CN116152748A - River and lake supervision method and system based on blue algae identification - Google Patents

River and lake supervision method and system based on blue algae identification Download PDF

Info

Publication number
CN116152748A
CN116152748A CN202310416654.1A CN202310416654A CN116152748A CN 116152748 A CN116152748 A CN 116152748A CN 202310416654 A CN202310416654 A CN 202310416654A CN 116152748 A CN116152748 A CN 116152748A
Authority
CN
China
Prior art keywords
blue algae
area
lake
river
image
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.)
Granted
Application number
CN202310416654.1A
Other languages
Chinese (zh)
Other versions
CN116152748B (en
Inventor
王晓宇
吴几
谢忱
洪磊
甘琳
范子武
蔡秋鹏
金义蓉
徐涛
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.)
Suzhou Water Resources And Water Affairs Information Dispatching Command Center
Nanjing Hydraulic Research Institute of National Energy Administration Ministry of Transport Ministry of Water Resources
Original Assignee
Suzhou Water Resources And Water Affairs Information Dispatching Command Center
Nanjing Hydraulic Research Institute of National Energy Administration Ministry of Transport Ministry of Water Resources
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 Suzhou Water Resources And Water Affairs Information Dispatching Command Center, Nanjing Hydraulic Research Institute of National Energy Administration Ministry of Transport Ministry of Water Resources filed Critical Suzhou Water Resources And Water Affairs Information Dispatching Command Center
Priority to CN202310416654.1A priority Critical patent/CN116152748B/en
Publication of CN116152748A publication Critical patent/CN116152748A/en
Application granted granted Critical
Publication of CN116152748B publication Critical patent/CN116152748B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious 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
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/54—Extraction of image or video features relating to texture
    • 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
    • G06V10/56—Extraction of image or video features relating to colour
    • 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
    • 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
    • G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A20/00—Water conservation; Efficient water supply; Efficient water use
    • Y02A20/20—Controlling water pollution; Waste water treatment

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a river and lake supervision method and system based on blue algae identification, and belongs to the technical field of river and lake management and control. Comprising the following steps: updating frame images continuously acquired in a specified time period according to a preset time interval to a blue algae image set; creating a blue algae identification model, inputting each newly added frame image in the blue algae image set into the blue algae identification model to obtain a uniform image, judging whether blue algae exists currently or not, acquiring real-time blue algae information if the blue algae exists currently, and triggering a blue algae prevention and control mechanism; and identifying whether pollutants exist in the river and lake water surface area in the video frame by using the pollution identification model, analyzing the pollutant types and triggering an environment control mechanism. According to the invention, the blue algae image is obtained from the monitoring video in real time by combining video monitoring with the image characteristics of the blue algae, the coverage rate or coverage area of the blue algae is obtained by dividing the blue algae image, the situation of the blue algae is known in time, and a corresponding blue algae prevention and control mechanism is executed, so that strong and high-timeliness control of the blue algae is realized.

Description

River and lake supervision method and system based on blue algae identification
Technical Field
The invention belongs to the technical field of river and lake management and control, and particularly relates to a river and lake supervision method and system based on blue algae identification.
Background
Inhibiting the onset of cyanobacterial bloom is the key point of controlling eutrophication in fresh water areas worldwide. Therefore, blue algae video monitoring is widely applied to observing and information collecting of the blue algae on the water surface and the shore of lakes of blue algae frequent lakes such as Tai lake, nest lake and the like, a video identification technology is adopted to directly acquire blue algae images from monitoring videos in real time, blue algae in the images are segmented, and then the coverage rate or coverage area of the blue algae (pixel points) is calculated, so that quantitative indexes can be provided for real-time early warning or decision making of the blue algae.
The conventional image segmentation algorithm is roughly divided into two steps: firstly, designing features according to the characteristics of a target, and then classifying or clustering pixel points according to the features. This requires that the object have more distinct characteristic information, including color features, contour features, or texture features, etc. For example, the threshold segmentation algorithm finds a reasonable threshold interval of the RGB channel according to the color characteristics of the target, and then classifies pixel points according to the interval. Blue algae lack significant feature information, however, making it very difficult to design effective features. First, blue algae lack distinct color characteristics. Blue algae have different colors in different growth stages, and blue algae in the growth period are blue green, but blue algae in the maturation period are yellow; and the blue algae has a color related to the concentration, and generally has a darker color when the concentration is high and a lighter color when the concentration is low. Secondly, blue algae lack obvious profile features: blue algae do not have a specific contour and cannot be segmented by contour like a human body or an automobile. Finally, a large amount of aquatic weeds are mixed in the water, tree reflection and artificial garbage can appear on the bank, and the difficulty is increased for the application of the traditional image segmentation algorithm in blue algae segmentation.
Disclosure of Invention
The invention provides a river and lake supervision method and a river and lake supervision system based on blue algae identification for solving the technical problems in the background technology.
The invention adopts the following technical scheme: a river and lake supervision method based on blue algae identification comprises the following steps:
acquiring basic data of a river and a lake, and laying a plurality of monitoring points based on the basic data; obtaining video frames of river and lake by utilizing video monitoring, and updating frame images continuously collected in a designated time period according to a preset time interval to a blue algae image set;
creating a blue algae identification model, inputting each newly added frame image in the blue algae image set into the blue algae identification model to obtain a uniform image, judging whether blue algae exists currently or not based on the uniform image, acquiring real-time blue algae information if the blue algae exists, and triggering a blue algae prevention and control mechanism: comparing and analyzing the current blue algae information with the historical blue algae information to obtain the control situation of the blue algae, updating and sending out real-time control measures based on the control situation;
identifying whether pollutants exist in the water surface area of the river and the lake in the video frame by using the pollution identification model, if so, further analyzing the pollutant types and triggering an environment management and control mechanism: and monitoring the position of the pollutant in real time, tracking, calling out a historical frame image corresponding to the current frame image, and searching a pollution source again by using a pollution identification model to obtain a generation record related to the pollutant.
In a further embodiment, the current cyanobacteria information comprises: the position information of the blue algae and the occupied area of the blue algae.
In a further embodiment, the specific step of determining whether blue algae is currently present based on the uniform image is as follows: the color characteristic and texture characteristic of the uniform image are obtained, and the color characteristic value of the ith pixel is calculated by adopting the following formulas
Figure SMS_1
And texture feature value +.>
Figure SMS_2
：
Figure SMS_3
The method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure SMS_4
Mean feature vector in color feature space for uniform image, +.>
Figure SMS_5
Pixel characteristic representing pixel i, +.>
Figure SMS_6
Is a mould;
Figure SMS_7
the method comprises the steps of carrying out a first treatment on the surface of the Wherein p represents the number of dimensions of the extracted texture feature, each number corresponding to a direction,/->
Figure SMS_8
Representing the texture value of pixel i in dimension p;
presetting a blue algae color characteristic threshold interval and a blue algae texture characteristic threshold interval: if the color characteristic value
Figure SMS_9
Texture feature value->
Figure SMS_10
At least one of the blue algae is not in the corresponding characteristic area, and the blue algae is not present in the current river and lake;
if the color characteristic value
Figure SMS_11
Belongs to blue algae color characteristic threshold value interval, and texture characteristic value +.>
Figure SMS_12
If the blue algae belongs to the blue algae texture feature threshold interval, the blue algae existing in the current river and lake is indicated, and the corresponding pixel i is listed as a blue algae area r;
the position of the blue algae area is the position information of the blue algae.
In a further embodiment, the calculation flow of the land occupation area of the blue algae is as follows: based on the blue algae area r, calculating the blue algae area by the following formula:
Figure SMS_13
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure SMS_14
Representing the image area of blue algae region r in a uniform image,/->
Figure SMS_15
Pixel area representing uniform image, +.>
Figure SMS_16
Is the actual area corresponding to the pixel area, +.>
Figure SMS_17
For perspective transformation coefficients->
Figure SMS_18
Is the included angle between the camera and the horizontal plane.
In a further embodiment, the uniform image is obtained as follows:
converting a frame image into a gray image, defining a gray threshold
Figure SMS_19
And->
Figure SMS_20
The method comprises the steps of carrying out a first treatment on the surface of the Acquiring region gray value +.>
Figure SMS_21
Based on the region gray value +.>
Figure SMS_22
Gray threshold->
Figure SMS_23
And->
Figure SMS_24
Dividing a frame image into a dark region, a normal region and a bright region;
based on the gray values of the normal areas, a gray compensation formula for the dark areas and the bright areas is obtained:
Figure SMS_25
；
in the method, in the process of the invention,
Figure SMS_27
gray value after illumination compensation for dark areas, < >>
Figure SMS_29
Is the gray value of the normal region,
Figure SMS_30
gray value after illumination compensation for bright area, +.>
Figure SMS_32
，/>
Figure SMS_34
Is the standard deviation of gray values of the normal region, +.>
Figure SMS_36
Is the standard deviation of the gray values of the dark areas, < >>
Figure SMS_38
，/>
Figure SMS_26
Is the gray average value of the normal area, +.>
Figure SMS_28
For the gray average value of the dark areas, +.>
Figure SMS_31
，/>
Figure SMS_33
Standard deviation of gray value for bright area, +.>
Figure SMS_35
，/>
Figure SMS_37
Is the gray average value of the bright area;
and matching the areas corresponding to the dark areas and the bright areas of the gray level image in the frame image to obtain corresponding dark areas and bright areas, and respectively processing the dark areas and the bright areas by using the gray level compensation formula to obtain uniform images with uniform luminosity.
In a further embodiment, the process of updating and issuing the real-time prevention and control measures based on the prevention and control conditions is as follows:
after comparing and analyzing the current blue algae information and the historical blue algae information, if at least one of the following triggering conditions is met, the prevention and control conditions are poor, and the prevention and control are required to be enhanced;
the triggering conditions are as follows:
Figure SMS_39
、/>
Figure SMS_40
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure SMS_41
Representing the number of blue algae areas acquired by the current frame, < + >>
Figure SMS_42
Representing the number of blue algae areas acquired by historical frames, < >>
Figure SMS_43
Representing the area of blue algae acquired by the current frame, +.>
Figure SMS_44
Representing the area of blue algae acquired by the history frame.
In a further embodiment, the contaminant types include: solid matter contamination and liquid matter contamination;
the identification steps of the pollutant type are as follows: presetting a water area color characteristic threshold interval and a water area texture characteristic threshold interval, and if the color characteristic value is
Figure SMS_45
Meanwhile, the blue algae color characteristic threshold value interval and the blue algae color characteristic threshold value interval do not belong to the water area color characteristic threshold value interval, and the texture characteristic value is +.>
Figure SMS_46
Meanwhile, if the pixel i does not belong to the water area texture feature threshold interval and the blue algae texture feature threshold interval, preliminarily judging the pixel i as a pollutant and endowing the pixel with a pixel tag j, and calculating the occupied area of the pixel tag j by adopting the following formula:
Figure SMS_47
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure SMS_48
Representing the image area of pixel label j in a uniform image,/-, and>
Figure SMS_49
pixel area representing uniform image, +.>
Figure SMS_50
Is the actual area corresponding to the pixel area, +.>
Figure SMS_51
For perspective transformation coefficients->
Figure SMS_52
Is the included angle between the camera and the horizontal plane;
when the blue algae is identified to exist and
Figure SMS_53
or +.>
Figure SMS_54
The method indicates that the lake water of the river and the lake is comprehensively polluted, and the pollution type which is primarily judged is liquid substance pollution; on the contrary, let(s)>
Figure SMS_55
Or->
Figure SMS_56
And then the solid pollution exists in the river and the lake.
In a further embodiment, the environment management mechanism comprises: water quality detection, water quality treatment, garbage removal and pollution source searching; further comprises: and executing early warning, warning and punishment on the behavior main body.
A river and lake supervision system for implementing a river and lake supervision method based on blue algae identification as described above, comprising:
the first module is arranged to acquire basic data of rivers and lakes, and a plurality of monitoring points are arranged based on the basic data; obtaining video frames of river and lake by utilizing video monitoring, and updating frame images continuously collected in a designated time period according to a preset time interval to a blue algae image set;
the second module is set to create a blue algae identification model, input every newly added frame image in the blue algae image set into the blue algae identification model to obtain a uniform image, judge whether blue algae exists currently based on the uniform image, acquire real-time blue algae information if the blue algae exists, and trigger a blue algae prevention and control mechanism: comparing and analyzing the current blue algae information with the historical blue algae information to obtain the control situation of the blue algae, updating and sending out real-time control measures based on the control situation;
a third module configured to identify whether a contaminant is present in the river/lake water surface area in the video frame using the contaminant identification model, and if so, to further analyze the type of contaminant and trigger an environmental control mechanism: and monitoring the position of the pollutant in real time, tracking, calling out a historical frame image corresponding to the current frame image, and searching a pollution source again by using a pollution identification model to obtain a generation record related to the pollutant.
The invention has the beneficial effects that: according to the invention, by utilizing video monitoring and combining with the image characteristics of blue algae, the blue algae image is obtained in real time from the monitoring video, the coverage rate or coverage area of the blue algae is obtained by dividing the blue algae image, the situation of the blue algae is known in time, and a corresponding blue algae prevention and control mechanism is executed based on the known situation of the blue algae, so that the blue algae is strongly and highly aged controlled.
Meanwhile, through video images, the pollution types such as garbage throwing and sewage dumping in the river and lake areas are recognized, pollution occurrence behaviors can be traced, a river and lake intelligent supervision platform is created, video analysis is used for replacing manual inspection, and behaviors of throwing garbage into a river channel and dumping sewage in disorder are captured in real time, so that the occurrence condition of blue algae is known in time.
And an intelligent video analysis platform is built, and information which is difficult to acquire by manpower is mined from mass storage videos, so that scientific basis is provided for subsequent event search prediction. The intelligent monitoring platform for the river and the lake is capable of effectively realizing the active monitoring of targets and behaviors such as engineering equipment, water quality and illegal behaviors of the river and the lake, forming the comprehensive monitoring of water, objects and people, providing scientific basis for decision optimization, trend prediction and intervention control, gradually realizing modernization and intellectualization of the river and the lake management and protection, and realizing an unattended operation mode, thereby improving the working efficiency, reducing the working strength, reducing the operation management cost and promoting the sustainable development of the river and the lake management.
Drawings
Fig. 1 is a flowchart of a blue algae intelligent identification method and a river and lake supervision method.
Fig. 2 is a view of frame images continuously acquired at predetermined time intervals within a specified period of time.
FIG. 3 (a) is an acquired frame image; FIG. 3 (b) is a drawing showing the blue algae region obtained by the blue algae identification model treatment.
Detailed Description
The present embodiment is further described below with reference to the drawings and examples of the specification.
Example 1
As shown in fig. 1, in order to solve the technical problems existing in the background technology, the embodiment provides an intelligent blue algae identification method and a river and lake supervision method, which comprise the following steps:
step one, obtaining basic data of rivers and lakes, and arranging a plurality of monitoring points based on the basic data; obtaining video frames of river and lake by utilizing video monitoring, and updating frame images continuously collected in a designated time period according to a preset time interval to a blue algae image set;
step two, creating a blue algae identification model, inputting every newly added frame image in the blue algae image set into the blue algae identification model to obtain a uniform image, judging whether blue algae exists currently or not based on the uniform image, acquiring real-time blue algae information if the blue algae exists, and triggering a blue algae prevention and control mechanism: comparing and analyzing the current blue algae information with the historical blue algae information to obtain the control situation of the blue algae, updating and sending out real-time control measures based on the control situation;
and step three, identifying whether pollutants exist in the water surface area of the river and the lake in the video frame by utilizing the pollution identification model, and homogenizing the required video frame in the embodiment, wherein the processing mode can refer to the step two. If further analysis of the contaminant type is present and then triggers the environmental regulatory mechanism: and monitoring the position of the pollutant in real time, tracking, calling out a historical frame image corresponding to the current frame image, and searching a pollution source again by using a pollution identification model to obtain a generation record related to the pollutant.
The basic data in the first step are: with respect to the geographical location of the river and lake, and the characteristics of the river and lake. In other words, based on the geographical position of the river and the lake and the characteristics of the river and the lake, the required monitoring points are distributed, corresponding numbers are given to the monitoring devices at each monitoring point, and each monitoring point is provided with a camera with a preset angle, so that later management and targeted data extraction are facilitated. By way of example, three monitoring points are selected in a typical area of a flat river, so that verification of a throwing object behavior recognition model is realized; and selecting three positions to construct video points in a typical area of the external pond and river, selecting one high-point construction video point in a water source area of the fishing ocean mountain, and selecting three positions to construct video points in a Yang Cheng lake area, so that the blue algae identification model is verified. It should be noted that the following three principles are considered when selecting the monitoring point:
(1) Representative principle: the monitoring points should be representative; if the monitoring purpose is to evaluate the influence of human activities (artificial throwing), point positions are required to be set in areas with intense human activities such as human gathering areas, production living areas, tourist areas and the like; the point positions are required to be arranged in the areas without shielding and with wide vision.
(2) Feasibility principle: on the premise of ensuring the aim of monitoring and the necessary monitoring precision and sample size, the practical applicability of the arrangement of the monitoring sample points is considered so as to obtain the most effective data with the least point positions, manpower, material resources and time investment.
(3) Economic applicability: the arrangement of the monitoring points should fully consider the surrounding basic conditions including but not limited to complete power supply facilities, convenient maintenance and the like so as to obtain the most abundant data with the least economic investment.
The blue algae occurrence time and the blue algae occurrence frequency of each region are different, and mainly depend on the environment and water quality of the region. And if the video data of each area needs to be acquired in real time and the data acquired in real time is analyzed in real time, the number of videos is huge, the included scenes are various, the related video analysis technology is complex, dynamic allocation of computing resources can be achieved only by uniformly planning and constructing in a platform mode, and the utilization efficiency of the computing resources is improved.
In this embodiment, it is configured that continuous collection is performed at predetermined time intervals over a specified period of time, for example, blue algae identification only requires analysis at 5-10 months per year, and only requires 7:30 minutes of short-term analysis per morning. The specified time period is correspondingly 5-10 months per year, and the predetermined time interval is 24 hours. In another real-time, the corresponding time period is configured and the corresponding time interval is set in combination with the time period of the blue algae burst in the past year of the local area. And the video analysis tasks with more multiple paths are completed by using relatively fewer video analysis calculation resources, so that the investment waste of the calculation resources is effectively reduced. As shown in fig. 2.
It should be noted that, due to the particularity of the regions of the rivers and lakes, the collected frame images are affected by certain natural conditions, such as uneven intensity and more shadows caused by building influence, tree influence and the like, so that the collected frame images have larger light intensity differences. When the river and the lake are directly irradiated by the sun, the light is relatively strong, and when the river and the lake are blocked by buildings, trees and the like, the light is relatively weak.
Therefore, in order to improve the recognition accuracy, the uniform image in this embodiment is acquired as follows: converting a frame image into a gray image, defining a gray threshold
Figure SMS_57
And->
Figure SMS_58
The method comprises the steps of carrying out a first treatment on the surface of the Acquiring region gray value +.>
Figure SMS_59
Based on the region gray value +.>
Figure SMS_60
Gray threshold->
Figure SMS_61
And->
Figure SMS_62
Dividing a frame image into a dark region, a normal region and a bright region;
based on the gray values of the normal areas, a gray compensation formula for the dark areas and the bright areas is obtained:
Figure SMS_63
；
in the method, in the process of the invention,
Figure SMS_64
gray value after illumination compensation for dark areas, < >>
Figure SMS_67
Is the gray value of the normal region,
Figure SMS_69
gray value after illumination compensation for bright area, +.>
Figure SMS_70
，/>
Figure SMS_73
Is the standard deviation of gray values of the normal region, +.>
Figure SMS_74
Is the standard deviation of the gray values of the dark areas, < >>
Figure SMS_75
，/>
Figure SMS_65
Is the gray average value of the normal area, +.>
Figure SMS_66
For the gray average value of the dark areas, +.>
Figure SMS_68
，/>
Figure SMS_71
Standard deviation of gray value for bright area, +.>
Figure SMS_72
，/>
Figure SMS_76
Is the gray average value of the bright area;
and matching the areas corresponding to the dark areas and the bright areas of the gray level image in the frame image to obtain corresponding dark areas and bright areas, and respectively processing the dark areas and the bright areas by using the gray level compensation formula to obtain uniform images with uniform luminosity.
By the technical scheme, the illumination non-uniformity of the frame image is subjected to uniformization treatment, namely gray level compensation, so that the color frame image with uniform illumination is obtained.
Based on the above description, the specific steps for judging whether blue algae exists in the uniform image currently are as follows:
the color characteristic and texture characteristic of the uniform image are obtained, and the color characteristic value of the ith pixel is calculated by adopting the following formulas
Figure SMS_77
And texture feature value +.>
Figure SMS_78
：
Figure SMS_79
The method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure SMS_80
Mean feature vector in color feature space for uniform image, +.>
Figure SMS_81
Pixel characteristic representing pixel i, +.>
Figure SMS_82
Is a mould;
Figure SMS_83
the method comprises the steps of carrying out a first treatment on the surface of the Wherein p represents the number of dimensions of the extracted texture feature, each number corresponding to a direction,/->
Figure SMS_84
Representing the texture value of pixel i in dimension p;
presetting a blue algae color characteristic threshold interval and a blue algae texture characteristic threshold interval: if the color characteristic value
Figure SMS_85
Texture feature value->
Figure SMS_86
If the blue algae does not belong to the corresponding characteristic region at the same time, the blue algae does not appear in the current river and lake;
if the color characteristic value
Figure SMS_87
Belongs to blue algae color characteristic threshold value interval, and texture characteristic value +.>
Figure SMS_88
If the blue algae belongs to the blue algae texture feature threshold interval, the blue algae existing in the current river and lake is indicated, and the corresponding pixel i is listed as a blue algae area r; the position of the blue algae area is the position information of the blue algae. As shown in fig. 3 (a) and (b), the white area in fig. 3 (b) is the blue algae area after screening by the method, and is consistent with the blue algae position in fig. 3 (a).
The blue algae monitoring aims at determining the position information of the blue algae on the river and the lake and acquiring the occupied area of the blue algae. Thus, current cyanobacteria information includes: the position information of the blue algae and the occupied area of the blue algae.
In this embodiment, the blue algae color feature threshold interval and the blue algae texture feature threshold interval are generated in advance based on blue algae characteristics, and are used for judging whether the pixel i in the uniform image belongs to the interval, and if both belong to the interval, the substance type corresponding to the pixel i is indicated to be blue algae. Therefore, whether the blue algae on the river and the lake appear in a piece or in a block, the blue algae can be effectively identified and the position information can be obtained, so that the blue algae can be effectively controlled.
Based on the method, the accuracy of the blue algae occupation area can be improved, and the calculation flow aiming at the irregular, continuous or discontinuous blue algae occupation area is as follows:
based on the blue algae area r, calculating the blue algae area by the following formula:
Figure SMS_89
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure SMS_90
Representing the image area of blue algae region r in a uniform image,/->
Figure SMS_91
Pixel area representing uniform image, +.>
Figure SMS_92
Is the actual area corresponding to the pixel area, +.>
Figure SMS_93
For perspective transformation coefficients->
Figure SMS_94
Is the included angle between the camera and the horizontal plane. When the camera is vertical to the water surface of the river and the lake, the camera is +.>
Figure SMS_95
。
In other words the first and second phase of the process,
Figure SMS_96
the camera is vertical to the horizontal plane, the obtained pixel area is in direct proportion to the actual area,
Figure SMS_97
indicating that the camera is level or at an angle to the water surface as shown. At the moment, blue algae is in a state of visual angle transformation in an image, the visual angle is tiled into a vertical state by utilizing perspective transformation, and coefficients obtained by utilizing perspective transformation are +.>
Figure SMS_98
And performing inverse perspective transformation. By the calculation formula, the occupied area of all blue algae on the river and the lake can be accurately obtained, and the calculation accuracy and the comprehensiveness are improved.
In a further embodiment, after comparing and analyzing the current blue algae information and the historical blue algae information, if at least one of the following triggering conditions is met, the prevention and control conditions are poor, and the prevention and control are required to be enhanced;
the triggering conditions are as follows:
Figure SMS_99
、/>
Figure SMS_100
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure SMS_101
Representing the number of blue algae areas acquired by the current frame, < + >>
Figure SMS_102
Representing the number of blue algae areas acquired by historical frames, < >>
Figure SMS_103
Representing the area of blue algae acquired by the current frame, +.>
Figure SMS_104
Representing the area of blue algae acquired by the history frame.
In other words, if at least one of the amplification of the dispersed regions of the blue algae region and the increase of the blue algae area is satisfied, the blue algae control force before that is insufficient, and thus, it is necessary to enhance the control. In a further embodiment, the blue algae prevention and control means comprises: under the condition of meeting the safety conditions such as flood prevention and the like, the water body replacement flow of the key water area is increased, the hydrodynamic condition is improved, and the growth of blue algae is inhibited; the workload of blue algae interception, algae water salvage, algae water treatment and the like in key water areas is increased, and the standing and the disposal are realized; when necessary, the algae water is salvaged, and the blue algae capturing operation is organized.
The embodiment also discloses that based on the blue algae prevention and control measures, analysis briefs on rivers and lakes are generated, and information contents such as blue algae bloom early warning times, time distribution, position distribution, processing record and the like in a certain time period are counted. The system intelligently analyzes early warning, recording and other result information, and ensures that a user timely and rapidly overall grasps the river blue algae management information.
In order to better control blue algae, the pollutant types in the third step include: solid matter contamination and liquid matter contamination;
the identification steps of the pollutant type are as follows: presetting a water area color characteristic threshold interval and a water area texture characteristic threshold interval, and if the color characteristic value is
Figure SMS_105
Meanwhile, the blue algae color characteristic threshold value interval and the blue algae color characteristic threshold value interval do not belong to the water area color characteristic threshold value interval, and the texture characteristic value is +.>
Figure SMS_106
Meanwhile, if the pixel i does not belong to the water area texture feature threshold interval and the blue algae texture feature threshold interval, preliminarily judging the pixel i as a pollutant and endowing the pixel with a pixel tag j, and calculating the occupied area of the pixel tag j by adopting the following formula:
Figure SMS_107
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure SMS_108
Representing the image area of pixel label j in the uniform image;
when the blue algae is identified to exist and
Figure SMS_109
or +.>
Figure SMS_110
The method indicates that the lake water of the river and the lake is comprehensively polluted, and the pollution type which is primarily judged is liquid substance pollution; on the contrary, let(s)>
Figure SMS_111
Or->
Figure SMS_112
And then the solid pollution exists in the river and the lake. In other words, if it is determined that blue algae is currently present, the area of the water area should be +.>
Figure SMS_113
The area occupied by the pollutant pixel tag j obtained by analysis is approximately equal to the water area and is about +.>
Figure SMS_114
The current water quality is poor, and the current water quality does not belong to a given threshold value interval, so that serious water quality pollution is shown. Similarly, if no blue algae is determined at present, the water area is Y, and the occupation area of the pollutant pixel tag j obtained by analysis is approximately equal to the water area +.>
Figure SMS_115
The current water quality is poor, and the current water quality does not belong to a given threshold value interval, so that serious water quality pollution is shown.
Otherwise, if the analysis results in a small footprint for contaminant pixel label j, it is an indication that a fixed contaminant is currently present. The environmental management mechanism therefore includes: water quality detection, water quality treatment, garbage removal and pollution source searching; further comprises: and executing early warning, warning and punishment on the behavior main body.
Example 2
In order to implement the method for monitoring and managing a river and a lake identified by blue algae in embodiment 1, this embodiment discloses a system for monitoring and managing a river and a lake, which includes:
the first module is arranged to acquire basic data of rivers and lakes, and a plurality of monitoring points are arranged based on the basic data; obtaining video frames of river and lake by utilizing video monitoring, and updating frame images continuously collected in a designated time period according to a preset time interval to a blue algae image set;
the second module is set to create a blue algae identification model, input every newly added frame image in the blue algae image set into the blue algae identification model to obtain a uniform image, judge whether blue algae exists currently based on the uniform image, acquire real-time blue algae information if the blue algae exists, and trigger a blue algae prevention and control mechanism: comparing and analyzing the current blue algae information with the historical blue algae information to obtain the control situation of the blue algae, updating and sending out real-time control measures based on the control situation;
a third module configured to identify whether a contaminant is present in the river/lake water surface area in the video frame using the contaminant identification model, and if so, to further analyze the type of contaminant and trigger an environmental control mechanism: and monitoring the position of the pollutant in real time, tracking, calling out a historical frame image corresponding to the current frame image, and searching a pollution source again by using a pollution identification model to obtain a generation record related to the pollutant.

Claims (9)

1. A river and lake supervision method based on blue algae identification is characterized by comprising the following steps:
acquiring basic data of a river and a lake, and laying a plurality of monitoring points based on the basic data; obtaining video frames of river and lake by utilizing video monitoring, and updating frame images continuously collected in a designated time period according to a preset time interval to a blue algae image set;
creating a blue algae identification model, inputting each newly added frame image in the blue algae image set into the blue algae identification model to obtain a uniform image, judging whether blue algae exists currently or not based on the uniform image, acquiring real-time blue algae information if the blue algae exists, and triggering a blue algae prevention and control mechanism: comparing and analyzing the current blue algae information with the historical blue algae information to obtain the control situation of the blue algae, updating and sending out real-time control measures based on the control situation;
identifying whether pollutants exist in the water surface area of the river and the lake in the video frame by using the pollution identification model, and if so, analyzing the types of the pollutants and triggering an environment management and control mechanism: and monitoring the position of the pollutant in real time, tracing, calling out a historical frame image corresponding to the current frame image, and searching a pollution source again by using a pollution identification model to obtain a generation record related to the pollutant.
2. The method for supervising a river and lake based on blue algae identification according to claim 1, wherein the current blue algae information comprises: the position information of the blue algae and the occupied area of the blue algae.
3. The river and lake supervision method based on blue algae identification according to claim 2, wherein the specific step of judging whether blue algae exists currently based on the uniform image is as follows: the color characteristic and texture characteristic of the uniform image are obtained, and the color characteristic value of the ith pixel is calculated by adopting the following formulas
Figure QLYQS_1
And texture feature value +.>
Figure QLYQS_2
：
Figure QLYQS_3
The method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure QLYQS_4
Mean feature vector in color feature space for uniform image, +.>
Figure QLYQS_5
Pixel characteristic representing pixel i, +.>
Figure QLYQS_6
Is a mould;
Figure QLYQS_7
the method comprises the steps of carrying out a first treatment on the surface of the Wherein p represents the dimension number of the extracted texture feature, each number corresponding toOne direction (I)>
Figure QLYQS_8
Representing the texture value of pixel i in dimension p;
presetting a blue algae color characteristic threshold interval and a blue algae texture characteristic threshold interval: if the color characteristic value
Figure QLYQS_9
Texture feature value->
Figure QLYQS_10
At least one of the blue algae is not in the corresponding characteristic area, and the blue algae is not present in the current river and lake;
if the color characteristic value
Figure QLYQS_11
Belongs to blue algae color characteristic threshold value interval, and texture characteristic value +.>
Figure QLYQS_12
If the blue algae belongs to the blue algae texture feature threshold interval, the blue algae existing in the current river and lake is indicated, and the corresponding pixel i is listed as a blue algae area r;
the position of the blue algae area is the position information of the blue algae.
4. The river and lake supervision method based on blue algae identification according to claim 2, wherein the calculation flow of the land occupation area of the blue algae is as follows: based on the blue algae area r, calculating the blue algae area by the following formula:
Figure QLYQS_13
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure QLYQS_14
Representing the image area of blue algae region r in a uniform image,/->
Figure QLYQS_15
Pixel area representing uniform image, +.>
Figure QLYQS_16
Is the actual area corresponding to the pixel area, +.>
Figure QLYQS_17
For perspective transformation coefficients->
Figure QLYQS_18
Is the included angle between the camera and the horizontal plane.
5. The river and lake supervision method based on blue algae identification according to claim 1, wherein the uniform image is obtained by the following method:
converting a frame image into a gray image, defining a gray threshold
Figure QLYQS_19
And->
Figure QLYQS_20
The method comprises the steps of carrying out a first treatment on the surface of the Acquiring region gray value +.>
Figure QLYQS_21
Based on the region gray value +.>
Figure QLYQS_22
Gray threshold->
Figure QLYQS_23
And->
Figure QLYQS_24
Dividing a frame image into a dark region, a normal region and a bright region;
based on the gray values of the normal areas, a gray compensation formula for the dark areas and the bright areas is obtained:
Figure QLYQS_25
；
in the method, in the process of the invention,
Figure QLYQS_27
gray value after illumination compensation for dark areas, < >>
Figure QLYQS_29
Gray value of normal region, +.>
Figure QLYQS_31
Gray value after illumination compensation for bright area, +.>
Figure QLYQS_32
，/>
Figure QLYQS_35
Is the standard deviation of gray values of the normal region, +.>
Figure QLYQS_36
Is the standard deviation of the gray values of the dark areas, < >>
Figure QLYQS_38
，/>
Figure QLYQS_26
Is the gray average value of the normal area, +.>
Figure QLYQS_28
For the gray average value of the dark areas, +.>
Figure QLYQS_30
，/>
Figure QLYQS_33
Standard deviation of gray value for bright area, +.>
Figure QLYQS_34
，/>
Figure QLYQS_37
Is the gray average value of the bright area;
and matching the areas corresponding to the dark areas and the bright areas of the gray level image in the frame image to obtain corresponding dark areas and bright areas, and respectively processing the dark areas and the bright areas by using the gray level compensation formula to obtain uniform images with uniform luminosity.
6. The river and lake supervision method based on blue algae identification according to claim 2, wherein the flow of updating and sending out real-time prevention and control measures based on the prevention and control conditions is as follows:
after comparing and analyzing the current blue algae information and the historical blue algae information, if at least one of the following triggering conditions is met, the prevention and control conditions are poor, and the prevention and control are required to be enhanced;
the triggering conditions are as follows:
Figure QLYQS_39
、/>
Figure QLYQS_40
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure QLYQS_41
Representing the number of cyanobacteria areas acquired by the current frame,
Figure QLYQS_42
representing the number of blue algae areas acquired by historical frames, < >>
Figure QLYQS_43
Representing the area of blue algae acquired by the current frame, +.>
Figure QLYQS_44
Representing the area of blue algae acquired by the history frame.
7. A method of monitoring a river or lake based on identification of cyanobacteria according to claim 3 wherein the contaminant types include: solid matter contamination and liquid matter contamination;
the identification steps of the pollutant type are as follows: presetting a water area color characteristic threshold interval and a water area texture characteristic threshold interval, and if the color characteristic value is
Figure QLYQS_45
Meanwhile, the blue algae color characteristic threshold value interval and the blue algae color characteristic threshold value interval do not belong to the water area color characteristic threshold value interval, and the texture characteristic value is +.>
Figure QLYQS_46
Meanwhile, if the pixel i does not belong to the water area texture feature threshold interval and the blue algae texture feature threshold interval, preliminarily judging the pixel i as a pollutant and endowing the pixel with a pixel tag j, and calculating the occupied area of the pixel tag j by adopting the following formula:
Figure QLYQS_47
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->
Figure QLYQS_48
Representing the image area of pixel label j in a uniform image,/-, and>
Figure QLYQS_49
pixel area representing uniform image, +.>
Figure QLYQS_50
Is the actual area corresponding to the pixel area, +.>
Figure QLYQS_51
For perspective transformation coefficients->
Figure QLYQS_52
Is a clamp between the camera and the horizontal planeA corner;
when the blue algae is identified to exist and
Figure QLYQS_53
or +.>
Figure QLYQS_54
The method indicates that the lake water of the river and the lake is comprehensively polluted, and the pollution type which is primarily judged is liquid substance pollution; on the contrary, let(s)>
Figure QLYQS_55
Or->
Figure QLYQS_56
And then the solid pollution exists in the river and the lake.
8. The river and lake supervision method based on blue algae identification of claim 7, wherein,
the environment management and control mechanism comprises: water quality detection, water quality treatment, garbage removal and pollution source searching; further comprises: and executing early warning, warning and punishment on the behavior main body.
9. River and lake supervision system for implementing a river and lake supervision method based on blue algae identification according to any one of claims 1 to 8, comprising:
the first module is arranged to acquire basic data of rivers and lakes, and a plurality of monitoring points are arranged based on the basic data; obtaining video frames of river and lake by utilizing video monitoring, and updating frame images continuously collected in a designated time period according to a preset time interval to a blue algae image set;
the second module is set to create a blue algae identification model, input every newly added frame image in the blue algae image set into the blue algae identification model to obtain a uniform image, judge whether blue algae exists currently based on the uniform image, acquire real-time blue algae information if the blue algae exists, and trigger a blue algae prevention and control mechanism: comparing and analyzing the current blue algae information with the historical blue algae information to obtain the control situation of the blue algae, updating and sending out real-time control measures based on the control situation;
a third module configured to identify whether a contaminant is present in the river/lake water surface area in the video frame using the contaminant identification model, and if so, to further analyze the type of contaminant and trigger an environmental control mechanism: and monitoring the position of the pollutant in real time, tracking, calling out a historical frame image corresponding to the current frame image, and searching a pollution source again by using a pollution identification model to obtain a generation record related to the pollutant.
CN202310416654.1A 2023-04-19 2023-04-19 River and lake supervision method and system based on blue algae identification Active CN116152748B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310416654.1A CN116152748B (en) 2023-04-19 2023-04-19 River and lake supervision method and system based on blue algae identification

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310416654.1A CN116152748B (en) 2023-04-19 2023-04-19 River and lake supervision method and system based on blue algae identification

Publications (2)

Publication Number Publication Date
CN116152748A true CN116152748A (en) 2023-05-23
CN116152748B CN116152748B (en) 2023-07-07

Family

ID=86373968

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310416654.1A Active CN116152748B (en) 2023-04-19 2023-04-19 River and lake supervision method and system based on blue algae identification

Country Status (1)

Country Link
CN (1) CN116152748B (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116503767A (en) * 2023-06-02 2023-07-28 合肥众安睿博智能科技有限公司 River course floater recognition system based on semantic image processing
CN116704197A (en) * 2023-08-07 2023-09-05 水利部珠江水利委员会水文局 Processing method and system for river and lake remote sensing image
CN117291910A (en) * 2023-11-23 2023-12-26 广东广宇科技发展有限公司 River and lake pollution discharge motion capturing method based on unmanned aerial vehicle image acquisition
CN117911878A (en) * 2024-03-20 2024-04-19 江西三思检测技术有限公司 Pollutant early warning method and device for intelligent environment-friendly housekeeper
CN118097562A (en) * 2024-04-18 2024-05-28 广东海洋大学 Remote monitoring method for seaweed proliferation condition
CN118212970A (en) * 2024-05-16 2024-06-18 水利部交通运输部国家能源局南京水利科学研究院 Blue algae prediction method and system based on time sequence and spatial distribution
CN118470651A (en) * 2024-07-10 2024-08-09 水利部交通运输部国家能源局南京水利科学研究院 River and lake dynamic sensing intelligent rod, sensing system and sensing method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102221551A (en) * 2011-06-02 2011-10-19 中国科学院计算技术研究所 Blue algae monitoring device and blue algae monitoring method
CN104616303A (en) * 2015-02-11 2015-05-13 西安电子科技大学 Ultraviolet image based water surface oil spill detection system and method
CN110108704A (en) * 2019-05-10 2019-08-09 合肥学院 A kind of automatic monitoring and pre-alarming method of cyanobacteria and its automatic monitoring and alarming system
CN115619998A (en) * 2022-11-30 2023-01-17 浙江绿维环境股份有限公司 Image-based sewage grade identification method and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102221551A (en) * 2011-06-02 2011-10-19 中国科学院计算技术研究所 Blue algae monitoring device and blue algae monitoring method
CN104616303A (en) * 2015-02-11 2015-05-13 西安电子科技大学 Ultraviolet image based water surface oil spill detection system and method
CN110108704A (en) * 2019-05-10 2019-08-09 合肥学院 A kind of automatic monitoring and pre-alarming method of cyanobacteria and its automatic monitoring and alarming system
CN115619998A (en) * 2022-11-30 2023-01-17 浙江绿维环境股份有限公司 Image-based sewage grade identification method and device

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
朱文涵 等: "苏州市河湖蓝藻智能识别监管平台建设设计", 《江苏水利》 *
罗艾娜 等: "基于深度神经网络的复杂光照下的蓝藻图片语义分割", 《计算机应用与软件》, vol. 35, no. 4 *

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116503767A (en) * 2023-06-02 2023-07-28 合肥众安睿博智能科技有限公司 River course floater recognition system based on semantic image processing
CN116503767B (en) * 2023-06-02 2023-09-22 合肥众安睿博智能科技有限公司 River course floater recognition system based on semantic image processing
CN116704197A (en) * 2023-08-07 2023-09-05 水利部珠江水利委员会水文局 Processing method and system for river and lake remote sensing image
CN116704197B (en) * 2023-08-07 2023-10-17 水利部珠江水利委员会水文局 Processing method and system for river and lake remote sensing image
CN117291910A (en) * 2023-11-23 2023-12-26 广东广宇科技发展有限公司 River and lake pollution discharge motion capturing method based on unmanned aerial vehicle image acquisition
CN117291910B (en) * 2023-11-23 2024-04-09 广东广宇科技发展有限公司 River and lake pollution discharge motion capturing method based on unmanned aerial vehicle image acquisition
CN117911878A (en) * 2024-03-20 2024-04-19 江西三思检测技术有限公司 Pollutant early warning method and device for intelligent environment-friendly housekeeper
CN117911878B (en) * 2024-03-20 2024-06-07 江西三思检测技术有限公司 Pollutant early warning method and device for intelligent environment-friendly housekeeper
CN118097562A (en) * 2024-04-18 2024-05-28 广东海洋大学 Remote monitoring method for seaweed proliferation condition
CN118097562B (en) * 2024-04-18 2024-07-05 广东海洋大学 Remote monitoring method for seaweed proliferation condition
CN118212970A (en) * 2024-05-16 2024-06-18 水利部交通运输部国家能源局南京水利科学研究院 Blue algae prediction method and system based on time sequence and spatial distribution
CN118470651A (en) * 2024-07-10 2024-08-09 水利部交通运输部国家能源局南京水利科学研究院 River and lake dynamic sensing intelligent rod, sensing system and sensing method
CN118470651B (en) * 2024-07-10 2024-10-11 水利部交通运输部国家能源局南京水利科学研究院 River and lake dynamic sensing intelligent rod, sensing system and sensing method

Also Published As

Publication number Publication date
CN116152748B (en) 2023-07-07

Similar Documents

Publication Publication Date Title
CN116152748B (en) River and lake supervision method and system based on blue algae identification
CN112149543B (en) Building dust recognition system and method based on computer vision
CN113155102A (en) River comprehensive treatment system
CN110334750A (en) Iron tower of power transmission line bolt corrosion degree image classification recognition methods
CN109255350A (en) A kind of new energy detection method of license plate based on video monitoring
CN110659546B (en) Illegal booth detection method and device
CN111814750A (en) Intelligent garbage classification method and system based on deep learning target detection and image recognition
CN110956183A (en) Asphalt pavement crack morphology extraction method
CN113989711B (en) Power distribution construction safety tool use identification method and system
CN116484259A (en) Urban pipe network defect position positioning analysis method and system
CN109521021A (en) A kind of nuclear power plant containment shell appearance inspecting system and method
CN107578414B (en) Method for processing pavement crack image
CN114509445B (en) Cluster analysis-based pollutant identification early warning method and system
CN103971367A (en) Hydrologic data image segmenting method
CN115439763A (en) Water engineering surface crack detection method based on YOLOv5 model
CN117151430B (en) Small watershed soil and water conservation treatment priority remote sensing evaluation method
CN117456353A (en) Management system of river and lake intelligent boundary pile
CN116611704A (en) System and method for monitoring fishery resources
CN114048953B (en) Intelligent wind control evaluation method based on multidimensional sensing and enterprise data quantification
CN104820818A (en) Fast recognition method for moving object
CN112818836B (en) Method and system for detecting personnel target of transformer substation scene
CN114241299A (en) Method and system for detecting garbage density of block chain water area based on genetic algorithm
CN114898291B (en) Visual monitoring method based on camera inspection path
CN105512608A (en) Method and system for determining man-made geomorphic feature change
Yasuno et al. River Surface Patch-wise Detector Using Mixture Augmentation for Scum-cover-index

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