CN117974962A - Depth learning-based riverway water level detection method without water rule - Google Patents
Depth learning-based riverway water level detection method without water rule Download PDFInfo
- Publication number
- CN117974962A CN117974962A CN202311379848.5A CN202311379848A CN117974962A CN 117974962 A CN117974962 A CN 117974962A CN 202311379848 A CN202311379848 A CN 202311379848A CN 117974962 A CN117974962 A CN 117974962A
- Authority
- CN
- China
- Prior art keywords
- water level
- water
- image
- coordinate system
- alarm device
- 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.)
- Pending
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 232
- 238000001514 detection method Methods 0.000 title claims abstract description 56
- 238000013507 mapping Methods 0.000 claims abstract description 20
- 238000013135 deep learning Methods 0.000 claims abstract description 7
- 238000000034 method Methods 0.000 claims description 19
- 230000035772 mutation Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 5
- 230000000877 morphologic effect Effects 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 241000196324 Embryophyta Species 0.000 abstract description 6
- 239000002699 waste material Substances 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000009434 installation Methods 0.000 abstract description 4
- 238000012423 maintenance Methods 0.000 abstract description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 2
- 239000012528 membrane Substances 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
- G06V10/225—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition based on a marking or identifier characterising the area
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- 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
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Measurement Of Levels Of Liquids Or Fluent Solid Materials (AREA)
Abstract
The invention relates to the technical field of machine vision water level detection, in particular to a water level detection method of a riverway without a water rule based on deep learning, which comprises the following steps: step one: setting a calibration object at a water level detection point, acquiring an image containing the calibration object from a camera, and obtaining a mapping relation between a pixel coordinate system and a world coordinate system through calibration. The invention solves the problems that the water gauge is stranded in the prior common technology, resources are wasted due to the arrangement of water gauges with different levels, weeds can grow around the field water gauge to shield the water gauge, so that the water gauge reading is inaccurate, the water gauge can continuously detect the water level, the waste of resources can be avoided, the maintenance cost of personnel and the cost of installing the water gauge are reduced, the position of detecting the water level is not only limited to a vertical lock chamber, a bridge pier, a pond and the like, but also can be used for a dam, a reservoir, a river bank line and the like with an inclined angle, and the invention has the advantages of simple and economical production, convenient installation and convenient use.
Description
Technical Field
The invention relates to the technical field of machine vision water level detection, in particular to a water level detection method of a riverway without a water rule based on deep learning.
Background
In order to meet the requirements of water level detection in different river channels, the problems of complex equipment installation, high maintenance cost, potential safety hazards and the like in a manual water level reading method and a water level gauge detection method are avoided, and with the continuous development of machine vision, the water level detection can be realized by utilizing a method of combining a water gauge with video monitoring.
The current common technology is that a target detection algorithm is used for positioning the position of a water gauge, an image segmentation algorithm is used for segmenting the water gauge and a water body, the water gauge is read to obtain the water level height, the method is good for a simple scene, however, the phenomenon that the water gauge is stranded at a place with a large water level fall occurs, and resource waste is caused by setting water gauges with different levels; moreover, weeds can grow around the field water gauge to shade the water gauge, so that the problem of inaccurate water gauge reading is caused; the manual weed removal workload is large, and the safety coefficient is low; the water gauge is loaded down with trivial details in the laying process.
Disclosure of Invention
The invention aims to provide a depth learning-based riverway water level detection method, which has the advantages of no need of high detection precision, convenience in installation and convenience in use, and solves the problems that the current common technology can cause the phenomenon of water gauge stranding, the water gauges at different levels can cause resource waste, weeds can grow around the field water gauge to shield the water gauge, and the water gauge reading is inaccurate.
In order to achieve the above purpose, the present invention provides the following technical solutions: a depth learning-based riverway water level detection method without a water rule comprises the following steps:
Step one: setting a calibration object at a water level detection point, acquiring an image containing the calibration object from a camera, and obtaining a mapping relation between a pixel coordinate system and a world coordinate system through calibration;
step two: dividing a water body area and a non-water body area of a target water area by utilizing the SAM model, and acquiring a binary image;
Step three: cutting out a neighborhood of a water level calibration position on the binary image, and obtaining a water level dividing line on the image through image processing;
step four: binarizing a single frame image obtained by pulling, obtaining an interested region near the water level, dividing the interested region into a plurality of sub interested regions, and calculating Laplace values;
step five: determining a region with a mutation of Laplace value, which is designed as a water line region, by comparing the region of interest of the sub;
Step six: determining the position of the water level line by comparing the third step with the fifth step, and calculating the water level height by the mapping relation;
step seven: setting alarm equipment at the early warning water level;
Step eight: and connecting the output signal of the alarm device with a remote terminal.
Preferably, in the first step, the vertex of the calibration object in the world coordinate system corresponds to the pixel coordinate of the vertex of the calibration object in the detection image, the bottom point in the world coordinate system corresponds to the bottom point in the pixel coordinate system, and meanwhile, the included angle between the calibration object and the normal direction of the water surface and the initial height of the water surface need to be measured, and the mapping relation between the pixel coordinate and the world coordinate system can be obtained through nonlinear solution.
Preferably, in the second step, the SAM segmentation model is used to segment the image, the precise segmentation of the water body and the non-water body is realized by providing the prompt point and the detection frame, the binary mask image is output, and the aperture generated when the image is segmented is removed by morphological closing operation.
Preferably, in the third step, the coordinate position of the water level detection point is calculated, the obtained frame of image is denoised, then binarized, the interested area including the water body is cut out according to the coordinate obtained in the calibration, the interested area is divided into a plurality of small areas with equal area, the Laplace value of each small area is calculated, and finally the area with abrupt change of the Laplace value is obtained.
Preferably, in the seventh step, a plurality of water level alarm devices are respectively arranged at different positions of the river channel, the distance between two adjacent water level alarm devices is 10-20m, a standby alarm device is arranged beside each water level alarm device, and alarm transmission signals of the two alarm devices are different.
Preferably, in the step eight, the alarm device is connected with the remote terminal through the wireless signal transceiver module, the remote terminal is arranged inside the working room, the alarm device is powered by the solar panel, and the solar panel is arranged above the water level alarm device and connected through the waterproof wire.
Preferably, in the sixth step, it is checked whether the coordinates of the water level detection point are in the abrupt region, if not, it indicates that the water level detection point may belong to the ghost region, and if the detection is wrong, the height of the water level is calculated through the mapping relationship between the pixel coordinates and the world coordinate system.
Preferably, the water level alarm device and the standby alarm device are both water level alarms, and the specific model is ZSB127-Z and BZ1201 respectively.
Compared with the prior art, the invention has the following beneficial effects:
The invention has the following advantages:
1. the problem of stranding of the water gauge can be solved, and the waste of resources can be avoided while the water level can be continuously detected;
2. The problem that weeds grow to shade the water gauge is solved, and the maintenance cost of personnel and the cost of installing the water gauge are reduced;
3. Because the included angle between the water gauge and the normal line of the water surface needs to be measured during calibration, the position for detecting the water level is not only limited to a vertical lock chamber, a bridge pier, a pond and the like, but also can be used for dams, reservoirs, river shorelines and the like with inclined angles;
4. the method only needs to calibrate the camera, and has the advantages of simple and economical production, convenient installation and convenient use;
5. utilize long-range water level detection warning, can in time send out the alarm after the water level reaches appointed height to long-range staff of reminding, conveniently handle, need not artifical land and survey, saved the manpower.
Drawings
FIG. 1 is a flow chart of the actual operation of the present invention;
FIG. 2 is a schematic representation of the region of the Laplace value mutation of the present invention;
fig. 3, 4 and 5 are schematic diagrams of different water level calibration according to the present invention.
Detailed Description
Referring to fig. 1-5, a method for detecting a riverway water level without a water rule based on deep learning includes the following steps:
Step one: setting a calibration object at a water level detection point, acquiring an image containing the calibration object from a camera, and obtaining a mapping relation between a pixel coordinate system and a world coordinate system through calibration;
step two: dividing a water body area and a non-water body area of a target water area by utilizing the SAM model, and acquiring a binary image;
Step three: cutting out a neighborhood of a water level calibration position on the binary image, and obtaining a water level dividing line on the image through image processing;
step four: binarizing a single frame image obtained by pulling, obtaining an interested region near the water level, dividing the interested region into a plurality of sub interested regions, and calculating Laplace values;
step five: determining a region with a mutation of Laplace value, which is designed as a water line region, by comparing the region of interest of the sub;
Step six: determining the position of the water level line by comparing the third step with the fifth step, and calculating the water level height by the mapping relation;
step seven: setting alarm equipment at the early warning water level;
Step eight: and connecting the output signal of the alarm device with a remote terminal.
In the first step, the vertex of the calibration object in the world coordinate system corresponds to the pixel coordinate of the vertex of the calibration object in the detection image, the bottom point in the world coordinate system corresponds to the bottom point in the pixel coordinate system, meanwhile, the included angle between the calibration object and the normal direction of the water surface and the initial height of the water surface need to be measured, and the mapping relation between the pixel coordinate and the world coordinate system can be obtained through nonlinear solution;
Dividing the image by using a SAM (sample membrane) dividing model, realizing accurate division of the water body and the non-water body by providing a prompt point and a detection frame, outputting a binary mask image, and removing pinholes generated during dividing the image by using morphological closing operation;
Thirdly, calculating the coordinate position of a water level detection point, denoising a frame of the extracted image, performing binarization treatment, cutting out an interested region containing water body according to the coordinate obtained during calibration, dividing the interested region into a plurality of small regions with equal areas, calculating the Laplace value of each small region, and finally obtaining a region with abrupt change of the Laplace value;
In the seventh step, a plurality of water level alarm devices are respectively arranged at different positions of a river channel, the distance between two adjacent water level alarm devices is 10-20m, a standby alarm device is arranged beside each water level alarm device, and alarm transmission signals of the two alarm devices are different;
in the eighth step, the alarm equipment is connected with a remote terminal through a wireless signal receiving and transmitting module, the remote terminal is arranged in a working room, the alarm equipment is powered by a solar panel, the solar panel is arranged above the water level alarm equipment, and the alarm equipment is connected through a waterproof wire;
In the sixth step, checking whether the coordinates of the water level detection point are in the mutation area, if not, indicating that the water level detection point possibly belongs to the reflection area, detecting errors, and if so, calculating to obtain the height of the water level through the mapping relation between the pixel coordinates and the world coordinate system;
The water level alarm device and the standby alarm device are water level alarms, and the specific model numbers are ZSB127-Z and BZ1201 respectively;
embodiment one:
a depth learning-based riverway water level detection method without a water rule comprises the following steps:
Step one: setting a calibration object at a water level detection point, acquiring an image containing the calibration object from a camera, and obtaining a mapping relation between a pixel coordinate system and a world coordinate system through calibration;
step two: dividing a water body area and a non-water body area of a target water area by utilizing the SAM model, and acquiring a binary image;
Step three: cutting out a neighborhood of a water level calibration position on the binary image, and obtaining a water level dividing line on the image through image processing;
step four: binarizing a single frame image obtained by pulling, obtaining an interested region near the water level, dividing the interested region into a plurality of sub interested regions, and calculating Laplace values;
step five: determining a region with a mutation of Laplace value, which is designed as a water line region, by comparing the region of interest of the sub;
Step six: determining the position of the water level line by comparing the third step with the fifth step, and calculating the water level height by the mapping relation;
step seven: setting alarm equipment at the early warning water level;
Step eight: and connecting the output signal of the alarm device with a remote terminal.
In the first step, the vertex of the calibration object in the world coordinate system corresponds to the pixel coordinate of the vertex of the calibration object in the detection image, the bottom point in the world coordinate system corresponds to the bottom point in the pixel coordinate system, meanwhile, the included angle between the calibration object and the normal direction of the water surface and the initial height of the water surface need to be measured, and the mapping relation between the pixel coordinate and the world coordinate system can be obtained through nonlinear solution;
Dividing the image by using a SAM (sample membrane) dividing model, realizing accurate division of the water body and the non-water body by providing a prompt point and a detection frame, outputting a binary mask image, and removing pinholes generated during dividing the image by using morphological closing operation;
Thirdly, calculating the coordinate position of a water level detection point, denoising a frame of the extracted image, performing binarization treatment, cutting out an interested region containing water body according to the coordinate obtained during calibration, dividing the interested region into a plurality of small regions with equal areas, calculating the Laplace value of each small region, and finally obtaining a region with abrupt change of the Laplace value;
the specific calculation method is as shown in fig. 2:
The water level calculating method is schematically shown in fig. 2:
Wherein alpha is the included angle between the calibration object and the normal vector of the water surface under the world coordinate system; beta is the included angle between the placement direction of the calibration object and the moving direction of the water level detection point; the pixel coordinate of the upper left corner of the calibration object is (x 1, y 1), the pixel coordinate of the lower right corner is (x 2, y 2), the actual length of the calibration object is L, the actual length corresponding to the unit pixel is m, the pixel of the water level is changed to n, the initial height of the water level is H, and the changed height is H;
m＝L/(y2-y1)*cos(β)
H＝n*m*cos(α)+h
the calibration object can be calibrated for a plurality of times, a plurality of m are obtained, the unit pixel segments correspond to the actual lengths, and a segment function is formed to achieve better precision;
1. When α=β=0, the corresponding is the case where the calibration object is perpendicular to the water surface and the camera is opposite to the calibration object;
2. when beta=0 and alpha is not equal to 0, the camera is opposite to the calibration object, and the calibration object is obliquely placed on the ground for calibration;
3. when beta is not equal to 0 and alpha is not equal to 0, the corresponding calibration object is vertical to the water surface, however, the camera is obliquely opposite to the calibration object, and the water level line is not horizontal;
When beta is not equal to 0 and alpha is not equal to 0, the example graph corresponds to the situation that the calibration object is not perpendicular to the water surface and the camera is obliquely opposite to the calibration object;
in the seventh step, a plurality of water level alarm devices are respectively arranged at different positions of a river channel, the distance between two adjacent water level alarm devices is 15m, a standby alarm device is arranged beside each water level alarm device, and alarm transmission signals of the two alarm devices are different;
in the eighth step, the alarm equipment is connected with a remote terminal through a wireless signal receiving and transmitting module, the remote terminal is arranged in a working room, the alarm equipment is powered by a solar panel, the solar panel is arranged above the water level alarm equipment, and the alarm equipment is connected through a waterproof wire;
In the sixth step, checking whether the coordinates of the water level detection point are in the mutation area, if not, indicating that the water level detection point possibly belongs to the reflection area, detecting errors, and if so, calculating to obtain the height of the water level through the mapping relation between the pixel coordinates and the world coordinate system;
The water level alarm device and the standby alarm device are water level alarms, and the specific model numbers are ZSB127-Z and BZ1201 respectively;
Embodiment two:
a depth learning-based riverway water level detection method without a water rule comprises the following steps:
Step one: setting a calibration object at a water level detection point, acquiring an image containing the calibration object from a camera, and obtaining a mapping relation between a pixel coordinate system and a world coordinate system through calibration;
step two: dividing a water body area and a non-water body area of a target water area by utilizing the SAM model, and acquiring a binary image;
Step three: cutting out a neighborhood of a water level calibration position on the binary image, and obtaining a water level dividing line on the image through image processing;
step four: binarizing a single frame image obtained by pulling, obtaining an interested region near the water level, dividing the interested region into a plurality of sub interested regions, and calculating Laplace values;
step five: determining a region with a mutation of Laplace value, which is designed as a water line region, by comparing the region of interest of the sub;
Step six: determining the position of the water level line by comparing the third step with the fifth step, and calculating the water level height by the mapping relation;
In the first step, the vertex of the calibration object in the world coordinate system corresponds to the pixel coordinate of the vertex of the calibration object in the detection image, the bottom point in the world coordinate system corresponds to the bottom point in the pixel coordinate system, meanwhile, the included angle between the calibration object and the normal direction of the water surface and the initial height of the water surface need to be measured, and the mapping relation between the pixel coordinate and the world coordinate system can be obtained through nonlinear solution;
Dividing the image by using a SAM (sample membrane) dividing model, realizing accurate division of the water body and the non-water body by providing a prompt point and a detection frame, outputting a binary mask image, and removing pinholes generated during dividing the image by using morphological closing operation;
Thirdly, calculating the coordinate position of a water level detection point, denoising a frame of the extracted image, performing binarization treatment, cutting out an interested region containing water body according to the coordinate obtained during calibration, dividing the interested region into a plurality of small regions with equal areas, calculating the Laplace value of each small region, and finally obtaining a region with abrupt change of the Laplace value;
the specific calculation method is as shown in fig. 2:
The water level calculating method is schematically shown in fig. 2:
Wherein alpha is the included angle between the calibration object and the normal vector of the water surface under the world coordinate system; beta is the included angle between the placement direction of the calibration object and the moving direction of the water level detection point; the pixel coordinate of the upper left corner of the calibration object is (x 1, y 1), the pixel coordinate of the lower right corner is (x 2, y 2), the actual length of the calibration object is L, the actual length corresponding to the unit pixel is m, the pixel of the water level is changed to n, the initial height of the water level is H, and the changed height is H;
m＝L/(y2-y1)*cos(β)
H＝n*m*cos(α)+h
the calibration object can be calibrated for a plurality of times, a plurality of m are obtained, the unit pixel segments correspond to the actual lengths, and a segment function is formed to achieve better precision;
1. When α=β=0, the corresponding is the case where the calibration object is perpendicular to the water surface and the camera is opposite to the calibration object;
2. when beta=0 and alpha is not equal to 0, the camera is opposite to the calibration object, and the calibration object is obliquely placed on the ground for calibration;
3. when beta is not equal to 0 and alpha is not equal to 0, the corresponding calibration object is vertical to the water surface, however, the camera is obliquely opposite to the calibration object, and the water level line is not horizontal;
When beta is not equal to 0 and alpha is not equal to 0, the example graph corresponds to the situation that the calibration object is not perpendicular to the water surface and the camera is obliquely opposite to the calibration object;
examples fig. 3, 4 and 5:
in the seventh step, a plurality of water level alarm devices are respectively arranged at different positions of a river channel, the distance between two adjacent water level alarm devices is 10m, a standby alarm device is arranged beside each water level alarm device, and alarm transmission signals of the two alarm devices are different;
In the eighth step, the alarm equipment is connected with a remote terminal through a wireless signal receiving and transmitting module, the remote terminal is arranged in a working room, the alarm equipment is powered through a solar panel, the solar panel converts direct current into alternating current through an inverter to power, and the solar panel is arranged above the water level alarm equipment and connected through a waterproof wire;
the water level alarm device and the standby alarm device are water level alarms, and the specific model is ZSB127-Z and BZ1201 respectively.
To sum up: according to the depth learning-based water level detection method for the riverway without the water gauge, through a water level detection scheme without the water gauge, water level detection of multiple scenes such as a vertical bridge pier, a lock chamber, a pond, an inclined reservoir, a dam, a riverway shoreline and the like can be realized through designing a water surface normal included angle, the Laplacian value of a region of interest is combined, the distinction of water body reflection and non-reflection regions is realized, the accuracy of a coordinate point of the water level line is corrected, the problem that the water gauge can be in ground due to the conventional common technology is solved, resource waste can be caused by setting water gauges with different levels, and weeds can grow around the field water gauge to shield the water gauge, so that the water gauge reading is inaccurate is solved.
Claims (8)
1. The depth learning-based riverway water level detection method without the water rule is characterized by comprising the following steps of:
Step one: setting a calibration object at a water level detection point, acquiring an image containing the calibration object from a camera, and obtaining a mapping relation between a pixel coordinate system and a world coordinate system through calibration;
step two: dividing a water body area and a non-water body area of a target water area by utilizing the SAM model, and acquiring a binary image;
Step three: cutting out a neighborhood of a water level calibration position on the binary image, and obtaining a water level dividing line on the image through image processing;
step four: binarizing a single frame image obtained by pulling, obtaining an interested region near the water level, dividing the interested region into a plurality of sub interested regions, and calculating Laplace values;
step five: determining a region with a mutation of Laplace value, which is designed as a water line region, by comparing the region of interest of the sub;
Step six: determining the position of the water level line by comparing the third step with the fifth step, and calculating the water level height by the mapping relation;
step seven: setting alarm equipment at the early warning water level;
Step eight: and connecting the output signal of the alarm device with a remote terminal.
2. The depth learning-based waterside river channel water level detection method of claim 1, wherein the method comprises the following steps: in the first step, the vertex of the calibration object in the world coordinate system corresponds to the pixel coordinate of the vertex of the calibration object in the detection image, the bottom point in the world coordinate system corresponds to the bottom point in the pixel coordinate system, meanwhile, the included angle between the calibration object and the normal direction of the water surface and the initial height of the water surface need to be measured, and the mapping relation between the pixel coordinate and the world coordinate system can be obtained through nonlinear solution.
3. The depth learning-based waterside river channel water level detection method of claim 1, wherein the method comprises the following steps: in the second step, the SAM segmentation model is utilized to segment the image, the accurate segmentation of the water body and the non-water body is realized by providing the prompt points and the detection frame, the binary mask image is output, and the small holes generated during the segmentation of the image are removed by morphological closed operation.
4. The depth learning-based waterside river channel water level detection method of claim 1, wherein the method comprises the following steps: in the third step, the coordinate position of the water level detection point is obtained through calculation, the obtained frame of image is subjected to binarization processing after denoising, the interested area including the water body is cut out according to the coordinate obtained during calibration, the interested area is divided into a plurality of small areas with equal areas, the Laplace value of each small area is calculated, and finally the area with abrupt change of the Laplace value is obtained.
5. The method for detecting the water level of the riverway without water rule based on deep learning according to claim 1, wherein the method comprises the following steps of: in the seventh step, a plurality of water level alarm devices are respectively arranged at different positions of the river channel, the distance between two adjacent water level alarm devices is 10-20m, a standby alarm device is arranged beside each water level alarm device, and alarm transmission signals of the two alarm devices are different.
6. The method for detecting the water level of the riverway without water rule based on deep learning according to claim 1, wherein the method comprises the following steps of: in the step eight, the alarm device is connected with the remote terminal through the wireless signal receiving and transmitting module, the remote terminal is arranged inside the working room, the alarm device is powered by the solar panel, the solar panel is arranged above the water level alarm device, and the alarm device is connected with the water level alarm device through the waterproof wire.
7. The method for detecting the water level of the riverway without water rule based on deep learning according to claim 1, wherein the method comprises the following steps of: in the sixth step, whether the coordinates of the water level detection point are in the mutation area is checked, if not, the water level detection point possibly belongs to the reflection area, and if the detection is wrong, the height of the water level is calculated through the mapping relation between the pixel coordinates and the world coordinate system.
8. The method for detecting the water level of the riverway without water rule based on deep learning according to claim 5, wherein the method comprises the following steps of: the water level alarm device and the standby alarm device are water level alarms, and the specific model is ZSB127-Z and BZ1201 respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311379848.5A CN117974962A (en) | 2023-10-23 | 2023-10-23 | Depth learning-based riverway water level detection method without water rule |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311379848.5A CN117974962A (en) | 2023-10-23 | 2023-10-23 | Depth learning-based riverway water level detection method without water rule |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117974962A true CN117974962A (en) | 2024-05-03 |
Family
ID=90856289
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311379848.5A Pending CN117974962A (en) | 2023-10-23 | 2023-10-23 | Depth learning-based riverway water level detection method without water rule |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117974962A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118411675A (en) * | 2024-06-27 | 2024-07-30 | 武汉大水云科技有限公司 | Water level identification method based on SAM large model |
-
2023
- 2023-10-23 CN CN202311379848.5A patent/CN117974962A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118411675A (en) * | 2024-06-27 | 2024-07-30 | 武汉大水云科技有限公司 | Water level identification method based on SAM large model |
| CN118411675B (en) * | 2024-06-27 | 2024-09-06 | 武汉大水云科技有限公司 | Water level identification method based on SAM large model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108759973B (en) | Water level measuring method | |
| CN102496232B (en) | Transmission facility monitoring method and system | |
| CN107833203A (en) | A kind of horizontal plane identification and water level method for real-time measurement based on image procossing | |
| Noto et al. | Low-cost stage-camera system for continuous water-level monitoring in ephemeral streams | |
| CN109186706B (en) | Method for early warning of urban rainfall flood inundated area | |
| CN108921165B (en) | Water level identification method based on water gauge image | |
| CN109764930B (en) | Water gauge water line visual detection method suitable for complex illumination conditions | |
| CN117974962A (en) | Depth learning-based riverway water level detection method without water rule | |
| CN204758083U (en) | River course water level alarm device based on image processing | |
| CN115060343B (en) | Point cloud-based river water level detection system and detection method | |
| CN107631782A (en) | A kind of level testing methods based on Harris Corner Detections | |
| CN115546235B (en) | Water level identification method, system and storage medium based on image segmentation | |
| CN115761682A (en) | Method and device for identifying travelable area based on laser perception and intelligent mine card | |
| CN104596484A (en) | Method of measuring drift ice density in ice flood season of Yellow River | |
| CN112509134A (en) | Tidal flat digital elevation model construction method and system | |
| CN113593040B (en) | Method and system for identifying abnormal vegetation growth diseases of dike revetment | |
| Utomo et al. | Early warning flood detector adopting camera by Sobel Canny edge detection algorithm method | |
| CN115761644A (en) | Transmission line foreign matter detection method based on deep learning and frame difference method | |
| CN110217360A (en) | A kind of Underwater Structure's Surface Flaws sniffing robot system and its working method | |
| CN106960437B (en) | Liquid level detection method for oil-immersed power transformer of power inspection robot | |
| CN118314693A (en) | Tower instability monitoring and early warning device based on image recognition and computer vision | |
| CN112990064A (en) | Dial pointer identification method based on color segmentation and probability model | |
| CN115578695B (en) | Water gauge water level machine vision detection method and device with free shooting visual angle | |
| Zhen et al. | Visual detection algorithm of water line based on feature fusion | |
| CN113822104B (en) | Artificial intelligence surface of water detecting system based on virtual scale of many candidates |
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 |