Method for measuring particle size and settling velocity of atmospheric particles in water body based on image recognition

Abstract

The invention discloses a method for measuring the particle size and sedimentation velocity of atmospheric particles in water bodies with different salinity based on image recognition, which comprises the following steps: leaching atmospheric particles collected on a sampling film in advance into a sedimentation column, and starting to record a video of the sedimentation process of the particles when the particles enter the sedimentation column. Using an image identification script written based on a python language to perform graying processing on the recorded video, performing particle identification according to the gray value of the particles, and calculating the settling velocity of the particles according to the displacement and time information of the front and rear frames of particles; and identifying the sizes of the particles in different directions by each frame of video information so as to estimate the geometric average particle size of the particles. And drawing a scatter diagram according to the identified statistical results of the particle sizes and the settling velocities of the plurality of particles. The measuring method has the characteristics of simple device, convenience, practicability, high measuring precision and the like. Meanwhile, the method can also be applied to laboratory determination of the particle size and the settling velocity of other micro particles (such as micro-plastics and planktons) in the water body.

Description

Technical Field

[0001] The present invention relates to a method for measuring the particle size of atmospheric particles in water and a method for determining the sedimentation velocity of atmospheric particles in water Background Art

[0002] The sedimentation of atmospheric particles has a significant impact on the material cycle in water. The sedimentation velocity and residence time of atmospheric particles in water affect whether nutrients such as nitrogen, phosphorus, and iron carried by atmospheric particles can be effectively and fully utilized by plankton in the water layer of limited depth. Therefore, it is crucial to clarify the sedimentation velocity of atmospheric particles in water. Atmospheric particles are relatively small in size, and currently, there is a lack of accurate and non-invasive methods for measuring the particle size of atmospheric particles and their sedimentation velocity in water Summary of the Invention

[0003] The specific problem to be solved by the present invention is to accurately, real-time, and synchronously measure the particle size and sedimentation velocity of atmospheric particles in water. Its purpose is to provide a non-invasive, image recognition-based synchronous measurement method for the particle size and sedimentation velocity of atmospheric particles, thereby helping to explore the residence time of atmospheric particles in the euphotic layer of the ocean, lakes, etc. after entering the water, and clarifying the impact of atmospheric sedimentation on the water environment

[0004] To achieve the above purpose, the present invention adopts the following technical solutions

[0005] A method for measuring the particle size and sedimentation velocity of atmospheric particles in water based on image recognition, and the measurement method is carried out according to the following steps

[0006] (1) Inject clear, transparent, and foreign object-free experimental water into the sedimentation column and let it stand for 12 hours. To simulate the real water environment, the experimental water can be configured according to the salinity of different water environments such as rivers, lakes, and oceans

[0007] (2) Set the resolution and recording frame rate of the camera

[0008] (3) Slowly wash the atmospheric particles pre-collected on the sampling membrane into the sedimentation column

[0009] (4) Fix the camera, adjust the camera focus, and shoot a clear video of the particles settling in the water

[0010] (5) Import the video file of the particle sedimentation into a script written in the Python language

[0011] (6) Perform gray-scale enhancement processing on the particle images of two consecutive frames before and after to obtain the binary image of the particles; identify and track individual particles according to the consistency of the gray values and sizes of adjacent two-frame particles

[0012] (7) For the particle-based binary image, measure the image sizes of the particles in each frame in three directions ( Figure 3 ), and obtain the geometric mean particle size by taking the geometric mean of the measurement results in the three directions;

[0013] (8) Calculate the instantaneous velocity of the particles at different times according to the displacement and time information of two consecutive frames of particles;

[0014] (9) Plot the statistical results of the particle sizes and sedimentation velocities of multiple particles as a scatter plot to visually record the relationship between particle size and sedimentation velocity.

[0015] Furthermore, the additional technical solutions are as follows.

[0016] The slide rail is made of aluminum alloy, and a rotating shaft is assembled in the middle so that the camera on the slide rail can be lifted slowly and stably.

[0017] The sedimentation column is made of high-transparency organic glass material, with a length, width and height of 20×20×50 cm.

[0018] The photographing device is a camera equipped with a macro lens. It is installed on the slide rail. After locking the focal length, the field of view range is 2×3.5 mm. It shoots the sedimentation video of micron-sized particles, with a video frame rate of 60 frames per second and a resolution of 1080p.

[0019] The LED lamp is placed on one side of the sedimentation column, at the same height as the camera. As an external light source, it irradiates the particles with strong light to make the particles photographed by the camera clearer.

[0020] The particle placement device is for atmospheric particles. It consists of a quartz filter membrane that has collected atmospheric particles and a wash bottle. Through the wash bottle, the atmospheric particles on the sampling membrane can be slowly rinsed into the sedimentation column.

[0021] The image recognition system is a script written in the Python language that can be used to process the particle sedimentation video. This script can grayscale the captured video, highlight the contours of the particles, and thus identify the images of the particles sinking in the video.

[0022] The method for measuring the particle size, taking the horizontal direction as an example, can obtain the true length of the particle image in the horizontal direction based on the resolution of the captured video and the number of pixels occupied by the particle image in the horizontal direction. Similarly, the lengths in the vertical direction and the direction inclined at 45° can also be measured. Starting from when the particle enters the camera's field of view until it leaves the field of view, the length of the particle image in the horizontal direction is measured frame by frame, and the average value of the horizontal particle lengths obtained for each frame is taken. Similarly, the average values of the particle image lengths in the vertical direction and the direction inclined at 45° can also be obtained. After obtaining the average lengths of the particle in the three directions, the geometric mean of the three sizes is taken to obtain the representative particle size. The method for measuring the particle sedimentation velocity is as follows: starting from when the particle enters the lens's field of view until it leaves the field of view, the instantaneous velocity of the particle at each frame at a specific height in the sedimentation column is measured, and the sum of the instantaneous velocities of each frame is averaged to obtain the representative sedimentation velocity of the particle at the specific height.

[0023] Compared with the prior art, the synchronous measurement method for the particle size and sedimentation velocity of atmospheric particles provided by the present invention has the following advantages: first, this method is non-invasive and microscopically measures, which can effectively avoid macroscopic interference and directly observe the particle size; second, this method can intuitively observe the sedimentation process of atmospheric particles in the water environment and obtain the instantaneous sedimentation velocity during the sedimentation process; third, this method can observe particles in the micron range and has high measurement accuracy for the particle size and sedimentation velocity; fourth, this method clearly and accurately counts the corresponding relationship between the particle size and sedimentation velocity during the sedimentation process of atmospheric particles. Description of the Drawings

[0024] Figure 1 is the flow chart of the implementation process of this method.

[0025] Figure 2 is the displacement diagram and instantaneous velocity of atmospheric particles in the sedimentation column at different times of this method.

[0026] Figure 3 is the schematic diagram for measuring the particle size of this method.

[0027] Figure 4 is the scatter diagram of particle size and sedimentation velocity of this method. Detailed Embodiments

[0028] The following further explains the detailed embodiments of the present invention.

[0029] Implement the method for measuring the particle size and sedimentation velocity of atmospheric particles in water based on image recognition provided by the present invention above. This method is carried out according to the following steps:

[0030] Step 1: Inject the experimental water body, which is clear, transparent and free of any foreign matter, into the sedimentation column and let it stand for 12 hours. To simulate the real water environment, the experimental water can be configured with different salinities according to different water environments such as rivers, lakes, and oceans.

[0031] Step 2: Set the resolution (1080p) and recording frame rate (60 frames) of the camera.

[0032] Step 3: Slowly wash the atmospheric particles pre-collected on the sampling membrane into the sedimentation column.

[0033] Step 4: Adjust the camera focus, turn on the LED light, place the LED light on one side of the sedimentation column, and keep it at the same height as the camera as an external light source. By irradiating the particles with strong light, take a clear video of the particles settling in the water body.

[0034] Step 5: Import the video of the particles settling into the script written in Python language.

[0035] Step 6: Perform grayscale enhancement processing on the particle images of two consecutive frames before and after to obtain the binary images of the particles; identify and track individual particles according to the consistency of the gray values and sizes of adjacent two-frame particles.

[0036] Step 7: Obtain the true length of the particle image based on the binary image of the particle and the resolution of the captured video and the number of pixels occupied by the particle image. Starting from when the particle enters the camera's field of view until it leaves the field of view, measure the length of the particle image in each frame in the vertical, horizontal, and 45° inclined directions. After averaging the lengths in the three directions respectively, then take the geometric mean of the averages of the lengths in the three directions to obtain the representative particle size.

[0037] Step 8: Calculate the instantaneous velocity of the particles at different times according to the displacement and time information of two consecutive frames of particles. Starting from when the particle enters the camera's field of view until it leaves the field of view, measure the instantaneous velocity of the particle at each frame at a specific height in the sedimentation column, sum up the instantaneous velocities of each frame and take the average to obtain the representative sedimentation velocity of the particle at the specific height.

[0038] Step 9: Plot the statistical results of the particle size and sedimentation velocity of each particle as a scatter plot to visually record the relationship between the particle size and sedimentation velocity.

[0039] Example 1

[0040] Determination test of the particle size and sedimentation velocity of atmospheric particles

[0041] 1. Test method

[0042] The test device mainly includes a slide rail, a sedimentation column, a camera equipped with a macro lens, white LED lights, and a particle placement device. The particle placement device includes a wash bottle and a sampling filter membrane collected with atmospheric particles.

[0043] Step 1: Inject clear, transparent and foreign-object-free experimental water into the sedimentation column and let it stand for 12 hours.

[0044] Furthermore, the experimental water can be configured according to the salinity of different water environments such as rivers, lakes, and oceans.

[0045] Step 2: Turn on the camera and set the resolution and recording frame rate of the camera.

[0046] Step 3: Slowly wash the atmospheric particles pre-collected on the sampling membrane into the sedimentation column using the wash bottle and let them settle freely.

[0047] Those skilled in the art should note that since the particles are on the sampling membrane, the sampling membrane should be washed slowly when using the wash bottle to avoid damage to the sampling membrane caused by human factors.

[0048] Step 4: Fix the camera, adjust the camera focal length, and shoot a clear video of the particles settling in the water body.

[0049] Those skilled in the art should note that since the particles are affected by gravity, buoyancy, and viscous force during the sedimentation process, the sedimentation speed of the particles will continue to increase in the initial stage of sedimentation. When it increases to a certain extent, the three forces are in balance, and then the particles will settle at a constant speed. Therefore, the particles should be allowed to settle stably before collecting images to avoid measurement errors caused by unstable sedimentation speed. In this experiment, at a depth of 15 cm from the water surface of the sedimentation column, the sedimentation speed of most particles reached stability.

[0050] Step 5: Import the video file of the particle sedimentation taken into the script written in the Python language.

[0051] Step 6: Run the script, measure the particle size and record the instantaneous sedimentation speed of the particles, and calculate the representative sedimentation speed. Plot it as a scatter plot.

[0052] II. Test Results

[0053] Appendix Figure 2 shows the tracking situation of each particle and the instantaneous sedimentation speed of each particle at different times of the atmospheric particles in the sedimentation column in this method. Appendix Figure 3 shows the schematic diagram of the particle size measurement of the particles in this method. Appendix Figure 4 is the scatter plot of the statistical results of the particles with different particle sizes and sedimentation speeds obtained by this method.

[0054] As mentioned above, this is just a simple application example of the present invention and does not impose any restrictions on the technical application scope of the present invention. Any changes or modifications made in accordance with the claims and the specification of the present invention shall fall within the scope covered by the patent of the present invention.

Claims

1. A method for measuring the particle size and sedimentation velocity of atmospheric particles in water bodies based on image recognition, including a sedimentation experimental device and a video processing system. The sedimentation experimental device consists of a slide rail, a sedimentation column, a camera with a macro lens, a white LED light, and a particle placement device. The video processing system is a particle tracking velocity measurement script independently developed based on the python language. The script can grayscale the captured video, highlight the outline of the particles, and thus identify the particles that are sinking in the video, and calculate and output the sedimentation velocity of the particles based on the change in the position of the particles in the two frames before and after the video. It can also be used to measure the particle size and shape of particles in water bodies. The measurement method is carried out according to the following steps: (1) Inject clear and transparent experimental water without any foreign matter into the sedimentation column and let it stand for 12 hours. To simulate the real water environment, the experimental water can be prepared according to the salinity of different water environments such as rivers, lakes, and oceans; (2) Set the camera resolution and recording frame rate; (3) Slowly eluting the atmospheric particles pre-collected on the sampling membrane into the sedimentation column; (4) Fix the camera, adjust the camera focus, and take clear videos of particles settling in the water; (5) Import the video file of particle sedimentation into a script written in Python; (6) grayscale enhancement processing is performed on the particle images of two consecutive frames to obtain a binary image of the particles; and a single particle is identified and tracked based on the consistency of the grayscale value and size of the particles in two adjacent frames; (7) Based on the binary image of the particles, the image size of the particles in each frame in three directions ( FIG. 3 ) is measured respectively, and the geometric mean of the measurement results in the three directions is taken to obtain the geometric mean particle size of the particles; (8) Calculating the instantaneous velocity of the particle at different times based on the displacement and time information of the particle in two consecutive frames; (9) The statistical results of particle size and sedimentation velocity of multiple particles are plotted as a scatter plot to intuitively record the relationship between particle size and sedimentation velocity.

2. The measuring method according to claim 1, characterized in that The slide rail is made of aluminum alloy, and a rotating shaft is installed in the middle, so that the camera on the slide rail can be raised and lowered slowly and stably.

3. The measuring method according to claim 1, characterized in that The sedimentation column is made of highly transparent organic glass material, and has a length, width and height of 20×20×50 cm.

4. The measuring method according to claim 1, characterized in that The camera with a macro lens is installed on the slide rail. After locking the focus, the field of view is 2×3.5mm. It can shoot the sedimentation video of micron-sized particles with a frame rate of 60 frames per second and a resolution of 1080p.

5. The measuring method according to claim 1, characterized in that The white LED light is placed on one side of the sedimentation column, consistent with the height of the camera, as an external light source. By illuminating the particles with strong light, the particles captured by the camera are clearer.

6. The measuring method according to claim 1, characterized in that The insertion device is composed of a quartz filter membrane for collecting atmospheric particles and a washing bottle. The atmospheric particles sampled on the membrane can be slowly eluted into the sedimentation column through the washing bottle.

7. The measuring method according to claim 1, characterized in that The image recognition system is a script written in Python that can be used to process particle sedimentation videos. The script can grayscale the captured video, highlight the outline of the particles, and thus identify the images of sinking particles.

8. The determination method as claimed in claim 1, wherein the method for measuring the particle size, taking the horizontal direction as an example, can obtain the real size of the particle in the horizontal direction by calculating the ratio of the number of pixels occupied by the particle image in the horizontal direction and the resolution of the video. Similarly, the length in the vertical direction and the direction inclined at 45° can also be obtained. From the moment the particle enters the field of view of the camera until the particle leaves the field of view, the length of the particle image in the horizontal direction is measured frame by frame, and the horizontal particle length of each frame is averaged. Similarly, the average value of the length of the particle image in the vertical direction and the direction inclined at 45° can also be obtained. After obtaining the average length of the particle in the three directions, the three dimensions are geometrically averaged to obtain a representative particle size.

9. The determination method as described in claim 1, wherein the method for measuring the particle settling velocity is to measure the instantaneous velocity of the particle in each frame within the camera field of view from the time the particle enters the camera field of view to the time the particle leaves the field of view, and take the average of the instantaneous velocity of each frame to obtain a representative settling velocity of the particle within the field of view.

Images