Waste classification processing device

Abstract

To classify waste such as trash in detail.SOLUTION: A waste classification processing device installed outdoors for estimating a type and an amount of waste conveyed by a belt conveyor and classifying and processing the waste, includes a sieving unit that sieves the waste conveyed by the belt conveyor, an estimation unit that applies halogen light to the sieved waste, extracts spectrum data of reflected light of the waste from an image captured by using a hyperspectral camera, and estimates the type and the amount of the sieved waste on the basis of the extracted spectrum data, and a classification unit that sorts the sieved waste and classifies the sorted waste for every type.SELECTED DRAWING: Figure 9

Description

[Technical Field]

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

[0002] In the above technical field, Patent Document 1 describes estimating values ​​indicating the characteristics of the garbage by irradiating the surface of the garbage in a garbage pit with an infrared laser and using a trained model that has learned values ​​indicating the characteristics of the garbage or the classification of the garbage through machine learning. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-183891 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the technology described in Patent Document 1 uses infrared rays to estimate the characteristics and quality of garbage in a room, and is not capable of classifying waste such as garbage in detail. [Means for solving the problem]

[0005] In order to achieve the above object, the waste sorting processing device according to the present invention comprises: A waste sorting processing device that is installed outdoors and estimates the type and amount of waste transported on a belt conveyor and sorts the waste, a screening unit that screens the waste transported on the belt conveyor; an estimation unit that irradiates the sifted waste with halogen light, extracts spectral data of the reflected light from the waste from an image captured using a hyperspectral camera, and estimates the type and amount of the sifted waste based on the extracted spectral data; a sorting unit that sorts the sifted waste and classifies it by type; Equipped with. [Effects of the Invention]

[0006] According to the present invention, waste such as garbage can be classified in detail. [Brief explanation of the drawings]

[0007] [Figure 1] 1 is a schematic side view showing a state in which an installation stand according to a preferred embodiment of the present invention is installed. [Figure 2] 1 is a schematic front view showing a state in which an installation stand according to a preferred embodiment of the present invention is installed. [Figure 3] FIG. 1 is a schematic diagram for explaining how to estimate waste using a hyperspectral camera installed using an installation stand according to a preferred embodiment of the present invention. [Figure 4] FIG. 10 is a front view showing a state in which a modified example of the installation stand according to the preferred embodiment of the present invention is installed. [Figure 5] 1A and 1B are diagrams for explaining the configuration of a frame portion of an installation stand according to a preferred embodiment of the present invention. [Figure 6] 10A and 10B are diagrams for explaining another configuration of the frame portion of the installation stand according to the preferred embodiment of the present invention. [Figure 7] 1 is a perspective view for explaining the configuration of a camera installation section of an installation stand according to a preferred embodiment of the present invention. FIG. [Figure 8] 1 is a front view for explaining the configuration of a camera installation section of an installation stand according to a preferred embodiment of the present invention. FIG. [Figure 9] 1 is a diagram illustrating the overall configuration of a waste classification processing apparatus including an estimation unit using an installation stand according to a preferred embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, embodiments of the present invention will be described in detail by way of example with reference to the drawings. However, the configurations, numerical values, processing flows, functional elements, etc. described in the following embodiments are merely examples, and are open to modification and alteration, and are not intended to limit the technical scope of the present invention to the following description.

[0009] An installation stand for a hyperspectral camera according to a preferred embodiment of the present invention will be described with reference to Figures 1 to 7. The installation stand 1 for the hyperspectral camera 11 is placed outdoors. The installation stand 1 is used to install the hyperspectral camera 11 at a predetermined position for estimating the type and amount of waste 13 transported on a belt conveyor 14. The waste 13 here includes disaster waste (paper, wood, plastic, etc.) generated in large quantities outdoors, mixed industrial waste, etc.

[0010] In recent years, climate change and natural disasters have become more frequent and severe, resulting in the continued generation of large amounts of disaster waste every year. Because the quality (type) and quantity of disaster waste varies depending on the type of disaster and the location of the disaster, it has become necessary to quickly grasp the type and amount of waste generated, as well as to optimize disposal methods according to the characteristics of each disaster.

[0011] In the underlying technology of this embodiment, there is a method using a hyperspectral camera as a way to easily identify the type and amount of waste, but there was no equipment available to use a hyperspectral camera, which is easily affected by ambient light, in outdoor waste disposal sites exposed to sunlight.

[0012] The configuration of the installation frame 1 will be described with reference to Figures 1 and 2. The installation frame 1 is installed in an outdoor waste disposal site or the like. The installation frame 1 is installed so that the frame unit 10 straddles the belt conveyor 14, and the hyperspectral camera 11 to be installed is positioned vertically above the flat belt of the belt conveyor 14. The installation frame 1 is installed at any position on the belt conveyor 14, but is installed downstream of a sieve 17 that sorts out the waste 13 to be transported by the belt conveyor 14. The sieve 17 is a device for removing fine waste 13 from the waste 13 and sorting out waste 13 of a certain size.

[0013] For example, a vibrating sieve or a rotary sieve can be used for sieving. Machines that perform these types of sieving include a vibrating sieving machine and a rotary sieving machine. A vibrating sieving machine (vibrating screen) that performs vibrating sieving is a machine that sieves input materials by vibrating a sieve mesh up and down. A rotary sieving machine (rotary sorter (trommel sorter)) that performs rotary sieving is a machine that sieves by rotating a cylindrical sieve surface.

[0014] The waste 13 sorted by the sieve 17 is transported on the belt conveyor 14. Then, a leveling jig 16 levels the height of the transported lumps of waste 13 to a constant height. The leveling jig 16 is installed at a predetermined height and horizontally relative to the flat belt in the conveying direction of the flat belt, just before the waste 13 placed on the flat belt passes under the frame unit 10. The leveling jig 16 is a rod-shaped (square rod) or plate-shaped jig for leveling the surface of the waste 13.

[0015] The leveling jig 16 has a length equal to or nearly equal to the width of the belt conveyor 14. The leveling jig 16 is provided at a position approximately 122 mm high from the flat belt surface of the belt conveyor 14. Therefore, the height (thickness) of the waste 13 after being leveled by the leveling jig 16 will be a maximum of 122 mm, although there will be some unevenness. The leveling jig 16 may be attached to the belt conveyor 14 side or the installation frame 1 side, for example.

[0016] In this way, by keeping the height of the mass of waste 13 constant, it becomes possible to stably estimate the type and amount of waste 13 using the hyperspectral camera 11 later. Note that the installation position and size of the leveling jig 16 are changed as appropriate depending on the belt conveyor 14 used.

[0017] With the installation stand 1 in place, the installation position, installation height, orientation, etc. of the hyperspectral camera 11 are adjusted so that the entire width of the flat belt of the belt conveyor 14 (direction perpendicular to the conveying direction) fits within the angle of view of the hyperspectral camera 11. The orientation, angle of view, focus, etc. of the hyperspectral camera 11 are adjusted while checking the image displayed on the monitor 18. The hyperspectral camera 11 and monitor 18 are connected directly or indirectly via wired or wireless communication. By adjusting the angle of view, etc. of the hyperspectral camera 11 in this way, it becomes possible to estimate the type and quantity of all of the waste 13 transported on the belt conveyor 14.

[0018] The height position of the hyperspectral camera 11 can be adjusted by a height adjustment mechanism 19 of the installation stand 1. That is, by adjusting the length of the height adjustment mechanism 19 provided on the legs of the installation stand 1, the height position of the upper surface of the installation stand 1 on which the hyperspectral camera 11 is installed can be adjusted, thereby adjusting the height position of the hyperspectral camera 11. Instead of using the height adjustment mechanism 19, the height of the installation stand 1 itself may be adjusted.

[0019] Also, here, the installation stand 1 is configured with three stacked frame sections 10, but instead of providing a height adjustment mechanism 19, the installation height of the hyperspectral camera 11 may be adjusted by the number of stacked frame sections 10.

[0020] Once the adjustment of the hyperspectral camera 11 is complete, the halogen light 12 is shone onto the waste 13 being transported by the belt conveyor 14, and the type and amount of waste 13 are identified based on the light reflected from the waste 13. A blackout curtain 15 is hung on the installation stand 1 to block external light such as sunlight, creating an environment in which only the halogen light from the halogen light 12 is irradiated onto the waste 13. The blackout curtain 15 is sized so that its end reaches a predetermined height position above the flat belt surface of the belt conveyor 14.

[0021] Similarly, a blackout curtain 15 is also hung over the hyperspectral camera 11. By hanging the blackout curtain 15 over the hyperspectral camera 11, it is possible to block external light such as sunlight from entering the camera lens. As a result, the only light that enters the camera lens of the hyperspectral camera 11 is the reflected halogen light from the halogen light 12, making it possible to stably estimate the type and amount of waste 13.

[0022] Here, the halogen lights 12 are fixed to the upper surface of the installation stand 1 (the top of the frame part 10) directly above the belt conveyor 14, facing the belt conveyor 14. The output of the halogen lights 12 is a rated luminous flux of 10,000 lm, and the halogen lights are earthquake-resistant 110V / 500W halogen bulbs. A total of four halogen lights 12 are installed at the four corners of the upper surface of the installation stand 1 with their light-emitting surfaces facing the belt conveyor 14, but the number of halogen lights installed is not limited to this. Furthermore, the inclination of the halogen lights 12 is adjusted so that the light-emitting surfaces of the lights can be irradiated with halogen light at an angle of approximately 30° to 45° with respect to the flat belt surface (horizontal plane) of the belt conveyor 14.

[0023] The waste 13 that flows down the conveyor belt is finally sorted by hand. The classification of the waste 13 that is sorted by hand is based on the acceptance conditions of the waste disposal facility, but if we refer to the "Disaster Waste Disposal Guidelines" issued by the Ministry of the Environment, for example, it is mainly divided into the following eight types (excluding residue that falls after sieving). In other words, the waste 13 is sorted by hand into eight types: "wood scraps," "combustible materials," "non-combustible materials," "scrap metal," "concrete debris," "sorted soil," "recycled materials," and "recycled crushed stone."

[0024] Next, with reference to Figure 3, we will explain how waste 13 is estimated using hyperspectral camera 11. Halogen light emitted from halogen light 12 is irradiated onto waste 13 flowing on belt conveyor 14. The halogen light irradiated onto waste 13 is reflected by the surface of waste 13, and the reflected halogen light enters hyperspectral camera 11.

[0025] The hyperspectral camera 11 can extract spectral data of reflected light for each pixel in a captured image, making it possible to estimate the type of waste 13 from the extracted spectral characteristics. Furthermore, the amount of waste 13 can be estimated based on the number of pixels in the captured image that have the same or similar spectral characteristics. In other words, the area of ​​each waste 13 can be derived based on the collection of pixels that have the same or similar spectral data, and the amount (volume) of each type of waste 13 can be estimated.

[0026] Here, the hyperspectral camera 11 is a type of camera that can precisely analyze the wavelength components (spectrum) of light and can measure the light intensity at a fine wavelength pitch (for example, 5 nm) for each pixel. Because the hyperspectral camera 11 is also a spectroscopic camera, the image captured by the hyperspectral camera contains wavelength information in addition to two-dimensional planar data in the x and y directions. An image with wavelength information added is called a data cube, and a data cube refers to an image in which a two-dimensional planar image in the x and y directions is layered for each spectral wavelength.

[0027] Wavelength information of more than 100,200 bands can be obtained from the image of the hyperspectral camera 11. Therefore, compared to a multispectral camera or the like, the hyperspectral camera 11 can acquire a much larger amount of wavelength data, and the spectral characteristics are clearly displayed, so a precise spectral pattern can be obtained and detailed analysis can be performed.

[0028] Since the spectral pattern (intensity distribution of reflected light) obtained varies depending on the type of waste 13 (e.g., disaster waste), if the spectral characteristics of the waste 13 can be extracted using the hyperspectral camera 11, the type of waste 13 can be accurately estimated.

[0029] A spectrum is the distribution intensity for each wavelength of light (electromagnetic waves), and can be expressed (defined) as a two-dimensional graph with wavelength on the horizontal axis and intensity on the vertical axis. Spectral expression is not limited to two-dimensional display; it can also be expressed (defined) as a vector, for example, as a Spectral Angle Mapper (SAM). In SAM, the target spectrum and image spectrum are expressed as vectors in n-dimensional spectral space. Therefore, the smaller the angle (spectral angle) between these two vectors, the greater the similarity. Using this principle, it is also possible to estimate the type of waste13.

[0030] The waste 13 is leveled by a leveling jig 16 to form a mass at a predetermined height above the surface of the belt conveyor 14 on which the waste 13 is placed. The hyperspectral camera 11 and the halogen light 12 are mounted at a predetermined height above the surface on which the waste 13 is placed. Since the distance between the hyperspectral camera 11 and the waste 13 and the distance between the halogen light 12 and the waste 13 are constant, the type of waste 13 can be reliably estimated. In the illustrated example, the distances from the flat belt surface of the belt conveyor 14 to the hyperspectral camera 11 and the halogen light 12 are equal, but these distances may be different. For example, the halogen light 12 may be positioned below the center of the frame unit 10. In this case, the inclination of the light-emitting surface of the halogen light 12 relative to the flat belt surface (horizontal plane) of the belt conveyor 14 is adjusted to an optimal state.

[0031] By using such an installation stand 1, it becomes possible to easily estimate the type and amount of waste 13 and sort the waste 13 based on the estimated type and amount, regardless of the location, such as an outdoor waste disposal site.

[0032] Next, a modified example of the installation stand will be described with reference to Fig. 4. This figure is a front view of the installation stand 2. The installation stand 2 has a two-tiered structure consisting of a lower frame section 40 and an upper frame section 41. The lower frame section 40 is a wide frame section, and the upper frame section 41 is a narrower frame section than the lower frame section 40. The installation stand 2 has a structure in which the upper frame section 41 is stacked two tiers above the lower frame section 40.

[0033] The upper frame section 41 has a width approximately equal to that of the belt conveyor 14. In contrast, the lower frame section 40 has a width that allows it to straddle the belt conveyor 14. In addition, a height adjustment mechanism 42 is provided on the legs of the lower frame section 40, which makes it possible to adjust the overall height of the installation stand 2. The upper frame section 41 may be attached so as to be fixed to the lower frame section 40 by welding or the like. Alternatively, a detachment mechanism may be provided on the lower ends of the legs of the upper frame section 41 so that the upper frame section 41 can be detachably attached to the lower frame section 40.

[0034] The structure of the frame section 10 (40, 41) will be described with reference to Figure 5. The frame section 10 (40, 41) is composed of vertical frame members 21 and horizontal frame members 22. The frame section 10 has a structure in which four vertical frame members 21 are arranged at the four corners, and four horizontal frame members 22 are arranged at one end of the vertical frame members 21 to connect the vertical frame members 21. The frame section 10 has braces on the side portions to provide strength in the lateral direction. The vertical frame members 21 and horizontal frame members 22 are pipe-shaped (cylindrical) members to reduce weight.

[0035] Furthermore, the vertical frame members 21 may be extendable. If the vertical frame members 21 themselves are extendable in this way, there is no need to provide the height adjustment mechanisms 19, 42. However, it is also possible to perform rough height adjustment by adjusting the length of the vertical frame members 21, and perform fine height adjustment by the height adjustment mechanisms 19, 42.

[0036] In addition, an attachment mechanism for detachably attaching the height adjustment mechanism 19 may be provided at at least one of both end portions of the vertical frame member 21. Furthermore, a connecting mechanism may be provided at both end portions of the vertical frame member 21 so that the frame member 10 can be connected to another frame unit 10. By using a common mechanism for the attachment mechanism and the connecting mechanism, the frame unit 10 can be used for various purposes. For example, if the frame unit 10 of the installation stand 1 is used as the lowest frame, the height adjustment mechanism 19 can be attached to one end portion, and the other end portion can be used to connect to another frame unit.

[0037] The vertical frame members 21 and horizontal frame members 22 are made of lightweight yet strong materials such as steel, aluminum, resin, or plastic, making it easy to move and carry the frame unit 10.

[0038] The frame unit 10 may be an assembly type member that is assembled by connecting the vertical frame members 21 and the horizontal frame members 22 at the installation site. By using assembly type members in this way, it becomes even easier to move and carry the unit.

[0039] Referring to Figure 6, another configuration of the frame unit 10 (40, 41) will be described. The frame unit 10 (40, 41) further has a leg horizontal frame member 23. The leg horizontal frame member 23 is a frame member that is longer than the horizontal frame member 22 and is connected to the end of the vertical frame member 21. The frame unit 10 having the leg horizontal frame member 23 can be used, for example, as the lowest frame of the frame unit 10 when the frame unit 10 is placed directly on the frame of the belt conveyor 14. When the frame unit 10 is placed directly on the belt conveyor 14, it is preferable to arrange the leg horizontal frame member 23 so that it is perpendicular to the transport direction of the waste 13 on the belt conveyor 14, but it may also be arranged so that it is parallel to the transport direction of the waste 13.

[0040] As shown in the figure, the relatively long leg horizontal frame members 23 are used as the contact points between the frame of the belt conveyor 14 and the frame of the lowest frame, allowing the frame to be stably placed on the belt conveyor 14. Furthermore, because the length of the leg horizontal frame members 23 is longer than the horizontal frame members 22, it is possible to accommodate belt conveyors 14 of various widths (sizes). For example, even when multiple belt conveyors 14 are used side by side, the frame unit 10 can be placed so as to cross the multiple belt conveyors 14. In this case, it is desirable to place the frame unit 10 via cushioning material to absorb vibrations so that the vibrations of the belt conveyor 14 are not transmitted.

[0041] 7 and 8, the configuration of the camera installation unit 30 will be described. The camera installation unit 30 has an installation base 31, a slide mechanism 32 (first slide mechanism), a frame 33, and a slide mechanism 36 (second slide mechanism).

[0042] The camera installation unit 30 is attached to the top of the frame unit 10. The hyperspectral camera 11 is attached to the camera installation unit 30. The camera installation unit 30 is a rectangular parallelepiped (cubic) member formed by combining horizontal and vertical square timbers (prismatic members). The hyperspectral camera 11 is attached to the space inside the camera installation unit 30 with its lens facing toward the belt conveyor 14. A plate-shaped member 37 is attached to one of the four side surfaces of the camera installation unit 30. That is, the top, bottom, and side surfaces of the camera installation unit 30, which are formed by combining square timbers, are open, but the side surface to which the plate-shaped member 37 is attached is closed. In this way, by providing the plate-shaped member 37 and closing one of the side surfaces, a space for installing the hyperspectral camera 11 is secured.

[0043] The frame 33 is a rectangular member composed of short-side members 34 and long-side members 35, on which the camera installation unit 30 and installation base 31 are placed. The short-side members 34 are rectangular members that form a side that is shorter than either the longer side of the width or the length of the frame 10. The long-side members 35 are members that form a side that is the same length as either the width or the length of the frame 10.

[0044] The short-side members 34 and the long-side members 35 are rectangular pillar-shaped members (square timbers). The frame portion 33 is formed by attaching the short-side members 34 from below to the lower ends of the long-side members 35. In other words, the frame portion 33 is formed by combining the short-side members 34 and the long-side members 35 in a grid pattern.

[0045] The installation base 31 is placed on a first slide mechanism 32 provided on the long side member 35 of the frame portion 33, and is slidable along the long side member 35. By providing the slide mechanism 32, the installation base 31 can slide along the long side member 35. In addition, a slide mechanism 36 is also provided on the short side member 34. By providing the slide mechanism 36, the camera installation unit 30 can slide along the horizontal frame member 22.

[0046] In this way, by providing the slide mechanisms 32 and 36, it becomes possible to slide the camera installation section 30 in the vertical and horizontal directions, and the position of the hyperspectral camera 11 can be freely determined.

[0047] Next, a waste classification processing device will be described with reference to FIG. 9. The waste classification processing device 100 is installed outdoors and is a device that estimates the type and amount of waste 13 transported on a belt conveyor 14 and classifies the waste 13. The waste classification processing device 100 has a sieving unit 101, an estimation unit 102, and a classification unit 103. The sieving unit 101 sieves the waste 13 transported on the belt conveyor 14. First, at a waste treatment site, the waste 13 is transported near the sieving unit 101 by a transport vehicle. The transported waste 13 is temporarily stored in one place and then loaded onto a belt conveyor separate from the belt conveyor 14, such as by a backhoe, and is automatically fed from the belt conveyor 14 into the sieving unit 101. The input waste 13 is sieved in the sieving unit 101.

[0048] In the sieving section 101, sieving is performed such that fine-grained waste 13 is sieved and rejected, and relatively large-sized waste 13 is removed and transported by the belt conveyor 14. As described above, the sieving section 101 is a machine such as a vibrating sieve or a rotary sieve, and sieves the waste 13 using these machines. In the sieving section 101, small-sized waste 13 is removed, and relatively large-sized waste 13 is removed and supplied to the belt conveyor 14. The sieved waste 13 supplied to the belt conveyor 14 is, for example, waste 13 that is large enough and heavy enough to be lifted by hand.

[0049] The estimation unit 102 uses the hyperspectral camera 11 to extract spectral data of the reflected light of the waste 13 from an image captured by irradiating the sifted waste 13 with halogen light, and estimates the type of the sifted waste 13 based on the extracted spectral data. The type of waste 13 is estimated based on the spectral characteristics of the reflected light of the halogen light, and the amount is estimated based on the amount of waste 13 transported by the flat belt and the proportion of each type. In the estimation unit 102, the hyperspectral camera 11 is installed at a predetermined position using installation stands 1 and 2. The transport speed of the belt conveyor 14 is also controlled in accordance with the shutter speed of the hyperspectral camera 11.

[0050] The estimation of the waste 13 in the estimation unit 102 is performed using, for example, machine learning using artificial intelligence (AI). Before the hyperspectral camera 11 is transported to a disposal site, it is trained on the waste 13 actually generated at the disposal site, etc., to generate an estimation model for estimating the waste 13. Machine learning using AI is performed in advance in a laboratory, etc. Then, the estimation model generated in this way is brought to the disposal site and used to estimate the type of waste 13, etc., thereby making it possible to estimate the type and amount of waste 13 with high accuracy. This makes it possible to estimate the amount of reusable material, combustible material, non-combustible material, etc., and to adjust the disposal site's acceptance status according to the type of waste 13.

[0051] The sorting unit 103 sorts the sifted waste 13 and classifies it by type. The sorting is performed, for example, by manual sorting. The sorted waste 13 is collected in sorting baskets or the like designated for each type. The sorting unit 103 may be an industrial robot or the like that can acquire the estimation result from the estimation unit 102 and automatically sort the waste 13 transported on the belt conveyor 14.

[0052] The waste classification processing device 100 may further include a collation unit. The collation unit collates the type and amount of waste 13 classified by the classification unit 103 with the type and amount of waste 13 estimated by the estimation unit 102. For example, the classification unit 103 can verify the type and amount of waste 13 estimated by the estimation unit 102 by measuring the weight of the waste 13 collected in baskets by type using a scale or the like.

[0053] According to this embodiment, waste such as garbage can be classified in detail. Furthermore, since a predetermined distance can be maintained between the hyperspectral camera and the waste, more accurate spectral analysis can be performed. Furthermore, since a halogen light is used to irradiate the waste with halogen light of a specific wavelength, spectral analysis can be performed easily and accurately, and the type and amount of waste can be accurately estimated. Furthermore, since the installation stand is movable, the system can be used in any outdoor location.

[0054] The present invention has been described above with reference to the embodiments. However, the present invention is not limited to the above-described embodiments and can be modified as appropriate. The configuration and details of the present invention can be modified in various ways that are understandable to those skilled in the art within the scope of the present invention. Furthermore, systems or devices that combine separate features included in each embodiment in any way are also included in the scope of the present invention. For example, a hyperspectral camera may be used during temporary storage to estimate the type and amount of waste 13 in advance, thereby predicting the amount of work that will be required thereafter. Furthermore, the classification unit 103 may be located in a separate location, and waste 13 that has passed through the estimation unit 102 may be temporarily stored therein. Alternatively, the sieving unit 101, estimation unit 102, and classification unit 103 may each be located in separate locations. [Explanation of symbols]

[0055] 1 Installation stand 2 Installation stand 10 Frame section 11 Hyperspectral camera 12 halogen lights 13 Waste 14 Conveyor Belt 15 Blackout 16 Leveling jig 17 Sieve 18 monitors 19 Height adjustment mechanism 21 Vertical frame members 22 Horizontal frame material 23 Leg horizontal frame material 30 Camera installation section 31 Installation base 32 Slide mechanism (first slide mechanism) 33 Frame 34 Short side members 35 Long side members 36 Slide mechanism (second slide mechanism) 37 Plate-shaped members 40 Lower frame section 41 Upper frame section 42 Height adjustment mechanism 100 Waste classification treatment equipment 101 Screening section 102 Estimation part 103 Classification Department

Claims

1. A waste sorting processing device that is installed outdoors and estimates the type and amount of waste transported on a belt conveyor and sorts the waste, a screening unit that screens the waste transported on the belt conveyor; an estimation unit that irradiates the sifted waste with halogen light, extracts spectral data of the reflected light from the waste from an image captured using a hyperspectral camera, and estimates the type and amount of the sifted waste based on the extracted spectral data; a sorting unit that sorts the sifted waste and classifies the waste by type; A waste classification processing device comprising:

2. The waste classification processing device described in claim 1 is provided with a rod-shaped or plate-shaped leveling jig that is positioned horizontally at a predetermined height relative to the belt conveyor, in a direction transverse to the conveying direction, at a position just before the waste placed on the belt conveyor passes under the frame portion in the conveying direction of the belt conveyor, and that levels the surface of the waste.

3. 3. The waste classification processing device according to claim 1, further comprising a collation unit that compares the type and amount of the waste classified by the classification unit with the type and amount of the waste estimated by the estimation unit.

4. The waste classification processing device according to claim 1 or 2, wherein the estimation unit further comprises an installation stand for installing the hyperspectral camera at a predetermined position.

5. 3. The waste classification processing apparatus according to claim 1, wherein the estimation unit estimates the type and amount of waste by using artificial intelligence having an estimation model generated in advance by machine learning.

Images