Atmospheric particulate monitoring method based on gray spot recognition

Abstract

The invention discloses an atmospheric particulate matter monitoring method based on gray spot recognition, which comprises the following steps: S1, arranging a monitoring camera on atmospheric particulate matter monitoring equipment for photographing particle gray spots on the section of an air port of a paper tape; s2, the particle gray spot picture is input into a gray spot recognition model, normal gray spots or abnormal gray spots are output, and the normal gray spots are regular circular section gray spots with uniform gray levels; and S3, under the condition that the output is abnormal gray spots, determining that the measurement result is abnormal, and under the condition that the output is normal gray spots, performing other processing. The monitoring of the particle gray spots of the paper tape can be used as an auxiliary measurement means for reflecting the abnormal condition of the measurement result, and the particle concentration and the atmospheric component abnormity can be estimated according to the measurement result as a reference or supplement of the beta-ray method measurement result.

Description

Technical Field

[0001] The invention relates to atmospheric particle monitoring equipment, and in particular to an atmospheric particle monitoring method. Background Art

[0002] The commonly used measurement methods for suspended particles in the atmosphere, such as PM2.5 particles with an aerodynamic equivalent diameter less than or equal to 2.5 microns and PM10 particles with an aerodynamic equivalent diameter ≤10 microns, are: weight method, beta ray absorption method, and micro-oscillation balance method. At present, most of my country's environmental protection base stations use the beta ray method.

[0003] Chinese patent document CN110455692A discloses a monitoring mechanism for atmospheric particulate matter concentration, which is composed of a sampling rod and a monitoring host. The sampling rod is used to collect and transport the atmospheric particulate matter into the monitoring host. The monitoring host includes a box, a paper tape motion unit, a flow control device, and a detection device. The monitoring device has an air port motion unit.

[0004] During use, the paper tape passes through the air port movement unit, and atmospheric particles are retained on the paper tape when passing through the paper tape. The detection device is provided with a β-ray radiation source and a β-ray receiving and measuring unit.

[0005] In the process of monitoring the concentration of atmospheric particulate matter at environmental protection base stations, it was found that the measurement results of some measuring instruments were occasionally obviously inaccurate. During the troubleshooting process, the inventors found that it was not entirely due to problems with the instruments themselves. The reasons were varied, such as failure of the air outlet paper tape seal, impurities in the air, etc.

[0006] In the process of seeking a solution to the problem, the inventor accidentally discovered that there are cases where the particle dust spots on the air port cross section of the paper tape with abnormal measurement results are crescent-shaped, while normally the particle dust spots on the air port cross section are perfect circles. Through analysis, it is believed that the air port and the paper tape are not completely compacted, resulting in air leakage.

[0007] The inventors also realized that under normal circumstances, the gray spots of atmospheric particulate matter will be evenly distributed on the circular cross-section of the air port, with different gray levels. In theory, the gray spots on the paper tape under other abnormal detection results may also have abnormal performance. For this reason, the paper tape with abnormal monitoring results was investigated and verified, and it was found that this was indeed the case. For example, there were abnormal measurement results caused by uneven gray levels of gray spots of particles, and abnormal measurement results corresponding to yellowish gray spots. Therefore, the abnormal measurement results can be monitored from the perspective of monitoring the gray spots on the paper tape as a supplement to the existing atmospheric particulate matter monitoring equipment or methods. Summary of the invention

[0008] The purpose of the present invention is to provide an atmospheric particulate matter monitoring method based on gray spot recognition, which can be used as an auxiliary means of measurement by monitoring the particle gray spots on a paper tape to reflect abnormal measurement results.

[0009] To this end, the present invention provides an atmospheric particulate matter monitoring method based on gray spot recognition, comprising the following steps: S1. On the atmospheric particulate matter monitoring equipment, a monitoring camera is arranged to take pictures of the particle gray spots on the air port cross section of the paper tape; S2. The particle gray spot photos are input into a gray spot recognition model, and normal gray spots or abnormal gray spots are output, wherein normal gray spots are gray spots with a perfect circular cross section, and gray spots of other shapes are abnormal gray spots; S3. When the output is an abnormal gray spot, it is determined that the measurement result is abnormal, wherein the gray spot recognition model is pre-trained.

[0010] The present invention can be used as an auxiliary measurement means to reflect abnormal measurement results by monitoring the particle gray spots on the paper tape, and can also estimate the particle concentration based on this as a reference for the measurement results of the β-ray method.

[0011] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 It is a flow chart of the atmospheric particulate matter monitoring method based on gray spot recognition of the present invention; Figure 2 is a schematic diagram of a paper tape with different gray spots of the present invention; Figure 3 It is a schematic diagram of the arrangement of the surveillance camera of the present invention. DETAILED DESCRIPTION

[0013] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0014] Figure 2 A schematic diagram of different gray spots on the paper tape is shown, such as Figure 2 As shown in the figure, from left to right, the gray spots are: normal gray spots, circular gray spots with light gray spots on the outside and dark gray spots on the inside, half-moon gray spots, and completely light gray spots. The last three are abnormal gray spots. Circular gray spots with light gray spots on the outside and dark gray spots on the inside represent that the air port cross section is not tightly sealed and leaking all around. Half-moon gray spots represent that half of the air port cross section is leaking. Light gray spots with the same color as the paper tape correspond to complete failure of the paper tape pressing.

[0015] Based on the discovery that abnormal measurement results are usually accompanied by abnormal gray spots on paper tape, the present invention provides an atmospheric particulate matter monitoring method based on gray spot recognition, such as Figure 1 As shown, the method includes the following steps S1-S3.

[0016] S1. A monitoring camera is provided on the atmospheric particulate matter monitoring device to take pictures of the particle spots on the air port cross section of the paper tape; S2, using the gray spot photo of the particles to input into the gray spot recognition model, and outputting normal gray spots or abnormal gray spots, wherein the normal gray spots are gray spots with a perfect circular cross section; S3. When the output is abnormal gray spots, determine that the measurement result is abnormal; when the output is normal gray spots, perform other processing.

[0017] When the output is normal gray spots, other processing includes: S31. When the output is a normal gray spot, extract the three primary color values ​​of the gray spot. If the three primary color values ​​exceed the threshold, it is determined that the atmospheric composition is abnormal.

[0018] S32, when the output is a normal gray spot, calculate the uniformity of the gray spot, divide the gray spot from the center to the periphery into several areas, calculate whether the grayscale of each area is uniform, and if it exceeds the threshold, determine that the measurement result is abnormal; S33. When the output is a normal gray spot, the gray value of the gray spot is extracted, and the particle concentration is estimated according to the gray value and concentration value relationship curve as a reference for the β-ray measurement result. As a special case, if the β-ray measurement result obviously exceeds the estimated concentration (for example, 1.3 times), the measurement result can be determined to be abnormal.

[0019] In the present invention, since the depth of the gray spots is gradually formed by atmospheric particles trapped on the paper tape, the concentration of atmospheric particles is related to the grayscale value of the gray spots. The relationship curve between the grayscale value of the gray spots and the particle concentration value can be obtained through experiments.

[0020] In the present invention, the color of the gray spots may not be gray, but may vary due to atmospheric pollutants, such as yellowish. The present invention extracts the grayscale value of the gray spots to avoid color influence.

[0021] In the present invention, the color of the paper tape used is assumed to be snow-white, which does not affect the gray spot grayscale extraction.

[0022] In the present invention, the gray spot recognition model is first trained by labeled samples. Samples are first collected, and the samples are granular gray spots on paper tapes. Images of normal gray spots and abnormal gray spots are first collected, and then labeled. After passing the test, a usable gray spot recognition model is obtained.

[0023] When the gray spot recognition model is used, the recognition result can be outputted by directly inputting data into the gray spot recognition model.

[0024] At the present stage of the present invention, since abnormal gray spots are in the minority and the amount of data is not large, it is sufficient to identify normal gray spots and abnormal gray spots. When enough data is obtained, the gray spots can be further classified, such as abnormal gray spot shape (non-circular cross-section), abnormal gray spot color (containing three primary colors and exceeding the threshold), and abnormal gray spot uniformity type (uneven grayscale, for example, there is a gray spot with light color in the entire circular cross-section, which may be a paper tape problem or a sealing problem).

[0025] The recognition camera of the present invention can be a miniature camera, such as a needle camera, which is arranged in the measurement cavity (such as above the paper tape) or outside the measurement cavity. In this case, there is no strict restriction on the size of the recognition camera. Figure 3 shown.

[0026] When the recognition camera is set in the measurement cavity, the gray spots can be photographed in real time to obtain an estimate of the particle concentration, which is faster than the beta-ray method.

[0027] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for monitoring atmospheric particulate matter based on gray spot recognition, characterized in that: The steps include: S1. A monitoring camera is provided on the atmospheric particulate matter monitoring device to take pictures of the particle spots on the air port cross section of the paper tape; S2, inputting the gray spot photo of particles into the gray spot recognition model, and outputting normal gray spots or abnormal gray spots, wherein normal gray spots are gray spots with perfect circular cross-sections, and gray spots of other shapes are abnormal gray spots; S3. When the output is abnormal gray spots, the measurement result is judged to be abnormal. Among them, the gray spot recognition model is pre-trained.

2. The atmospheric particulate matter monitoring method based on gray spot recognition according to claim 1 is characterized in that: When the output is a normal gray spot, the gray value of the gray spot is extracted, and the particle concentration is estimated based on the relationship curve between the gray value and the concentration value.

3. The atmospheric particulate matter monitoring method based on gray spot recognition according to claim 1 is characterized in that: When the output is a normal gray spot, the three primary color values ​​of the gray spot are extracted. If the three primary color values ​​exceed the threshold, it is determined that the atmospheric composition is abnormal.

4. The atmospheric particulate matter monitoring method based on gray spot recognition according to claim 1 is characterized in that: When the output is a normal gray spot, calculate the uniformity of the gray spot, divide the gray spot from the center to the periphery into several areas, and calculate whether the grayscale of each area is uniform. If it exceeds the threshold, the measurement result is judged to be abnormal.

5. The atmospheric particulate matter monitoring method based on gray spot recognition according to claim 1 is characterized in that: The identification camera is placed on one side of the top surface of the paper tape in the measuring cavity and / or is placed outside the measuring cavity to monitor the gray spots on the paper tape moving out of the measuring cavity.

6. The atmospheric particulate matter monitoring method based on gray spot recognition according to claim 1 is characterized in that: The grayscale recognition model is a convolutional neural network, and the training samples of the model include gray spot images and labels, where normal gray spots are 0 and abnormal gray spots are 1.

7. The atmospheric particulate matter monitoring method based on gray spot recognition according to claim 1 is characterized in that: Gray spot images judged as abnormal gray spots are collected as samples, and after re-labeling, the gray spot recognition model is trained.

8. The atmospheric particulate matter monitoring method based on gray spot recognition according to claim 1 is characterized in that: The measurement result is the particle concentration measurement result using the β-ray method.

9. The atmospheric particulate matter monitoring method based on gray spot recognition according to claim 1 is characterized in that: The gray spot recognition model is also used to identify the types of gray spot abnormalities, wherein the types of gray spot abnormalities include: gray spot shape abnormalities, gray spot color abnormalities, and gray spot uniformity abnormalities.

Images