Detection equipment and dust concentration detection method

Abstract

The invention provides detection equipment and a dust concentration detection method. The detection device comprises: a housing having a detection cavity and an air inlet hole which are communicated with each other, the air inlet hole being used for conveying to-be-detected gas into the detection cavity; a linear laser; an image acquisition device; a particle counter; the detection device is arranged in the detection cavity and is used for detecting the dust mass concentration of the to-be-detected gas; the processing module is connected with the image acquisition device, the particle counter and the detection device, the processing module has a first working mode and a second working mode, and when the processing module is in the first working mode, the processing module establishes a support vector machine model according to the laser surface image, a count value of the particle counter and a detection value of the detection device; and when the processing module is in the second working mode, the processing module calculates the dust mass concentration of the to-be-detected gas according to the laser surface image and the count value based on a support vector machine model. The dust mass concentration detection device effectively solves the problem that a dust mass concentration detection device in the prior art cannot perform real-time detection.

Description

Technical Field

[0001] The present invention relates to the technical field of dust concentration detection, and in particular to a detection device and a dust concentration detection method. Background Art

[0002] At present, fine dust particles suspended in the air can not only cause respiratory diseases in humans, but also reduce visibility in cities and regions and cause building corrosion. Therefore, the measurement of dust mass concentration in workplaces and urban environments is crucial to people's health and daily life.

[0003] In the prior art, dust mass concentration detection usually uses a dust mass concentration sensor, which can only detect the average dust mass concentration within a certain time range, and does not detect the real-time concentration value in the current area.

[0004] However, for some mine workplaces with strong ventilation, the dust mass concentration in the air changes more dramatically. The difference between the detection method of the dust mass concentration sensor and the actual dust mass concentration in the mine workplace is large, which seriously affects the personal health of the workers. Summary of the invention

[0005] The main purpose of the present invention is to provide a detection device and a dust concentration detection method to solve the problem that the dust mass concentration detection device in the prior art cannot perform real-time detection.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present invention, a detection device is provided for detecting the dust mass concentration of a gas to be detected, and the detection device comprises: a housing having a detection cavity and an air inlet hole which are interconnected, the air inlet hole being used to convey the gas to be detected into the detection cavity; a line laser which is arranged in the detection cavity and is used to emit a line laser; an image acquisition device, wherein the central axis of a camera of the image acquisition device is arranged at a first angle to a laser surface of the line laser, so as to acquire an image of the laser surface of the line laser; and a particle counter which is arranged in the detection cavity and is used to detect the dust mass concentration of a gas to be detected. Dust of different particle sizes is counted; a detection device is arranged in the detection chamber, and the detection device is used to detect the dust mass concentration of the gas to be detected; a processing module, the processing module is connected to the image acquisition device, the particle counter and the detection device, the processing module has a first working mode and a second working mode, when the processing module is in the first working mode, the processing module establishes a support vector machine model according to the laser surface image, the counting value of the particle counter and the detection value of the detection device; when the processing module is in the second working mode, the processing module calculates the dust mass concentration of the gas to be detected according to the laser surface image and the counting value based on the support vector machine model.

[0007] Furthermore, the detection device also includes a light absorbing structure, which is arranged on the cavity wall of the detection cavity to absorb the light beam in the detection cavity; wherein at least part of the light absorbing structure is made of light absorbing material.

[0008] Furthermore, the shell also has a detection port connected to the detection cavity, and the image acquisition device is arranged outside the detection cavity. The detection device also includes: a blocking structure, which is arranged at the detection port to block the detection port; wherein at least part of the blocking structure is made of transparent material, and the image acquisition device captures the laser surface image of the straight line laser located in the detection cavity through the part of the blocking structure.

[0009] Furthermore, the blocking structure includes: a blocking body made of a transparent material; a cylindrical connecting portion, the first end of which extends into the detection cavity, and the blocking body is arranged on the first end; an annular stopper, which is arranged on the second end of the cylindrical connecting portion, and the annular stopper is used to limit the stop with at least part of the outer surface of the shell; wherein at least part of the image acquisition device extends into the inner cavity of the cylindrical connecting portion via the second end.

[0010] Furthermore, the shell also has an air outlet, which is connected to the air inlet through the detection cavity. The detection equipment also includes: a first tubular structure, which is arranged at the air outlet; a suction device, which is connected to the detection cavity through the first tubular structure, and the suction device extracts at least part of the gas in the detection cavity to the outside of the shell, so that the gas to be detected outside the shell flows into the detection cavity through the air inlet; wherein a filter is arranged in the first tubular structure.

[0011] According to another aspect of the present invention, a dust concentration detection method is provided for detecting the dust mass concentration of a gas to be detected. The dust concentration detection method is applicable to the above-mentioned detection device, and the dust concentration detection method comprises:

[0012] Step S1: operating an image detection device of a detection device to collect a laser surface image of a line laser emitted by a line laser;

[0013] Step S2: Obtain the average grayscale value, peak signal-to-noise ratio and count value of the laser surface image, and input the average grayscale value, peak signal-to-noise ratio and count value into the support vector machine model to obtain the required dust mass concentration.

[0014] Furthermore, in step S2, the method for establishing a support vector machine model includes:

[0015] Step S21: Suck a preset volume of gas in the detection chamber to the outside of the housing through a suction device of the detection device;

[0016] Step S22: operating the image detection device to collect a plurality of sample laser surface images of the line laser emitted by the line laser;

[0017] Step S23: the processing module obtains the sample average grayscale value and the sample peak signal-to-noise ratio of each sample laser surface image, the processing module obtains the sample count value of the particle counter corresponding to each sample laser surface image when it is collected, and the processing module obtains the sample detection value of the dust concentration sensor corresponding to each sample laser surface image when it is collected; wherein the sample laser surface image and the sample average grayscale value, sample peak signal-to-noise ratio, sample count value and sample detection value corresponding to the sample laser surface image form a set of sample data;

[0018] Step S24: The processing module selects at least one set of sample data, and uses the sample average grayscale value, sample peak signal-to-noise ratio and sample count value in the sample data as input variables, and the sample detection value in the sample data as output variables, and uses the support vector machine to simulate the functional relationship between the input variables and the output variables to establish a support vector machine model.

[0019] Furthermore, the method for establishing a support vector machine model also includes:

[0020] Step S25: The processing module selects at least another set of sample data, and inputs the sample average grayscale value, sample peak signal-to-noise ratio and sample count value in the set of sample data into the support vector machine model to obtain the predicted dust mass concentration value A1; wherein, if the sample detection value A2 in the set of sample data and the predicted dust mass concentration value A1 satisfy: 1.08A2≤A1≤0.95A2, the support vector machine model is judged to be qualified.

[0021] Furthermore, the method for obtaining the average grayscale value and peak signal-to-noise ratio of the laser surface image includes:

[0022] Step S26: the processing module converts the laser surface image into a grayscale image, and intercepts the central scattering area in the grayscale image;

[0023] Step S27: filtering the central scattering area based on a median filtering algorithm to obtain a filtered image matrix;

[0024] Step S28: Obtain the coordinates of each pixel point in the image matrix to be filtered, and calculate the peak signal-to-noise ratio according to the coordinates of each pixel point. The formula for calculating the peak signal-to-noise ratio is:

[0025]

[0026] Wherein, M is the length value of the central scattering area, N is the width value of the central scattering area, R is the horizontal coordinate and vertical coordinate of the pixel point of the central scattering area before filtering, and F is the horizontal coordinate and vertical coordinate of the pixel point of the central scattering area after filtering.

[0027] Furthermore, the method for obtaining the average grayscale value and peak signal-to-noise ratio of the laser surface image also includes:

[0028] Step S29: performing binarization processing on the central scattering area based on the Otsu binarization method;

[0029] Step S30: performing normalization processing on the binarized central scattering region based on the maximum-minimum normalization method to obtain a scattering image matrix;

[0030] Step S31: performing a Hadamard product on the scattered image matrix and the filtered image matrix to obtain a finished grayscale image;

[0031] Step S32: The processing module obtains the required average grayscale value according to the finished grayscale image.

[0032] According to the technical solution of the present invention, the detection device is used to detect the dust mass concentration of the gas to be detected. The shell of the detection device has a detection cavity and an air inlet that are interconnected. The air inlet is used to transport the gas to be detected into the detection cavity. The straight-line laser is arranged in the detection cavity to emit the straight-line laser. The central axis of the camera of the image acquisition device is arranged at a first angle with the laser surface of the straight-line laser to collect the laser surface image of the straight-line laser. The particle counter is arranged in the detection cavity to count dust of different particle sizes in the gas to be detected. The detection device is arranged in the detection cavity to detect the dust mass concentration of the gas to be detected. The processing module is connected to the image acquisition device, the particle counter and the detection device. The processing module has a first working mode and a second working mode. When the processing module is in the first working mode, the processing module establishes a support vector machine model according to the laser surface image, the count value of the particle counter and the detection value of the detection device; when the processing module is in the second working mode, the processing module calculates the dust mass concentration of the gas to be detected according to the laser surface image and the count value based on the support vector machine model. In this way, the detection equipment can establish a support vector machine model based on the laser surface image, the count value of the particle counter and the detection value of the detection device. After the support vector machine model is established, the staff only needs to continuously deliver the gas to be detected into the detection chamber through the air inlet. The control module can calculate the real-time dust mass concentration of the gas to be detected based on the support vector machine model by using the laser surface image collected in real time by the image acquisition device and the real-time count value of the particle counter, thereby solving the problem that the dust mass concentration detection device in the prior art cannot perform real-time detection and ensuring the personal health of the staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] 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:

[0034] Figure 1 It shows a schematic diagram of the overall structure of an embodiment of a detection device according to the present invention;

[0035] Figure 2 A flow chart of an embodiment of a dust concentration detection method according to the present invention is shown.

[0036] The above drawings include the following reference numerals:

[0037] 10. Shell; 11. Detection cavity; 12. Air inlet; 13. Air outlet; 14. Detection port; 20. Linear laser; 30. Image acquisition device; 40. Particle counter; 50. Detection device; 60. Processing module; 70. Light absorption structure; 80. Sealing structure; 81. Sealing body; 82. Cylindrical connecting part; 83. Annular stopper; 90. First tubular structure; 100. Suction device; 110. Filter element; 120. Second tubular structure; 121. First sub-segment; 122. Second sub-segment. DETAILED DESCRIPTION

[0038] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0039] It should be noted that, unless otherwise specified, all technical and scientific terms used in this application have the same meanings as commonly understood by ordinary technicians in the technical field to which this application belongs.

[0040] In the present invention, unless otherwise specified, the directional words used, such as "up" and "down", usually refer to the directions shown in the drawings, or to the vertical, perpendicular or gravity directions; similarly, for ease of understanding and description, "left" and "right" usually refer to the left and right shown in the drawings; "inside" and "outside" refer to the inside and outside relative to the outline of each component itself, but the above directional words are not used to limit the present invention.

[0041] In order to solve the problem in the prior art that the dust mass concentration detection device in the prior art cannot perform real-time detection, the present application provides a detection device and a dust concentration detection method.

[0042] like Figure 1As shown, the detection device is used to detect the dust mass concentration of the gas to be detected. The detection device includes a housing 10, a straight line laser 20, an image acquisition device 30, a particle counter 40, a detection device 50 and a processing module 60. The housing 10 has a detection cavity 11 and an air inlet 12 that are interconnected. The air inlet 12 is used to transport the gas to be detected into the detection cavity 11. The straight line laser 20 is arranged in the detection cavity 11, and the straight line laser 20 is used to emit a straight line laser. The central axis of the camera of the image acquisition device 30 is arranged at a first angle with the laser surface of the straight line laser, so as to collect the laser surface image of the straight line laser. The particle counter 40 is arranged in the detection cavity 11, so as to count the dust of different particle sizes in the gas to be detected. The detection device 50 is arranged in the detection cavity 11, and the detection device 50 is used to detect the dust mass concentration of the gas to be detected. The processing module 60 is connected to the image acquisition device 30, the particle counter 40 and the detection device 50. The processing module 60 has a first working mode and a second working mode. When the processing module 60 is in the first working mode, the processing module 60 establishes a support vector machine model according to the laser surface image, the count value of the particle counter 40 and the detection value of the detection device 50; when the processing module 60 is in the second working mode, the processing module 60 calculates the dust mass concentration of the gas to be detected according to the laser surface image and the count value based on the support vector machine model.

[0043] The technical solution of this embodiment is applied, and the detection device is used to detect the dust mass concentration of the gas to be detected. The housing 10 of the detection device has a detection cavity 11 and an air inlet 12 that are interconnected. The air inlet 12 is used to transport the gas to be detected into the detection cavity 11. The straight-line laser 20 is arranged in the detection cavity 11 to emit a straight-line laser. The central axis of the camera of the image acquisition device 30 is arranged at a first angle to the laser surface of the straight-line laser to collect the laser surface image of the straight-line laser. The particle counter 40 is arranged in the detection cavity to count the dust of different particle sizes in the gas to be detected. The device 50 is arranged in the detection chamber to detect the dust mass concentration of the gas to be detected. The processing module 60 is connected to the image acquisition device 30, the particle counter 40 and the detection device 50. The processing module 60 has a first working mode and a second working mode. When the processing module 60 is in the first working mode, the processing module 60 establishes a support vector machine model according to the laser surface image, the count value of the particle counter 40 and the detection value of the detection device 50; when the processing module 60 is in the second working mode, the processing module 60 calculates the dust mass concentration of the gas to be detected according to the laser surface image and the count value based on the support vector machine model. In this way, the detection equipment can establish a support vector machine model based on the laser surface image, the count value of the particle counter 40 and the detection value of the detection device 50. After the support vector machine model is established, the staff only needs to continuously deliver the gas to be detected into the detection chamber 11 through the air inlet 12. The control module can calculate the real-time dust mass concentration of the gas to be detected based on the support vector machine model by using the laser surface image collected in real time by the image acquisition device 30 and the real-time count value of the particle counter 40, thereby solving the problem that the dust mass concentration detection device in the prior art cannot perform real-time detection and ensuring the personal health of the staff.

[0044] In this embodiment, the image acquisition device 30 is an industrial camera.

[0045] The industrial camera uses a 4K high-definition USB industrial camera with a C-mount and a manually adjustable aperture and focus lens.

[0046] In this embodiment, the line laser 20 can emit line-shaped laser light, which can form a laser surface.

[0047] Specifically, the line laser 20 adopts a conductor laser, the output laser wavelength is 650nm, the working power is 10mW, and its structure includes a laser diode, a metal shell, a lens, a constant power circuit board, and a high-quality conductor with a DC plug / red and black wires.

[0048] In this embodiment, the central axis of the camera head of the industrial camera is set at 90° to the laser surface of the straight line laser to avoid deflection and tilt of the image collected by the industrial camera, which affects the calculation results of subsequent steps, thereby improving the detection accuracy of the detection equipment.

[0049] In this embodiment, the particle counter 40 is an optical particle counter, which can measure the amount of dust with different particle sizes.

[0050] In this embodiment, the detection device 50 is a dust concentration sensor. Figure 1 As shown, the detection device 50 also includes a light absorbing structure 70, which is arranged on the cavity wall of the detection cavity 11 to absorb the light beam in the detection cavity 11. Among them, at least part of the light absorbing structure 70 is made of light absorbing material. In this way, the above-mentioned arrangement can fully reduce the brightness in the detection cavity 11 to ensure that the line laser emitted by the line laser has a high contrast with other areas in the detection cavity 11, thereby ensuring that the image acquisition device 30 can clearly and completely photograph and collect the laser surface of the line laser, so as to ensure the calculation accuracy of the subsequent steps, thereby improving the detection accuracy of the detection equipment.

[0051] In this embodiment, the light absorbing structure 70 is a light absorbing cloth.

[0052] Optionally, the light-absorbing cloth is bonded to the cavity wall of the detection cavity 11 .

[0053] In this embodiment, the dust concentration sensor is capable of detecting the mass concentration of particles of different particle sizes, including particles of 0.3 microns, 0.5 microns, 1.0 microns, 2.5 microns, 5.0 microns, and 10 microns.

[0054] like Figure 1 As shown, the housing 10 also has a detection port 14 connected to the detection chamber 11, the image acquisition device 30 is arranged outside the detection chamber 11, and the detection device 50 also includes a blocking structure 80, which is arranged at the detection port 14 to block the detection port 14. Among them, at least part of the blocking structure 80 is made of transparent material, and the image acquisition device 30 acquires the laser surface image of the straight line laser located in the detection chamber 11 through the part of the blocking structure 80. In this way, the blocking structure 80 made of transparent material can separate the image acquisition device 30 and the detection chamber 11 while realizing the image acquisition function of the image acquisition device 30, so as to prevent the tiny dust in the gas to be detected from directly contacting the image acquisition device 30, which causes the image acquisition device 30 to be dusted or even damaged, thereby extending the service life of the image acquisition device 30.

[0055] like Figure 1As shown, the blocking structure 80 includes a blocking body 81, a cylindrical connecting portion 82 and an annular stopper 83, and the blocking body 81 is made of a transparent material. The first end of the cylindrical connecting portion 82 extends into the detection cavity 11, and the blocking body 81 is arranged on the first end. The annular stopper 83 is arranged on the second end of the cylindrical connecting portion 82, and the annular stopper 83 is used to limit and stop with at least part of the outer surface of the shell 10. Among them, at least part of the image acquisition device 30 extends into the inner cavity of the cylindrical connecting portion 82 through the second end. In this way, the above-mentioned arrangement reduces the difficulty of disassembly and assembly by the staff through the limit stop between the annular stopper 83 and the shell 10, that is, the staff only needs to extend the cylindrical connecting portion 82 into the detection port 14 until the annular stopper 83 and the shell 10 limit stop, and then it can be determined that the blocking structure 80 is installed in place. At the same time, the cylindrical connecting portion 82 can protect the image acquisition device 30 on the one hand to extend the service life of the image acquisition device 30; on the other hand, it can reduce the distance between the acquisition end of the image acquisition device 30 and the laser surface of the straight-line laser, so that the image acquired by the image acquisition device 30 is clearer and more complete.

[0056] like Figure 1 As shown, the housing 10 also has an air outlet 13, which is connected to the air inlet 12 through the detection chamber 11. The detection device also includes a first tubular structure 90 and a suction device 100, and the first tubular structure 90 is arranged at the air outlet 13. The suction device 100 is connected to the detection chamber 11 through the first tubular structure 90, and the suction device 100 extracts at least part of the gas in the detection chamber 11 to the outside of the housing 10, so that the gas to be detected outside the housing 10 flows into the detection chamber 11 through the air inlet 12. Among them, a filter 110 is arranged in the first tubular structure 90. In this way, the above arrangement, on the one hand, makes the setting position of the suction device 100 more flexible and diverse through the first tubular structure 90, so as to adapt to different working conditions and usage requirements, and also improves the processing flexibility of the staff; on the other hand, the filter element 110 located in the first tubular structure 90 can filter and adsorb the dust in the gas, which can not only prevent the dust from staying in the detection chamber 11 and causing the detection result of the detection equipment to be too large, thereby improving the accuracy of the detection result of the detection equipment, but also prevent the dust from entering the suction device 100 with the gas and causing the suction device 100 to be damaged, thereby extending the service life of the suction device 100. At the same time, since the detection equipment adopts passive air intake (that is, the gas is sucked through the suction device 100, so that the air pressure in the detection chamber 11 is reduced, and then the gas to be detected in the outside is sucked in), the gas is distributed more evenly in the detection chamber 11, further improving the detection accuracy of the detection equipment.

[0057] In this embodiment, the suction device 100 is a suction pump.

[0058] Optionally, the filter element 110 is a filter membrane.

[0059] like Figure 1 As shown, the detection device also includes a second tubular structure 120, which is arranged at the air inlet 12, and the detection cavity 11 is connected to the outside through the second tubular structure 120. Among them, the second tubular structure 120 includes a first sub-segment 121 and a second sub-segment 122 connected to each other, and the first sub-segment 121 and the second sub-segment 122 are arranged at a second angle. In this way, the second tubular structure 120 composed of the first sub-segment 121 and the second sub-segment 122 connected to each other can block the external light beam while realizing the air intake of the air inlet 12, so as to prevent the external light beam from entering the detection cavity 11 through the air inlet 12 and affecting the quality of the image collected by the image acquisition device 30, thereby improving the detection accuracy of the detection device. At the same time, the above-mentioned setting makes the angle between the first sub-segment 121 and the second sub-segment 122 more flexible and diverse, so as to adapt to different working conditions and usage requirements, and also improve the processing flexibility of the staff.

[0060] In this embodiment, the first sub-tube segment 121 is arranged near the air inlet 12, and the particle counter 40 is arranged on the inner wall of the first sub-tube segment 121 near the air inlet 12 to detect the number of particles of different particle sizes in the gas passing through the air inlet 12.

[0061] In this embodiment, the particle counter 40 can detect particles of different sizes including 0.3 microns, 0.5 microns, 1.0 microns, 2.5 microns, 5.0 microns, and 10 microns.

[0062] In this embodiment, the first sub-segment 121 and the second sub-segment 122 are arranged at 90 degrees.

[0063] It should be noted that the value of the second angle between the first sub-segment 121 and the second sub-segment 122 is not limited thereto, and can be adjusted according to working conditions and usage requirements. Optionally, the value of the second angle between the first sub-segment 121 and the second sub-segment 122 is 60°, 70°, 80°, 100°, 110°, or 120°.

[0064] The present application also provides a method for detecting dust concentration, such as Figure 2 As shown, the dust concentration detection method is used to detect the dust mass concentration of the gas to be detected. The dust concentration detection method is applicable to the above-mentioned detection equipment. The dust concentration detection method includes:

[0065] Step S1: operating the image detection device 50 of the detection equipment to collect a laser surface image of a line laser emitted by the line laser 20;

[0066] Step S2: Obtain the average grayscale value, peak signal-to-noise ratio and count value of the laser surface image, and input the average grayscale value, peak signal-to-noise ratio and count value into the support vector machine model to obtain the required dust mass concentration.

[0067] In this embodiment, since the image collected by the image detection device 50 and the count value of the particle counter 40 are both instantaneous values, the average grayscale value and the peak signal-to-noise ratio obtained by the processing module 60 are also instantaneous values, so that the detection equipment can perform real-time calculation of the dust mass concentration of the gas to be detected based on the support vector machine model, which solves the problem in the prior art that the dust mass concentration detection device in the prior art cannot perform real-time detection.

[0068] Specifically, in step S2, the method for establishing a support vector machine model includes:

[0069] Step S21: The suction device 100 of the detection device sucks a preset volume of gas in the detection chamber 11 to the outside of the housing 10;

[0070] Specifically, while inhaling, a preset volume of gas will also enter the housing 10 through the air inlet 12 .

[0071] Step S22: operating the image detection device 50 to collect a plurality of sample laser surface images of the line laser emitted by the line laser 20;

[0072] Step S23: the processing module 60 obtains the sample average grayscale value and the sample peak signal-to-noise ratio of each sample laser surface image, the processing module 60 obtains the sample count value of the particle counter 40 corresponding to each sample laser surface image when it is collected, and the processing module 60 obtains the sample detection value of the dust concentration sensor corresponding to each sample laser surface image when it is collected; wherein the sample laser surface image and the sample average grayscale value, sample peak signal-to-noise ratio, sample count value and sample detection value corresponding to the sample laser surface image form a set of sample data;

[0073] Step S24: The processing module 60 selects at least one set of sample data, and uses the sample average grayscale value, sample peak signal-to-noise ratio and sample count value in the sample data as input variables, and the sample detection value in the sample data as output variables, and uses a support vector machine to simulate the functional relationship between the input variables and the output variables to establish a support vector machine model (actually a prediction model).

[0074] Specifically, the method for establishing a support vector machine model also includes:

[0075] Step S25: The processing module 60 selects at least another set of sample data, and inputs the sample average grayscale value, sample peak signal-to-noise ratio and sample count value in the set of sample data into the support vector machine model to obtain the predicted dust mass concentration value A1; wherein, if the sample detection value A2 in the set of sample data satisfies the predicted dust mass concentration value A1: 1.08A2≤A1≤0.95A2, then the support vector machine model is judged to be qualified. In this way, the above setting provides a verification basis for whether the support vector machine model is accurate, so as to ensure the accuracy of the establishment of the support vector machine model, thereby improving the detection accuracy of the detection equipment.

[0076] Specifically, if the support vector machine model does not satisfy the numerical relationship between the sample detection value A2 and the predicted dust mass concentration value A1, the support vector machine model is judged to be unqualified, and the processing module repeats step S24 and re-establishes the support vector machine model.

[0077] Specifically, the method for obtaining the average grayscale value and peak signal-to-noise ratio of the laser surface image includes:

[0078] Step S26: the processing module 60 converts the laser surface image into a grayscale image, and intercepts the central scattering area in the grayscale image;

[0079] Step S27: filtering the central scattering area based on a median filtering algorithm to obtain a filtered image matrix;

[0080] Step S28: Obtain the coordinates of each pixel point in the image matrix to be filtered, and calculate the peak signal-to-noise ratio according to the coordinates of each pixel point. The formula for calculating the peak signal-to-noise ratio is:

[0081]

[0082] Wherein, M is the length value of the central scattering area, N is the width value of the central scattering area, Ri,j is the horizontal coordinate and vertical coordinate of the pixel point of the central scattering area before filtering, and Fi,j is the horizontal coordinate and vertical coordinate of the pixel point of the central scattering area after filtering.

[0083] Specifically, the method for obtaining the average grayscale value and peak signal-to-noise ratio of the laser surface image also includes:

[0084] Step S29: performing binarization processing on the central scattering area based on the Otsu binarization method;

[0085] Step S30: performing normalization processing on the binarized central scattering region based on the maximum-minimum normalization method to obtain a scattering image matrix;

[0086] Step S31: performing a Hadamard product on the scattered image matrix and the filtered image matrix to obtain a finished grayscale image;

[0087] Step S32: the processing module 60 obtains the required average grayscale value according to the finished grayscale image.

[0088] Specifically, binarization of the central scattering area using the Otsu binarization method can remove pixels that do not contain scattering information to improve subsequent prediction accuracy.

[0089] Specifically, in step S1, before operating the image detection device 50 to collect the laser surface image of the line laser emitted by the line laser 20, the dust concentration detection method further includes:

[0090] The gas in the detection chamber 11 is sucked out of the housing 10 by the suction device 100 of the detection device. In this way, the above arrangement makes the gas to be detected in the detection chamber 11 in a continuous flow process, that is, the external gas to be detected can continuously flow into the detection chamber 11, and because the detection device can detect the instantaneous dust mass concentration of the gas to be detected, the detection value of the detection device can accurately reflect the dust mass concentration of the external gas to be detected.

[0091] Specifically, in step S21, the size of the preset volume is consistent with the volume size of the detection chamber 11. In this way, the above setting can ensure that the detection chamber 11 is filled with the gas to be detected, so as to ensure the detection accuracy of the detection device; on the other hand, after the preset volume of the gas to be detected is introduced, the suction device 100 immediately stops working to ensure that the gas in the detection chamber 11 is in a settled state where the flow stops, thereby ensuring that the detection result of the detection device 50 is relatively accurate, thereby improving the detection accuracy of the detection device.

[0092] Specifically, the support vector machine model obtained by the above method is actually a prediction model, that is, after obtaining the above model, the detection device 50 is no longer used, and only the average grayscale value, peak signal-to-noise ratio and particle count values ​​of different particle sizes are used as input variables to obtain the corresponding dust mass concentration. Although a conventional sensor (detection device 50) that can only measure the average dust mass concentration within a certain time range is used in the training process of the prediction model, the staff can intentionally control the dust mass concentration of the gas introduced into the detection equipment to remain unchanged (that is, the average dust mass concentration within a certain time range is equal to the instantaneous value), so that the predicted value of the prediction model is the corresponding instantaneous value (the average grayscale value, peak signal-to-noise ratio and particle count values ​​of different particle sizes are all instantaneous values).

[0093] From the above description, it can be seen that the above embodiments of the present invention achieve the following technical effects:

[0094] The detection device is used to detect the dust mass concentration of the gas to be detected. The shell of the detection device has a detection cavity and an air inlet that are interconnected. The air inlet is used to transport the gas to be detected into the detection cavity. The straight-line laser is arranged in the detection cavity to emit the straight-line laser. The central axis of the camera of the image acquisition device is arranged at a first angle to the laser surface of the straight-line laser to collect the laser surface image of the straight-line laser. The particle counter is arranged in the detection cavity to count dust of different particle sizes in the gas to be detected. The detection device is arranged in the detection cavity to detect the dust mass concentration of the gas to be detected. The processing module is connected to the image acquisition device, the particle counter and the detection device. The processing module has a first working mode and a second working mode. When the processing module is in the first working mode, the processing module establishes a support vector machine model according to the laser surface image, the count value of the particle counter and the detection value of the detection device; when the processing module is in the second working mode, the processing module calculates the dust mass concentration of the gas to be detected according to the laser surface image and the count value based on the support vector machine model. In this way, the detection equipment can establish a support vector machine model based on the laser surface image, the count value of the particle counter and the detection value of the detection device. After the support vector machine model is established, the staff only needs to continuously deliver the gas to be detected into the detection chamber through the air inlet. The control module can calculate the real-time dust mass concentration of the gas to be detected based on the support vector machine model by using the laser surface image collected in real time by the image acquisition device and the real-time count value of the particle counter, thereby solving the problem that the dust mass concentration detection device in the prior art cannot perform real-time detection and ensuring the personal health of the staff.

[0095] Obviously, the above-described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0096] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0097] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0098] The above description is only a preferred 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, equivalent 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 detection device for detecting the dust mass concentration of a gas to be detected, characterized in that: The detection equipment comprises: A housing (10) having a detection cavity (11) and an air inlet (12) which are interconnected, wherein the air inlet (12) is used to transport a gas to be detected into the detection cavity (11); A word-line laser (20) is arranged in the detection cavity (11), and the word-line laser (20) is used to emit a word-line laser; An image acquisition device (30), wherein a central axis of a camera of the image acquisition device (30) is arranged at a first angle to a laser surface of the line laser, so as to acquire an image of the laser surface of the line laser; A particle counter (40), the particle counter (40) being arranged in the detection chamber (11) and used for counting dust particles of different particle sizes in the gas to be detected; A detection device (50) is arranged in the detection chamber (11), and the detection device (50) is used to detect the dust mass concentration of the gas to be detected; A processing module (60), the processing module (60) is connected to the image acquisition device (30), the particle counter (40) and the detection device (50), the processing module (60) has a first working mode and a second working mode, when the processing module (60) is in the first working mode, the processing module (60) establishes a support vector machine model according to the laser surface image, the count value of the particle counter (40) and the detection value of the detection device (50); when the processing module (60) is in the second working mode, the processing module (60) calculates the dust mass concentration of the gas to be detected based on the support vector machine model according to the laser surface image and the count value.

2. The detection device according to claim 1, characterized in that: The detection device (50) further comprises a light absorbing structure (70), wherein the light absorbing structure (70) is arranged on the cavity wall of the detection cavity (11) so as to absorb the light beam in the detection cavity (11); wherein at least a portion of the light absorbing structure (70) is made of a light absorbing material.

3. The detection device according to claim 1, characterized in that: The housing (10) further comprises a detection port (14) connected to the detection chamber (11); the image acquisition device (30) is arranged outside the detection chamber (11); and the detection device (50) further comprises: A blocking structure (80), the blocking structure (80) being arranged at the detection port (14) for blocking the detection port (14); wherein at least a portion of the blocking structure (80) is made of a transparent material, and the image acquisition device (30) acquires a laser surface image of a line laser located in the detection cavity (11) through the portion of the blocking structure (80).

4. The detection device according to claim 3, characterized in that: The blocking structure (80) comprises: A blocking body (81) made of a transparent material; a cylindrical connecting portion (82), wherein a first end of the cylindrical connecting portion (82) extends into the detection cavity (11), and the blocking body (81) is arranged on the first end; an annular stopper (83), arranged on the second end of the cylindrical connecting portion (82), the annular stopper (83) being used for limiting and stopping with at least part of the outer surface of the housing (10); Wherein, at least a portion of the image acquisition device (30) extends into the inner cavity of the cylindrical connecting portion (82) via the second end.

5. The detection device according to claim 1, characterized in that: The housing (10) further comprises an air outlet (13), wherein the air outlet (13) is connected to the air inlet (12) through the detection cavity (11), and the detection device further comprises: A first tubular structure (90) is arranged at the air outlet (13); A suction device (100), the suction device (100) being connected to the detection chamber (11) via the first tubular structure (90), the suction device (100) extracting at least part of the gas in the detection chamber (11) to the outside of the shell (10), so that the gas to be detected outside the shell (10) flows into the detection chamber (11) through the air inlet hole (12); wherein a filter element (110) is provided in the first tubular structure (90).

6. A dust concentration detection method for detecting the dust mass concentration of a gas to be detected, characterized in that: The dust concentration detection method is applicable to the detection equipment described in any one of claims 1 to 5, and the dust concentration detection method comprises: Step S1: operating the image detection device (50) of the detection equipment to collect a laser surface image of a line of laser light emitted by a line of laser light (20); Step S2: obtaining the average grayscale value, the peak signal-to-noise ratio and the count value of the particle counter (40) of the laser surface image, and inputting the average grayscale value, the peak signal-to-noise ratio and the count value into a support vector machine model to obtain the required dust mass concentration.

7. The dust concentration detection method according to claim 6, characterized in that: In step S2, the method for establishing the support vector machine model includes: Step S21: using the suction device (100) of the detection device to suck a preset volume of gas in the detection chamber (11) to the outside of the housing (10); Step S22: operating the image detection device (50) to collect a plurality of sample laser surface images of a line of laser light emitted by the line laser (20); Step S23: the processing module (60) obtains the sample average grayscale value and the sample peak signal-to-noise ratio of each of the sample laser surface images, the processing module (60) obtains the sample count value of the particle counter (40) corresponding to each of the sample laser surface images when collected, and the processing module (60) obtains the sample detection value of the dust concentration sensor corresponding to each of the sample laser surface images when collected; wherein the sample laser surface image and the sample average grayscale value, sample peak signal-to-noise ratio, sample count value and sample detection value corresponding to the sample laser surface image form a set of sample data; Step S24: The processing module (60) selects at least one group of the sample data, and uses the sample average grayscale value, sample peak signal-to-noise ratio and sample count value in the sample data as input variables, and uses the sample detection value in the sample data as output variables, and uses a support vector machine to simulate the functional relationship between the input variables and the output variables to establish the support vector machine model.

8. The dust concentration detection method according to claim 7, characterized in that: The method for establishing the support vector machine model also includes: Step S25: The processing module (60) selects at least another group of the sample data, and inputs the sample average grayscale value, sample peak signal-to-noise ratio and sample count value in the group of sample data into the support vector machine model to obtain a predicted dust mass concentration value A1; wherein, if the sample detection value A2 in the group of sample data satisfies the predicted dust mass concentration value A1: 1.08A2≤A1≤0.95A2, the support vector machine model is judged to be qualified.

9. The dust concentration detection method according to claim 6, characterized in that: The method for obtaining the average grayscale value and peak signal-to-noise ratio of the laser surface image includes: Step S26: the processing module (60) converts the laser surface image into a grayscale image, and intercepts a central scattering area in the grayscale image; Step S27: filtering the central scattering area based on a median filtering algorithm to obtain a filtered image matrix; Step S28: Obtain the coordinates of each pixel point in the filter image matrix to calculate the peak signal-to-noise ratio according to the coordinates of each pixel point. The formula for calculating the peak signal-to-noise ratio is: Among them, M is the length value of the central scattering area, N is the width value of the central scattering area, R(i, j) is the horizontal and vertical coordinates of the pixel point of the central scattering area before filtering, and F(i, j) is the horizontal and vertical coordinates of the pixel point of the central scattering area after filtering.

10. The dust concentration detection method according to claim 9, characterized in that: The method for obtaining the average grayscale value and peak signal-to-noise ratio of the laser surface image also includes: Step S29: performing binarization processing on the central scattering area based on Otsu binarization method; Step S30: performing normalization processing on the binarized central scattering region based on the maximum-minimum normalization method to obtain a scattering image matrix; Step S31: performing a Hadamard product on the scattered image matrix and the filtered image matrix to obtain a finished grayscale image; Step S32: the processing module (60) obtains the average grayscale value according to the finished grayscale image.

Images