Waste sorting method and waste sorting apparatus

Abstract

To provide a waste sorting method that can efficiently and accurately sort waste into plastics, combustibles, or non-combustibles.SOLUTION: A waste sorting method includes: an imaging step of imaging waste A with a hyperspectral camera 2 and generating waste spectral data representing a spectrum of a near infrared ray of the waste A; a first sorting step of determining whether the waste A is plastic on the basis of a correlation in a specific wavelength range between first reference data which represents a spectrum of a near infrared ray of the plastics and the waste spectral data; and a second sorting step of determining whether the waste A is combustible or non-combustible on the basis of a correlation between the second reference data being the spectrum of the near infrared ray of at least one of combustibles and non-combustibles and the waste spectral data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a waste sorting method and a waste sorting device. [Background technology]

[0002] When a large amount of waste is generated during a disaster, it is necessary to efficiently separate and dispose of the waste by material. For example, in Patent Document 1, the waste is photographed with a hyperspectral camera, and the spectral data of the waste obtained from the image is compared with the spectral data prepared in advance for each type of material to determine the material of the waste. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2021-137738 A Summary of the Invention [Problem to be solved by the invention]

[0004] When disposing of a large amount of waste, the disposal costs are estimated by determining the approximate ratio of plastics, combustibles, and non-combustibles, which have different disposal methods. The above-mentioned Patent Document 1 judges the material of the waste, but does not separate the waste into plastics, combustibles, and non-combustibles according to the disposal method. In addition, since the reflectance of plastics varies greatly depending on the wavelength band in the near-infrared spectrum, when plastics are identified using the near-infrared spectrum, they are easily mistaken for combustible or non-combustible materials. SUMMARY OF THE PRESENT EMBODIMENT An object of the present invention is to provide a method and apparatus for sorting waste materials, which can solve the above-mentioned problems and efficiently and accurately sort waste materials into plastics, combustible materials, and non-combustible materials. [Means for solving the problem]

[0005] In order to solve the above-mentioned problems, a first invention is a waste sorting method comprising: an imaging process of photographing waste with a hyperspectral camera and creating waste spectrum data showing the near-infrared spectrum of the waste; a first sorting process of determining whether the waste is a plastic or not based on a correlation between first standard data showing the near-infrared spectrum of plastics and the waste spectrum data in a specific wavelength band; and a second sorting process of determining whether the waste is combustible or non-combustible based on a correlation between second standard data, which is the near-infrared spectrum of at least one of combustible and non-combustible materials, and the waste spectrum data.

[0006] In order to solve the above problem, a second invention is a waste sorting device comprising a hyperspectral camera that photographs waste and a determination device that acquires a near-infrared spectrum of the waste from the hyperspectral camera. The determination device creates waste spectrum data that indicates the near-infrared spectrum of the waste, and determines whether the waste is a plastic based on a correlation in a specific wavelength band between first reference data that indicates the near-infrared spectrum of plastics and the waste spectrum data, and determines whether the waste is a combustible material or a non-combustible material based on a correlation between second reference data that is a near-infrared spectrum of at least one of combustible material and non-combustible material and the waste spectrum data.

[0007] In the present invention, the first reference data is a near-infrared spectrum of a material to be disposed of as plastics, and is preferably an average value of near-infrared spectrum data of multiple types of materials to be disposed of as plastics. In the present invention, the near-infrared spectrum of combustible material in the second standard data is the near-infrared spectrum of a material disposed of as combustible material, and is preferably an average value of near-infrared spectrum data of multiple types of materials disposed of as combustible material. In the present invention, the near-infrared spectrum of non-combustible material in the second reference data is the near-infrared spectrum of a material disposed of as non-combustible material, and is preferably an average of the near-infrared spectra of multiple types of materials disposed of as non-combustible material.

[0008] In the present invention, first, it is determined whether the waste is plastic or not. At this time, by obtaining a correlation between spectra in a wavelength band where the characteristics of the near-infrared spectrum of plastics are prominent, it is possible to accurately determine whether the waste is plastic or not. Then, if the waste is not plastic, it is determined whether the waste is combustible or non-combustible. At this time, since the characteristics of the near-infrared spectrum of combustible materials and the near-infrared spectrum of non-combustible materials are significantly different, the waste can be accurately sorted into combustible materials or non-combustible materials. In this way, in the present invention, the waste can be efficiently and accurately sorted into plastics, combustible materials, or non-combustible materials depending on the disposal method.

[0009] In the first sorting step, it is preferable to determine whether the waste is a plastic or not based on a correlation coefficient between the first reference data and the waste spectrum data. In this configuration, the correlation between the first reference data and the waste spectrum data can be easily obtained. In the first sorting step, it is preferable to determine whether the waste is a plastic or not based on a correlation between the first reference data and the waste spectrum data within a wavelength band of 1000 to 1095 nm, 1200 to 1295 nm, 1400 to 1495 nm, or 1550 to 1645 nm. In each of the above wavelength bands, the near-infrared spectral characteristics of plastics are prominent, so that the waste can be accurately determined to be a plastic. In the second sorting step, it is preferable to determine whether the waste is combustible or non-combustible based on the correlation between the second reference data and the waste spectrum data in the wavelength band range of 1200 to 1595 nm. In each of the wavelength bands, the difference between the near-infrared spectrum of combustible materials and the near-infrared spectrum of non-combustible materials is prominent, so that it is possible to accurately determine whether the waste is combustible or non-combustible. Effect of the Invention

[0010] The waste sorting method and sorting device of the present invention first determine whether the waste is plastic or not based on the near-infrared spectrum in a specific wavelength band, thereby enabling the waste to be efficiently and accurately sorted into plastics, combustibles, or non-combustibles depending on the disposal method. [Brief description of the drawings]

[0011] [Figure 1] 1 is a configuration diagram showing a sorting device according to an embodiment of the present invention. [Diagram 2] FIG. 4 is a diagram showing first reference data and second reference data according to the embodiment of the present invention. [Diagram 3] 1 is a table showing correlation coefficients between first reference data and near-infrared spectra of plastics for each wavelength band in an embodiment of the present invention. [Figure 4] 1 is a table showing the percentage of plastics that fall within the range of correlation coefficients in an embodiment of the present invention. [Diagram 5] 1 is a table showing the sorting performance of plastics in an embodiment of the present invention. [Figure 6] 1 is a flowchart of a waste sorting method according to an embodiment of the present invention. [Figure 7] FIG. 13 is a diagram showing another embodiment of the present invention, and is a configuration diagram in which a drone equipped with a hyperspectral camera is used. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail with reference to the accompanying drawings. FIG. 1 is a configuration diagram showing a sorting device according to an embodiment of the present invention. The sorting device 1 of this embodiment is a device for sorting waste A shown in FIG. 1 into plastics, combustible materials, or non-combustible materials according to a disposal method, and is equipped with a hyperspectral camera 2 and a determination device 3. The hyperspectral camera 2 photographs the waste A and records a spectrum showing the near-infrared reflectance for each wavelength band.

[0013] The determination device 3 is a computer including a CPU, a RAM, a ROM, a display, etc. Each process of the determination device 3 is realized by the CPU executing a program stored in the ROM or the like. The determination device 3 creates waste spectrum data from the near-infrared spectrum of the waste A acquired from the hyperspectral camera 2. The waste spectrum data is data that pairs wavelength and reflectance, and can be expressed, for example, as a line graph with the horizontal axis representing wavelength and the vertical axis representing reflectance.

[0014] FIG. 2 is a diagram showing the first reference data and the second reference data according to the embodiment of the present invention. A storage unit (not shown) of the determination device 3 stores first reference data and second reference data that are created in advance. The first reference data was created by photographing several types of materials (e.g., acrylic, polystyrene, polyethylene, etc.) disposed of as plastics with hyperspectral camera 2, averaging the near-infrared spectra of each material, and plotting the relationship between the wavelength of the averaged spectrum and reflectance in a line graph (see Figure 2). The second reference data includes combustible matter spectrum data and incombustible matter spectrum data (see FIG. 2). The combustible material spectrum data was obtained by photographing several types of materials disposed of as combustible materials (e.g., wood chips (raw lumber, plywood, etc.), paper, cloth, etc.) with hyperspectral camera 2, averaging the near-infrared spectrum of each material, and then plotting the relationship between the wavelength of the averaged spectrum and the reflectance in a line graph (see Figure 2). The non-combustible material spectrum data was obtained by photographing multiple types of materials (e.g., concrete, stone, boards, metals, etc.) disposed of as non-combustible material with hyperspectral camera 2, averaging the near-infrared spectrum of each material, and then plotting the relationship between the wavelength of the averaged spectrum and reflectance in a line graph (see Figure 2).

[0015] The determination device 3 determines whether or not the waste is a plastic based on the correlation between the first reference data and the waste spectrum data in a specific wavelength band. The determination device 3 of this embodiment calculates a correlation coefficient between the first reference data and the waste spectrum data in a specific wavelength band, and if the correlation is equal to or greater than a reference value, determines that waste A is a plastic. In the determination device 3 of this embodiment, the correlation coefficient between the first reference data and the waste spectrum data is calculated using the following formula 1.

number

[0016] In this embodiment, the correlation coefficient between the first reference data and the waste spectrum data is calculated within the wavelength bands of 1000 to 1095 nm, 1200 to 1295 nm, 1400 to 1495 nm, and 1550 to 1645 nm. 2, in each of the wavelength bands, the characteristics of the first reference data are prominent. Specifically, the slope of the line graph of the first reference data is much greater than in other regions.

[0017] FIG. 3 is a table showing correlation coefficients between the first reference data and near-infrared spectra of sample plastics (thermal insulation, blue tarpaulin) for each wavelength band in an embodiment of the present invention. As shown in the table in Figure 3, it is clear that in the wavelength bands of 1000 to 1095 nm (regions A and B), 1200 to 1295 nm (regions E and F), 1400 to 1495 nm (regions I and J), and 1550 to 1645 nm (regions L and M), the correlation coefficient between the first standard data and the near-infrared spectra of the sample plastics is 0.4 or more.

[0018] FIG. 4 is a table showing the percentage of plastics that fall within the range of correlation coefficients in an embodiment of the present invention. As shown in the table in Figure 4, the percentage of plastics with a correlation coefficient of 1.0 to 0.4 out of the total plastics contained in the waste under test is 66.3%, and the percentage of plastics with a correlation coefficient of 1.0 to 0.2 is 84.8%. In other words, it is clear that the majority of plastics are included if the correlation coefficient is in the range of 1.0 to 0.2. 5 is a table showing the sorting performance of plastics in an embodiment of the present invention. In the table, "67 cases" indicates that the number of wastes with correlation coefficients in the range of 1.0 to 0.4 was 67. As shown in the table in Figure 5, the identification performance for plastics was 91% when the correlation coefficient was in the range of 1.0 to 0.4 (i.e., the rate at which non-plastic waste was mistakenly identified as plastic was 9%), and 86% when the correlation coefficient was in the range of 1.0 to 0.2. It is clear that the identification performance for plastics is high when the correlation coefficient is 0.4 or higher.

[0019] The determination device 3 shown in Figure 1 determines whether waste A is a plastic or not, and if waste A is not a plastic, determines whether waste A is combustible or non-combustible based on the correlation between the second reference data and the waste spectrum data. The determination device 3 of this embodiment calculates a correlation coefficient between the combustible material spectrum data and the waste material spectrum data in a specific wavelength band, and also calculates a correlation coefficient between the incombustible material spectrum data and the waste material spectrum data. Then, in the determination device 3, if the correlation coefficient between the combustible material spectrum data and the waste material spectrum data is greater than the correlation coefficient between the incombustible material spectrum data and the waste material spectrum data, the waste material A is determined to be combustible. Furthermore, in the determination device 3, when the correlation coefficient between the combustible material spectrum data and the waste material spectrum data is equal to or less than the correlation coefficient between the incombustible material spectrum data and the waste material spectrum data, the waste material A is determined to be incombustible. The correlation coefficient between the combustible matter spectrum data and the waste matter spectrum data and the correlation coefficient between the non-combustible matter spectrum data and the waste matter spectrum data are calculated using Equation 1, which was used to calculate the correlation coefficient between the first reference data and the waste matter spectrum data described above. Furthermore, the determination device 3 calculates a correlation coefficient between the second reference data and the waste spectrum data within the wavelength band range of 1200 to 1595 nm. As shown in Fig. 2, the difference in characteristics between the combustibles spectrum data and the non-combustibles spectrum data is evident in the above-mentioned wavelength band. Specifically, in the above-mentioned wavelength band, the combustibles spectrum data has a tendency to have a lower reflectance as the wavelength increases, whereas the non-combustibles spectrum data has a tendency to have a higher reflectance as the wavelength increases.

[0020] Next, a method for sorting waste using the above-mentioned sorting device 1 will be described. FIG. 6 is a flowchart of a waste sorting method according to an embodiment of the present invention, and the following description will refer to the flowchart of FIG. 6 as appropriate.

[0021] (Filming process) 1, waste A is photographed by the hyperspectral camera 2 (step S10). In this embodiment, one piece of waste A is photographed by the hyperspectral camera 2. The hyperspectral camera 2 records a spectrum showing the reflectance of near-infrared light for each wavelength band. The determination device 3 obtains the near-infrared spectrum of the waste A from the hyperspectral camera 2, and creates waste spectrum data indicating the near-infrared spectrum of the waste A.

[0022] (First sorting process) The determination device 3 calculates a correlation coefficient between the first reference data, which indicates the near-infrared spectrum of plastics, and the waste spectrum data within specific wavelength bands (in this embodiment, 1000 to 1095 nm, 1200 to 1295 nm, 1400 to 1495 nm, and 1550 to 1645 nm) (step S20). In the judgment device 3, if the correlation coefficient in each wavelength band is equal to or greater than a reference value (0.4 in this embodiment) (YES in step S30), the judgment device 3 judges that the waste A is a plastic, and sorts the waste A into one to be disposed of as a plastic (step S31).

[0023] (Second sorting process) In the first sorting step, if the waste A is not determined to be plastic (NO in step S30), the determination device 3 calculates a correlation coefficient between the combustible material spectrum data and the waste material spectrum data in the second reference data (step S41).The determination device 3 also calculates a correlation coefficient between the incombustible material spectrum data and the waste material spectrum data in the second reference data (step S42). In the judgment device 3, if the correlation coefficient between the combustible material spectrum data and the waste material spectrum data is greater than the correlation coefficient between the non-combustible material spectrum data and the waste material spectrum data (YES in step S50), the judgment device 3 judges that waste A is combustible and selects waste A to be disposed of as combustible (step S51). Furthermore, in the judgment device 3, if the correlation coefficient between the combustible material spectrum data and the waste material spectrum data is equal to or less than the correlation coefficient between the non-combustible material spectrum data and the waste material spectrum data (NO in step S50), the judgment device 3 judges that the waste A is non-combustible, and selects the waste A to be disposed of as non-combustible (step S52).

[0024] In the above-described waste A sorting method and sorting device 1, first, it is determined whether the waste A shown in Fig. 1 is a plastic or not. At this time, by calculating the correlation coefficient between spectra in a wavelength band where the near-infrared spectrum characteristics of plastics are prominent, it is possible to accurately determine whether the waste A is a plastic or not. Furthermore, if the waste A is not a plastic, it is accurately determined whether the waste A is combustible or non-combustible based on the near-infrared spectrum of combustible materials and the near-infrared spectrum of non-combustible materials. In this manner, the method for sorting waste A and the sorting device 1 of the present embodiment enable accurate sorting into plastics, combustible materials, and non-combustible materials. In the waste A sorting method and sorting device 1 of this embodiment, waste A is sorted into plastics, combustible materials, or non-combustible materials depending on the disposal method, so the cost of disposing of waste A can be calculated more quickly than if the type of material of waste A was detected.

[0025] Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and can be appropriately modified without departing from the spirit of the present invention. FIG. 7 is a diagram showing another embodiment of the present invention, which is a configuration diagram in which a drone equipped with a hyperspectral camera is used. In the sorting device 1 of this embodiment, as shown in Fig. 1, one waste A is photographed by the hyperspectral camera 2, but as shown in Fig. 7, the hyperspectral camera 2 may be mounted on a drone D or installed on a tower or ceiling to photograph multiple wastes A with the hyperspectral camera 2. In this case, the sorting method of this embodiment is used to determine the ratio of plastics, combustibles, and non-combustibles in the multiple wastes A by determining the plastics, combustibles, and non-combustibles in each part of the image data photographed by the hyperspectral camera 2.

[0026] In the second sorting step of this embodiment, the determination device 3 shown in Fig. 1 compares the correlation coefficient between the combustible material spectrum data and the waste material spectrum data with the correlation coefficient between the incombustible material spectrum data and the waste material spectrum data to determine whether the waste material A is combustible or incombustible. However, it is also possible to calculate only the correlation coefficient between the combustible material spectrum data and the waste material spectrum data, and determine the waste material A as combustible if the correlation coefficient is equal to or greater than a preset reference value, and determine the waste material A as incombustible if the correlation coefficient is less than the reference value.

[0027] In the second sorting step of this embodiment, the determination device 3 may determine that the waste A is unknown if the correlation coefficient between the combustible material spectrum data and the waste material spectrum data is less than a preset first threshold value and the correlation coefficient between the incombustible material spectrum data and the waste material spectrum data is less than a preset second threshold value. If the waste A is determined to be unknown, the worker visually checks the waste A and inputs the determination result to the determination device 3. [Explanation of symbols]

[0028] 1. Sorting device 2. Hyperspectral camera 3 Judgment device A. Waste D. Drone

Claims

1. an imaging process of imaging the waste with a hyperspectral camera and creating waste spectrum data showing the near-infrared spectrum of the waste; a first sorting step of determining whether the waste is a plastic based on a correlation between first reference data showing a near-infrared spectrum of plastics and the waste spectrum data in a specific wavelength band; a second sorting step of determining whether the waste is combustible or non-combustible based on a correlation between second reference data, which is a near-infrared spectrum of at least one of combustible and non-combustible materials, and the waste spectrum data; A waste sorting method comprising the steps of:

2. The waste sorting method according to claim 1, characterized in that in the first sorting process, it is determined whether the waste is a plastic or not based on a correlation coefficient between the first reference data and the waste spectrum data.

3. The waste sorting method according to claim 1, characterized in that in the first sorting step, it is determined whether the waste is a plastic or not based on a correlation between the first reference data and the waste spectrum data within at least one wavelength band range of 1000 to 1095 nm, 1200 to 1295 nm, 1400 to 1495 nm and 1550 to 1645 nm.

4. The waste sorting method according to claim 1, characterized in that in the second sorting process, it is determined whether the waste is combustible or non-combustible based on a correlation between the second reference data and the waste spectrum data within a wavelength band range of 1200 to 1595 nm.

5. A hyperspectral camera that photographs waste materials, A determination device that acquires a near-infrared spectrum of the waste from the hyperspectral camera, The determination device includes: creating waste spectrum data showing a near-infrared spectrum of the waste; Based on a correlation between first reference data showing a near-infrared spectrum of plastics and the waste spectrum data in a specific wavelength band, it is determined whether the waste is the plastic, and A waste sorting device characterized by determining whether the waste is combustible or non-combustible based on a correlation between second reference data, which is a near-infrared spectrum of at least one of combustible and non-combustible materials, and the waste spectrum data.

Images