CN111416964B - Remote image intelligent identification method for hydraulic engineering deformation - Google Patents
Remote image intelligent identification method for hydraulic engineering deformation Download PDFInfo
- Publication number
- CN111416964B CN111416964B CN202010385449.XA CN202010385449A CN111416964B CN 111416964 B CN111416964 B CN 111416964B CN 202010385449 A CN202010385449 A CN 202010385449A CN 111416964 B CN111416964 B CN 111416964B
- Authority
- CN
- China
- Prior art keywords
- hydraulic engineering
- dangerous case
- image
- camera
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- 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
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Alarm Systems (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a remote image intelligent identification method for hydraulic engineering deformation, S1, data acquisition; s2, the data processing terminal equipment receives the video images transmitted by the camera in real time, automatically judges whether a hydraulic engineering dangerous case occurs in the current monitored area after the video images are processed by an image intelligent recognition algorithm, marks the dangerous case area by colors, estimates the area of the dangerous case and stores screenshots before and after the dangerous case occurs; and S3, the dangerous case comprehensive management platform of the cloud server is used for real-time monitoring, field historical video playback, field dangerous case information query and daily safety inspection and management of the monitored hydraulic engineering field, and automatically generates a field dangerous case report according to the related image and data information. Compared with the traditional manual inspection mode, the hydraulic engineering safety maintenance and inspection work realizes the remote and visual operation, reduces the investment of human resources, reduces the careless omission in the inspection of dangerous cases, and can effectively early warn the dangerous cases on site in real time.
Description
Technical Field
The invention relates to a hydraulic engineering deformation identification method, in particular to a remote image intelligent identification method for hydraulic engineering deformation.
Background
At present, the evaluation of the dangerous case stability of hydraulic engineering such as river protection engineering is mainly realized by root stone detection in non-flood season and artificial inspection in flood season, and no effective engineering safety monitoring measures exist. The national flood prevention third-stage planning and the intelligent water conservancy project construction carried out by the water conservancy department put forward higher requirements on river protection project management. However, in actual work, due to the fact that the river protection engineering bank line is long, the field conditions are variable, investment in inspection labor cost is large, in addition, manual inspection is time-consuming and labor-consuming, and careless omission easily occurs, and serious lag and hidden danger exist in river bank protection early warning work.
Disclosure of Invention
The invention aims to provide a remote image intelligent identification method for hydraulic engineering deformation.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention discloses a remote image intelligent identification method for hydraulic engineering deformation, which comprises the following steps:
s1, data acquisition: acquiring a video image of a monitored hydraulic engineering site in real time through a camera and transmitting the video image to data processing terminal equipment;
s2, the data processing terminal equipment receives the video images transmitted by the camera in real time, automatically judges whether a hydraulic engineering dangerous case occurs in the current monitored area after the video images are processed by an image intelligent recognition algorithm, marks the dangerous case area by colors, estimates the area of the dangerous case and stores screenshots before and after the dangerous case occurs, and the image intelligent recognition algorithm is as follows:
s21, denoising the video stream by using median filtering, wherein the filter type is a rectangular filter, and the filter size = 7;
s22, using KNN background modeling, comparing a new pixel value at a certain position of the image with historical information (including pixel values of previous frames and judgment of whether pixel points are foreground or background) of the pixel value, and if the difference between the pixel values is within a set threshold value, considering that the new pixel value is matched with the historical information and classified as background, otherwise, classifying as foreground, and obtaining a foreground image; in the method, the modeling historical duration =7, and the clustering threshold = 20;
s23, denoising the foreground image by using a graphical closed operation method, wherein an operation kernel = 3 x 3 rectangle, and then extracting a connected region from the denoised foreground image;
s24, calculating image characteristics in each communication area, wherein in the method, HS histogram characteristics and HOG characteristics are selected, and the calculation parameters of the HS histogram characteristics are set as bin = 180; the HOG characteristic calculation parameters are set as follows: image size =64 × 128, sliding window size =64 × 128, Block size =16 × 16, Block _ Stride = (8,8), Cell size =8 × 8, histogram bin = 9;
s25, judging each connected region by using an SVM (support vector machine) model, removing the region with the judgment result of environmental interference, reserving other connected regions, calculating the area of the connected region, and uploading the area to a cloud server in real time;
s3, the comprehensive dangerous case management platform of the cloud server is used for real-time monitoring, field historical video playback, field dangerous case information query and daily safety inspection and management of the monitored hydraulic engineering field; and the basic-level patrol personnel manually recheck the dangerous case information through the comprehensive dangerous case management platform and automatically generate a field dangerous case report by the related image and data information.
The cameras are two digital cameras, one is a gunlock camera and is used for algorithm observation, pixels are more than or equal to 400 ten thousand, infrared night vision of more than 100 meters is supported, and zooming is 8 times; and the other camera is a ball camera of the ball machine and is used for daily inspection of the monitored hydraulic engineering site.
As the monitored hydraulic engineering site is generally located in a remote area and cannot ensure the power supply of a site power grid and a wired network, the camera and the data processing terminal equipment are powered by a solar storage battery arranged on the monitored hydraulic engineering site, the storage battery is a colloid storage battery, the capacity is 800AH, and the equipment can normally work for 3 days in continuous rainy days; the power of the solar panel is 700w, and the storage battery is fully filled in 4-6 hours under the condition of good illumination; and the communication between the camera and the data processing terminal equipment and the communication between the data processing terminal equipment and the cloud server are realized by adopting a wireless 4G network.
The advantages of the invention are embodied in the following aspects:
1. compared with the traditional manual inspection mode, the system realizes remote and visual hydraulic engineering safety maintenance and inspection work, reduces the investment of human resources, reduces careless omission in inspection of dangerous cases, and can perform real-time effective early warning on the dangerous cases on site.
2. The system rate is earlier applied to hydraulic engineering deformation detection area with artificial intelligence and machine learning technique, has realized hydraulic engineering safety monitoring's intellectuality, accords with the development theory of water conservancy department "wisdom water conservancy".
3. The data processing part is realized in an edge computing mode, data are computed locally, network delay is reduced, and communication cost is saved.
4. The system does not depend on wired power supply network, the whole set of system is simple and convenient to install, is suitable for being installed and deployed in various outdoor and remote hydraulic engineering areas, and has better economical efficiency and practicability.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The following describes embodiments of the present invention in detail with reference to the drawings, which are implemented on the premise of the technical solution of the present invention, and detailed embodiments and specific operation procedures are provided, but the scope of the present invention is not limited to the following embodiments.
As shown in fig. 1, the remote image intelligent identification method for hydraulic engineering deformation of the present invention is characterized in that: the method comprises the following steps:
s1, data acquisition: acquiring a video image of a monitored hydraulic engineering site in real time through a camera and transmitting the video image to data processing terminal equipment; the cameras are two digital cameras, one is a gunlock camera and is used for algorithm observation, pixels are more than or equal to 400 ten thousand, infrared night vision of more than 100 meters is supported, and zooming is 8 times; the other camera is a ball camera of the ball machine and is used for daily inspection of the monitored hydraulic engineering site;
s2, the data processing terminal equipment receives the video images transmitted by the camera in real time, automatically judges whether a hydraulic engineering dangerous case occurs in the current monitored area after the video images are processed by an image intelligent recognition algorithm, marks the dangerous case area by colors, estimates the area of the dangerous case and stores screenshots before and after the dangerous case occurs, and the image intelligent recognition algorithm is as follows:
s21, denoising the video stream by using median filtering, wherein the filter type is a rectangular filter, and the filter size = 7;
s22, using KNN background modeling, the basic idea is as follows: comparing a new pixel value at a certain position of the image with historical information (including the pixel values of previous frames and the judgment of whether pixel points are foreground or background) of the pixel value, and if the difference between the pixel values is within a set threshold value, considering that the new pixel value is matched with the historical information and classified as background, otherwise, classifying as foreground, and obtaining a foreground image; in the method, the modeling historical duration =7, and the clustering threshold = 20;
s23, denoising the foreground image by using a graphical closed operation method, wherein an operation kernel = 3 x 3 rectangle, and then extracting a connected region from the denoised foreground image;
s24, calculating image characteristics in each communication area, wherein in the method, HS histogram characteristics and HOG characteristics are selected, and the calculation parameters of the HS histogram characteristics are set as bin = 180; the HOG characteristic calculation parameters are set as follows: image size =64 × 128, sliding window size =64 × 128, Block size =16 × 16, Block _ Stride = (8,8), Cell size =8 × 8, histogram bin = 9;
s25, judging each connected region by using an SVM (support vector machine) model, removing the region with the judgment result of environmental interference, reserving other connected regions, calculating the area of the connected region, and uploading the area to a cloud server in real time;
s3, the comprehensive dangerous case management platform of the cloud server is used for real-time monitoring, field historical video playback, field dangerous case information query and daily safety inspection and management of the monitored hydraulic engineering field; and the basic-level patrol personnel manually recheck the dangerous case information through the comprehensive dangerous case management platform and automatically generate a field dangerous case report by the related image and data information.
As the monitored hydraulic engineering site is generally located in a remote area and cannot ensure the power supply of a site power grid and a wired network, the camera and the data processing terminal equipment are powered by a solar storage battery arranged on the monitored hydraulic engineering site, the storage battery is a colloid storage battery, the capacity is 800AH, and the equipment can normally work for 3 days in continuous rainy days; the power of the solar panel is 700w, and the storage battery is fully filled in 4-6 hours under the condition of good illumination; and the communication between the camera and the data processing terminal equipment and the communication between the data processing terminal equipment and the cloud server are realized by adopting a wireless 4G network.
Claims (2)
1. A remote image intelligent identification method for hydraulic engineering deformation is characterized in that: the method comprises the following steps:
s1, data acquisition: acquiring a video image of a monitored hydraulic engineering site in real time through a camera and transmitting the video image to data processing terminal equipment; the cameras are two digital cameras, one is a gunlock camera and is used for algorithm observation, pixels are more than or equal to 400 ten thousand, infrared night vision of more than 100 meters is supported, and zooming is 8 times; the other camera is a ball camera of the ball machine and is used for daily inspection of the monitored hydraulic engineering site;
s2, the data processing terminal equipment receives the video images transmitted by the camera in real time, automatically judges whether a hydraulic engineering dangerous case occurs in the current monitored area after the video images are processed by an image intelligent recognition algorithm, marks the dangerous case area by colors, estimates the area of the dangerous case and stores screenshots before and after the dangerous case occurs, and the image intelligent recognition algorithm is as follows:
s21, denoising the video stream by using median filtering, wherein the filter type is a rectangular filter, and the filter size = 7;
s22, using KNN background modeling, comparing a new pixel value at a certain position of the image with historical information of the pixel value, if the difference between the pixel values is within a set threshold value, considering that the new pixel value is matched with the historical information and classified as background, otherwise, classifying as foreground, and obtaining a foreground image; in the method, the modeling historical duration =7, and the clustering threshold = 20; the historical information comprises pixel values of the previous frames and judgment of whether pixel points are foreground or background;
s23, denoising the foreground image by using a graphical closed operation method, wherein an operation kernel = 3 x 3 rectangle, and then extracting a connected region from the denoised foreground image;
s24, calculating image characteristics in each communication area, wherein in the method, HS histogram characteristics and HOG characteristics are selected, and the calculation parameters of the HS histogram characteristics are set as bin = 180; the HOG characteristic calculation parameters are set as follows: image size =64 × 128, sliding window size =64 × 128, Block size =16 × 16, Block _ Stride = (8,8), Cell size =8 × 8, histogram bin = 9;
s25, judging each connected region by using an SVM model, removing the region with the judgment result of environmental interference, reserving other connected regions, calculating the area of the connected region, and uploading the area to a cloud server in real time;
s3, the comprehensive dangerous case management platform of the cloud server is used for real-time monitoring, field historical video playback, field dangerous case information query and daily safety inspection and management of the monitored hydraulic engineering field; and the basic-level patrol personnel manually recheck the dangerous case information through the comprehensive dangerous case management platform and automatically generate a field dangerous case report by the related image and data information.
2. The remote image intelligent identification method for hydraulic engineering deformation according to claim 1, characterized in that: the camera and the data processing terminal equipment are powered by a solar storage battery installed on a monitored hydraulic engineering site, the storage battery is a colloid storage battery, the capacity is 800AH, the power of a solar panel is 700w, and the communication between the camera and the data processing terminal equipment, the communication between the data processing terminal equipment and the cloud server is realized by adopting a wireless 4G network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010385449.XA CN111416964B (en) | 2020-05-09 | 2020-05-09 | Remote image intelligent identification method for hydraulic engineering deformation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010385449.XA CN111416964B (en) | 2020-05-09 | 2020-05-09 | Remote image intelligent identification method for hydraulic engineering deformation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111416964A CN111416964A (en) | 2020-07-14 |
| CN111416964B true CN111416964B (en) | 2021-08-24 |
Family
ID=71495015
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010385449.XA Active CN111416964B (en) | 2020-05-09 | 2020-05-09 | Remote image intelligent identification method for hydraulic engineering deformation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111416964B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112733798A (en) * | 2021-01-22 | 2021-04-30 | 上海高诚智能科技有限公司 | Flood control emergency management system and method for dike facilities |
| CN117496678B (en) * | 2024-01-02 | 2024-06-21 | 广州市声讯电子科技股份有限公司 | Emergency broadcast image-text alarm method, alarm system and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005292685A (en) * | 2004-04-05 | 2005-10-20 | Seiko Epson Corp | Liquid crystal device, projector, and electronic equipment |
| CN102006455A (en) * | 2009-09-03 | 2011-04-06 | 复旦大学 | Intelligent watercourse monitoring system based on video |
| CN104165668A (en) * | 2014-08-18 | 2014-11-26 | 深圳市亿维锐创科技有限公司 | Water conservancy monitoring method for automatic pre-warning of water conservancy risks and system thereof |
| CN104751603A (en) * | 2015-04-10 | 2015-07-01 | 招商局重庆交通科研设计院有限公司 | Rockfall monitoring and early warning system and rockfall monitoring and early warning method |
| CN105447463A (en) * | 2015-11-23 | 2016-03-30 | 江苏方天电力技术有限公司 | Camera-crossing automatic tracking system for transformer station based on human body feature recognition |
| CN107145851A (en) * | 2017-04-28 | 2017-09-08 | 西南科技大学 | Constructions work area dangerous matter sources intelligent identifying system |
| CN107958309A (en) * | 2017-12-07 | 2018-04-24 | 河海大学 | A kind of high dam multi-point deformetion condition combines forecasting procedure |
-
2020
- 2020-05-09 CN CN202010385449.XA patent/CN111416964B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005292685A (en) * | 2004-04-05 | 2005-10-20 | Seiko Epson Corp | Liquid crystal device, projector, and electronic equipment |
| CN102006455A (en) * | 2009-09-03 | 2011-04-06 | 复旦大学 | Intelligent watercourse monitoring system based on video |
| CN104165668A (en) * | 2014-08-18 | 2014-11-26 | 深圳市亿维锐创科技有限公司 | Water conservancy monitoring method for automatic pre-warning of water conservancy risks and system thereof |
| CN104751603A (en) * | 2015-04-10 | 2015-07-01 | 招商局重庆交通科研设计院有限公司 | Rockfall monitoring and early warning system and rockfall monitoring and early warning method |
| CN105447463A (en) * | 2015-11-23 | 2016-03-30 | 江苏方天电力技术有限公司 | Camera-crossing automatic tracking system for transformer station based on human body feature recognition |
| CN107145851A (en) * | 2017-04-28 | 2017-09-08 | 西南科技大学 | Constructions work area dangerous matter sources intelligent identifying system |
| CN107958309A (en) * | 2017-12-07 | 2018-04-24 | 河海大学 | A kind of high dam multi-point deformetion condition combines forecasting procedure |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111416964A (en) | 2020-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108109385B (en) | System and method for identifying and judging dangerous behaviors of power transmission line anti-external damage vehicle | |
| CN111047818A (en) | Forest fire early warning system based on video image | |
| CN100543765C (en) | Method for monitoring instruction based on computer vision | |
| CN207820108U (en) | A kind of transmission line long-distance video monitoring system | |
| CN102013147B (en) | High voltage power transmission tower intelligent anti-theft method for supervising and device | |
| CN102638013B (en) | The target image identification power transmission state monitoring system of view-based access control model attention mechanism | |
| CN111416964B (en) | Remote image intelligent identification method for hydraulic engineering deformation | |
| CN104394361A (en) | Pedestrian crossing intelligent monitoring device and detection method | |
| CN102446390A (en) | Method and system for safety detection and early warning of monitoring areas near power transmission lines | |
| CN104966372A (en) | Multi-data fusion forest fire intelligent recognition system and method | |
| CN105611244A (en) | Method for detecting airport foreign object debris based on monitoring video of dome camera | |
| CN112327698B (en) | Flood disaster early warning system and method based on Internet of things | |
| CN106651855A (en) | Image monitoring and shooting method for automatic identification and alarming of hidden troubles of power transmission line channel | |
| CN103400117B (en) | A kind of personnel in the pit's positioning and tracing method based on computer vision | |
| CN109872483A (en) | A kind of invasion warning photoelectric monitoring system and method | |
| CN103945197A (en) | Electric power facility external damage prevention warming scheme based on video motion detecting technology | |
| CN112541439A (en) | Intelligent video monitoring method and system for power plant | |
| CN106934796A (en) | High-speed belt conveyor rock slag video analytic system and method that rock tunnel(ling) machine is carried | |
| CN110691224A (en) | Transformer substation perimeter video intelligent detection system | |
| CN107977531B (en) | A kind of ground resistance flexible measurement method based on image procossing and mathematical model | |
| CN115171054A (en) | Fishing administration patrol method and system | |
| CN109345787A (en) | A kind of anti-outer damage monitoring and alarming system of the transmission line of electricity based on intelligent image identification technology | |
| CN116597394A (en) | Railway foreign matter intrusion detection system and method based on deep learning | |
| Zhen et al. | Transmission tower protection system based on Internet of Things in smart grid | |
| CN117935486A (en) | Flood early warning method and system for field river in arid region |
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 |