Classification system for sequentially classifying LED lighting waste in mixed LED lighting waste according to housing type

Abstract

The objective of the present invention is to provide a system for sequentially identifying, according to each process, and classifying various types of LED lighting waste, having different types of housings, on a conveyor all at the same time. A classification system for sequentially classifying LED lighting waste in mixed LED lighting waste according to housing type according to the present invention for solving the above problem is characterized by comprising: a conveyor for moving a plurality of pieces of mixed LED lighting waste having different types of housings; and a plurality of sorting units which are arranged on the conveyor, identify specific preset pieces of LED lighting waste from images, acquired through a plurality of cameras, of the mixed LED lighting waste moving on the conveyor, and pick up the identified pieces of LED lighting waste from the conveyor and transfer same, wherein the sorting units are arranged spaced a predetermined distance apart from each other on the conveyer process line to enable a sorting process, and each sorts LED lighting waste having a different type of housing.

Description

A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

[0001] The present invention relates to a classification system for LED waste lighting, and more particularly, to a classification system for sequentially classifying LED waste lighting by housing type from mixed LED waste lighting that can be sequentially identified and classified by each process at a time on a conveyor.

[0002] Generally, used fluorescent lamps that have reached the end of their lifespan are hazardous waste containing hazardous substances such as mercury, and if incinerated or buried, they cause serious environmental pollution problems.

[0003] Therefore, in order to prevent harmful gases such as mercury and harmful dust that are harmful to the human body from being released into the atmosphere, automatic recycling system equipment is provided that first cuts the base socket, which is the metal terminal part at both ends of the waste fluorescent lamp, with a cutter, then shreds it and classifies the metal terminals, and secondly, separates the finely crushed waste fluorescent lamp glass powder in a heating device to classify harmful dust and vaporized mercury and other harmful gases, and at this time, vaporized mercury and other harmful gases are captured in a mercury distillation device and recycled.

[0004] Meanwhile, the use of LED lighting has been rapidly increasing recently as lighting that consumes less electricity and does not generate mercury waste.

[0005] Depending on the shape of the housing, these LED lights can be divided into bulb-type LED lights, tubular LED lights, etc. Currently, LED lights with various types of housings are being produced, and each type of LED light has a different connection structure between the parts.

[0006] According to data released by the Ministry of Environment, the expected domestic LED waste lighting emissions are expected to increase from 163,000 tons in 2020 to 723,000 tons by 2030.

[0007] In this situation, the current system for collecting LED mixed waste lighting that has reached the end of its service life, the classification standards, and the preprocessing technology for recycling are insufficient, so they are being landfilled, incinerated, or exported overseas. Landfilling or incineration leads to environmental load problems, and exporting overseas leads to the problem of metal resources used in LED mixed waste lighting being exported.

[0008] In particular, since preprocessing technologies such as classification, disassembly, LED chip separation, and precious metal sorting that take into account the shape of the current LED lighting housing or the type of lighting have not been established, only the socket part of the mixed LED waste lighting is separated, and the remaining parts are crushed in bulk, there is a problem that fine powders such as plastic and glass other than metal components are mixed together, making sorting and sorting difficult.

[0009] Therefore, it is necessary to establish a system for collecting and recycling mixed LED waste lighting, and to build an automated plant for recycling mixed LED waste lighting by applying optimal dismantling and separation technology to the classified mixed LED waste lighting.

[0010] In addition, although the conventional classification system provides a system for classifying a single type of LED lighting, there is a need to configure a system for classifying each individually, and there is a need for many improvements in terms of time, cost, and efficiency.

[0011] In particular, in the case of bulb-type LED lights, a large amount of silicone adhesive is used, making manual dismantling difficult, and if not sorted, a lot of contaminants may accumulate inside the dismantling device when dismantling other types of LED lights.

[0012] Furthermore, downlights are easier to disassemble than bulbs, but manual disassembly is challenging, requiring crushing to sort the internal LED packages. Furthermore, many tubular products are designed to replace existing fluorescent lamps, and thus are made of glass rather than plastic. This necessitates a separate disassembly process to sort the LED packages.

[0013] Therefore, there is a need for a system that can easily and quickly classify LED waste lights with various types of housings into a single process system.

[0014] [Prior Art Literature]

[0015] [Patent Document 1] Republic of Korea Patent Publication No. 10-2222176 (Published: February 24, 2021)

[0016] [Patent Document 2] Republic of Korea Patent Publication No. 10-1540707 (Published: July 24, 2015)

[0017] The purpose of the present invention to solve the above-described problem is to provide a system that sequentially identifies and classifies various types of LED waste lights having different housing shapes for each process at once on a conveyor.

[0018] In addition, the purpose of the present invention is to provide an LED waste lighting classification system that enables fast and efficient processing, greatly improves productivity, and can continuously process various types of waste lighting.

[0019] In addition, an object of the present invention is to provide an LED waste lighting classification system that can improve quality management by classifying waste lighting with a lower error rate than manual human work and consistent standards.

[0020] In addition, it is an object of the present invention to provide an LED waste lighting classification system that can reduce costs and has relatively low maintenance and operating costs after the initial investment cost.

[0021] In order to solve the above-described problem, a classification system for sequentially classifying LED waste lights by housing type in mixed LED waste lights according to the present invention comprises: a conveyor for moving a plurality of LED mixed waste lights having different housing types; and a plurality of sorting units arranged on the conveyor, the sorting units identifying a preset specific LED waste light from an image of the LED mixed waste light moving on the conveyor acquired through a plurality of cameras, and picking up and transporting the identified LED waste light from the conveyor, wherein the sorting units are arranged on the conveyor process line at a predetermined distance apart from each other so that a sorting process can be performed, and are characterized in that they sort LED waste lights having different housing types.

[0022] In addition, in a classification system for sequentially classifying LED waste lights by housing type in a mixed LED waste light according to the present invention, the sorting unit is characterized by including: a camera unit that is disposed on the conveyor and acquires an image of the LED mixed waste light moving along the path of the conveyor using a plurality of cameras; an LED waste light identification unit that recognizes and selects any one of the LED waste lights from the image acquired by the camera unit using vision recognition technology and acquires location information; and a pickup and transfer unit that receives location information of the LED waste light identified by the LED waste light identification unit and picks up and transfers the identified LED waste light from a conveyor belt.

[0023] In addition, in the classification system for sequentially classifying LED waste lighting according to the housing type in the mixed LED waste lighting according to the present invention, the sorting unit is characterized in that it is sequentially arranged spaced apart on the conveyor process line so as to sort the LED mixed waste lighting in order of size and recycling value.

[0024] In addition, in a classification system for sequentially classifying LED waste lights by housing type in a mixed LED waste light according to the present invention, the LED waste light identification unit is characterized by including: a data collection unit for collecting the acquired image data; a preprocessing unit for preprocessing the collected image data; a waste light recognition unit for identifying any one LED waste light preset among the LED mixed waste lights using a deep learning model learned in advance based on the image data preprocessed in the preprocessing unit; and a location calculation unit for calculating location information of the LED waste light identified in the waste light recognition unit.

[0025] In addition, in a classification system for sequentially classifying LED waste lighting according to a housing type in a mixed LED waste lighting according to the present invention, the deep learning model is characterized by being one of a YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector), and Faster R-CNN model.

[0026] In addition, in a classification system for sequentially classifying LED waste lights by housing type in a mixed LED waste light according to the present invention, the LED waste light identification unit is characterized in that it produces information using a stereo vision technique based on synchronized image data captured by at least two cameras with different shooting angles.

[0027] In addition, in a classification system for sequentially classifying LED waste lights by housing type in a mixed LED waste light according to the present invention, the pickup and transfer unit is characterized by including: a data receiving unit for receiving location information from the waste light identification unit; a pickup control unit for calculating pickup control information based on the received location information; and a robot arm-based pickup unit for picking up LED waste lights and moving them to a transfer storage unit using the control information.

[0028]

[0029] And, in order to solve the above-described problem, a classification system for sequentially classifying LED waste lights by housing type in mixed LED waste lights according to the present invention comprises: a conveyor for moving a plurality of LED mixed waste lights having different housing types; a plurality of sorting operation units provided on the conveyor, the plurality of units including a camera unit having a plurality of cameras for photographing the LED mixed waste lights moving on the conveyor to obtain image data, and a pickup and transport unit for picking up and transporting the LED waste lights according to received sorting control information; and a central sorting control unit for generating sorting control information including identification information for identifying a pre-specified LED waste light, location information of the identified LED waste light, and pickup control information based on the image data received from each of the sorting operation units, and transmitting the sorting control information to the sorting operation units, wherein a plurality of the sorting operation units are arranged on the conveyor process line at a predetermined distance apart from each other so that a sorting process can be performed, and the central sorting control unit is characterized in that it produces sorting control information for each LED waste light having a different housing type for each of the sorting operation units.

[0030] In addition, in the classification system for sequentially classifying LED waste lighting according to the housing type in the mixed LED waste lighting according to the present invention, the sorting operation unit is characterized in that it is sequentially arranged spaced apart on the conveyor process line so that the LED mixed waste lighting is sorted in order of size and recycling value.

[0031] In addition, in a classification system for sequentially classifying LED waste lighting according to a housing type in a mixed LED waste lighting according to the present invention, the central selection control unit is characterized in that it produces information using a stereo vision technique based on synchronized image data captured by at least two cameras with different shooting angles.

[0032] In addition, in a classification system for sequentially classifying LED waste lights by housing type in a mixed LED waste light according to the present invention, the central selection control unit is characterized by including: a data collection unit for collecting the acquired image data; a preprocessing unit for preprocessing the collected image data; a waste light recognition unit for identifying any one LED waste light preset among the LED mixed waste lights using a deep learning model learned in advance based on the image data preprocessed in the preprocessing unit; and a location calculation unit for calculating location information of the LED waste light identified in the waste light recognition unit.

[0033] Specific details of other embodiments are included in the “Specific Details for Carrying Out the Invention” and the attached “Drawings.”

[0034] The advantages and / or features of the present invention and the methods for achieving them will become clear with reference to the various embodiments described in detail below together with the accompanying drawings.

[0035] However, the present invention is not limited to the configuration of each embodiment disclosed below, but may be implemented in various different forms, and each embodiment disclosed in this specification is provided only to ensure that the disclosure of the present invention is complete and to fully inform a person having ordinary skill in the art to which the present invention pertains of the scope of the present invention, and it should be understood that the present invention is defined only by the scope of each claim of the claims.

[0036] A classification system that sequentially classifies LED waste lighting by housing type in such mixed LED waste lighting can achieve the following effects.

[0037] First, according to the present invention, a LED waste lighting classification system is provided that enables fast and efficient processing compared to manual classification, significantly improves productivity, and can continuously process various types of waste lighting.

[0038] Second, according to the present invention, an LED waste lighting classification system is provided that can improve quality management by classifying waste lighting with a lower error rate than manual human work and consistent standards using artificial intelligence and a computer vision system.

[0039] Third, according to the present invention, a LED waste lighting classification system is provided that can reduce manpower costs through an automated system and has relatively low maintenance and operating costs after the initial investment cost.

[0040] Fourth, according to the present invention, an LED waste lighting classification system is provided that can contribute to environmental protection by facilitating the recycling process through an effective classification system, and can reduce waste of resources and increase the recycling rate through accurate classification.

[0041] Fifth, according to the present invention, an LED waste lighting classification system is provided that can collect and analyze data obtained during a processing process and continuously improve the performance of the system by utilizing the collected data.

[0042] Sixth, according to the present invention, an LED waste lighting classification system is provided that can be applied to waste lighting of various shapes and sizes, and the system can be expanded or upgraded as needed.

[0043] Seventh, according to the present invention, an LED waste lighting classification system is provided that can reduce the risk to humans working with hazardous substances or conditions and create a safer and healthier working environment.

[0044] Figure 1 is a schematic diagram of the configuration and process of a classification system that sequentially classifies LED waste lighting according to housing type in a mixed LED waste lighting according to an embodiment of the present invention.

[0045] FIG. 2 is a block diagram showing a classification system for sequentially classifying LED waste lights by housing type in a mixed LED waste light according to an embodiment of the present invention.

[0046] FIG. 3 is a schematic diagram of the configuration and process of a classification system that sequentially classifies LED waste lighting according to housing type in a mixed LED waste lighting system, as another embodiment of the present invention.

[0047] The best embodiment for carrying out the present invention comprises a conveyor for moving a plurality of LED mixed waste lights having different shapes of housings; and a plurality of sorting units arranged on the conveyor for identifying a preset specific LED waste light from an image of the LED mixed waste light moving on the conveyor acquired through a plurality of cameras, and for picking up and transporting the identified LED waste light from the conveyor, wherein the sorting units are arranged at a predetermined distance from each other on the conveyor process line so that a sorting process can be performed, and a sorting system for sequentially sorting LED waste lights by housing shape from mixed LED waste lights that sort LED waste lights having different shapes of housings is provided.

[0048] Before describing the present invention in detail, it should be understood that the terms or words used in this specification should not be interpreted as being unconditionally limited to their usual or dictionary meanings, and that the inventor of the present invention may appropriately define and use the concepts of various terms in order to explain his or her invention in the best possible manner, and further, that these terms or words should be interpreted as meanings and concepts that are consistent with the technical idea of ​​the present invention.

[0049] That is, it should be noted that the terms used in this specification are only used to describe preferred embodiments of the present invention, and are not intended to specifically limit the contents of the present invention, and that these terms are defined in consideration of various possibilities of the present invention.

[0050] Additionally, it should be noted that in this specification, singular expressions may include plural expressions unless the context clearly indicates a different meaning, and similarly, even if expressed in plural, may include a singular meaning.

[0051] Throughout this specification, whenever a component is described as "including" another component, it may mean that the component may further include any other component, rather than excluding any other component, unless specifically stated otherwise.

[0052] Furthermore, when a component is described as being "inside or connected to" another component, it should be understood that the component may be installed in direct connection with or in contact with the other component, may be installed spaced apart from the other component by a certain distance, and if installed spaced apart from the other component by a certain distance, there may be a third component or means for fixing or connecting the component to the other component, and the description of this third component or means may be omitted.

[0053] On the other hand, if a component is described as being "directly connected" or "directly connected" to another component, it should be understood that no third component or means exists.

[0054] Likewise, other expressions that describe the relationship between components, such as "between" and "directly between", or "adjacent to" and "directly adjacent to", should be interpreted as having the same meaning.

[0055] Additionally, it should be noted that the terms “one side,” “the other side,” “one side,” “the other side,” “first,” “second,” etc. in this specification, if used, are used to clearly distinguish one component from another component, and that the meaning of the component is not limited by such terms.

[0056] In addition, terms related to position, such as “upper,” “lower,” “left,” and “right,” etc., in this specification, if used, should be understood to indicate relative positions of the corresponding components in the corresponding drawings, and unless absolute positions are specified for these positions, these position-related terms should not be understood to refer to absolute positions.

[0057] In addition, in this specification, when specifying the drawing numbers for each component of each drawing, the same component has the same drawing number even if the component is shown in a different drawing, that is, the same reference number indicates the same component throughout the specification.

[0058] In the drawings attached to this specification, the size, position, connection relationship, etc. of each component constituting the present invention may be described with some exaggeration, reduction, or omission in order to sufficiently clearly convey the idea of ​​the present invention or for convenience of explanation, and therefore the proportions or scales may not be strict.

[0059] In addition, in the following description of the present invention, a detailed description of a configuration that is judged to unnecessarily obscure the gist of the present invention, for example, a known technology including a prior art, may be omitted.

[0060]

[0061] Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the drawings.

[0062] FIG. 1 is a schematic diagram of the configuration and process of a classification system for sequentially classifying LED waste lights by housing type in a mixed LED waste light according to an embodiment of the present invention, FIG. 2 is a diagram showing the block configuration of a classification system for sequentially classifying LED waste lights by housing type in a mixed LED waste light according to an embodiment of the present invention, and FIG. 3 is a schematic diagram of the configuration and process of a classification system for sequentially classifying LED waste lights by housing type in a mixed LED waste light according to another embodiment of the present invention.

[0063] As illustrated in FIG. 1, a classification system for sequentially classifying LED waste lighting according to housing type in a mixed LED waste lighting according to an embodiment of the present invention comprises a conveyor (100) for a waste lighting sorting process and a plurality of sorting units spaced apart from each other by process on the conveyor (100), wherein the sorting units are spaced apart from each other by a certain distance on the process line of the conveyor (100) so that each sorting process is possible, and are characterized in that they sort LED waste lighting having different housing types.

[0064] In the case where various types of LED waste lights having different sizes and shapes are input and sorted at once according to an embodiment of the present invention, a continuous process on a conveyor (100) can be performed, and a classification system is provided that individually identifies each type using vision recognition technology and performs the tasks of sorting, picking up, and transporting the LED waste lights identified at each process stage in batches.

[0065] More specifically, as shown in FIGS. 1 and 2, a conveyor (100) system can be configured to perform a continuous process, and a structure can be formed in which a plurality of sorting units are arranged at a certain distance apart on the conveyor (100) line to sort each type of LED waste light.

[0066] Here, a selection unit is disposed at each process stage of the conveyor (100), and a plurality of systems may be provided to identify a specific LED waste light set in advance from an image of LED mixed waste light moving on the conveyor (100) acquired through a plurality of cameras, and to pick up and transport the identified LED waste light from the conveyor (100).

[0067] More specifically, as illustrated in FIG. 2, the sorting unit may be configured to include a camera unit (210) that is disposed on a conveyor (100) and acquires images of LED mixed waste lights moving along a conveyor (100) path using a plurality of cameras, an LED waste light identification unit (230) that recognizes and selects one of the LED mixed waste lights using a vision recognition technology from the image acquired by the camera unit (210) and acquires location information, and a pickup and transfer unit (250) that receives location information of the waste lights identified by the LED waste light identification unit (230) and picks up and transfers the identified LED waste lights from the conveyor (100) belt.

[0068] The camera unit (210) may be composed of multiple cameras, and is preferably installed so as to be able to capture mixed LED waste lighting from both sides of the conveyor (100) belt or from different shooting angles, and may be configured as a device capable of accurately aligning the cameras according to control and correcting lens distortion or other errors.

[0069] The reason the cameras are installed at different shooting angles is to form a stereo image in order to derive the exact location of a specific LED light from the captured image footage.

[0070] Additionally, it is desirable to have at least two cameras capture images simultaneously so that the images acquired are synchronized from different angles to acquire image data.

[0071] Here, the LED waste light identification unit (230) may be configured to include a data collection unit (231), a preprocessing unit (232), a waste light recognition unit (233), a location calculation unit (234), and a selection control information generation unit (235), as illustrated in FIG. 2.

[0072] The data collection unit (231) may be a configuration that collects the acquired image data, and the preprocessing unit (232) may be a configuration that preprocesses the collected image data.

[0073] Here, the data collection unit (231) may be a device equipped with a communication device that receives and stores images captured in real time by the camera unit (210), and the preprocessing unit (232) may be a computing device configured with software that performs preprocessing processes such as noise removal, contrast adjustment, and size adjustment on the collected image data.

[0074] In addition, the closed light recognition unit (233) may be configured to identify any one LED closed light set in advance among the LED mixed closed lights using a pre-learned deep learning model based on image data preprocessed in the preprocessing unit (232).

[0075] In this way, the closed light recognition unit (233) is configured to perform the task of identifying and classifying a preset specific object, and performs the process of identifying and classifying an object within an image based on features extracted from image data preprocessed by the data preprocessing unit (232).

[0076] Object recognition technology like this has advanced further with the development of deep learning, especially convolutional neural networks (CNNs), which can automatically learn image features and accurately identify objects based on them.

[0077] A CNN object recognition process can be structured to include a) a learning phase where a network is trained using a large dataset of labeled images, during which the network learns the features of various objects, and b) a recognition phase where the trained network is used to identify and classify objects in new images.

[0078] In addition, the deep learning model applied to the classification system for sequentially classifying LED waste lights by housing type in the mixed LED waste lights according to the embodiment of the present invention may be any one of the YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector), and Faster R-CNN models.

[0079] YOLO (You Only Look Once) is a model that provides very fast speed for identifying and classifying objects by looking at an image only once, making it very suitable for real-time object recognition. The latest versions, YOLOv4 and YOLOv5, have the advantage of being faster and more accurate than previous versions.

[0080] SSD (Single Shot MultiBox Detector) is another model that offers high speed and high accuracy, can effectively detect objects of various sizes, and has the advantage of being suitable for real-time processing.

[0081] Faster R-CNN is a highly accurate model consisting of two parts: one for generating object suggestions and the other for classification. This model is advantageous in areas where accuracy is crucial, such as in the embodiments of the present invention.

[0082] It is preferable that the LED waste light identification unit (230) derives information using a stereo vision technique based on synchronized image data captured by at least two cameras with different shooting angles.

[0083] As described above, synchronized image data is acquired by simultaneously capturing images using at least two cameras, and based on the acquired data, the position calculation unit (234) can accurately calculate the position of a specific LED backlight using a stereo vision technique.

[0084] In addition, the location calculation unit (234) extracts features (e.g., borders, shapes, colors, etc.) from the images acquired from the data preprocessing unit (232) described above, and finds and matches identical features between the two extracted images. This is because it provides a reference point necessary for calculating the 3D location.

[0085] The 3D position calculation process used in the position calculation unit (234) may be configured to include a) a triangulation step for calculating 3D coordinates of the LED waste light by performing triangulation based on the positions and angles of the two cameras and the distance between matched feature points, and b) a coordinate conversion step for converting the calculated 3D coordinates into a coordinate system of a conveyor belt (100) or a robot arm.

[0086] And, as shown in FIGS. 1 and 2, it is preferable that the sorting unit be sequentially arranged on the conveyor (100) process line so that the LED mixed waste lighting is sorted in order of size and recycling value.

[0087] To increase the efficiency of such a sorting process, the sorting order can be a very important factor. That is, in a continuous sorting process using a conveyor (100) by inputting various waste LED lights at once, the arrangement order of the sorting section, i.e., which waste LED lights will be sorted first, is a very important issue.

[0088] This directly impacts the efficiency and performance of the entire process. Key factors to consider in this regard include:

[0089] 1) Physical Characteristics: The sorting order can be determined based on physical characteristics of the discarded LED lights, such as size, shape, and weight. For example, it may be more efficient to sort and process larger or heavier lights first.

[0090] 2) Ease of Processing: Some waste LED lights may be easier to separate and process than others, and processing these lights first can speed up the overall process.

[0091] 3) Recycling Value: Some waste LED lights may have a higher recycling value than others, and these lights can be prioritized for recycling to maximize recycling efficiency.

[0092] 4) Process Bottlenecks: Since certain types of waste LED lights can cause process bottlenecks, dealing with these lights first can help keep the overall process flowing smoothly.

[0093] 5) The sorting order can be adjusted based on the capacity and processing capabilities of the sorting system. That is, if the system can process a specific type of waste LED lighting more efficiently, it is desirable to prioritize sorting that type.

[0094] By comprehensively considering these factors and determining the sorting order, we can increase the efficiency of the overall process, optimize resources, and minimize environmental impact.

[0095] And, as shown in FIGS. 1 and 2, the pickup transport unit (250) may be configured to include a data receiving unit (251) that receives location information from the LED waste light identification unit (230), a pickup control unit (253) that calculates pickup control information based on the received location information, and a robot arm-based pickup unit (255) that picks up the LED waste light using the control information and moves it to the transfer storage unit.

[0096] That is, the pickup unit (255) can use a three-axis multi-joint robot arm of the X, Y, and Z axes or a horizontal multi-joint robot, a SCARA (Selective Compliance Assembly Robot Arm) robot.

[0097] As for the configuration of the pickup unit (255), the robot's pickup device for capturing the identified LED waste lights may be configured as a gripper device or an adsorption system. It is preferable to install a customized device so as to increase pickup efficiency according to the size, shape, appearance, and weight of the LED waste lights specified in advance at each process stage on the conveyor (100).

[0098] That is, in the case of LED waste lights that are difficult to pick up with a gripper, such as bulb-shaped LED waste lights with a hemispherical or bulb-shaped shape, it is desirable to pick them up using a suction system.

[0099] And, as shown in FIGS. 1 and 2, the pickup transfer unit (250) may be configured to include a data receiving unit (251) that receives location information from the waste light identification unit, a pickup control unit (253) that calculates pickup control information based on the received location information, and a robot arm-based pickup unit (255) that picks up the LED waste light through the control information and moves it to the transfer storage unit.

[0100] In this way, the pickup transport unit (250) receives control information including location information of the LED waste light identified through the LED waste light identification unit (230) described above, and controls the robot arm-based pickup transport unit (250) to pick up the identified LED waste light and transport it to a desired location.

[0101] Looking at a more specific process flow, as illustrated in Fig. 1, when mixed LED waste lighting having various types of housings is first fed into a sorting input unit (110), a first camera unit (210) equipped with at least two cameras in a first identification process stage (111) photographs the mixed LED waste lighting and acquires image data.

[0102] The first LED waste light identification unit (230) receives image data from the first camera unit (210), identifies a pre-specified first LED waste light based on the image data, and calculates and generates location information of the identified first LED waste light.

[0103] Then, the mixed LED waste light is transferred to the first pickup transfer unit (250a) in the first pickup transfer process stage (112) where the first LED waste light identification unit (230a) receives location information and control information of the first LED waste light, and based on this, the first LED waste light is picked up and moved to a desired location, and then the first process is performed, which leads to the dismantling process.

[0104] As the conveyor (100) belt continues to move, the mixed LED waste light is captured by the second camera unit (210b) in the second identification process stage (113) and an image is acquired, and the second LED waste light identification unit (230b) calculates and generates position information and control information of the second LED waste light.

[0105] In the second pickup transfer process stage (114), the second pickup transfer unit (250b) receives the generated location information and control information, picks up the corresponding second LED waste light, moves it to the desired location, and then performs the second process, which leads to the dismantling process.

[0106] The conveyor (100) belt continues to move, and in the third identification process stage (115), the third camera unit (210c) photographs the mixed LED waste light and acquires an image, and the third LED waste light identification unit (230c) identifies the third LED waste light and calculates and generates position information and control information.

[0107] Then, in the third pickup process stage (116), the third pickup transfer unit (250c) receives the generated location information and control information, picks up the corresponding third LED waste light, moves it to the desired location, and then the third process leading to the dismantling process is performed.

[0108] The above illustrates the first to third processes, but the order and number of processes are not limited thereto, and can be changed to perform the optimal process depending on the efficiency, characteristics, and environment of the process.

[0109] And, as shown in FIG. 3, as another embodiment of the present invention, a classification system for sequentially classifying LED waste lights by housing type in mixed LED waste lights can be configured to include a conveyor (100), a sorting operation unit (260), and a central sorting control unit (270).

[0110] Here, the sorting operation unit (260) may be configured as a system comprising a plurality of cameras, a camera unit (211) that is arranged on a conveyor (100) and obtains image data by photographing LED mixed waste lighting moving on the conveyor (100) using a plurality of cameras, and a pickup and transport unit (215) that picks up and transports LED waste lighting according to the received sorting control information.

[0111] The central selection control unit (270) may be configured to generate selection control information including identification information for identifying a pre-specified LED waste light, location information of the identified LED waste light, and pickup control information based on image data received from each selection work unit (260), and transmit the information to the selection work unit (260).

[0112] Here, a plurality of sorting work units (260) are arranged at a certain distance apart on the conveyor (100) process line so that each sorting process can be performed, and a central sorting control unit (270) that is connected to the sorting work unit (260) by communication can produce sorting control information for each LED waste light having a different type of housing for each sorting work unit (260a, 260b, 260c).

[0113] The embodiment illustrated in FIG. 3 is different from the embodiment of FIG. 1 in that a computing device that performs the function of an LED waste light identification unit (230a, 230b, 230c) that analyzes an image acquired from a camera unit (210) to identify a specific LED waste light in each process stage and calculates the location of the identified LED waste light is individually provided in each process stage, in that all photographed image data acquired from a plurality of sorting work units (260a, 260b, 260c) are received, and a central sorting control unit (270) identifies the LED waste light in each process stage, generates location information and control information of the identified LED waste light, and transmits them back to the sorting work unit (260) to perform the sorting process.

[0114] Below, the process of photographing and picking up and transporting the LED waste light of the sorting work unit (260) and the sorting process of the central sorting control unit (270) include functions and configurations that are the same or similar to those of the embodiment of FIG. 1, so a detailed description thereof will be omitted.

[0115] Looking at a more specific process flow, as illustrated in FIG. 3, when mixed LED waste lighting having various types of housings, such as a straight-through type, a bulb type, and a downlight type, is first fed into a selection input unit (110), a first camera unit (211a) equipped with at least two cameras in a first identification process stage (111) photographs the mixed LED waste lighting and acquires image data.

[0116] The central selection control unit (270) receives image data from the first camera unit (211a), identifies a pre-specified first LED closed light based on the image data, and calculates and generates location information of the identified first LED closed light.

[0117] Then, the mixed LED waste light is transferred to the first pick-up transfer unit (215a) in the first pick-up transfer process stage (112) and receives the location information and control information of the first LED waste light from the central sorting control unit (270), and based on this, the first process is performed to pick up the corresponding first LED waste light and move it to the desired location, and then proceed to the dismantling process.

[0118] As the conveyor (100) belt continues to move, the mixed LED waste light is captured by the second camera unit (211b) in the second identification process stage (113) and an image is acquired, and the central selection control unit (270) calculates and generates position information and control information of the second LED waste light.

[0119] The second pickup transport unit (215b) receives location information and control information generated from the central selection control unit (270), picks up the corresponding second LED waste light, moves it to the desired location, and then performs the second process leading to the dismantling process.

[0120] The conveyor belt (100) continues to move, and the mixed LED waste light is captured by the third camera unit (211c) in the third identification process stage (115) and an image is acquired, and the central selection control unit (270) identifies the third LED waste light and calculates and generates position information and control information.

[0121] Then, the third pickup transport unit (215c) receives the location information and control information generated from the central selection control unit (270), picks up the corresponding third LED waste light, moves it to the desired location, and then performs the third process leading to the dismantling process.

[0122] The above illustrates the first to third processes, but the order and number of processes are not limited thereto, and can be changed to perform the optimal process depending on the efficiency, characteristics, and environment of the process.

[0123] In this way, the embodiment of FIG. 1, which independently performs the identification and pickup / transfer work of LED waste lights in each process stage in a classification system that sequentially classifies LED waste lights by housing type in the above-described mixed LED waste lights, and the embodiment of FIG. 3, which has one central selection control unit (270) to comprehensively produce or generate identification and control information of LED waste lights in each process stage, have the following advantages.

[0124]

[0125] Individual screening system (example of Fig. 1)

[0126] 1) Distributed processing: Since each selection unit processes images independently, the load across the entire system is distributed. This improves the stability and reliability of the entire system.

[0127] 2) Flexibility: Since each sorting unit operates independently, problems in one unit have less impact on other units, and each sorting unit can be upgraded or adjusted individually.

[0128] 3) Reduced latency: Since image processing and position calculation are performed directly in each sorting unit, the time required to transmit data to the central processing system is reduced.

[0129]

[0130] Central screening system (example of Fig. 3)

[0131] 1) Centralized management: Using a single, integrated system allows you to efficiently manage and coordinate the entire process.

[0132] 2) Cost-effectiveness: A centralized system can reduce initial installation and maintenance costs compared to installing separate computing systems in each sorting unit.

[0133] 3) Consistent performance and updates: Maintains consistent performance of the entire system and is easy to update and maintain.

[0134] As described above, each embodiment can vary depending on the system's purpose, budget, required processing speed, and stability requirements. While standalone systems offer independent operation and flexibility for each sorting unit, integrated systems offer consistent management and cost-effectiveness for the entire system. Therefore, the system best suited to a specific environment and requirements can be selected and applied.

[0135]

[0136] Above, although some examples have been given and various preferred embodiments of the present invention have been described, the description of the various embodiments described in the “Specific Details for Carrying Out the Invention” section is merely exemplary, and those skilled in the art to which the present invention pertains will readily understand that they can carry out various modifications of the present invention or carry out equivalent implementations of the present invention based on the above description.

[0137] In addition, since the present invention can be implemented in various other forms, the present invention is not limited by the above description, and the above description is provided only to make the disclosure of the present invention complete and to fully inform a person having ordinary skill in the art to which the present invention belongs of the scope of the present invention, and it should be understood that the present invention is defined only by each claim of the claims.

[0138] The present invention provides an LED waste lighting classification system that enables fast and efficient processing compared to manual sorting, significantly improves productivity, and can continuously process various types of waste lighting; provides an LED waste lighting classification system that uses artificial intelligence and a computer vision system to classify waste lighting with a lower error rate than manual human work and improves quality control by classifying waste lighting with consistent standards; provides an LED waste lighting classification system that can reduce manpower costs through an automation system and has relatively low maintenance and operating costs after an initial investment; provides an LED waste lighting classification system that can contribute to environmental protection by facilitating the recycling process through an effective sorting system; provides an LED waste lighting classification system that can reduce waste of resources and increase the recycling rate through accurate sorting; provides an LED waste lighting classification system that can collect and analyze data obtained during the processing and continuously improve the performance of the system by utilizing the collected data; provides an LED waste lighting classification system that can be applied to waste lighting of various shapes and sizes and that can expand or upgrade the system as needed; provides an LED waste lighting classification system that can reduce the risk of humans working with hazardous materials or conditions and create a safer and healthier working environment.

Claims

1. A conveyor for moving a plurality of LED mixed waste lights having different shapes of housings, and A method of manufacturing a light source comprising: a plurality of sorting units arranged on the conveyor, the method comprising: identifying a preset specific LED light source from an image of LED mixed light moving on the conveyor acquired through a plurality of cameras; and picking up and transporting the identified LED light source from the conveyor; The above selection section, The LED waste lights are arranged at a certain distance apart from each other on the process line of the conveyor so that each sorting process can be performed, and each has a different type of housing, characterized in that they are sorted. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

2. In paragraph 1, The above selection section, A camera unit which is arranged on the conveyor and acquires image data of LED mixed waste lighting moving along the path of the conveyor using a plurality of cameras; An LED waste light identification unit that recognizes and selects one LED waste light among the LED mixed waste lights from image data acquired by the camera unit using vision recognition technology and acquires location information; and It is characterized by including a pick-up and transport unit that receives location information of the LED waste light identified by the LED waste light identification unit and picks up and transports the identified LED waste light from a conveyor belt. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

3. In paragraph 1, The above sorting unit is characterized in that it is sequentially arranged on the conveyor process line so that the LED mixed waste lighting is sorted in order of size and recycling value. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

4. In paragraph 2, The above LED waste light identification part is, A data collection unit that collects image data acquired by the above camera unit; A preprocessing unit that preprocesses the image data collected by the data collection unit; A closed light recognition unit that identifies one preset LED closed light among the LED mixed closed lights using a pre-learned deep learning model based on the image data preprocessed in the above preprocessing unit; and It is characterized by including a location calculation unit that calculates location information of the LED waste light identified by the above waste light recognition unit. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

5. In paragraph 4, The above deep learning model is, Featuring one of the YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector) and Faster R-CNN models. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

6. In paragraph 2, The above LED waste light identification part is, A stereo vision technique characterized by producing information based on synchronized image data captured from at least two cameras with different shooting angles. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

7. In paragraph 2, The above pickup transport unit, A data receiving unit that receives location information from the above-mentioned closed light identification unit; A pickup control unit that calculates pickup control information based on received location information; and It is characterized by including a robot arm-based pickup unit that picks up LED waste lights and moves them to a transfer storage unit through the above control information. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

8. A conveyor for moving multiple LED mixed waste lights having different types of housings; A camera unit, which is arranged on the conveyor and has a plurality of cameras, captures LED mixed waste lighting moving on the conveyor and obtains image data including image data; A plurality of sorting work units including a pick-up and transport unit that picks up and transports LED waste lights according to the received sorting control information; and A central sorting control unit is included that generates sorting control information including identification information for identifying a pre-specified LED waste light based on image data received from each of the above sorting work units, location information of the identified LED waste light, and pickup control information, and transmits the information to the sorting work units; The above-mentioned sorting work units are arranged in multiple units spaced apart at a certain distance on the process line of the conveyor so that each sorting process can be performed, and the central sorting control unit is characterized in that it produces sorting control information for each LED waste light having a different type of housing for each sorting work unit. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

9. In paragraph 8, The above sorting work unit is characterized in that it is sequentially arranged on the conveyor process line so that the LED mixed waste lighting is sorted in order of size and recycling value. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

10. In paragraph 8, The above central selection control unit, A stereo vision technique characterized by producing information based on synchronized image data captured from at least two cameras with different shooting angles. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

11. In paragraph 8, The above central selection control unit, A data collection unit that collects image data acquired by the above camera unit; A preprocessing unit for preprocessing the image data collected by the genital data collection unit; A closed light recognition unit that identifies one LED closed light set in advance among the LED mixed closed lights using a pre-learned deep learning model based on the image data preprocessed in the above preprocessing unit; and It is characterized by including a location calculation unit that calculates location information of the LED waste light identified by the above waste light recognition unit. A classification system that sequentially classifies LED waste lighting by housing type in mixed LED waste lighting.

Images