Automatic plastic sorting method using light spectrum

Abstract

The present invention relates to an automatic plastic sorting method using a light spectrum, characterized by comprising: an input step of inputting a plastic article and placing it on a jig; a visible light irradiation step of irradiating and transmitting visible light onto the plastic article; a shooting step of dispersing the visible light transmitted through the plastic article by wavelength and then photographing to form an image; a first spectrum analysis step of analyzing the spectrum of the transmitted visible light using the image to determine the material of the plastic article; and a sorting step of classifying the plastic article into different collection bins according to its material.

Description

Technology Field

[0001] The present invention relates to an automatic plastic classification method using a light spectrum, and more specifically, to an automatic plastic classification method using a light spectrum that can automate separate collection by automatically classifying the type of plastic by measuring the value of the spectrum change that occurs as light emitted from a light source passes through the plastic. Background Technology

[0002] With the rapid progress of industrialization, modern society is seeing a rise in consumer markets based on mass production and mass consumption activities, and consequently, the generation of various types of waste is increasing rapidly.

[0003] As such waste is a major cause of environmental pollution, interest in the collection and utilization of recyclables is increasing as one of the active efforts to reduce it.

[0004] Among the household waste generated in typical households, recyclable items mainly include waste paper, glass bottles, cans, plastics, and metals. Of these, plastics are known as a major culprit of environmental pollution because, while their usage is gradually increasing due to their distribution in various container forms, their natural decomposition rate is significantly slow.

[0005] Currently, plastic (PET) bottles are required to be separated by law. However, the separation process of recycling participants and the current systems of collection and recycling companies are considerably insufficient to achieve the intent of policies aimed at increasing the recycling rate of waste containers.

[0006] According to the current form of waste separation, waste generators separate items such as PET bottles using recycling bins provided within apartment complexes; however, this system is currently only operated in large apartment complexes, and there is a limitation in that it is difficult to use these bins in general single-family residential areas.

[0007] Furthermore, due to the inconvenience of separate waste collection, a significant number of waste generators are illegally discarding PET bottles on the streets or throwing them into landfill bags along with general waste.

[0008] In addition, plastics must be sorted and collected separately according to the materials used, but there was a problem in that it was difficult for ordinary consumers to perform this sorting based on material. Prior art literature

[0009] Korean Patent Registration No. 10-2133671 The problem to be solved

[0010] The objective of the present invention, which aims to solve the aforementioned problems, is to provide an automatic plastic sorting method using a light spectrum that analyzes the material of a plastic item when it is inserted, automatically sorts it, and separates and disposes of it in a designated collection bin.

[0011] In addition, another objective of the present invention is to provide an automatic plastic classification method using a light spectrum that can perform plastic classification quickly and accurately by identifying the plastic material using visible light and then additionally analyzing near-infrared light if it is unclear.

[0012] In addition, another objective of the present invention is to provide an automatic plastic classification method using a light spectrum that can improve the accuracy of classifying plastics using visible light by storing data measured by visible light and near-infrared light, respectively, and matching them. means of solving the problem

[0013] The automatic plastic sorting method using the spectrum of light according to the present invention for solving the above problem is characterized by comprising: an input step of inputting a plastic article and placing it on a jig; a visible light irradiation step of irradiating and transmitting visible light onto the plastic article; a shooting step of dispersing the visible light transmitted through the plastic article by wavelength and then photographing to form an image; a first spectrum analysis step of analyzing the spectrum of the transmitted visible light using the image to determine the material of the plastic article; and a sorting step of classifying the plastic article into different collection bins according to its material.

[0014] In addition, the automatic plastic classification method using the light spectrum of the present invention is characterized by further including, after the first spectrum analysis step, a near-infrared irradiation step of irradiating and transmitting near-infrared rays to the plastic article, and a second spectrum analysis step of converting the near-infrared rays transmitted to the plastic article into a digital signal and generating a spectrum of near-infrared rays to determine the material of the plastic article.

[0015] In addition, the first spectrum analysis step of the automatic plastic classification method using the light spectrum of the present invention is characterized by converting the image into grayscale, calculating the average brightness of each pixel column representing the intensity by wavelength to generate spectrum data in the form of a one-dimensional array, mapping the reference pixel position using the input reference wavelength, and performing polynomial curve fitting to convert the pixel value into an actual wavelength value.

[0016] In addition, the first spectrum analysis step of the automatic plastic classification method using the light spectrum of the present invention is characterized by determining the material of the plastic article by utilizing the spectral intensity for each wavelength band generated as visible light is transmitted or absorbed according to the material of the plastic article.

[0017] In addition, the second spectrum analysis step of the automatic plastic classification method using the light spectrum of the present invention is characterized by deriving a major absorption peak according to the molecular structure in the near-infrared spectrum of the plastic article and determining the material of the plastic article using the derived major absorption peak. Effects of the invention

[0018] As described above, according to the automatic plastic sorting method using the light spectrum of the present invention, when a plastic item is input, the material of the item is analyzed and then automatically sorted and separated for disposal into a designated collection bin.

[0019] In addition, according to the automatic plastic classification method using the light spectrum of the present invention, plastic classification can be performed quickly and accurately by identifying the plastic material using visible light and then additionally analyzing near-infrared light if it is unclear.

[0020] In addition, according to the automatic plastic classification method using the light spectrum of the present invention, by storing data measured by visible light and near-infrared light respectively and matching them, there is an effect of improving the accuracy of classifying plastics using visible light. Brief explanation of the drawing

[0021] FIG. 1 is a flowchart illustrating an automatic plastic sorting method using a light spectrum according to the present invention. FIG. 2 is a schematic diagram showing the overall configuration of an automatic plastic sorting method using a light spectrum according to the present invention. Specific details for implementing the invention

[0022] The specific features and advantages of the present invention will be described in detail below with reference to the accompanying drawings. Prior to this, if it is determined that a detailed description of the functions and configurations related to the present invention may unnecessarily obscure the essence of the invention, such detailed description will be omitted.

[0023] The present invention relates to an automatic plastic classification method using a light spectrum, and more specifically, to an automatic plastic classification method using a light spectrum that can automate separate collection by automatically classifying the type of plastic by measuring the value of the spectrum change that occurs as light emitted from a light source passes through the plastic.

[0024] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0025] FIG. 1 is a flowchart showing an automatic plastic classification method using a light spectrum according to the present invention, and FIG. 2 is a configuration diagram showing the overall configuration of an automatic plastic classification method using a light spectrum according to the present invention.

[0026] As illustrated in FIGS. 1 and 2, the automatic plastic sorting method using the spectrum of light according to the present invention is characterized by comprising: an input step (S10) of inputting a plastic article and placing it on a jig (100); a visible light irradiation step (S20) of irradiating and transmitting visible light to the plastic article; a shooting step (S30) of dispersing the visible light transmitted to the plastic article by wavelength and then photographing to form an image; a first spectrum analysis step (S40) of analyzing the spectrum of the transmitted visible light using the image to determine the material of the plastic article; and a sorting step (S70) of classifying the plastic article into different collection bins (610) according to the material of the plastic article.

[0027] The first spectrum analysis step (S40) is characterized by converting the image to grayscale, calculating the average brightness of each pixel column representing the intensity by wavelength to generate spectrum data in the form of a one-dimensional array, mapping the reference pixel position using the input reference wavelength, and performing polynomial curve fitting to convert the pixel value into an actual wavelength value.

[0028] The first spectrum analysis step (S40) is characterized by determining the material of the plastic article using the spectral intensity for each wavelength band generated as visible light is transmitted or absorbed according to the material of the plastic article.

[0029] The input step (S10) is a step for inputting plastic items to be classified into an inspection location, and is formed so that the label-removed plastic items are mounted on a jig (100).

[0030] At this time, a hopper is formed on the upper part of the jig (100) so that multiple plastic items can be stacked, and the hopper is capable of discharging the stacked plastic items one by one so that only one of them can be mounted on the jig (100).

[0031] The jig (100) is formed in a plate shape so that a plastic article can be mounted on its upper surface, and it is preferable to form a concave groove in the center so that the plastic article can be kept in a fixed state so that the spherical or bottle-shaped plastic article does not move.

[0032] In addition, when moving plastic items discharged from the hopper to the jig (100), the mounting direction may be configured to be changed according to the shape of the plastic items using a multi-axis robot arm or a transfer.

[0033] The visible light irradiation step (S20) is a step for allowing visible light to pass through a plastic article by irradiating visible light onto a plastic article placed on a jig (100).

[0034] At this time, the light source (400) used a halogen lamp (410) to provide a continuous and stable spectrum in the visible light region (400~750nm), and the halogen lamp (410) is suitable for effectively deriving spectral differences between plastic materials by providing excellent color temperature stability and a wide spectral distribution.

[0035] When visible light is transmitted through a plastic article, some of the light sources (400) are reflected from the surface, some pass through, and some are absorbed internally, so that only light sources (400) of a specific wavelength range can pass through the plastic article.

[0036] The shooting step (S30) is a step for generating an image by separating visible light transmitted through a plastic article into wavelength bands using a diffraction spectrometer (300) and capturing the separated wavelengths with a camera (200).

[0037] The diffraction spectrometer (300) is configured to precisely separate each wavelength constituting light and is suitable for capturing minute differences in the spectrum of light according to the material of the plastic article, and a visible light spectrometer (310) is used to classify wavelengths for visible light.

[0038] By using a diffraction spectrometer (300) with high spectral resolution and sensitivity, reliable data for distinguishing various plastic materials can be obtained.

[0039] The camera (200) captures visible light separated by wavelength in real time through a diffraction spectrometer (300) to create an image, and to reduce image processing, a region of interest (ROI) is set based on the location of the plastic item and the light path of the spectrometer, and the image can be edited so that only the image of the region of interest is processed.

[0040] The first spectrum analysis step (S40) is a step for determining the material of a plastic article using a captured image, and analyzes the visible light spectrum using pixels of the region of interest of the image.

[0041] Visible light separated by wavelength bands through a diffraction spectrometer (300) is output in different colors, and the region of interest can be limited to the color portion generated by the wavelength band of the visible light, and the captured image is generated so that the colors are arranged differently in the horizontal direction.

[0042] In other words, wavelengths are arranged sequentially in the horizontal direction of the image, and only identical wavelengths can exist in the column direction.

[0043] The control unit (500) converts the image into grayscale to reduce distortion due to color and increases calculation efficiency, and calculates the average brightness of pixels arranged in the column direction of the image, and the average brightness of each calculated column is generated as spectrum data in the form of a one-dimensional array.

[0044] Spectrum data generated by the control unit (500) has a higher intensity as the average brightness of each column increases, and to calibrate the spectrum, a reference wavelength is input from the user and polynomial curve fitting is performed by mapping it to the reference pixel position.

[0045] Polynomials play the role of converting pixel values ​​into actual wavelength values, enabling the provision of accurate spectral data.

[0046] Through this, the control unit (500) can visualize spectrum data in real time and save the analyzed data to a file so that it can be used in a classification algorithm for plastic items subsequently analyzed.

[0047] In addition, the control unit (500) stores standard visible light spectrum data that is pre-set for each plastic material, and the standard visible light spectrum data is stored in groups according to the material.

[0048] When the control unit (500) completes the analysis of the visible light spectrum of the plastic article, it compares the measured visible light spectrum with the standard visible light spectrum to determine the material.

[0049] At this time, if the measured visible light spectrum matches the standard visible light spectrum or the similarity is 80% or more, it is determined that it matches the material of the standard visible light spectrum, and the material of the plastic article can be estimated.

[0050] In other words, by comparing the intensity of specific wavelengths of the transmitted visible light spectrum according to the material of the plastic article, it becomes possible to estimate the material of the plastic article.

[0051] The classification step (S70) is a step for moving plastic items to a collection bin (610) corresponding to the material of the plastic item determined in the first spectrum analysis step (S40) and discharging them.

[0052] Plastic items discharged through the sorting step (S70) are configured to be moved via a conveyor belt, and a plurality of collection bins (610) are formed on the side of the conveyor belt, which are independently separated according to the plastic material.

[0053] Additionally, the conveyor belt is provided with a plurality of sorters (600) that discharge plastic items into collection bins (610), and when it reaches a collection bin (610) that matches the material of the plastic item, the sorter (600) rotates so that the plastic item is struck into the collection bin (610) and collected in the collection bin (610).

[0054] Depending on the need, the classifier (600) may also use a robot arm or transfer to classify and discharge plastic items by material.

[0055] In addition, after the first spectrum analysis step (S40), the method further includes a near-infrared irradiation step (S50) for irradiating and transmitting near-infrared rays to a plastic article, and a second spectrum analysis step (S60) for converting the near-infrared rays transmitted to the plastic article into a digital signal and generating a spectrum of near-infrared rays to determine the material of the plastic article.

[0056] In addition, the second spectrum analysis step (S60) is characterized by deriving major absorption peaks according to molecular structure in the near-infrared spectrum of the plastic article and determining the material of the plastic article using the derived major absorption peaks.

[0057] When comparing the measured visible light spectrum with the standard visible light spectrum, if there is a low match or similarity, it is difficult to estimate the material of the plastic article, so a secondary analysis using near-infrared light is performed.

[0058] To this end, the light source (400) uses a near-infrared lamp (420) that irradiates near-infrared light, and the diffraction spectrometer (300) is additionally formed with a near-infrared spectrometer (320) capable of separating the wavelengths of near-infrared light.

[0059] For spectrum analysis using visible light and near-infrared light, the halogen lamp (410) and the near-infrared lamp (420) can be formed so that their positions can be varied toward the center of the plastic article placed on the jig (100).

[0060] That is, in the first spectrum analysis step (S40), the position of the halogen lamp (410) is varied so that it is positioned at the center of one side of the plastic article, and in the second spectrum analysis step (S60), the position of the near-infrared lamp (420) is varied so that it is positioned at the center of one side of the plastic article.

[0061] In addition, the diffraction spectrometer (300) is also formed with a visible light spectrometer (310) and a near-infrared spectrometer (320), and in order to analyze the visible light spectrum, the visible light spectrometer (310) can be positioned at the center of the opposite side of the plastic article, and in order to analyze the near-infrared spectrum, the near-infrared spectrometer (320) can be positioned at the center of the opposite side of the plastic article.

[0062] Accordingly, the light source (400) and the diffraction spectrometer (300) can be positioned to match the plastic article, and can be controlled so that the position is partially varied according to the contamination, damage, or labeling status of the plastic article being photographed by the camera (200) as needed, allowing the light source (400) to be transmitted to the normal area to acquire a spectrum.

[0063] When near-infrared rays penetrate a plastic article, the internal chemical properties based on the molecular structure and composition of the plastic material can be clearly identified, and through near-infrared spectrum analysis, the limitations of information obtained in the visible light region can be compensated for and the internal material properties can be precisely identified.

[0064] The near-infrared irradiation step (S50) is a step for positioning a near-infrared lamp (420) and a near-infrared spectrometer (320) at the center of one side and the other side of a plastic article mounted on a jig (100), and then allowing wavelengths in the near-infrared region to be transmitted to the plastic article through the near-infrared lamp (420).

[0065] When near-infrared rays are irradiated and pass through a plastic article, the transmitted near-infrared rays can acquire near-infrared spectrum data by a near-infrared spectrometer (320), and the near-infrared spectrometer (320) uses FT-IR (Fourier Transform Infrared Spectroscopy).

[0066] FT-IR is a sensor capable of precisely analyzing the major chemical components and bonding structures of plastic materials through high sensitivity and resolution, and can contribute to further enhancing the reliability of plastic classification.

[0067] The second spectrum analysis step (S60) is a step for identifying the plastic material through the main absorption peak using the near-infrared spectrum of the plastic article measured through FT-IR.

[0068] Through FT-IR, the spectrum of near-infrared rays transmitted through a plastic article can be used to measure the major absorption peaks according to the plastic molecular structure.

[0069] For example, PS (polystyrene) exhibits unique vibrational peaks due to its aromatic ring structure, and the main absorption peak is at 1600 cm⁻¹. -1 Stretching vibration of the benzene ring (C=C) at 1500 cm -1 Asymmetric vibration of the benzene ring at 700 cm -1 ~ 800 cm -1 There is a characteristic in which asymmetric bending vibrations of the benzene ring occur.

[0070]

[0071] [ PS Near-infrared data values ​​]

[0072] PET (polyethylene terephthalate) exhibits a unique spectral pattern due to its ester groups (C=O) and aromatic structure, with a major absorption peak at 1730 cm⁻¹. -1 Stretching vibration of the ester group (C=O) at 1240 cm -1 ~ 1100 cm -1 Stretching vibration of the CO bond at 3000 cm -1 There is a characteristic in which stretching vibrations of the CH bond occur in the vicinity.

[0073]

[0074] [ PET Near-infrared Data Values ​​]

[0075] HDPE (High-density Polyethylene) exhibits a relatively simple spectrum due to its simple hydrocarbon structure, and the main absorption peak is at 2915 cm⁻¹. -1 , 2849 cm -1 Stretching vibration of the CH bond at 1465 cm -1 , 720 cm -1 There is a characteristic where bending vibration of the CH2 group occurs.

[0076]

[0077] [ HDPE Near-Infrared Data Values ​​]

[0078] The control unit (500) can identify unique absorption peaks reflecting the molecular structural characteristics of each plastic article through the near-infrared spectrum of PET, PS, and HDPE, and compares the main absorption peaks of the measured near-infrared spectrum with a standard near-infrared spectrum set according to the plastic material.

[0079] At this time, if the major absorption peaks of the measured near-infrared spectrum and the standard near-infrared spectrum match or the similarity is 80% or more, it is determined that it matches the material of the corresponding standard near-infrared spectrum, and the material of the plastic article can be estimated.

[0080] In other words, by comparing the major absorption peaks of the transmitted near-infrared spectrum according to the material of the plastic article, it becomes possible to estimate the material of the plastic article.

[0081] Through the second spectrum analysis step (S60), the material of the plastic article that could not be identified in visible light can be identified a second time, thereby enabling stable material classification.

[0082] As described above, according to the automatic plastic classification method using the light spectrum of the present invention, when a plastic item is input, its material is analyzed and it is automatically classified and separated for disposal in a designated collection bin. Furthermore, by identifying the plastic material using visible light and additionally analyzing near-infrared light if it is unclear, plastic classification can be performed quickly and accurately. Additionally, by storing data measured by visible light and near-infrared light respectively and matching them, the accuracy of classifying plastics using visible light can be improved.

[0083] As described above, although the present invention has been explained with reference to preferred embodiments, those skilled in the art may implement the present invention with various modifications or variations without departing from the technical spirit and scope described in the claims of the present invention. Accordingly, the scope of the present invention should be interpreted by the claims described to include such many variations. Explanation of the symbols

[0084] 100 : Jig 200 : Camera 300: Diffraction spectrometer 310 : Visible light spectrometer 320 : Near-infrared spectrometer 400 : Light source 410 : Halogen lamp 420 : Near-infrared lamp 500 : Control unit 600 : Classifier 610 : Collection box S10: Input stage S20: Visible light irradiation stage S30: Shooting stage S40: First spectrum analysis step S50: Near-infrared irradiation stage S60: Second spectrum analysis step S70: Classification stage

Claims

Claim 1 A method for automatically classifying plastics using a light spectrum, characterized by comprising: an input step of inputting a plastic item and settling it on a jig; a visible light irradiation step of irradiating and transmitting visible light onto the plastic item; a shooting step of dispersing the visible light transmitted through the plastic item by wavelength and then photographing to form an image; a first spectrum analysis step of analyzing the spectrum of the transmitted visible light using the image to determine the material of the plastic item; and a classification step of classifying the plastic item into different collection bins according to its material. Claim 2 A method for automatically classifying plastics using a light spectrum, characterized in that, in claim 1, after the first spectrum analysis step, it further comprises: a near-infrared irradiation step of irradiating and transmitting near-infrared rays to the plastic article; and a second spectrum analysis step of converting the near-infrared rays transmitted to the plastic article into a digital signal and generating a spectrum of near-infrared rays to determine the material of the plastic article. Claim 3 A method for automatically classifying plastics using a light spectrum according to claim 1, wherein the first spectrum analysis step is characterized by converting the image into grayscale, calculating the average brightness of each pixel column representing wavelength-specific intensity to generate spectrum data in the form of a one-dimensional array, mapping the reference pixel position using an input reference wavelength, and performing polynomial curve fitting to convert the pixel value into an actual wavelength value. Claim 4 A method for automatically classifying plastics using a light spectrum, wherein, in claim 1, the first spectrum analysis step determines the material of the plastic article by utilizing the spectral intensity for each wavelength band generated as visible light is transmitted or absorbed according to the material of the plastic article. Claim 5 A method for automatically classifying plastics using a light spectrum, wherein, in claim 2, the second spectrum analysis step derives a major absorption peak according to the molecular structure in the near-infrared spectrum of the plastic article and determines the material of the plastic article using the derived major absorption peak.

Images