPU as the processor 21 may be a single core GPU or a multi-core GPU. Alternatively, the processor 21 may be a processor group formed by a plurality of GPUs, and the plurality of processors are coupled to each other through one or more buses. In the alternative, the processor may be another type of processor, and the embodiment of the invention is not limited.
Memory 22 may be used to store computer program instructions as well as various types of computer program code for performing aspects of the present invention. Optionally, the memory includes, but is not limited to, random access memory (random access memory, RAM), read-only memory (ROM), erasable programmable read-only memory (erasable programmable read only memory, EPROM), or portable read-only memory (compact disc read-only memory, CD-ROM) for associated instructions and data.
The input means 23 are for inputting data and/or signals and the output means 24 are for outputting data and/or signals. The input device 23 and the output device 24 may be separate devices or may be an integral device.
It will be appreciated that in embodiments of the present invention, the memory 22 may not only be used to store relevant instructions, but embodiments of the present invention are not limited to the specific data stored in the memory.
It will be appreciated that fig. 4 shows only a simplified design of an electronic device. In practical applications, the electronic device may further include other necessary elements, including but not limited to any number of input/output devices, processors, memories, etc., and all video parsing devices capable of implementing the embodiments of the present invention are within the scope of the present invention.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein. It will be further apparent to those skilled in the art that the descriptions of the various embodiments of the present invention are provided with emphasis, and that the same or similar parts may not be described in detail in different embodiments for convenience and brevity of description, and thus, parts not described in one embodiment or in detail may be referred to in description of other embodiments.
In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present invention, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted across a computer-readable storage medium. The computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line (digital subscriber line, DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital versatile disk (DIGITAL VERSATILEDISC, DVD)), or a semiconductor medium (e.g., a Solid State Disk (SSD)), or the like.
Those of ordinary skill in the art will appreciate that implementing all or part of the above-described method embodiments may be accomplished by a computer program to instruct related hardware, the program may be stored in a computer readable storage medium, and the program may include the above-described method embodiments when executed. And the aforementioned storage medium includes: a read-only memory (ROM) or a random-access memory (random access memory, RAM), a magnetic disk or an optical disk, or the like.

Claims (7)

1. An air quality monitoring and management method, characterized in that the method comprises:
acquiring temperature and humidity data in the sewer through a sensor, and acquiring an environment image of the sewer through a camera;
extracting features of the environment image, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features as target features; the feature extraction of the environment image comprises the following steps:
Establishing a plurality of target detection frames, and respectively extracting the characteristics of the multi-channel color characteristics of a plurality of environmental images of the sewer by utilizing each target detection frame to generate a plurality of characteristic sub-graphs;
Calculating the intersection ratio between any two target detection frames to be used as the correlation between any two target detection frames; splicing the feature sub-graphs according to the correlation relationship to obtain a target feature graph;
Extracting color features, texture features, spectrum features and spatial distribution features in the target feature map;
Fusing the temperature and humidity data with target characteristics, and inputting the fused temperature and humidity data into a trained prediction model to obtain a sewer air quality prediction result; training the predictive model, comprising:
Acquiring a plurality of temperature and humidity data obtained by monitoring in a historical scene and an environment image corresponding to the temperature and humidity as training samples;
Extracting features of the environment images in the training samples, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features;
fusing the temperature and humidity data with the feature quantity with the maximum correlation to train the GRU network model with the self-attention mechanism;
Constructing a loss function by using root mean square error, screening out secondary correlation feature quantities from the fitted features if the value of the loss function is larger than a first threshold value, reconstructing fusion feature iteration training the GRU network model, and generating a prediction model until the value of the loss function is smaller than the first threshold value;
And generating an air purification instruction according to the air quality prediction result, and processing the sewer by using the air purification instruction.
2. The air quality monitoring and management method according to claim 1, wherein the fitting the feature of dimension reduction according to the gradient-lifted tree algorithm comprises:
Assuming that the feature dimensions have m in total and the category number is N, calculating the coefficient of the node's base under the decision tree:
In the method, in the process of the invention, The coefficient of the foundation for the node v under the kth decision tree; The proportion of the category l under the node v is calculated;
calculating the variation of the coefficient of the characteristic x z before and after the node v branches
In the method, in the process of the invention,AndRespectively the coefficient of the foundation of 2 new nodes after branching;
defining the node set of the feature x z under the decision tree k as Q, the importance of x z under the kth decision tree is:
wherein z' is a node in the node set Q;
assuming that T trees are co-generated during training, the importance of feature x z in all trees, V z, is:
In the method, in the process of the invention, The feature importance corresponding to each tree;
and (3) performing normalization operation to obtain a final importance score I GBz of the feature x z:
Where V w is the importance of the feature dimension w.
3. The air quality monitoring management method according to claim 1, wherein the expression of the self-attention mechanism is:
where Q *,K*,V* is the find, key and value, Q *K*T is the find similarity to the key, and d k is the key dimension, respectively.
4. The air quality monitoring and management method according to claim 1, characterized in that the method further comprises:
Taking the difference value between the original input data and the GRU network output predicted value as a training set, training a model constructed based on an SVR algorithm, and obtaining an error prediction model;
and inputting the characteristics of the temperature and humidity data and the target characteristics after fusion into an error prediction model, and compensating a prediction result output by the prediction model by using the error prediction model to obtain a final air quality prediction result.
5. An air quality monitoring management system, the system comprising:
the acquisition unit is used for acquiring temperature and humidity data in the sewer through the sensor and acquiring an environment image of the sewer through the camera;
The feature extraction unit is used for extracting features of the environment image, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features to serve as target features; the feature extraction of the environment image comprises the following steps:
Establishing a plurality of target detection frames, and respectively extracting the characteristics of the multi-channel color characteristics of a plurality of environmental images of the sewer by utilizing each target detection frame to generate a plurality of characteristic sub-graphs;
Calculating the intersection ratio between any two target detection frames to be used as the correlation between any two target detection frames; splicing the feature sub-graphs according to the correlation relationship to obtain a target feature graph;
Extracting color features, texture features, spectrum features and spatial distribution features in the target feature map;
The prediction unit is used for fusing the temperature and humidity data with target characteristics, inputting the temperature and humidity data into a trained prediction model, and obtaining a predicted result of the air quality of the sewer; training the predictive model, comprising:
Acquiring a plurality of temperature and humidity data obtained by monitoring in a historical scene and an environment image corresponding to the temperature and humidity as training samples;
Extracting features of the environment images in the training samples, performing feature dimension reduction on the extracted features by using a PCA algorithm, fitting the dimension-reduced features according to a gradient lifting tree algorithm, and screening out feature quantities with the largest correlation from the fitted features;
fusing the temperature and humidity data with the feature quantity with the maximum correlation to train the GRU network model with the self-attention mechanism;
Constructing a loss function by using root mean square error, screening out secondary correlation feature quantities from the fitted features if the value of the loss function is larger than a first threshold value, reconstructing fusion feature iteration training the GRU network model, and generating a prediction model until the value of the loss function is smaller than the first threshold value;
and the purification control unit is used for generating an air purification instruction according to the air quality prediction result and processing the sewer by using the air purification instruction.
6. An electronic device, comprising: a processor and a memory for storing computer program code comprising computer instructions which, when executed by the processor, the electronic device performs the air quality monitoring management method of any of claims 1 to 4.
7. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program comprising program instructions which, when executed by a processor of an electronic device, cause the processor to perform the air quality monitoring management method of any of claims 1 to 4.
CN202410225414.8A 2024-02-29 2024-02-29 Air quality monitoring and management method and system Active CN117975220B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410225414.8A CN117975220B (en) 2024-02-29 2024-02-29 Air quality monitoring and management method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410225414.8A CN117975220B (en) 2024-02-29 2024-02-29 Air quality monitoring and management method and system

Publications (2)

Publication Number Publication Date
CN117975220A CN117975220A (en) 2024-05-03
CN117975220B true CN117975220B (en) 2024-08-30

Family

ID=90849761

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410225414.8A Active CN117975220B (en) 2024-02-29 2024-02-29 Air quality monitoring and management method and system

Country Status (1)

Country Link
CN (1) CN117975220B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11222217B1 (en) * 2020-08-14 2022-01-11 Tsinghua University Detection method using fusion network based on attention mechanism, and terminal device

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108387262B (en) * 2018-01-03 2020-05-01 江苏大学 Greenhouse information automatic monitoring method based on suspension type sliding rail platform
CN207797248U (en) * 2018-05-14 2018-08-31 天津膜术工场环保科技有限公司 The public space air cleaning intelligent equipment for calculating analysis is acquired based on big data
CN110823291A (en) * 2019-11-27 2020-02-21 山东建筑大学 Method and system for indoor temperature and humidity environment monitoring in buildings based on K-means clustering algorithm
CN111859672B (en) * 2020-07-22 2021-03-02 南京工业职业技术大学 Intelligent construction method for ecological environment of strawberry and bee composite facility
CN113657042B (en) * 2021-09-01 2024-11-05 中国人民解放军国防科技大学 A method for air quality parameter prediction based on multi-scale attention mechanism
CN113947573A (en) * 2021-10-07 2022-01-18 桂林理工大学 Underground pipeline defect detection method based on improved Faster R-CNN
CN114358401A (en) * 2021-12-21 2022-04-15 湖北工业大学 Road icing identification method fusing image and meteorological environment data
US12055464B2 (en) * 2022-04-14 2024-08-06 Pipesense, Llc Systems and methods for improved pipeline leak detection
CN114912707B (en) * 2022-06-01 2023-06-30 郑州大学 Air quality prediction system and prediction method based on multi-mode fusion
CN115188429A (en) * 2022-07-27 2022-10-14 华东理工大学 Catalytic cracking unit key index modeling method integrating time sequence feature extraction
CN115729287A (en) * 2022-11-04 2023-03-03 中国人民解放军火箭军工程大学 A warehouse environment monitoring system and method
CN116051496A (en) * 2023-01-04 2023-05-02 淮阴工学院 Real-time sewer defect detection method
CN116229678A (en) * 2023-03-09 2023-06-06 中国人民公安大学 Fatigue driving detection warning system, method and automobile

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11222217B1 (en) * 2020-08-14 2022-01-11 Tsinghua University Detection method using fusion network based on attention mechanism, and terminal device

Also Published As

Publication number Publication date
CN117975220A (en) 2024-05-03

Similar Documents

Publication Publication Date Title
Okamoto et al. Plant classification for weed detection using hyperspectral imaging with wavelet analysis
Ji et al. A novel deep learning approach for anomaly detection of time series data
CN113298058A (en) Water quality prediction inversion method and system based on hyperspectral image of unmanned aerial vehicle
CN116952654B (en) Environment monitoring and early warning system for administrative supervision
CN118365497A (en) Smart city environment monitoring method and device and electronic equipment
CN117009785A (en) Security monitoring method, device, server and system
Cheng et al. Water quality monitoring method based on TLD 3D fish tracking and XGBoost
Liso et al. A review of deep learning based anomaly detection strategies in Industry 4.0 focused on application fields, sensing equipment and algorithms
CN117975220B (en) Air quality monitoring and management method and system
Vijayan et al. A universal foreground segmentation technique using deep-neural network
Feng et al. High-efficiency progressive transmission and automatic recognition of wildlife monitoring images with WISNs
Yan et al. A fast and effective image preprocessing method for hot round steel surface
CN118230159A (en) A method for detecting temporal changes in remote sensing mining images based on dual-branch deep supervision
You et al. Tampering detection and localization base on sample guidance and individual camera device convolutional neural network features
CN117809118A (en) Visual perception identification method, device and medium based on deep learning
CN115909144A (en) Method and system for detecting abnormity of surveillance video based on counterstudy
Ghosh et al. Robust autonomous detection of the defective pixels in detectors using a probabilistic technique
CN114782895A (en) Multi-temporal monitoring image anti-interference abnormal detection model training method, device and detection method
CN117978620B (en) Abnormality detection method and system for electronic distribution frame
Jöchl et al. Deep learning image age approximation-what is more relevant: Image content or age information?
CN117197134B (en) Defect detection method, device, equipment and storage medium
CN119129857B (en) Water quality prediction method and software system for water supply network based on spatiotemporal graph neural network
Elharrouss et al. PDC-ViT: source camera identification using pixel difference convolution and vision transformer
CN119723441A (en) Substation monitoring and identifying method and device, electronic equipment and storage medium
Cheng et al. The fusion strategy of multimodal learning in image and text recognition

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