Rapid micro-plastic detection device and method based on backward polarization imaging

Through the detection device and one-dimensional data classification model based on backpolarization imaging, the problems of low accuracy and time-consuming in microplastic detection are solved, and the rapid and accurate identification and classification of microplastics in water bodies are realized, and the detection process is simplified. , improves detection efficiency and accuracy.

CN118937239AActive Publication Date: 2024-11-12TSINGHUA SHENZHEN INTERNATIONAL GRADUATE SCHOOL

Patent Information

Application Number
CN202411045430.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-11-12
Estimated Expiration
2044-08-01

AI Technical Summary

Technical Problem

The existing microplastic detection methods have problems such as low accuracy, long time consumption, complex operation and difficulty in quickly identifying and classification, especially in the rapid and accurate detection of microplastics in water environments.

Method used

Using a detection device based on backpolarization imaging, polarized light in a specific polarization state is generated through a polarization state generator, backscattered polarized light is received by a polarization state receiver, and polarized pixel points are classified in combination with one-dimensional data classification model 1D-ResNet. , realize rapid detection and visual identification of microplastics.

Benefits of technology

Significantly improve the efficiency and accuracy of detection, avoid complex image preprocessing and threshold segmentation, enable measurement of microplastics of different shapes and thicknesses, and do not require slices or staining preprocessing, simplify the detection process and achieve fast and accurate microplastic detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118937239A_ABST
    Figure CN118937239A_ABST
Patent Text Reader

Abstract

The invention relates to a micro-plastic rapid detection device and method based on backward polarization imaging, and the device comprises a polarization state generator which generates polarized light to irradiate micro-plastic in a water body; the polarization state receiver is used for receiving the polarization information of the back scattering polarized light and acquiring a polarization image containing a polarization parameter; the polarization image detection and recognition device is used for splitting the polarization image into a plurality of polarization pixel points, each polarization pixel point corresponds to a one-dimensional vector containing a plurality of elements, and each element value in the one-dimensional vector represents the polarization value of the corresponding polarization pixel point in a plurality of different polarization states, so that the complete polarization characteristic of each polarization pixel point is captured; classifying each polarization pixel point through a one-dimensional data classification model 1D-ResNet to obtain a classification label of each polarization pixel point; and a classification result is mapped back to an image space again, so that rapid detection and visual identification of the water body micro-plastics are realized. Compared with the existing detection technology, the micro-plastic particles in the water body can be detected more quickly and accurately.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of optical measurement technology, and in particular to a device and method for rapid detection of microplastics based on backward polarization imaging. Background Art

[0002] Microplastics are usually defined as plastic particles with a particle size of less than 5 mm, which mainly come from the breakage and degradation of plastic products. In recent years, microplastics, as a new type of man-made pollutant, has become one of the major environmental problems in the world, and its distribution in water bodies has received the focus of scientific community, society and the public. Due to factors such as human emissions and water transportation, the content of microplastics in the aquatic environment has continued to increase, and it is difficult to collect and clean up. Microplastics are easily ingested by aquatic organisms, which has a lasting negative impact on the health of organisms. Since they cannot be excreted from the body through metabolism, microplastics will continue to accumulate in the food chain, causing serious damage to aquatic ecosystems. At the same time, microplastics will also be ingested by the human body through the food chain and drinking water, which may cause a series of health problems and pose a potential threat to human life. Therefore, it is urgent to develop detection methods and devices for microplastics, realize rapid and accurate statistics of the abundance and distribution of microplastics, strengthen the monitoring and management of microplastics in the aquatic environment, and provide strong data support for real-time monitoring of microplastic pollution, clear emission sources, and precise water treatment.

[0003] Microplastic samples in water bodies are generally obtained through sampling and pretreatment. The detection methods for pretreated microplastics include visual method, spectroscopy, thermal analysis, etc. The visual method uses the naked eye or a microscope to observe the particles, count and record their color, morphology and other information. The visual method is simple to operate, but it cannot judge the composition of microplastics and has a low accuracy rate. In addition, since the light source of the traditional projection microscope cannot penetrate thicker samples, the visual method has higher requirements for samples and often requires cumbersome pretreatment operations. The spectroscopy method uses Fourier transform infrared spectroscopy, Raman spectroscopy and other technologies to perform spectral detection and analysis of samples, which has the characteristics of non-destructive and high accuracy. The spectroscopy method takes a long time, so the visual method is usually combined with the spectroscopy method. The suspected microplastic particle samples are first selected by visual method, and then analyzed by spectroscopy. This can improve the detection accuracy of microplastics, but the use of multiple instruments not only tests the professional skills of the operator, but also has high detection costs and is difficult to reduce time. The thermal analysis method thermally cracks microplastic polymers to generate characteristic pyrolysis spectra, which can analyze the chemical composition of microplastics with high accuracy, but it is too time-consuming and cannot be used in rapid measurement scenarios.

[0004] CN116840120A discloses a flow polarization imaging and classification device for suspended particles in water, comprising a flow liquid inlet system, a collection module and a processing module, wherein the collection module comprises a detection chamber, a flash lamp, a polarizer and an image-polarization detection module; the flow liquid inlet system is connected to the detection chamber and is used to inject a sample to be tested into the detection chamber; the flash lamp is used to emit a non-polarized light beam with uniform intensity; the polarizer is used to adjust the non-polarized light beam emitted by the flash lamp to a predetermined polarization state and keep it constant during the sampling process; the image-polarization detection module is used to collect the original image of the sample to be tested in the detection chamber at a fixed frequency; the processing module is used to process the original image to obtain a Stokes image, and simultaneously obtain the polarization information and morphological information of the suspended particles in the sample to be tested according to the Stokes image. However, the device has limitations and shortcomings in achieving rapid and accurate detection and identification of microplastic particles. When extracting and classifying the characteristic information of microplastics from the original polarization images, it is difficult to automatically analyze and identify the unique optical properties of microplastic particles efficiently and accurately. It lacks the ability to quickly and accurately identify and classify microplastic particles in polarization images, and requires more complex image preprocessing and threshold segmentation steps.

[0005] Therefore, there is still an urgent need to develop new solutions to detect microplastic particles more accurately and quickly.

[0006] It should be noted that the information disclosed in the above background technology section is only used for understanding the background of the present application, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the invention

[0007] The main purpose of the present invention is to solve the problems existing in the above-mentioned background technology and to provide a device and method for rapid detection of microplastics based on backward polarization imaging.

[0008] To achieve the above object, the present invention adopts the following technical solutions:

[0009] A microplastic rapid detection device based on backward polarization imaging, comprising:

[0010] A polarization state generator is used to generate polarized light of a specific polarization state to irradiate microplastic samples in water bodies;

[0011] A polarization state receiver, used for receiving polarization information of backscattered polarized light to obtain a polarization image including polarization parameters;

[0012] A polarization image detection and identification device is configured to: split the polarization image into multiple polarization pixel points, each polarization pixel point corresponds to a one-dimensional vector containing multiple elements, and each element value in the one-dimensional vector represents the polarization value of the corresponding polarization pixel point under multiple different polarization states, thereby capturing the complete polarization characteristics of each polarization pixel point; classify each polarization pixel point through a one-dimensional data classification model 1D-ResNet to obtain a classification label for each polarization pixel point; remap the classification result back to the image space to achieve rapid detection and visual identification of each microplastic sample.

[0013] A method for rapid detection of microplastics using the microplastics rapid detection device comprises:

[0014] The polarization state generator generates polarized light of a specific polarization state to illuminate microplastic samples in the water body;

[0015] The polarization state receiver receives polarization information of the backscattered polarized light to obtain a polarization image including polarization parameters;

[0016] The polarization image detection and recognition device splits the polarization image into multiple polarization pixels, each polarization pixel corresponds to a one-dimensional vector containing multiple elements, and each element value in the one-dimensional vector represents the polarization value of the corresponding polarization pixel under multiple different polarization states, thereby capturing the complete polarization characteristics of each polarization pixel; through the one-dimensional data classification model 1D-ResNet, each polarization pixel is classified to obtain a classification label for each polarization pixel; the classification result is remapped back to the image space to achieve rapid detection and visual identification of each microplastic sample.

[0017] The present invention has the following beneficial effects:

[0018] The present invention provides an innovative device and method for rapid detection of microplastics based on backward polarization imaging. In the present invention, a polarization state generator generates polarized light of a specific polarization state and irradiates microplastics in the water body, and a polarization state receiver receives the backscattered polarized light to obtain a polarization map containing polarization parameters, and the polarization image is split into multiple polarization pixels to form a one-dimensional vector, and each pixel is classified by a one-dimensional data classification model 1D-ResNet, thereby realizing pixel-level recognition and classification of microplastic particles. The solution of the present invention not only avoids complex image preprocessing and threshold segmentation, but also realizes rapid detection and visual recognition by remapping the classification results to the image space, significantly improving the efficiency and accuracy of the detection. In addition, the present invention allows the measurement of microplastics of different shapes and thicknesses without the need for preprocessing such as slicing and staining, further simplifying the detection process, making it a rapid and accurate detection technology with practical application value for environmental microplastic pollution problems.

[0019] Other beneficial effects of the embodiments of the present invention will be further described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Schematic diagram of the optical path structure of an embodiment of the present invention.

[0021] Figure 2 4 is a flow chart of a detection algorithm according to an embodiment of the present invention.

[0022] Figure 3 Schematic diagram of the optical path structure of a preferred embodiment of the present invention.

[0023] Figure 4 This is a flow chart of a method for implementing microplastic pixel classification according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following is a detailed description of the embodiments of the present invention. It should be emphasized that the following description is only exemplary and is not intended to limit the scope and application of the present invention.

[0025] It should be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or indirectly connected to the other element. In addition, connection can be used for fixing as well as for coupling or communication.

[0026] It should be understood that the orientation or position relationship indicated by terms such as "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0027] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0028] See also Figures 1 to 4, an embodiment of the present invention provides a microplastic rapid detection device based on backward polarization imaging, comprising: a polarization state generator, used to generate polarized light of a specific polarization state to irradiate microplastic samples in water bodies; a polarization state receiver, used to receive polarization information of backscattered polarized light to obtain a polarization image containing polarization parameters, wherein the polarization parameters may include a Stokes vector, a Mueller matrix and / or a polarization quantity derived therefrom; a polarization image detection and identification device, which is configured to: utilize pixel splitting technology to decompose the polarization image into multiple polarization pixel points, each polarization pixel point corresponds to a one-dimensional vector containing multiple elements, the length of the vector is determined by the measured polarization information depth (i.e., the number of polarization states), and each element value in the one-dimensional vector represents the polarization value of the corresponding polarization pixel point under multiple different polarization states, thereby capturing the complete polarization characteristics of each polarization pixel point; utilize a one-dimensional data classification model 1D-ResNet to classify each polarization pixel point to obtain a classification label for each polarization pixel point; and remap the classification result back to the image space to achieve rapid detection and visual identification of each microplastic sample.

[0029] The present invention uses a backward polarization imaging device to obtain polarization images in the field of view, and uses an innovative rapid detection method to identify and classify microplastic particles at the pixel level, thereby achieving rapid detection and visual identification of millimeter-level and submillimeter-level microplastic samples. In particular, unlike traditional image classification schemes, the present invention uses an innovative pixel classification method to address the problem that microplastic particles have obvious morphological differences, a wide range of particle sizes, and complex composition of particles in real scenes, and that traditional detection methods such as image preprocessing and image segmentation are difficult to achieve rapid and accurate measurement. The present invention innovatively proposes the idea of ​​overall decomposition, splitting the polarization image into several polarization pixels containing polarization information, using the polarization information contained in each pixel for identification and classification, and remapping the classification results back to the image space, to quickly and accurately achieve detection and visual identification of microplastics.

[0030] like Figure 3 As shown, in a preferred embodiment, the polarization state generator includes a collimating lens, a linear polarizer and a quarter wave plate, which are used to form incident light with different polarization states.

[0031] like Figure 3As shown, in a preferred embodiment, the polarization state receiver includes a collimating lens, a non-polarized beam prism, a quarter wave plate and at least two cameras for acquiring Stokes images and Mueller images. Preferably, the polarization state receiver is configured so that the backscattered light from the sample is collected by the collimating lens along the normal direction of the sample surface to form a parallel light beam again, which is then divided into two light beams with the same polarization state in a 50:50 ratio by the non-polarized beam prism, wherein the transmitted light is collected by one camera after passing through the fixed quarter wave plate, and the reflected light is collected by another camera, and the polarization images taken by the two cameras are processed to obtain the Stokes image of the microplastic sample. Preferably, the Stokes image under different polarization states of incidence can be obtained by rotating the angle of the quarter wave plate at least four times, and then the corresponding Mueller image can be calculated.

[0032] like Figure 1 As shown, in some embodiments, the angle between the polarization state generator and the polarization state receiver can be dynamically adjusted to achieve optimal measurement.

[0033] In a preferred embodiment, the polarization image detection and recognition device maps the obtained classification results to the original image space through color coding; wherein the polarization image detection and recognition device is configured with a predefined color mapping table for corresponding different classification labels to specific colors, so as to achieve intuitive distinction and visual recognition of the background and different types of plastics.

[0034] An embodiment of the present invention also provides a method for rapid detection of microplastics based on backward polarization imaging, which uses the detection device of any of the aforementioned embodiments to achieve rapid detection and visual identification of each microplastic sample.

[0035] The specific embodiments of the present invention are further described below.

[0036] Figure 1 This is a schematic diagram of the optical path structure of the microplastic rapid detection device. The light source can generate a light beam with a specific polarization state through the polarization state generator, and the incident light beam is incident obliquely on the microplastic sample to form a uniformly illuminated area. Subsequently, the scattered light of the sample obtains the image information of the measured sample through the polarization state receiver; the image information is transmitted to the signal processing unit, and the polarization image of the microplastic can be obtained through image analysis. Using algorithms such as image processing, feature parameter extraction, and classification recognition, the quantity, particle size, morphology, color, composition, and other information of different microplastic samples in a field of view can be obtained, and rapid detection of microplastics can be achieved.

[0037] (1) Rapid measurement of polarized images of microplastics. Microplastic particles have obvious morphological differences. Traditional projection imaging cannot penetrate thicker microplastics and must be pre-processed by slicing and staining. Therefore, the present invention designs a device based on backward polarization imaging, which can obtain backward polarized images of microplastics without tedious pre-processing of samples. Figure 1 As shown, the polarization state generator can generate polarized light of a specific polarization state to achieve continuous and stable irradiation of microplastic samples; the polarization state receiver can receive the polarization information of the backscattered polarized light, and obtain the Stokes vector, Mueller matrix or the polarization quantity derived from the image through the signal processing unit to meet the requirements of accurate identification and detection of microplastics.

[0038] Polarization is one of the inherent properties of light. Polarization measurement not only has the optical characteristics of non-contact, non-destructive, and high-resolution. The polarization state of light can be represented by a 4x1 Stokes (S) vector.

[0039]

[0040] Where I represents the light intensity, Q, U and V are polarization components. Q represents the horizontal component IH and the vertical component I V The intensity difference, U represents the 45° linear component I of the light 45 With 135° line component I 135 The intensity difference, V represents the right-handed component of light I R With the left-hand component I L Since the propagation direction of light can be maintained without changing the polarization state when using polarizers and wave plates, the polarization optical device can be compatible with the traditional non-polarization measurement optical path. When the incident polarized light hits the object, the polarization state S of the incident light in After linear transformation, the polarization state S of the scattered light becomes out , this linear transformation is expressed as follows:

[0041] S out =M×S in #(2)

[0042] Where M is a 4×4 Mueller matrix:

[0043]

[0044] The 16 elements of the Mueller matrix contain all the polarization characteristics of suspended particles. Polarized scattered light is very sensitive to the optical properties (including refractive index, absorption, birefringence, dichroism, etc.), morphology (particle size, shape, orientation, arrangement, order, etc.) and microstructure (surface morphology, internal substructure) of objects, especially to the ultra-optical resolution structural features below the subwavelength scale, which can express the rich microstructural information of particles. The Mueller matrix, which can fully describe the polarized optical properties of microplastics, can achieve a detailed classification of microplastic components and quantitatively characterize the microstructural characteristics of microplastics.

[0045] During the measurement process, when the light beam irradiates the microplastic sample, it will produce strong surface reflection, resulting in overexposure and reduced image quality, which will seriously interfere with the measurement of the polarization image of the microplastic. Therefore, by controlling the angle between the polarization state generator and the polarization state receiver, the imaging result will not be distorted. The backward oblique incident angle θ of the device can also be dynamically adjusted to achieve optimal measurement. In addition, the instrument can be measured and calibrated to eliminate the errors of the external environment and the instrument itself to ensure the consistency of the measurement.

[0046] (2) Rapid detection of polarized images of microplastics. Different from traditional image classification methods, the present invention adopts an innovative pixel classification method. Microplastic particles have obvious morphological differences and a wide range of particle sizes. In addition, the composition of particles in real scenes is complex. It is difficult to achieve rapid and accurate measurement using traditional detection methods such as image preprocessing and image segmentation. The innovative method of the present invention adopts the idea of ​​overall decomposition, splitting the polarized image into several polarized pixels containing polarization information, and using the polarization information contained in each pixel for identification and classification.

[0047] The rapid detection algorithm process of the present invention is as follows Figure 2 As shown, firstly, the polarization parameters (Stokes vector, Mueller matrix or polarization quantity derived therefrom) measured by the device are input, and the image is split into pixels containing polarization information, i.e., several one-dimensional vectors. The vector length is determined by the depth of the measured polarization information. Subsequently, a one-dimensional classification model is used to classify these pixels to obtain the type of each pixel. Finally, the classification results are remapped into the image space to visually identify the types of microplastics in different areas of the image. This method decouples microplastic classification from image recognition, thereby reducing the interference of microplastic morphology on recognition and classification. In addition, this method directly inputs the original polarization image measured by the device, and does not require image preprocessing, image segmentation and other operations like traditional image recognition methods, which greatly improves the detection efficiency and realizes rapid and accurate detection of microplastics.

[0048] Examples

[0049] (1) Rapid measurement of polarization images of microplastics. Figure 3is an implementation of the device. The polarization state generator is composed of a collimating lens, a linear polarizer and a quarter wave plate. The LED light source forms a parallel light beam through the collimating lens 1, and then passes through the linear polarizer and the rotatable quarter wave plate in turn to obtain incident light of different polarization states. The illumination beam is incident on the sample obliquely, forming an almost uniform illumination area with a diameter of 5 cm on the sample. The polarization state receiver is composed of a collimating lens, a non-polarized beam prism, a quarter wave plate and two cameras. The backscattered light from the sample is collected by the collimating lens 2 along the normal direction of the sample surface to form a parallel light beam again. Subsequently, it is then divided into two beams of light with the same polarization state in a ratio of 50:50 by the non-polarized beam prism. Among them, the transmitted light is collected by camera 1 after passing through the fixed quarter wave plate, and the reflected light is directly collected by camera 2. The polarization images taken by the two cameras can obtain the Stokes image of the microplastic sample after image processing by the image processing unit. To obtain the Mueller image, it is necessary to rotate the angle of the quarter-wave plate 1 at least four times to obtain the Stokes image under different polarization states, and then calculate the Mueller image under the field of view by formula 2. In practical applications, the polarization image of microplastics can be flexibly obtained according to the requirements of measurement speed and measurement accuracy.

[0050] In order to avoid the influence of reflected light from the sample surface on the measurement, it is necessary to control the angle θ between the polarization state generator and the polarization state receiver. Due to the different surface properties of different microplastic samples, the reflection ability of incident light is different. The angle θ can be changed during measurement to achieve the optimal measurement of the polarization image of microplastics. In order to reduce the instrument error and the interference of the external environment, this method uses air as a standard sample to calibrate the system before measuring the microplastic sample.

[0051] (2) Rapid detection of microplastics using polarization images. Figure 4 As an implementation of this detection method, pixel splitting is combined with a one-dimensional data classification model (1D-ResNet) to achieve rapid and accurate detection of microplastic samples. Specifically, the device first performs Mueller matrix imaging on the sample and obtains images of 16 matrix array elements, each of which has a size of n×m. The image information of this 16×n×m dimension is input into the algorithm, and each pixel of the image is split and processed to obtain a one-dimensional vector of (n×m), and the vector length of each pixel is 16. After extracting all polarized pixels in the image, the 1D-ResNet model is used for pixel classification, and the specific labels of each pixel in t categories are obtained. After all pixels are classified, they are remapped to the original n×m image space to obtain a visual classification result, and the background and plastic, as well as different types of plastics, are visually identified according to the colors of different labels to achieve rapid detection of each microplastic sample.

[0052] In summary, the present invention proposes a rapid detection device and method for microplastics based on backward polarization imaging. In order to better meet the detection needs of microplastic samples of different morphologies and thicknesses, the device of the present invention measures the polarization image of the sample, and the incident light is incident obliquely to the sample to form a uniformly illuminated area. The sample can be measured without pretreatment methods such as slicing and staining, which greatly improves the detection efficiency; in particular, the method of the present invention does not require pretreatment and threshold segmentation, but first splits the polarization image into several polarization pixels, and classifies the pixels containing polarization information, and then remaps the classification results to the image space, thereby realizing rapid detection and visual identification of microplastics. The present invention splits the polarization image into pixels containing polarization information, that is, several one-dimensional vectors. The vector length is determined by the measured polarization information depth. Subsequently, a one-dimensional classification model is used to classify these pixels to obtain the type of each pixel. Finally, the classification results are remapped to the image space, and the types of microplastics in different areas of the image are visually identified. The present invention decouples microplastic classification from image recognition, thereby reducing the interference of microplastic morphology on recognition and classification. In addition, the original polarization image measured by the direct input device of the present invention does not require image preprocessing, image segmentation and other operations like traditional image recognition methods, which greatly improves the detection efficiency and realizes rapid and accurate detection of microplastics.

[0053] The above content is a further detailed description of the present invention in combination with specific / preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, without departing from the concept of the present invention, it can also make several substitutions or modifications to these described embodiments, and these substitutions or modifications should be regarded as belonging to the protection scope of the present invention. In the description of this specification, the description of the reference terms "an embodiment", "some embodiments", "preferred embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in combination with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In the absence of mutual contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples. Although the embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions and modifications can be made herein without departing from the scope of protection of the patent application.

Claims

1. A rapid detection device for microplastics based on backward polarization imaging, characterized in that: include: A polarization state generator is used to generate polarized light of a specific polarization state to illuminate microplastics in water bodies; A polarization state receiver, used for receiving polarization information of backscattered polarized light to obtain a polarization image including polarization parameters; A polarization image detection and identification device is configured to: split a polarization image into multiple polarization pixel points, each polarization pixel point corresponds to a one-dimensional vector containing multiple elements, and each element value in the one-dimensional vector represents the polarization value of the corresponding polarization pixel point under multiple different polarization states, thereby capturing the complete polarization characteristics of each polarization pixel point; classify each polarization pixel point through a one-dimensional data classification model 1D-ResNet to obtain a classification label for each polarization pixel point; remap the classification result back to the image space to achieve rapid detection and visual identification of microplastics.

2. The microplastic rapid detection device according to claim 1, characterized in that: The polarization state generator comprises a collimating lens, a linear polarizer and a quarter wave plate, and is used for forming incident light of different polarization states.

3. The microplastic rapid detection device according to claim 1 or 2, characterized in that: The polarization state receiver includes a collimating lens, a non-polarized beam prism, a quarter wave plate and at least two cameras for acquiring a Stokes image and a Mueller image.

4. The microplastic rapid detection device according to claim 3, characterized in that: The polarization state receiver is configured so that the backscattered light from the sample is collected by the collimating lens along the normal direction of the sample surface to form a parallel light beam again, which is then divided into two light beams with the same polarization state in a 50:50 ratio by the non-polarized beam prism, wherein the transmitted light is collected by one camera after passing through the fixed quarter-wave plate, and the reflected light is collected by another camera, and the polarization images taken by the two cameras are processed to obtain the Stokes image of the microplastic sample.

5. The microplastic rapid detection device according to claim 4, characterized in that: By rotating the angle of the quarter wave plate at least four times, Stokes images under different polarization states are obtained, and then the corresponding Mueller images are calculated.

6. The microplastic rapid detection device according to any one of claims 1 to 5, characterized in that: The angle between the polarization state generator and the polarization state receiver is dynamically adjusted to achieve optimal measurement.

7. The microplastic rapid detection device according to any one of claims 1 to 6, characterized in that: The polarized image detection and recognition device maps the obtained classification results to the original image space through color coding; wherein the polarized image detection and recognition device is configured with a predefined color mapping table for corresponding different classification labels to specific colors, so as to achieve intuitive distinction and visual recognition of the background and different types of plastics.

8. The microplastic rapid detection device according to any one of claims 1 to 7, characterized in that: The polarization parameters of the polarization image include the Stokes vector, the Mueller matrix and / or the polarization quantity derived therefrom.

9. A method for rapid detection of microplastics using the microplastics rapid detection device according to any one of claims 1 to 8, characterized in that: include: The polarization state generator generates polarized light of a specific polarization state to illuminate microplastics in the water; The polarization state receiver receives polarization information of the backscattered polarized light to obtain a polarization image including polarization parameters; The polarization image detection and recognition device splits the polarization image into multiple polarization pixels, each polarization pixel corresponds to a one-dimensional vector containing multiple elements, and each element value in the one-dimensional vector represents the polarization value of the corresponding polarization pixel under multiple different polarization states, thereby capturing the complete polarization characteristics of each polarization pixel; through the one-dimensional data classification model 1D-ResNet, each polarization pixel is classified to obtain a classification label for each polarization pixel; the classification result is remapped back to the image space to achieve rapid detection and visual identification of microplastics.

10. The rapid detection method for microplastics according to claim 9, characterized in that: It also includes calibrating the system by measuring with air as a standard sample before measurement.

Citation Information

Patent Citations

  • Polarimetric SAR (Synthetic Aperture Radar) image classification method based on SDIT (Secretome-Derived Isotopic Tag) and SVM (Support Vector Machine)

    CN103824084A

  • Method and device for measuring particles by polarized light scattering

    CN104089855A

  • De-scattering imaging method based on Mueller matrix

    CN111122452A

  • Suspended particulate matter Mueller matrix rapid measurement method and device

    CN113218876A

  • Polarization image identification and pixel level classification method

    CN115953629A

Cited By

  • Underwater micro-plastic detection method and system and storage medium

    CN119354891A

  • Image digital processing method based on infrared polarized light imaging

    CN119693599A