KR102222177B1 - Recycling classification system using ai - Google Patents

Abstract

The present invention relates to a recycling classification system using AI, which comprises: a capturing unit generating an image of continuously supplied waste; a storage unit storing information of the waste which contains a shape, the image, and a color of the waste; and an analysis unit analyzing similarity by using the image of the waste generated by the capturing unit and the information of the waste stored in the storage unit and generating classification information of the continuously supplied waste. The classification information contains information about a type, a contamination degree, the color, and existence of a cover and a label of the waste.

Description

Recycling classification system using AI {RECYCLING CLASSIFICATION SYSTEM USING AI}

The present invention relates to a recycling classification system using AI, and more specifically, the time required for waste classification by replacing the process of repair and selection depending on the conventional manpower according to a specific classification criterion from randomly supplied waste. It is related to a recycling classification system using AI that shortens the value and can more easily separate waste through automatic waste recognition using AI.

A large amount of waste is discharged every day from homes, factories, and restaurants, and many people separate and dispose of food waste among general waste, recyclable waste, and food waste.

However, still recyclable garbage is mixed with general garbage and discharged, and manpower is directly input to separate it.

In addition, companies that use plastic bottles or glass bottles are importing recycled waste from foreign countries because recycled waste is not properly separated and is disposed of as general waste.

Various inventions have been made in order to solve these problems in Korea, and in particular, methods, devices, and systems for separating PET bottles have been applied and registered as patents.

For example, Registration Patent No. 10-1270354 "Waste Recycling Classification System"

There is an advantage in that the waste can be automatically classified according to the material by photographing the waste.

However, even if the waste is classified, there is a problem that a cleaning process or a removal process is required afterwards depending on whether foreign substances are included in the waste or a lid or label is included.

Korean Patent Registration No. 10-1270354 (Registration Date, 2013.05.27)

The technical problem to be achieved by the present invention is to reduce the time required for waste sorting and to reduce the time required for waste sorting by replacing the process of repair and selection depending on the conventional manpower according to a specific classification criterion from randomly supplied waste. We intend to provide a recycling classification system using AI that can more easily separate waste through recognition.

In order to solve this problem, a recycling classification system using AI according to an embodiment of the present invention includes an image pickup unit for generating an image of continuously supplied waste; A storage unit storing information on the waste including the shape, image, and color of the waste; And an analysis unit that analyzes similarity by using the image of the waste generated by the imaging unit and information on the waste stored in the storage unit and generates classification information on the continuously supplied waste; including, the classification information May include information on the type of waste, degree of contamination, color, presence or absence of lids and labels.

The analysis unit may continuously update the classification information to the storage unit to analyze the image of the waste generated by the imaging unit and the updated information in the storage unit in real time.

The types of waste may include PET bottles, cans, glass bottles, paper, paper packs, styrofoam, general plastics, and general waste.

When the type of waste is any one of PET bottles, glass bottles, paper, paper packs, styrofoam and general plastics, the analysis unit additionally generates pollution level information of the waste, and the pollution level information is generated by the analysis unit at the imaging unit. It may be information obtained by analyzing whether there are foreign substances attached to the inside or outside of the waste by using the image of the waste and information on the waste stored in the storage unit.

When the type of waste is any one of PET bottles, glass bottles, paper, and general plastics, the analysis unit additionally generates color information of the waste, and the color information is transparent, white, green, brown, yellow, blue, and It may be information on any one or more colors of red.

When the type of waste is either a PET bottle or a glass bottle, the analysis unit additionally generates information about the presence or absence of a lid and a label of the waste, and the analysis unit additionally generates an image of the waste generated by the imaging unit and the storage unit. By comparing the information on the stored waste with each other, it is possible to determine the presence or absence of the lid and label of the waste.

The analysis unit may include: a recognition unit for recognizing waste in the image generated by the image pickup unit; And a classification unit that calculates a matching ratio using information on the waste recognized by the recognition unit and the waste stored in the storage unit and classifies the waste.

When the matching ratio between the shape of the waste recognized by the recognition unit and the shape of the waste stored in the storage unit is calculated to be greater than or equal to a preset standard, the classification unit may classify it as one of the types of the waste.

When the classification unit calculates a matching ratio of the image of the waste recognized by the recognition unit and the image of the waste stored in the storage unit using the sorted waste type to be less than a preset standard, it is determined that the recognized waste contains foreign matter. The degree of contamination can be classified.

The analysis unit optical unit for irradiating light to the supplied waste; And a condensing unit for condensing light after the light irradiated from the optical unit is transmitted to the supplied waste, wherein the sorting unit determines the color of the supplied waste by using the light condensed by the condensing unit. Can be classified.

In addition to the technical problems of the present invention mentioned above, other features and advantages of the present invention will be described below or will be clearly understood by those of ordinary skill in the art from such technology and description.

According to the present invention as described above has the following effects.

By generating classification information by using the information on the waste stored in the storage unit, the analysis unit, which is an embodiment of the present invention, wastes that can be recycled from the waste supplied at a time can be easily classified even if they are not directly classified by humans. .

In addition, by classifying the waste by including the pollution level in the classification information, even if it is classified as a recyclable waste, it is possible to classify the recycled waste that needs a cleaning process later, thus reducing the time to sort the waste that needs cleaning and reducing additional costs. have.

By classifying waste by including whether or not a label is included in the classification information, even if it is classified as a recyclable waste, it is possible to classify the recycled waste that needs a label removal process later. Accordingly, it is possible to shorten the time for sorting waste and reduce additional costs.

The present invention can recognize the shape and image of the waste supplied from the recognition unit. By using this to compare the information of the waste stored in the storage unit, the classification unit can classify the type of waste only by the shape of the waste recognized by the recognition unit.

In addition, the degree of pollution can be classified using the image of the waste recognized by the recognition unit, so that recyclable waste can be easily classified from the supplied waste.

In particular, in the present invention, by actively using AI for waste classification, not only can the time required for sorting waste be significantly reduced, but also the classification accuracy can be greatly improved.

In addition, other features and advantages of the present invention may be newly recognized through embodiments of the present invention.

1 is a block diagram showing a recycling classification system using AI according to an embodiment of the present invention.
2 is a classification table showing classification information according to an embodiment of the present invention.
3 is a block diagram illustrating an analysis unit of FIG. 1.
4 is a block diagram showing the analysis unit of FIG. 1 according to another embodiment of the present invention.

In the present specification, in adding reference numerals to elements of each drawing, it should be noted that only the same elements have the same number as possible, even if they are indicated on different drawings.

Meanwhile, the meaning of the terms described in the present specification should be understood as follows.

Singular expressions are to be understood as including plural expressions unless clearly defined differently in context, and the scope of rights should not be limited by these terms.

It is to be understood that terms such as "comprises" or "have" do not preclude the presence or addition of one or more other features, numbers, steps, actions, elements, or combinations thereof.

Hereinafter, preferred embodiments of the present invention designed to solve the above problem will be described in detail with reference to the accompanying drawings.

In addition, wastes input to the recycling classification system using AI according to an embodiment of the present invention refer to plastic, PET bottles, glass bottles, glass, paper, styrofoam, and general waste, and will be described as wastes hereinafter. do.

1 is a block diagram showing a recycling classification system using AI according to an embodiment of the present invention, and FIG. 2 is a classification table showing classification information according to an embodiment of the present invention.

Referring to FIG. 1, the recycling classification system 10 using AI includes an imaging unit 100 for generating an image of waste, a storage unit 200 for storing information on the waste, and generating classification information for the waste. It may include an analysis unit 300.

The imaging unit 100 is a device that generates an image of waste that is continuously supplied, and may be a camera.

The camera is an example for carrying out the present invention, and it is natural that the camera may be implemented with a lens, an infrared sensor, or various types of sensors that can recognize images of waste or barcodes contained in waste.

The imaging unit 100 can not only generate an image for each of a large amount of waste that is input to the recycling classification system 10 using the AI of the present invention, but also generate an image of continuously supplied waste in real time. I can.

The storage unit 200 may store information on waste.

Information on waste may include the shape, image and color of the waste.

For example, if the waste is a PET bottle or glass bottle, it may include a shape and image in a complete state before the waste is supplied, an image in a complete state with a label removed, an image and color with a lid removed, and so on. In the case of paper, cartons, styrofoam or general plastic, the shape and color of the waste may be included.

However, since this is an example for carrying out the present invention, various information such as a brand name and grade for waste may be stored according to the user's use and convenience.

The analysis unit 300 may analyze the similarity by using the image of the waste generated by the imaging unit 100 and the information on the waste stored in the storage unit 200 and generate classification information on the continuously supplied waste. .

The classification information may include the type of waste, the degree of contamination, the color, the presence or absence of lids and labels. This example is shown in FIG. 2.

Referring to FIG. 2, the classification information generated by the analysis unit 300 may be information divided into a large classification, a medium classification, and three small classifications of the type of waste, the degree of pollution, the color, the presence or absence of a lid and a label.

The major classification may be a type of waste, and may include PET bottles, aluminum cans, glass bottles, paper, paper cartons, styrofoam, general plastics, and general garbage.

Although the major classification is divided into eight types of waste, it will be natural for those skilled in the art that the type of waste can be added according to the user's use and convenience.

The analysis unit 300 may analyze similarity by using the image of the waste generated by the imaging unit 100 and the image of the waste stored in the storage unit 200 and classify the type of waste.

In addition, the analysis unit 300 may analyze the image of the waste generated by the imaging unit 100 in real time, and continuously store the analyzed result and images of the waste used for the analysis in the storage unit 200. That is, the data of the storage unit 200 used to determine the type of waste may be continuously updated.

In addition, not only the entire image of the waste or the image of the intact form, but also an image showing only part of the waste or an image of damaged waste can be used as a classification criterion. Therefore, it is possible to reduce the time required for sorting while continuously improving the accuracy of waste sorting.

For example, when information on a Coca-Cola bottle, which is a PET bottle, is stored in the storage unit 200, when the Coca-Cola bottle is supplied as waste, first, the imaging unit 100 generates an image of the Coca-Cola bottle.

The analysis unit 300 may analyze the similarity between the information on the Coca-Cola bottle stored in the storage unit 200 and the image of the Coca-Cola bottle generated through the imaging unit 100.

At this time, the analysis unit 300 may preset a similarity value between the image of the waste generated by the imaging unit 100 and the information on the waste stored in the storage unit 200, and the user can You can enter any value.

If the similarity is calculated above a preset value, the analysis unit 300 may classify the Coca-Cola bottle as a PET bottle.

In addition, the storage unit 200 may continuously store the similarity analyzed by comparing the Coca-Cola bottle image generated by the analysis unit 300 with the Coca-Cola bottle image stored in the storage unit 200.

Using the stored similarity, the analysis unit 300 continuously updates the image of the Coca-Cola bottle, and the analysis unit 300 is a Coca-Cola bottle generated by the imaging unit 100 through newly updated information in the storage unit 200. Coca-Cola bottles can be classified as PET bottles with only some images of.

In this way, the analysis unit 300 generates classification information using the information on the waste stored in the storage unit 200, so that the waste that can be recycled from the waste supplied at a time can be easily classified even if it is not directly classified by humans. I can.

The medium classification may be a pollution degree, and the pollution degree may be a classification according to the presence or absence of foreign substances contained in the supplied waste.

The analysis unit 300 may additionally generate information on the degree of contamination when the type of waste is any one of PET bottles, glass bottles, paper, paper packs, styrofoam, and general plastics.

In detail, using the image of the waste generated by the imaging unit 100 and the information on the waste stored in the storage unit 200, the analysis unit 300 analyzes whether foreign substances are contained inside or outside the image of the waste, Pollution degree information can be generated.

Continuing to describe the Coca-Cola bottle, as described above, the imaging unit 100 generates an image of the Coca-Cola bottle.

Then, the analysis unit 300 may compare the image of the Coca-Cola bottle generated by the imaging unit 100 with the image of the Coca-Cola bottle stored in the storage unit 200.

If a foreign substance is contained inside or outside the image of the Coca-Cola bottle generated by the imaging unit 100, the color of the Coca-Cola bottle may appear partially dark or in a different color.

When the color of the Coca-Cola bottle is partially dark or in a different color, and the image and similarity of the Coca-Cola bottle stored in the storage unit 200 are analyzed to be low, the analysis unit 300 may generate contamination level information indicating that a foreign substance is contained.

At this time, the analysis unit 300 may preset a similarity value between the image of the waste generated by the imaging unit 100 and the information on the waste stored in the storage unit 200, and the user can You can enter any value.

Accordingly, when the similarity is calculated to be less than or equal to a preset value, the analysis unit 300 may generate contamination level information as a foreign substance in the Coca-Cola bottle.

By classifying the waste by including the degree of pollution in the classification information, even if it is classified as a recyclable waste, it is possible to classify the recycled waste that needs a cleaning process afterwards, thereby shortening the time to classify the waste that needs cleaning and reducing additional costs.

The first of the three subcategories may be for color, and the color may include any one or more of transparent, white, green (including mint), brown, yellow (including light green), blue (including light blue), and red. have.

The analysis unit 300 may additionally generate color information when the type of waste is any one of a PET bottle, a glass bottle, paper, and general plastic.

By using the image of the waste generated by the imaging unit 100 and the information on the waste stored in the storage unit 200, the matching color may be analyzed, and information on the color may be generated.

For example, the analysis unit 300 may compare an image of a Coca-Cola bottle generated by the imaging unit 100 with color information about a Coca-Cola bottle stored in the storage unit 200 to analyze whether they match.

In detail, the analysis unit 300 includes a color included in the image in a complete state from which the label has been removed among information on the Coca-Cola bottle stored in the storage unit 200 and an image of the Coca-Cola bottle generated by the imaging unit 100 Colors included in can be compared.

If it matches the Coca-Cola bottle of a transparent color, the analysis unit 300 may generate color information by classifying the Coca-Cola bottle into a transparent color.

Among the recyclable wastes, PET bottles, glass bottles, paper and general plastics may have different processing processes after they are classified as recycling according to their color.

In particular, in the case of PET bottles, the cost of transparent PET bottles is high as a processing process of coating different colors to transparent colors can be added, and the cost of PET bottles is low because the processing process of coating different colors to dark colors is not added. It can be an element.

The remaining two subclasses may be related to the presence or absence of a lid and a label in the waste, and may be classification according to whether or not a lid and label are included in the waste.

When the type of waste is either a PET bottle or a glass bottle, the analysis unit 300 may additionally generate information on the presence or absence of a lid and a label of the waste.

At this time, the analysis unit 300 generates classification information on whether a lid or label is included in the waste by using the image of the waste generated by the imaging unit 100 and the information on the waste stored in the storage unit 200 can do.

For example, the analysis unit 300 compares the image of the waste generated in the imaging unit 100 with the image in a complete state stored in the storage unit 200, and if a lid or label is included, the lid or label is included. It can be classified as

Conversely, if a lid or label is not included in the image of the waste generated by the imaging unit 100, it may be classified as not including a lid or label.

By classifying the waste by including this, even if it is classified as a recyclable waste, it is possible to classify the recycled waste that needs a label removal process later. Accordingly, it is possible to shorten the time for sorting waste and reduce additional costs.

2 is exemplary classification information as described above, and may be selected or appropriately added or changed according to the needs of the user.

For example, in addition to the type of waste, pollution degree, color, the presence or absence of a lid and label, the weight of the waste, and the material of the waste may be added.

3 is a block diagram illustrating an analysis unit of FIG. 1.

Referring to FIG. 3, the analysis unit 300 includes a recognition unit 310 for recognizing waste and a classification unit 320 for classifying waste using a matching ratio between the recognized waste and the waste stored in the storage unit 200. can do.

The recognition unit 310 may recognize waste in the image generated by the imaging unit 100, and the waste recognized by the recognition unit 310 may be a shape and an image.

The shape is the shape of the waste, and the image may be a photograph that entirely includes the color, label, and foreign matter of the waste.

If the entire image of the waste is included in the image generated by the imaging unit 100, the recognition unit 310 may recognize the entire image of the waste.

In contrast, when the entire image of the waste is not included in the image generated by the imaging unit 100, the recognition unit 310 may recognize only a part of the waste included in the image.

The classification unit 320 may classify the waste by calculating a matching ratio using the waste recognized by the recognition unit 310 and the information on the waste stored in the storage unit 200.

In detail, the classification unit 320 may classify the type of waste using the shape of the waste recognized by the recognition unit 310 and information on the waste stored in the storage unit 200.

The classification of the waste type by the classification unit 320 may be classification using a ratio in which the shape of the waste recognized by the recognition unit 310 matches the shape of the waste stored in the storage unit 200.

If the shape of the waste recognized by the recognition unit 310 is calculated as a matching ratio greater than or equal to a predetermined standard with the shape of the waste stored in the storage unit 200, the classification unit 320 determines that a PET bottle, an aluminum can, and a glass bottle , Paper, cartons, styrofoam, general plastics, and general garbage.

Information on PET bottles, aluminum cans, glass bottles, paper, paper packs, styrofoam and general plastics may be stored as waste information in the storage unit 200, and if not stored as information, the classification unit 320 Can be classified as general trash.

Hereinafter, a case in which information on a Coca-Cola bottle, which is a PET bottle, is stored in the storage unit 200, and the shape matching ratio is set to 50% for classifying the type of waste will be described as an example.

The recognition unit 310 may recognize the shape of the Coca-Cola bottle from the image of the Coca-Cola bottle generated by the imaging unit 100. The classification unit 320 may classify the supplied Coca-Cola bottle as a PET bottle when a matching ratio of 50% or more with the shape of the Coca-Cola bottle stored in the storage unit 200 is calculated.

In other words, when the entire image of the waste is included in the image generated by the imaging unit 100 and the recognition unit 310 recognizes the entire image of the waste, the shape of the generated Coca-Cola bottle and the stored in the storage unit 200 The shape for the Coca-Cola bottle will match more than 50%, and it will be easily classified as a PET bottle.

On the other hand, even if the entire image of the waste is not included in the image generated by the imaging unit 100, if only 50% or more of the shape of the generated Coca-Cola bottle matches the shape of the Coca-Cola bottle stored in the storage unit 200, the PET bottle It could be classified as

The classification unit 320 continuously compares the image of the Coca-Cola bottle generated by the imaging unit 100 with the shape of the Coca-Cola bottle stored in the storage unit 200, and when the matching ratio is 50% or more, calculated as 50% or more. The image of the Coca-Cola bottle may be continuously updated in the storage unit 200.

If the image of the Coca-Cola bottle generated by the imaging unit 100 is generated with only some of the images, the classification unit 320 may extract and analyze the shape of the Coca-Cola bottle matched with the updated information.

Even at this time, if the matching ratio of each image and shape is calculated to be 50% or more, the classification unit 320 may classify the Coca-Cola bottle as a PET bottle with only images of some Coca-Cola bottles generated by the imaging unit 100.

As the data of the storage unit 200 is continuously updated by the classification unit, the time for determining the type of waste may be drastically reduced.

On the other hand, after the type of the supplied waste is classified, the classification unit 320 uses this to determine the ratio at which the image of the waste recognized by the recognition unit 310 and the image of the waste stored in the storage unit 200 are matched. If it is calculated below the set standard, the degree of contamination can be classified as having foreign substances.

In the case where the pollution degree of waste is classified in the storage unit 200, the image matching ratio is set to 90%.

If a Coca-Cola bottle classified as a PET bottle contains foreign substances, the color of a portion of the Coca-Cola bottle may appear dark or may appear in a color other than the original color of the Coca-Cola bottle. Then, the recognition unit 310 recognizes the image of the waste as it is.

The classification unit 320 matches the image of the Coca-Cola bottle stored in the storage unit 200 with the image of the Coca-Cola bottle containing foreign substances, and when the matching ratio of the image is calculated to be less than 90%, it analyzes that there is a foreign substance and classifies the degree of contamination. can do.

In addition, the classification unit 320 uses the image of the waste recognized by the recognition unit 310 and the image of the waste stored in the storage unit 200 after the type of waste is classified, and whether or not there is a color, a lid, and a label. Can be classified.

The classification unit 320 may also update information on the presence or absence of a color, a lid, and a label to the storage unit 200 by using the information continuously updated in the storage unit 200 while classifying the type of waste.

The classification unit 320 can easily determine the color of the waste, the presence of a lid, and a label by using the updated information in the storage unit 200 even if the recognition unit 310 recognizes a partial image of the waste or the image of the damaged waste. So you can accurately sort recyclable waste

The reference matching ratios preset in the classification unit 320 may be changed according to the user's use and convenience.

As described above, the information on the stored waste is stored in the storage unit 200 as an image including the color of the waste door, an image not including the lid and label, and an image in a complete state including both the lid and label, This may be used to match the image of the waste recognized by the recognition unit 310.

For example, if the image of the waste recognized by the recognition unit 310 is a green cider bottle, the classification unit 320 may match the green cider bottles stored in the storage unit 200 with each other, and the supplied waste is colored green. Can be classified.

In addition, the classification unit 320 matches the cider bottle recognized by the recognition unit 310 with a cider bottle that does not contain a lid stored in the storage unit 200 or a cider bottle with a lid so that the lid is not included at the entrance portion. If it matches an image that is not included, it can be classified as not including the lid.

4 is a block diagram showing the analysis unit of FIG. 1 according to another embodiment of the present invention.

Referring to FIG. 4, the analysis unit 300 may further include an optical unit 330 and a condensing unit 340.

The optical unit 330 may irradiate light to the supplied waste, and the irradiated light may be light including visible light.

Visible light refers to light that can be seen by the human eye, and light reflected or transmitted from an object is seen in color.

In this way, since the color is affected by light, the color may vary depending on the environment of the supplied waste. In addition to generating color information from the image of the waste generated by the imaging unit 100, the present invention further includes an optical unit 330 and a condensing unit 340 to clearly distinguish the color of the waste. It is possible to directly irradiate the light onto and distinguish the color of the waste through reflected or transmitted light.

The condensing unit 340 may collect light after the light irradiated from the optical unit 330 is transmitted to the supplied waste.

The classification unit 320 may classify the color of the supplied waste by using the light collected by the condensing unit 340.

Accordingly, even if the environment in which the waste is supplied is changed, the color of the waste can be clearly distinguished through the optical unit 330 and the condensing unit 340, thereby improving accuracy in classifying recycled waste.

The present invention recognizes the shape and image of the waste supplied from the recognition unit 310 as it is, and compares the information with the information of the waste stored in the storage unit 200 by using the same, the classification unit 320 is the recognition unit 310 ), the type of waste can be classified using only the shape of the waste recognized.

In addition, the degree of pollution can be classified using the image of the waste recognized by the recognition unit 310, so that recyclable waste can be easily classified from the supplied waste.

The present invention described above is not limited to the above-described embodiments and the accompanying drawings, and that various substitutions, modifications and changes are possible within the scope of the technical spirit of the present invention, in the technical field to which the present invention pertains. It will be obvious to those of ordinary skill.

10: Recycling classification system using AI
100: imaging unit
200: storage unit
300: analysis unit

Claims (10)

An imaging unit that generates an image of waste that is continuously supplied;
A storage unit storing information on the waste including the shape, image, and color of the waste; And
Including; an analysis unit for analyzing the similarity by using the image of the waste generated by the imaging unit and the information on the waste stored in the storage unit and generating classification information on the continuously supplied waste; including,
The types of waste are classified into PET bottles, cans, glass bottles, paper, paper packs, styrofoam, general plastics, and general garbage,
In the PET bottle and the glass bottle, pollution degree information, color information, and information on the presence or absence of a lid and a label are generated,
Pollution degree information and color information are generated for the paper and the general plastic,
Pollution degree information is generated for the paper pack and the styrofoam,
The analysis unit calculates a matching ratio by using a recognition unit for recognizing waste in the image generated by the image pickup unit and information on the waste recognized by the recognition unit and the waste stored in the storage unit, and classifies the waste. Including the sorting unit, waste is classified by type, and among the sorted waste types, any one or more of pollution level information, color information, lid information, and label information is matched for PET bottles, glass bottles, paper, cartons, styrofoam, and general plastics,
When the matching ratio between the shape of the waste recognized by the recognition unit and the shape of the waste stored in the storage unit is calculated to be greater than or equal to a preset standard, the classification unit classifies as one of the types of waste,
When the classification unit calculates a matching ratio of the image of the waste recognized by the recognition unit and the image of the waste stored in the storage unit using the sorted waste type to be less than a preset standard, it is determined that there is a foreign substance in the recognized waste. Recycling classification system using AI to classify pollution levels. The method of claim 1,
The analysis unit continuously updates the classification information to the storage unit to analyze the image of the waste generated by the image pickup unit and the updated information in the storage unit in real time. delete The method of claim 1,
The pollution level information is the information obtained by analyzing the presence of foreign substances attached to the inside or outside of the waste by the analysis unit using the image of the waste generated by the imaging unit and the information on the waste stored in the storage unit. Recycling classification system using AI characterized. The method of claim 1,
The color information is information on one or more colors of transparent, white, green, brown, yellow, blue, and red. The method of claim 1,
The analysis unit compares the image of the waste generated by the imaging unit with the information on the waste stored in the storage unit to determine the presence or absence of a lid and a label of the waste. delete delete delete The method of claim 1,
The analysis unit includes an optical unit that directly irradiates light onto the supplied waste; And
It further includes a condensing unit for condensing the light after the light irradiated from the optical unit is transmitted to the supplied waste,
The classification unit recycling classification system utilizing AI, characterized in that the color of the supplied waste is clearly classified by using the light collected by the condensing unit.

Images