River image flow measurement method based on automatic tracer particle delivery system
By designing an automatic tracer particle delivery system and image processing algorithm, the problems of insufficient automatic tracer particle delivery and identification were solved, and efficient and accurate measurement of river flow velocity was achieved.
Patent Information
- Application Number
- CN202511128537.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-13
- Publication Date
- 2025-09-12
- Estimated Expiration
- 2045-08-13
AI Technical Summary
In existing river image flow measurement methods, the automated delivery and recognition of tracer particles are insufficient, especially when there is a lack of identifiable textures on the natural water surface, which affects the measurement accuracy.
An automatic tracer particle delivery system was designed, including a ball bin system and a control system. The concave roller of the ball was driven by a stepper motor, and the automatic delivery of tracer particles was achieved by using a convex block structure. The flow rate was calculated by combining camera shooting and image processing algorithm.
It has realized the automated delivery of tracer particles and high-precision flow velocity measurement, significantly improving the automation and intelligence level of river image flow measurement, and enhancing the efficiency and accuracy of flow measurement.
Smart Images

Figure CN120629628A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of river flow monitoring, and in particular relates to a river flow measurement method based on an automatic tracer particle delivery system. Background Art
[0002] River velocity measurement plays a fundamental and strategic role in multiple fields, including water conservancy projects, water environment management, and flood prevention and disaster reduction. In supporting water resource management and scheduling, flow velocity, combined with water level data, can be used to calculate river flow, forming the foundation for water resource accounting. This provides accurate data support for basin water use scheduling, reservoir regulation, and water rights allocation, and provides a data basis for determining river ecological flows and maintaining ecosystem health. In ensuring flood control safety, flow measurement during floods can provide insights into flow velocity changes, peak flood arrival times, and flow capacity.
[0003] Compared with traditional flow measurement methods (such as current meters, ADCPs, flow weirs, etc.), image flow measurement technology has the following significant advantages: it is non-contact, the equipment is set up on the shore or in the air, does not disturb the water body, and does not interfere with the flow field; it is highly operational and suitable for flood seasons, rapids, dangerous areas, etc., and has been widely used in river measurement.
[0004] Tracer particles are crucial in river flow measurement using imagery. Essentially, image flow measurement calculates flow velocity by tracking the characteristics of water flow. The placement of tracer particles can artificially enhance water surface texture or features, making flow trajectories more visible and improving algorithm recognition and tracking accuracy. In real-world scenarios, natural water surfaces often lack recognizable textures (e.g., clear water, cloudy skies, strong light reflections), which can severely impact measurement accuracy. Currently, the most commonly used tracer method involves manually placing paper, foam balls, wood chips, or plastic sheets into the water. This method cannot be automated, and recognition effectiveness needs to be improved. Summary of the Invention
[0005] Purpose of the invention: The technical problem to be solved by the present invention is to provide an automatic tracer particle delivery system.
[0006] The technical problem that the present invention also aims to solve is to provide a river flow measurement method based on an automatic tracer particle delivery system.
[0007] Technical solution: In order to solve the above technical problems, the present invention provides an automatic tracer particle delivery system, the automatic tracer particle delivery system includes a ball bin system and a control system, the ball bin system includes a ball bin and a plurality of tracer particles placed therein, the control system includes a drive system and a rolling ball bin with an upper end opening connected thereto, the rolling ball bin includes a rolling ball bin bottom plate, a rolling ball concave wheel and a flange cover plate, a first protrusion is provided on the inner wall of the rolling ball bin bottom plate, a rolling ball concave wheel is provided with a rolling ball groove and a second protrusion, and the rolling ball concave wheel is driven by the drive system relative to the rolling ball bin Rotate, a hose interface is provided on the flange cover, a hose is connected to the hose interface, and the hose interface is set corresponding to the position of the first protrusion. When the tracer particle falls into the ball rolling groove, the driving system drives the ball rolling wheel to rotate. When the ball rolling groove rotates to the corresponding position of the first protrusion, the first protrusion will push the tracer particle to move toward the hose interface. When the ball rolling wheel continues to rotate to the position corresponding to the second protrusion and the tracer particle, the second protrusion will push the tracer particle out of the hose interface. At this point, the tracer particle is pushed out of the hose interface to realize the automatic delivery of the tracer particle.
[0008] The ball warehouse is a structure with an upper and lower circular warehouse, and supporting legs are provided below the ball warehouse.
[0009] The driving system automatically releases a number of tracer particles by controlling the number of rotations.
[0010] The drive system comprises a stepper motor and a speed reducer. As the number of balls in the hose increases, the number of squeezed balls also increases, so a speed reducer is installed at the output shaft end of the stepper motor to increase the squeezing force. The stepper motor can then be accurately controlled to achieve the purpose of automatic ball throwing. The hose can also be rotated on the water surface, with the balls directly squeezed onto the water surface to serve as tracer particles for surface image flow measurement. The tracer particles, i.e., spheres, suitable for the system of the present invention have a diameter of approximately 3 cm and are made of a material that floats in water. The size of the ball bin can be adjusted and designed according to the number of balls required. The circular bin structure below the ball bin facilitates the balls to automatically fall into the ball bin due to their own weight. The ball bin of the present invention is installed below the ball bin, and a ball roller is installed inside it and is driven by a stepper motor to rotate within the ball bin. The function is that each rotation of the ball roller squeezes a ball into the hose interface.
[0011] The present invention also includes a river image flow measurement method based on an automatic tracer particle delivery system, comprising the following steps: (1) placing the automatic tracer particle delivery system on the bank of a river where the flow velocity needs to be measured or on a bridge over the river; (2) Place the hose on the surface, middle or bottom of the river where the flow rate needs to be measured, start the drive device, and the tracer particles in the hose will be squeezed out of the hose one by one as the number increases; (3) The tracer particles follow the water flow and the camera captures the tracer particle motion image on the water flow surface. Then, the computer captures two frames of color particle images at regular intervals. The time interval between the images is ; (4) The characteristic color RGB value of the tracer particle is represented as (Rp, Gp, Bp), and the RGB value of the pixel in the image is represented as (Ri, Gi, Bi). According to the difference between the color of the tracer particle and the background color of the water flow, the RGB difference between the two is calculated. ; (5) Set the RGB difference threshold T and perform binary segmentation on the tracer particle image using the following formula, where G is the grayscale value of the pixel in the tracer particle image: ; (6) After binary segmentation of the two frames of tracer particle images, morphological processing is performed using corrosion and dilation, followed by region marking, and the horizontal and vertical coordinates (Ip, Jp) of the center position of the tracer particle in the image are calculated. According to the camera position and the height of the water surface, the relationship between the horizontal and vertical coordinates of the image and the actual horizontal and vertical coordinates at the scene is calibrated, and the horizontal and vertical coordinates (Ip, Jp) of the center position of the tracer particle in the image are converted to the actual horizontal and vertical coordinates (Xp, Yp); (7) The actual coordinates of the tracer particles in the two frames of particle images are (Xp1, Yp1) and (Xp2, Yp2), respectively. The flow velocity at each particle location can be expressed as: ; Vx and Vy represent the flow velocity components in the x and y directions.
[0012] The tracer particles are in a standard spherical shape and are made of a material with a density lower than that of water, and the surface of the spheres is coated with a water-based environmentally friendly paint with reflective properties.
[0013] Wherein, the material is wood, and the water-based environmentally friendly paint is red, yellow or blue water-based environmentally friendly paint.
[0014] The rivers include plain rivers or mountain rivers.
[0015] The camera is installed on a bridge at a height of 6 to 8 meters directly above the river surface, and the camera resolution is above 12 million pixels.
[0016] The method automatically places tracer particles at different locations in the river, and automatically measures the flow velocity and direction at the locations where the tracer particles are located, thereby realizing fully automatic image flow measurement.
[0017] Working principle of the present invention: The present invention realizes the flow velocity measurement method of river images by using tracer particles and their automatic delivery system. When the ball (tracer particle) in the upper ball bin system naturally falls from below into the rolling groove of a ball-rolling concave wheel due to its own weight, as the stepper motor drives the rolling concave wheel to rotate, the ball will be carried by the first protrusion to the direction of the hose interface. When it approaches the hose interface position, due to the eccentric guidance effect of the second protrusion, the ball will move Figure 3 The hose on the right side of the diagram is squeezed. Each rotation of the ball roller pushes a sphere into the hose connector. This cycle repeats, and as the number of spheres increases, they are squeezed out of the hose one by one. If the hose is placed on the riverbed, the spheres will be squeezed out of the hose and float to the surface, serving as tracer particles for surface current measurement. Alternatively, the hose can be rotated above the water surface, with the spheres being squeezed directly to the surface, serving as tracer particles for surface current measurement. As the number of spheres in the hose increases, the number of spheres squeezed also increases. Therefore, a speed reducer is installed at the output end of the stepper motor shaft to increase the squeezing force. This allows the stepper motor to accurately control the number of revolutions, achieving automatic ball delivery.
[0018] Beneficial Effects: Compared with existing technologies, the advantages of the present invention are that the tracer particles designed in the present invention fully consider optical properties, environmental friendliness, and applicability, significantly improving particle imaging. The flow measurement method of the automatic tracer particle delivery system designed in the present invention can achieve automatic delivery at the river bottom and river surface, and can automatically deliver particles for image-based flow measurement methods such as floating depth flow measurement, surface flow velocity, and flow field measurement, significantly improving the automation and intelligence level of river image flow measurement. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 is a schematic diagram of the control system; Figure 2 This is a front cross-sectional schematic diagram of the tracer particle automatic delivery system; Figure 3 Schematic diagram of the side section of the automatic tracer particle delivery system; 1-tracer particle, 2-ball bin, 3-leg, 4-ball concave wheel, 5-ball bin, 6-flange cover, 7-hose interface, 8-hose, 9-stepping motor, 10-reducer, 11-ball groove, 12-first bump, 13-second bump. DETAILED DESCRIPTION
[0020] The present invention is further described below through specific embodiments. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the principles of the present invention, and these should also be regarded as falling within the scope of protection of the present invention.
[0021] Example 1 Flow velocity measurement of plain rivers 1. Construction of the automatic tracer particle delivery system like Figures 1-3 First, build a control system. The control system includes a drive system and a ball bin 5 with an upper opening connected thereto. The drive system includes a stepper motor 9 and a reducer 10. The ball bin 5 includes a ball bin bottom plate, a ball concave wheel 4 and a flange cover 6. A first convex block 12 is provided on the inner wall of the ball bin bottom plate. A ball concave wheel 4 is provided with a ball groove 11 and a second convex block 13. The ball concave wheel 4 rotates relative to the ball bin 5 under the drive of the drive system. A hose interface 7 is provided on the flange cover 6. The hose interface 7 A hose 8 is connected to the hose interface 7, and the hose interface 7 is configured to correspond to the first protrusion 12. When the tracer particle falls into the ball rolling groove 11, the drive system drives the ball rolling wheel 4 to rotate. When the ball rolling groove 11 rotates to the position corresponding to the first protrusion 12, the first protrusion 12 will push the tracer particle toward the hose interface 7. When the ball rolling wheel 4 continues to rotate until the second protrusion 13 corresponds to the tracer particle, the second protrusion 13 pushes the tracer particle out of the hose interface 7. Next, an automatic tracer particle delivery system is constructed. The automatic tracer particle delivery system includes a ball bin system and a control system. The ball bin system includes a ball bin 2 and a tracer particle 1 placed in the ball bin 2. The ball bin 2 is a top-down round bin structure with four legs 3 below the ball bin 2. Considering the tracer particle's tracking ability and the convenience of automatic delivery, the tracer particle is designed as a standard spherical shape. The diameter can be adjusted according to the camera installation height to ensure that the tracer particle can be clearly identified in the captured image. Considering river environmental protection requirements, a wooden material (with a density lower than water) was selected. To improve particle identification, a water-based, environmentally friendly paint with a color that is highly distinguishable from the river water (such as red, yellow, blue, etc.) and has good reflective properties was selected to uniformly color the tracer particles and ensure good imaging quality. The tracer particles (hereinafter referred to as spheres) in this invention have a diameter of approximately 3 cm and are made of a material that floats in water. The size of the ball bin 2 can be adjusted based on the number of spheres required. The circular bin structure below the ball bin 2 facilitates the automatic drop of the spheres into the ball bin due to their own weight. The ball bin 5 is installed directly below the ball bin 2. A ball roller cam 4 is installed inside the bin and rotates within the bin 5 by a stepper motor 9. With each rotation of the ball roller cam 4, a sphere is squeezed into the hose interface 7 of the hose 8. The multiple balls in the upper ball bin naturally fall from the bottom into the ball rolling groove 11 of the ball rolling concave wheel 4 due to their own weight. As the stepper motor 9 drives the ball rolling concave wheel 4 to rotate, the ball will be carried by the first protrusion 12 to roll towards the direction of the hose interface 7. When it is close to the position of the hose interface 7, the ball will move towards the direction of the hose interface 7 due to the eccentric guiding effect of the second protrusion 13. Figure 3The ball is squeezed into the hose 8 on the right side of the diagram. That is, with each rotation of the ball-rolling concave wheel 4, a ball is squeezed into the hose connector 7. This cycle repeats, and as the number of balls inside the hose increases, they are squeezed out of the hose 8 one by one. If the hose 8 is placed on the riverbed, the balls will be squeezed out of the hose 8 and float to the surface, serving as tracer particles for surface current measurement. The hose 8 can also be rotated above the water surface, with the balls being squeezed directly to the surface, serving as tracer particles for surface current measurement.
[0022] 2. Plain river speed measurement 1) Plain rivers usually have a relatively gentle flow, are about 60 meters wide, and have an average depth of about 3 meters. A camera with a resolution of 12 megapixels is installed on a bridge about 6 meters above the river surface. An automatic tracer particle delivery system is placed on the shore. A hose connected to the system via a delivery pipeline automatically delivers tracer particles from the riverbed at a high flow rate in the middle of the cross-section. As the number of tracer particles increases, they are squeezed out of the hose one by one. 2) The tracer particles follow the water flow and the camera captures the tracer particle motion image on the water flow surface. Then the computer captures two frames of color particle images at regular intervals. The time interval between the images is ; 3) The color of the tracer particles is significantly different from the background color of the water flow. The characteristic color RGB value of the tracer particles is represented by (Rp, Gp, Bp), and the RGB value of the pixel in the image is represented by (Ri, Gi, Bi). The RGB difference between the two is calculated. ; 4) Set the RGB difference threshold T to 20 and perform binary segmentation on the particle image using the following formula, where G is the grayscale value of the pixel in the particle image: ; 5) After binary segmentation of the two particle images, morphological processing using corrosion and dilation is performed, followed by region labeling. The horizontal and vertical coordinates (Ip, Jp) of the center position of each particle in the image are calculated. Based on the camera position and the height of the water surface, the relationship between the horizontal and vertical coordinates of the image and the actual horizontal and vertical coordinates at the scene is calibrated, and the horizontal and vertical coordinates of the particle center position in the image (Ip, Jp) are converted to the actual horizontal and vertical coordinates (Xp, Yp) at the scene. 6) The actual coordinates of each particle in the two frames of particle images are (Xp1, Yp1) and (Xp2, Yp2) respectively. The flow velocity at the location of each particle can be expressed as: Vx = ,Vy= ; Vx and Vy represent the velocity components of the flow in the x and y directions; 7) According to the above steps, the flow velocity and direction at the particle location are automatically measured, thereby realizing fully automatic image flow measurement.
[0023] Using ADCP on-site flow measurement as a standard, velocity measurement was compared with artificial placement of foam boards as tracer particles. Five measurement points were randomly selected and defined as measurement points 1 to 5. The measurement results are shown in Table 1.
[0024] Table 1 Comparison of flow velocity measurements (perpendicular to the cross section)
[0025] Field verification has shown that manual foam placement, including on-site production and placement, generally takes more than 30 seconds. The tracer particles of the present invention can be conveniently and quickly delivered by an automatic delivery system, taking no more than 3 seconds to complete the automatic delivery, significantly improving efficiency and accuracy by more than 1.2%, thereby enhancing the efficiency and accuracy of river image flow measurement.
[0026] Example 2: Flow velocity measurement of mountain rivers The river is approximately 120 meters wide and has an average depth of approximately 8 meters. The camera is installed on a bridge approximately 8 meters above the river surface. The camera has a resolution of 12 megapixels. During floods, due to the high flow rate, the automatic tracer particle delivery system of Example 1 is installed above the bridge through which the river passes. The tracer particles are delivered from above the river to the river surface by the automatic tracer particle delivery system. The same particle image velocimetry algorithm as in Example 1 is used to automatically obtain the surface flow velocity. The measured value of the rotor velocimeter at the on-site hydrological station is used as the standard and compared with the manually delivered plastic sheets as tracer particles. Five measuring points are randomly selected and defined as measuring points 1 to 5. The measurement results are shown in Table 2.
[0027] Table 2 Comparison of flow velocity measurements (perpendicular to the cross section)
[0028] Field verification has shown that the manual placement of plastic sheets, including on-site production and placement, generally takes more than 45 seconds. The tracer particles of the present invention can be conveniently and quickly placed automatically by the automatic placement system, taking no more than 3 seconds to complete the automatic placement, significantly improving efficiency and accuracy by more than 1.6%, thereby improving the efficiency and accuracy of river image flow measurement.
Claims
1. A tracer particle automatic delivery system, characterized in that: The automatic tracer particle delivery system includes a ball bin system and a control system. The ball bin system includes a ball bin and a plurality of tracer particles placed therein. The control system includes a drive system and a rolling ball bin with an upper end connected thereto. The rolling ball bin includes a rolling ball bin bottom plate, a rolling ball concave wheel and a flange cover plate. A first protrusion is provided on the inner wall of the rolling ball bin bottom plate, and a rolling ball concave wheel is provided with a rolling ball groove and a second protrusion. The rolling ball concave wheel rotates relative to the rolling ball bin under the drive of the drive system. A hose interface is provided on the flange cover plate, and a hose is connected to the hose interface. The hose interface is arranged corresponding to the position of the first protrusion. When the tracer particle falls into the rolling ball groove, the drive system drives the rolling ball concave wheel to rotate. When the rolling ball groove rotates to a position corresponding to the first protrusion, the first protrusion pushes the tracer particle toward the hose interface. When the rolling ball concave wheel continues to rotate to a position corresponding to the second protrusion and the tracer particle, the second protrusion pushes the tracer particle out of the hose interface. At this point, the tracer particle is pushed out of the hose interface to realize automatic delivery of the tracer particle.
2. The automatic tracer particle delivery system according to claim 1, characterized in that: The driving system automatically releases a number of tracer particles by controlling the number of rotations.
3. The automatic tracer particle delivery system according to claim 1, characterized in that: The driving system comprises a stepping motor and a speed reducer.
4. A river image flow measurement method based on an automatic tracer particle delivery system, based on the automatic tracer particle delivery system according to any one of claims 1 to 3, characterized in that: The following steps are involved: (1) Place the tracer particle automatic delivery system on the bank of the river where the flow velocity needs to be measured or on a bridge passing over the river; (2) Place the hose on the surface, middle or bottom of the river where the flow rate needs to be measured, start the drive system, and the tracer particles in the hose will be squeezed out of the hose one by one as the number increases; (3) The tracer particles follow the water flow and the camera captures the tracer particle motion image on the water flow surface. Then, the computer captures two frames of color particle images at regular intervals. The time interval between the images is ; (4) The characteristic color RGB value of the tracer particle is represented as (Rp, Gp, Bp), and the RGB value of the pixel in the image is represented as (Ri, Gi, Bi). According to the difference between the color of the tracer particle and the background color of the water flow, the RGB difference between the two is calculated. ; (5) Set the RGB difference threshold T and perform binary segmentation on the tracer particle image using the following formula, where G is the grayscale value of the pixel in the tracer particle image: ; (6) After binary segmentation of the two frames of tracer particle images, morphological processing is performed using corrosion and dilation, followed by region marking, and the horizontal and vertical coordinates (Ip, Jp) of the center position of the tracer particle in the image are calculated. According to the camera position and the height of the water surface, the relationship between the horizontal and vertical coordinates of the image and the actual horizontal and vertical coordinates at the scene is calibrated, and the horizontal and vertical coordinates (Ip, Jp) of the center position of the tracer particle in the image are converted to the actual horizontal and vertical coordinates (Xp, Yp); (7) The actual coordinates of the tracer particles in the two frames of particle images are (Xp1, Yp1) and (Xp2, Yp2), respectively. The flow velocity at each particle location can be expressed as: ; Vx and Vy represent the flow velocity components in the x and y directions.
5. The river image flow measurement method of the tracer particle automatic delivery system according to claim 4 is characterized in that: The tracer particles are in a standard spherical shape and are made of a material with a density lower than that of water. The surface of the spheres is coated with a water-based environmentally friendly paint with reflective properties.
6. The river image flow measurement method of the tracer particle automatic delivery system according to claim 5, characterized in that: The material is wood, and the water-based environmentally friendly paint is red, yellow or blue water-based environmentally friendly paint.
7. The river image flow measurement method of the tracer particle automatic delivery system according to claim 4, characterized in that: The method automatically releases tracer particles at different locations in the river, and automatically measures the flow velocity and direction at the locations where the tracer particles are located.
8. The river image flow measurement method of the tracer particle automatic delivery system according to claim 4 is characterized in that: The rivers include plain rivers or mountain rivers.
9. The river image flow measurement method of the tracer particle automatic delivery system according to claim 4, characterized in that: The camera is installed on a bridge at a height of 6 to 8 meters above the river surface, and the camera resolution is above 12 million pixels.
Citation Information
Patent Citations
Tracer particle release apparatus
CN105403729A
Water tank PIV (particle image velocimetry) particle sowing device
CN119714792A
Tracer particle release structure and device for microgravity combustion flow field PIV (particle image velocimetry) measurement
CN120313860A
Automatic tracer particle throwing device in water tank PIV experiment
CN215640037U
Measuring apparatus for flow velocity
JP1995027780A