CN114943917B - Algorithm for visually identifying aeration quantity of aerobic tank of sewage plant - Google Patents
Algorithm for visually identifying aeration quantity of aerobic tank of sewage plant Download PDFInfo
- Publication number
- CN114943917B CN114943917B CN202210522952.4A CN202210522952A CN114943917B CN 114943917 B CN114943917 B CN 114943917B CN 202210522952 A CN202210522952 A CN 202210522952A CN 114943917 B CN114943917 B CN 114943917B
- Authority
- CN
- China
- Prior art keywords
- aeration
- picture
- training
- video
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F3/00—Biological treatment of water, waste water, or sewage
- C02F3/006—Regulation methods for biological treatment
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F3/00—Biological treatment of water, waste water, or sewage
- C02F3/02—Aerobic processes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/75—Clustering; Classification
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- 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
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- 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
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W10/00—Technologies for wastewater treatment
- Y02W10/10—Biological treatment of water, waste water, or sewage
Abstract
The invention discloses an algorithm for visually identifying aeration quantity of an aerobic tank of a sewage plant, which comprises the following steps: firstly, constructing an input sample, including data acquisition and pretreatment, secondly, constructing a training network, adjusting network parameters to obtain an optimal model for identifying the aeration amount of the whole picture, wherein a neural network frame based on HRNet is specifically adopted, the whole network is connected with subnets with high resolution to low resolution in parallel instead of recovering resolution through a low-to-high process, and thirdly, checking the feasibility of the model.
Description
Technical Field
The invention relates to the technical field of visual identification analysis, in particular to an algorithm for visually identifying aeration quantity of an aerobic tank of a sewage plant.
Background
The capping and buried construction of the sewage treatment plant can not timely know the condition of the whole tank in a manual inspection mode, and the aeration tank with the aeration cost of more than 50% of the whole plant running cost is a main process for removing organic matters in a biochemical section.
However, at present, the aeration quantity of the aeration tank is monitored and regulated only by one aeration quantity input value, a sewage plant with an accurate aeration system is not manufactured, the aeration tank of the whole plant only has a blower quantity value, the monitoring and judging system at the rear end is not available, the aeration conditions of different areas of the aeration tank can not be known, if the aeration system is in a problem, the phenomena of weak aeration, overexposure, uneven aeration and the like can be caused, and further regional anaerobic and anoxic environments are caused, the treatment efficiency of the process section is reduced, and the quality of effluent is not up to standard.
In addition, whether the aeration tank has problems or not is observed by visual inspection of inspection staff at present, the randomness is large, the error is large, the accurate aeration quantity of each area of the aeration tank can not be analyzed and calculated, and a systematic and intelligent monitoring system can not be formed.
Disclosure of Invention
The invention aims to provide an algorithm for visually identifying aeration quantity of an aerobic tank of a sewage plant, which is used for independently analyzing videos of all areas of an aeration tank, giving out fitting aeration quantity of all areas and analysis results which are compared with normal aeration quantity, and giving out result judgment of whether all areas are normal or not.
The aim of the invention can be achieved by the following technical scheme:
an algorithm for visually identifying aeration quantity of an aerobic tank of a sewage plant comprises the following steps:
step one: constructing an input sample data set, comprising video acquisition and preprocessing, and acquiring an aeration picture by framing the video to acquire picture aeration in the aeration picture;
step two: constructing a training network, adjusting network parameters, and obtaining an optimal model for identifying the picture aeration quantity in an aeration picture by adopting a neural network frame based on an HRNet network;
step three: and (3) through an optimal model, the acquired aeration picture passes through a model test, and aeration value output is carried out.
As a further scheme of the invention: the video acquisition and preprocessing in the first step comprises the following steps:
s1: setting up a multipoint fixed monitoring mode for a sewage treatment plant, setting monitoring points in an aeration tank, and erecting an acquisition camera to acquire a process video so as to acquire a related process video;
s2: carrying out framing treatment on the related process video, and screening the treated pictures to obtain aeration pictures;
s3: and acquiring aeration quantity at all times by utilizing the SQL data set, reading the aeration quantity at the time in the aeration picture, and marking the aeration quantity at the time.
As a further scheme of the invention: the obtaining of the optimal model in the second step comprises the following steps:
w1: putting the data set obtained in the step one into a YOLOV5 training network model, and iterating the video based on a Pytorch frame by training the network model;
w2: establishing a cross entropy loss function;
w3: adding a head of classification output, the input is s response maps: { X1, X2, …, xs };
the output is an s-response map: { X1, X2, …, xs }, where its resolution and width are the same as the input maps, each output map is an aggregation of the input maps, i.e.:
the cross-stage switching unit has an additional output map ys+1, i.e
Y s+1 ＝a(Y S ,S+1)；
Function a (X i K) consists of downsampling Xi of resolution i to resolution k, downsampling being convolved with stride 3*3;
w4: and carrying out a plurality of rounds of training on the data set to obtain the optimal model of the training.
As a further scheme of the invention: the YOLOV5 training network model in W1 adopts Adam optimizer to optimize the training network model, the optimizer is SGD, the training parameters are 512x512, the initial learning rate is 0.05, and the activation function is relu.
As a further scheme of the invention: the number of video iterations in W1 is not less than 100.
As a further scheme of the invention: step three, judging the state of the identified result output through a model test;
if the state is normal, outputting an aeration value;
abnormal state, the abnormal state is displayed.
The invention has the beneficial effects that:
(1) The invention provides a visual recognition aeration fitting algorithm, which is characterized in that the aeration amount of each area of an aeration tank is quantitatively analyzed by a visual recognition method to give accurate aeration amount data, and the operation state of the aeration tank which is rough and fuzzy is ended;
(2) The algorithm vision recognition aeration quantity fitting algorithm has universality, can be applied to different sewage plants by adjusting model parameters after short-term training by applying the established algorithm aiming at different sewage plants, and is suitable for different environments and conditions;
(3) According to the invention, the fitting aeration quantity of a plurality of areas is used for reflecting the whole tank condition of the aeration tank, each area is thinned, operators can find abnormal areas in time, and targeted fixed-point overhaul is realized;
(4) The algorithm analysis result of the invention forms a systematic database which is used as a reference index for monitoring the long-term running state of the aeration tank of the sewage plant, integrates and analyzes with other monitoring data, and assists the energy conservation and emission reduction of the sewage plant.
Drawings
The invention is further described below with reference to the accompanying drawings.
Fig. 1 is a schematic diagram of the flow chart of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but 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.
Referring to fig. 1, the invention is an algorithm for visually identifying aeration rate of an aerobic tank of a sewage plant, by fitting the aeration rate values of a picture where the aeration rate is located, outputting an actual value, and judging whether the actual state of the area accords with a normal state;
the system comprises: video processing, statistical analysis, data fitting, model building and result output module;
the video processing is carried out on the collected video of the aeration tank, firstly, a monitoring real-time video stream is obtained through an upper computer, the upper computer system stores the real-time video stream, the real-time video stream is recorded and stored in real time in a DVR hard disk video recording mode, the DVR segmented video is obtained according to 10S/segment segmentation, the 10S/segment is obtained, the segments are overlapped in time, and H.246-MP 4 format conversion is carried out on the segmented video respectively;
the picture screening obtains available aeration pictures through framing treatment of the video by using OPENCV;
the statistical analysis can judge the data transmitted in real time on site, divide the condition of the video into two conditions of normal and abnormal, and classify the judgment of the video state, so that the condition of unnecessary detection of the aeration concentration state is preferentially eliminated, the frequency of model training is reduced, and the training precision is improved;
the real-time aeration quantity is recorded by using SQL (structured query language) in data fitting, a database of ventilation quantity is built, the data can be conveniently inquired and used, and the aeration quantity of a usable picture at a specific moment is collected and counted;
since the quantity of aeration is determined according to the quantity of bubbles in the picture, each actual X has a fixed Y, and the quantity of aeration at the moment is determined by adopting a linear regression mode, namely the formula is as follows:
y＝Wx+b
w is weight, x is an input image, b is offset, aeration fitting is carried out on the aeration quantity of the input image, a model for realizing aeration is finally formed, and the model is brought into an HRNet network;
model building and result output are carried out by putting the aeration quantity of the picture at each moment into an HRNet model to obtain a linear regression model, training is carried out for 100 rounds in total, a loss function is continuously reduced, and finally a model capable of estimating the aeration quantity of the picture is obtained and is brought into the HRNet network;
outputting the aeration value at the moment, and judging whether the moment is in a normal aeration state by judging the aeration value at the moment:
if yes, outputting an aeration value at the moment;
if not, the output is in an abnormal state.
If the input is a video file, the picture of each second is stored, the aeration amount of the second is output, the aeration values identified by a single second are added at the end by using an average, and the value identified by the normal state is divided to obtain an average value of the aeration values. Thus, the aeration amount of the whole video is obtained, the accuracy of the whole recognition process is improved, and the aeration amount at a certain moment is regressed by using an image sequence.
The visual recognition algorithm for the aeration quantity of the aerobic tank of the sewage plant comprises the following steps:
step one: constructing an input sample data set, comprising video acquisition and preprocessing, and acquiring an aeration picture by framing the video to acquire picture aeration in the aeration picture;
the video acquisition and preprocessing steps are as follows:
s1: setting up a multipoint fixed monitoring mode for a sewage treatment plant, setting monitoring points in an aeration tank, and erecting an acquisition camera to acquire a process video so as to acquire a related process video;
s2: carrying out framing treatment on the related process video, and screening the treated pictures to obtain aeration pictures;
s3: collecting aeration quantity at all moments by utilizing an SQL data set, reading the aeration quantity at the moment in an aeration picture, and marking the aeration quantity at the moment;
step two: constructing a training network, adjusting network parameters, and obtaining an optimal model for identifying the picture aeration quantity in an aeration picture by adopting a neural network frame based on an HRNet network;
w1: putting the data set obtained in the step one into a YOLOV5 training network model, and iterating the video based on a Pytorch frame by training the network model;
wherein, the YOLOV5 training network model in W1 adopts an Adam optimizer to optimize the training network model, the optimizer is SGD, the training parameters are 512x512, the initial learning rate is 0.05, and the activation function is relu;
wherein the number of video iterations in W1 is not less than 100;
w2: establishing a cross entropy loss function;
w3: adding a head of classification output, the input is s response maps: { X1, X2, …, xs };
the output is an s-response map: { X1, X2, …, xs }, where its resolution and width are the same as the input maps, each output map is an aggregation of the input maps, i.e.:
the cross-stage switching unit has an additional output map ys+1, i.e
Y s+1 ＝a(Y S ,S+1)；
Function a (X i K) consists of downsampling Xi of resolution i to resolution k, downsampling being convolved with stride 3*3;
wherein one stride 3×3 convolution of stride2 is used for 2×downsampling and a stride 3×3 convolution of two consecutive stride2 is used for 4×downsampling); if i=k, a (..) is just one identification (identity) connection, thenThe following steps: a (X) i ,k)＝X i ；
W4: carrying out a plurality of rounds of training on the data set to obtain an optimal model of the training;
step three: the acquired aeration pictures pass through a model test through an optimal model, and the state judgment is carried out on the recognized result output through the model test;
if the state is normal, outputting an aeration value;
abnormal state, the abnormal state is displayed.
The foregoing describes one embodiment of the present invention in detail, but the description is only a preferred embodiment of the present invention and should not be construed as limiting the scope of the invention. All equivalent changes and modifications within the scope of the present invention are intended to be covered by the present invention.
Claims (4)
1. An algorithm for visually identifying aeration quantity of an aerobic tank of a sewage plant is characterized by comprising the following steps:
step one: constructing an input sample data set, comprising video acquisition and preprocessing, and acquiring an aeration picture by framing the video to acquire picture aeration in the aeration picture;
step two: constructing a training network, adjusting network parameters, and obtaining an optimal model for identifying the picture aeration quantity in an aeration picture by adopting a neural network frame based on an HRNet network;
step three: through an optimal model, the acquired aeration picture passes through a model test, and aeration value output is carried out;
the video acquisition and preprocessing in the first step comprises the following steps:
s1: setting up a multipoint fixed monitoring mode for a sewage treatment plant, setting monitoring points in an aeration tank, and erecting an acquisition camera to acquire a process video so as to acquire a related process video;
s2: carrying out framing treatment on the related process video, and screening the treated pictures to obtain aeration pictures;
s3: collecting aeration quantity at all moments by utilizing an SQL data set, reading the aeration quantity at the moment in an aeration picture, and marking the aeration quantity at the moment;
the obtaining of the optimal model in the second step comprises the following steps:
w1: putting the data set obtained in the step one into a YOLOV5 training network model, and iterating the video based on a Pytorch frame by training the network model;
w2: establishing a cross entropy loss function;
w3: adding a head of classification output, the input is s response maps: { X1, X2, …, xs };
the output is an s-response map: { X1, X2, …, xs }, where its resolution and width are the same as the input maps, each output map is an aggregation of the input maps, i.e.:
the cross-stage switching unit has an additional output map ys+1, i.e
Y s+1 ＝a(Y S ,S+1)；
Function a (X i K) consists of downsampling Xi of resolution i to resolution k, downsampling being convolved with stride 3*3;
w4: and carrying out a plurality of rounds of training on the data set to obtain the optimal model of the training.
2. The method for visually identifying aeration rate of an aerobic tank of a sewage plant according to claim 1, wherein a YOLOV5 training network model in W1 adopts an Adam optimizer to optimize the training network model, the optimizer is SGD, training parameters are 512x512, initial learning rate is 0.05, and an activation function is relu.
3. The method for visually recognizing aeration quantity of an aerobic tank of a sewage plant according to claim 1, wherein the number of video iterations in W1 is not less than 100.
4. The method for visually identifying aeration rate of an aerobic tank of a sewage plant according to claim 1, wherein in the third step, the status of the identified result output is judged by a model test;
if the state is normal, outputting an aeration value;
abnormal state, the abnormal state is displayed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210522952.4A CN114943917B (en) | 2022-05-13 | 2022-05-13 | Algorithm for visually identifying aeration quantity of aerobic tank of sewage plant |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210522952.4A CN114943917B (en) | 2022-05-13 | 2022-05-13 | Algorithm for visually identifying aeration quantity of aerobic tank of sewage plant |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114943917A CN114943917A (en) | 2022-08-26 |
| CN114943917B true CN114943917B (en) | 2023-09-19 |
Family
ID=82907874
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210522952.4A Active CN114943917B (en) | 2022-05-13 | 2022-05-13 | Algorithm for visually identifying aeration quantity of aerobic tank of sewage plant |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114943917B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115417492B (en) * | 2022-08-30 | 2023-06-20 | 同济大学建筑设计研究院(集团)有限公司 | Advanced oxidation system based on underwater vision and control method |
| CN116135797B (en) * | 2023-04-19 | 2023-07-04 | 江苏海峡环保科技发展有限公司 | Intelligent control system for sewage treatment |
| CN116693075B (en) * | 2023-07-27 | 2023-11-21 | 杭州回水科技股份有限公司 | Aeration device of activated carbon biological filter |
| CN117466501B (en) * | 2023-12-28 | 2024-03-29 | 山东公用环保科技集团有限公司 | Sewage aeration method with deodorization function |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103499982A (en) * | 2013-09-30 | 2014-01-08 | 北京工业大学 | Self-organization control method of sewage treatment process |
| CN112101454A (en) * | 2020-09-15 | 2020-12-18 | 成都明控科技有限公司 | Sewage aeration analysis method |
| CN113408321A (en) * | 2020-03-16 | 2021-09-17 | 中国人民解放军战略支援部队信息工程大学 | Real-time target detection method and device for lightweight image and video data |
| CN114049590A (en) * | 2021-11-10 | 2022-02-15 | 安徽大学 | Video-based ski-jump analysis method |
| CN114180733A (en) * | 2021-11-02 | 2022-03-15 | 合肥中盛水务发展有限公司 | Sewage aeration amount detection and aeration control system based on video analysis algorithm |
| CN114359851A (en) * | 2021-12-02 | 2022-04-15 | 广州杰赛科技股份有限公司 | Unmanned target detection method, device, equipment and medium |
-
2022
- 2022-05-13 CN CN202210522952.4A patent/CN114943917B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103499982A (en) * | 2013-09-30 | 2014-01-08 | 北京工业大学 | Self-organization control method of sewage treatment process |
| CN113408321A (en) * | 2020-03-16 | 2021-09-17 | 中国人民解放军战略支援部队信息工程大学 | Real-time target detection method and device for lightweight image and video data |
| CN112101454A (en) * | 2020-09-15 | 2020-12-18 | 成都明控科技有限公司 | Sewage aeration analysis method |
| CN114180733A (en) * | 2021-11-02 | 2022-03-15 | 合肥中盛水务发展有限公司 | Sewage aeration amount detection and aeration control system based on video analysis algorithm |
| CN114049590A (en) * | 2021-11-10 | 2022-02-15 | 安徽大学 | Video-based ski-jump analysis method |
| CN114359851A (en) * | 2021-12-02 | 2022-04-15 | 广州杰赛科技股份有限公司 | Unmanned target detection method, device, equipment and medium |
Non-Patent Citations (1)
| Title |
|---|
| Ke Sun etc.."Deep High-Resolution Representation Learning for Human Pose Estimation".《arXiv:1902.09212v1》.2019,第1-12页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114943917A (en) | 2022-08-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114943917B (en) | Algorithm for visually identifying aeration quantity of aerobic tank of sewage plant | |
| CN108765404B (en) | A kind of road damage testing method and device based on deep learning image classification | |
| CN111400547B (en) | Human-computer cooperation video anomaly detection method | |
| CN110738127A (en) | Helmet identification method based on unsupervised deep learning neural network algorithm | |
| CN110955288B (en) | Environment supervision system based on big data | |
| CN112288597A (en) | Energy consumption online anomaly detection method based on hierarchical clustering and histogram algorithm | |
| CN115372571A (en) | Intelligent water environment monitoring system | |
| CN116843955A (en) | Microorganism classification and identification method and system based on computer vision | |
| CN113469938B (en) | Pipe gallery video analysis method and system based on embedded front-end processing server | |
| KR20140117959A (en) | Apparatus and method for provide of causative factors for stage of outflow water of wastewater treatment plant | |
| CN116168356B (en) | Vehicle damage judging method based on computer vision | |
| CN113485218A (en) | Wisdom thing allies oneself with supervision platform based on 5G | |
| CN117351702A (en) | Intelligent traffic management method based on adjustment of traffic flow | |
| CN116913047A (en) | River basin agriculture non-point source pollutant tracing method and early warning system | |
| CN116452358A (en) | Intelligent agriculture management system based on Internet of things | |
| CN115219472B (en) | Method and system for quantitatively identifying multiple pollution sources of mixed water body | |
| CN116363426A (en) | Automatic detection method for health state of highway antiglare shield | |
| CN115659056A (en) | Accurate matching system of user service based on big data | |
| CN110097552A (en) | A kind of automatic division method of mouse prefrontal lobe neuron two-photon fluorescence image | |
| CN110084125B (en) | Agricultural insurance investigation technical method based on deep learning | |
| CN113033623A (en) | Pollution source identification method and system based on ultraviolet-visible absorption spectrum | |
| CN113311081A (en) | Pollution source identification method and device based on three-dimensional liquid chromatography fingerprint | |
| CN112132088A (en) | Inspection point location missing inspection identification method | |
| Dara et al. | Detection of coordinate based accident-prone areas on road surface using machine learning methods | |
| CN116468205B (en) | Method and system for monitoring environment-friendly detection quality of motor vehicle |
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 |