WO2021201251A1 - Method for analyzing composition of electronic/electrical apparatus component layer, method for processing electronic/electrical apparatus component layer, device for analyzing composition of electronic/electrical apparatus component layer, and device for processing electronic/electrical apparatus component layer - Google Patents

Method for analyzing composition of electronic/electrical apparatus component layer, method for processing electronic/electrical apparatus component layer, device for analyzing composition of electronic/electrical apparatus component layer, and device for processing electronic/electrical apparatus component layer Download PDF

Info

Publication number
WO2021201251A1
WO2021201251A1 PCT/JP2021/014225 JP2021014225W WO2021201251A1 WO 2021201251 A1 WO2021201251 A1 WO 2021201251A1 JP 2021014225 W JP2021014225 W JP 2021014225W WO 2021201251 A1 WO2021201251 A1 WO 2021201251A1
Authority
WO
WIPO (PCT)
Prior art keywords
electronic
electrical equipment
scraps
equipment component
composition
Prior art date
Application number
PCT/JP2021/014225
Other languages
French (fr)
Japanese (ja)
Inventor
智也 後田
寿文 河村
Original Assignee
Ｊｘ金属株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ｊｘ金属株式会社 filed Critical Ｊｘ金属株式会社
Publication of WO2021201251A1 publication Critical patent/WO2021201251A1/en

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • B07C5/04—Sorting according to size
    • B07C5/10—Sorting according to size measured by light-responsive means
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B09—DISPOSAL OF SOLID WASTE; RECLAMATION OF CONTAMINATED SOIL
    • B09B—DISPOSAL OF SOLID WASTE NOT OTHERWISE PROVIDED FOR
    • B09B5/00—Operations not covered by a single other subclass or by a single other group in this subclass
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
    • G01N21/27—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands using photo-electric detection ; circuits for computing concentration
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume

Definitions

  • the present invention relates to a method for analyzing the composition of electronic / electrical equipment parts waste, a method for processing electronic / electrical equipment parts waste, a composition analysis device for electronic / electrical equipment parts waste, and an electronic / electrical equipment parts waste processing apparatus.
  • Patent Document 1 electrical / electronic equipment component scraps containing copper are incinerated, crushed to a predetermined size or less, and the crushed electrical / electronic equipment component scraps are smelted with copper. It is described that it is processed in a furnace.
  • electronic / electrical equipment parts scraps include parts scraps of various shapes and types, and the raw material composition changes depending on the difference in supply source and the like.
  • the raw material composition of electronic and electrical equipment parts scraps is currently evaluated in advance by visual judgment and chemical analysis, and the results are evaluated for the operation management and operating conditions of the sorting process. It is reflected in the setting of.
  • the present inventors examined the use of image recognition technology as a method for grasping the raw material composition other than visual inspection, and conducted various studies. However, since various parts are mixed in the target electronic / electrical equipment parts waste, the composition is greatly different depending on the situation of the supplier, and the shape is also various. Therefore, the electronic / electrical equipment parts waste by image recognition is used. Individual identification may not be performed accurately.
  • the present disclosure can improve the image recognition accuracy, and can efficiently analyze the composition of the component waste in the electronic / electrical equipment component waste in a short time regardless of the individual experience and skill.
  • the method for analyzing the composition of electronic / electrical equipment component scraps is, in one embodiment, from an image taken from an image of a raw material containing a plurality of electronic / electrical equipment component scraps including a plurality of component types.
  • the learning data used for learning the machine learning system includes raw materials for composition analysis. If the information is reflected, the certainty of the machine learning system is set to the first threshold, and if the information of the raw material to be analyzed for composition is not reflected in the learning data, it is lower than the first threshold.
  • This is a method for analyzing the composition of electronic / electrical equipment component waste which includes extracting electronic / electrical equipment component waste by setting a second threshold value.
  • the method for treating electronic / electrical equipment component waste according to the embodiment of the present invention includes a sorting step of selecting a specific component type from a plurality of component types based on the composition analysis result.
  • -It is a method of disposing of scraps of electrical equipment parts.
  • the electronic / electrical equipment component scrap composition analyzer is obtained from an image taken from an image of a raw material containing a plurality of electronic / electrical equipment component scraps including a plurality of component types.
  • a learning device used to learn machine learning systems which is a composition analysis device for electronic and electrical equipment component scraps that extracts electronic and electrical equipment component scraps and analyzes the composition using image recognition processing using a machine learning system.
  • the certainty of the machine learning system is set to the first threshold value, and the information of the raw material to be the composition analysis is not reflected in the learning data.
  • it is a composition analysis device for electronic / electrical equipment component waste including a processing device for extracting electronic / electrical equipment component waste by setting a second threshold value lower than the first threshold value.
  • the electronic / electrical equipment component waste processing apparatus includes an imaging device that images a plurality of electronic / electrical equipment component waste including a plurality of component types, and an image captured image.
  • a learning device used to learn machine learning systems which is a composition analysis device for electronic and electrical equipment component scraps that extracts electronic and electrical equipment component scraps and analyzes the composition using image recognition processing using a machine learning system.
  • the certainty of the machine learning system is set to the first threshold value, and the information of the raw material to be the composition analysis is not reflected in the learning data.
  • a composition analysis device provided with a processing device for extracting electronic / electrical equipment component scraps by setting a second threshold value lower than the first threshold, and electrons based on the composition analysis result analyzed by the composition analysis device.
  • -It is a processing device for electronic / electrical equipment parts waste equipped with a sorter for sorting specific parts waste from electrical equipment parts waste.
  • the image recognition accuracy can be improved, and the composition of component scraps in electronic / electrical device scraps can be efficiently analyzed in a short time regardless of individual experience and skill. It is possible to provide a method for analyzing the composition of parts scraps, a method for treating electronic / electrical equipment parts waste, an electronic / electrical equipment parts waste composition analysis device, and an electronic / electrical equipment parts waste treatment device.
  • FIG. 2A is a photograph showing an example of an image in which a recognition frame is attached to electronic / electrical equipment component scraps (boards) existing in the captured image
  • FIG. 2B is an electron having the recognition frame attached.
  • -It is a photograph showing an example in which electrical equipment component scraps are arranged by component type (board / plastic).
  • It is a flowchart which shows an example of the image analysis processing of the electronic / electrical equipment component waste. It is a table showing an example of comparison between the result of estimating the area and the measured value of the area by extracting each of the substrate and the plastic as electronic / electrical equipment from the captured image. It is a table which shows the example of the comparison of the recognition result with respect to the measured value at the time of extracting the electronic / electrical equipment component scrap from the captured image by using the machine learning system which concerns on embodiment of this invention.
  • the electronic / electrical equipment component waste processing apparatus acquires an image of a raw material containing a plurality of electronic / electrical equipment component waste including a plurality of component types.
  • a composition analysis device 10 and a composition analysis device for electronic / electrical device component scraps that extract electronic / electrical device component scraps and perform composition analysis using a possible image pickup device 12 and image recognition processing using a machine learning system. It is provided with a sorting machine 13 that sorts specific parts scraps from electronic / electrical equipment parts scraps based on the composition analysis result analyzed by 10.
  • the "electronic / electrical equipment component waste” in the present embodiment is waste crushed electronic / electrical equipment such as waste home appliances / PCs and mobile phones, and is crushed to an appropriate size after being collected. Point to.
  • the crushing for making electronic / electrical equipment parts waste may be performed by the processor himself, or may be crushed in the city and purchased.
  • the crushing method is not limited to a specific device, and may be a shearing method or an impact method, but crushing that does not impair the shape of parts is desirable as much as possible. Therefore, equipment belonging to the category of crushers intended for fine crushing is not included.
  • Electronic / electrical equipment component scraps consist of multiple component types such as plastics (synthetic resins) used for substrates and housings, metal pieces, copper wire scraps, capacitors, IC chips, etc., depending on the processing purpose. Can be further classified. Although not limited to the following, in the present embodiment, electronic / electrical equipment component scraps crushed to a particle size of 50 mm or less can be suitably treated. The lower limit of the particle size is not particularly limited, but is 5 mm or more, more typically 10 mm or more, and further 15 mm or more.
  • the composition analysis device 10 includes a processing device 100 that performs image analysis processing of captured images according to the composition analysis algorithm according to the present embodiment, a storage device 110 that stores information necessary for various processes, an input device 120, and a display device 130. And.
  • the composition analysis device 10 is configured to be able to transmit the processing result of the processing device 100 to the server 15 or the sorting machine 14 connected via the network 11 via the network 11.
  • the processing device 100 performs image recognition processing on the captured image by using a machine learning system using the learning data, and analyzes the composition of electronic / electrical equipment component scraps in the captured image. For machine learning, various analysis software for image recognition using deep learning or the like can be used.
  • the storage device 110 stores information necessary for processing by the processing device 100.
  • a plurality of component types that is, a substrate, a plastic, a metal piece, a copper wire scrap, a condenser, an IC chip, and others (connector, film-shaped component scrap), etc. , Covered wire scraps, etc.), preferably 7 or more types, and further 10 or more types of extracted data, and for assigning a recognition frame to the extracted electronic / electrical equipment parts scraps.
  • Various composition analysis information including the above are stored.
  • the extraction information used in the extraction process includes, for example, shape information for extracting geometric shapes such as contours of electronic / electrical equipment component scraps, color information for extracting the color of electronic / electrical equipment component scraps, and electronics. -Includes rotating images of electronic / electrical equipment parts scraps, etc., such as background information for extracting background information such as colors of background images around electrical equipment parts scraps and shadows due to unevenness.
  • the processing device 100 extracts a plurality of electronic / electrical equipment component scraps from the captured image by machine learning using the learning data stored in the storage device 110, and classifies them by classifying them according to the plurality of component types. do.
  • the training data includes various information for detecting the characteristics of electronic / electrical equipment component scraps based on the past image analysis results, and new data is input via the input device 120 to learn the machine learning system. By updating (reflecting) the learning data used in the above, the recognition accuracy of electronic / electrical equipment component scraps is improved.
  • the processing device 100 uses it for learning a machine learning system when performing an extraction process of electronic / electrical equipment component waste from the captured image of the new raw material.
  • the certainty of the machine learning system is set to the first threshold value, and the information of the raw material to be the composition analysis target is reflected in the learning data. If not, the second threshold is set to be lower than the first threshold.
  • the recognition target of the electronic / electrical equipment component waste can be expanded by setting the second threshold value lower than the first threshold value. Therefore, the recognition target range can be relaxed. As a result, the extraction rate of electronic / electrical equipment parts waste can be improved even for a new raw material, and the accuracy of the composition analysis of the raw material can be improved.
  • the certainty of the present embodiment indicates the degree of certainty of the extraction result of the electronic / electrical equipment component waste by the machine learning system included in the processing device 100 of FIG.
  • the first threshold value is preferably 0.2 to 0.5, more preferably 0.25 to 0.4.
  • the second threshold value is preferably 0.01 to 0.1, more preferably 0.02 to 0.09.
  • the processing device 100 assigns a recognition frame to the electronic / electrical equipment component scraps extracted according to the set values of the first and second threshold values of the certainty.
  • the recognition frame may be a minimum frame that borders the outline of the electronic / electrical equipment component waste, or a recognition frame including an image of the background around the electronic / electrical equipment component waste and the electronic / electrical equipment component waste is added. May be good.
  • a recognition frame including an image of the background around the electronic / electrical equipment component scraps it is preferable to give a circumscribed figure circumscribing the outline of the electronic / electrical equipment parts scraps as the recognition frame.
  • the shape of the inscribed figure may have any shape such as a circumscribed rectangle, a circumscribed circle, and a circumscribed polygon.
  • FIG. 2A is a photograph showing an example in which a substrate is extracted as scraps of electronic / electrical equipment parts from a captured image.
  • the upper part of the paper surface of FIG. 2B shows an example of a substrate, and the lower part of the paper surface shows an example of a plastic extract.
  • the processing device 100 further measures or estimates the area of the electronic / electrical equipment component waste existing in the given recognition frame.
  • an area measurement tool capable of measuring the area based on the captured image can be used.
  • the processing device 100 at least obtains information on the area ratio of electronic / electrical equipment component waste with respect to the recognition frame stored in the storage device 110. Using the component type area ratio data, the total area of electronic / electrical equipment component scraps with recognition frames is estimated for each of a plurality of component types.
  • the part type area ratio data includes the shape information of the outline of the electronic / electrical equipment component scrap, the area of the recognition frame, the area of the electronic / electrical equipment component scrap and the background area occupied in the recognition frame, and the electronic / electrical equipment with respect to the recognition frame.
  • Various information such as information on the area ratio of parts scraps and position information can be included.
  • the processing device 100 further multiplies the estimation result of the total area of electronic / electrical equipment component scraps for each component type by the estimated weight per unit area for each of a plurality of predetermined component types to obtain a plurality of component types.
  • the weight ratios of the plurality of component types are analyzed, respectively, and thereby the composition of the electronic / electrical equipment component scraps existing in the captured image is analyzed.
  • the estimated weight per unit area of the plurality of component types can be set in advance by the operator via the input device 120 or the like according to the operation result.
  • the estimated weight of the substrate scraps is, for example, 2.0 g / cm 2 , and the assumed plastics.
  • the weight can be set to 1.5 g / cm 2 and the other parts can be set to 1.0 g / cm 2.
  • the processing device 100 is calculated from the number (pieces) of the parts constituting the extracted part type for each extracted part type, and the estimation result and the number of the above area.
  • Other physical properties such as the average particle size and the weight ratio of the elements constituting each component type can also be analyzed and output to the display device 130 or the like.
  • the processing device 100 further generates information on the operating conditions of the sorter for sorting the plurality of part types based on the analysis result of the weight ratio of the plurality of part types.
  • the sorter there are various sorters such as a picking machine, a color sorter, a metal sorter, an eddy current sorter, a wind power sorter, and a sieving machine.
  • the processing device 100 from the analysis results of a plurality of component types, the processing device 100 generates operating conditions for a color sorter that selects, for example, a substrate and plastic, and stores the generated operating conditions in the storage device 110.
  • the operating conditions stored in the storage device 110 are output to the sorters 13 and 14, and the sorters 13 and 14 can perform the sorting process according to the output operating conditions.
  • the processing device 100 may acquire the position information of each of the electronic / electrical equipment component scraps extracted from the captured image of the electronic / equipment component scraps. Then, the position information may be output to the specific sorters 13 and 14 for extracting the position of the specific scraps of electronic / electrical equipment parts and sorting them.
  • the substrate and the metal piece cannot be separated by specific sorters 13 and 14 such as a metal sorter, but if the position information can be obtained individually from the image information, they can be sorted by the sorters 13 and 14 having a picking function. Will be able to.
  • the processing device 100 can further learn the learning data required for the extraction process and the guess process by machine learning.
  • the processing device 100 has characteristics for each of a plurality of component types such as a substrate, plastic, metal pieces, and copper wire scraps, for example, in the present embodiment, the color, shape, recognition frame, and area ratio of the component scraps for each component type. It is possible to learn the characteristics of information such as relationships based on the input of hundreds to tens of thousands of sheets of data, and to improve the accuracy of extraction processing and estimation processing.
  • the machine learning system of the processing device 100 further learns the characteristics.
  • the processing device 100 includes information on the outline of the electronic / electrical equipment component waste not extracted by the extraction process, the electronic / electrical equipment component waste not extracted by the extraction process, and an image of the background.
  • a new learning model can be created by machine learning or the like in response to input of information on a new recognition frame including the above, information on the area ratio of electronic / electrical equipment component scraps with respect to the recognition frame, and the like.
  • the weight ratio of each component type of electronic / electrical equipment component waste can be quantified and evaluated by image processing of the captured image. can.
  • the sorting conditions such as whether the raw material should be physically sorted or incinerated by a kiln furnace or the like.
  • the recognition result of the captured image for example, when the ratio of plastic in the raw material is large, it is possible to review the purchase conditions based on the weight ratio of plastic and adjust the heat load of the kiln furnace. ..
  • step S100 the captured image is acquired.
  • the captured image may be an captured image input via the input device 120 or the network 11, or the captured image obtained by the imaging device 12 may be used.
  • step S101 it is determined whether or not the information of the raw material to be the composition analysis target included in the captured image is already reflected in the learning data used for learning the machine learning system. The determination may be made manually by the operator via the input device 120, or may be made automatically by the processing device 100 with reference to the data of the storage device 110.
  • step S102 If the information of the raw material to be analyzed for composition captured in the captured image is reflected in the learning data used for learning the machine learning system, the process proceeds to step S102, and the certainty of the machine learning system is the first threshold value. Is set to. If the information of the raw material to be the composition analysis target is not reflected in the learning data, the process proceeds to step S103, is set to a second threshold value lower than the first threshold value, and proceeds to step S104.
  • steps S104 to S107 image recognition processing using a machine learning system is performed.
  • the processing apparatus 100 removes the electronic / electrical equipment component scraps existing in the captured image for each component type (for example, substrate, plastic, metal piece, copper) based on the set certainty of the machine learning system. Classify into 7 categories of wire scraps, capacitors, IC chips, and other parts) and extract. The classification result can be displayed by the display device 130 or the like.
  • step S105 the processing apparatus 100 assigns a recognition frame including an image of the background around the electronic / electrical equipment component scraps to the plurality of electronic / electrical equipment component scraps extracted from the captured image for each of the plurality of component types. do.
  • step S106 the processing device 100 attaches a recognition frame to each of a plurality of parts types based on the part type area ratio data having information on the area ratio of the electronic / electrical equipment parts waste with respect to the recognition frame. Estimate the area (total area) of electrical equipment parts scraps.
  • step S107 the processing apparatus 100 multiplies the area estimation result and the assumed weight per unit area of the plurality of component types to analyze the weight ratios of the plurality of component types, thereby performing the electrons in the captured image.
  • Analyze the composition of electrical equipment component scraps For example, by multiplying the total area of each part type in the image by the weight per unit area, the total weight of the part scraps can be roughly calculated. By comparing the total weight of the component scraps for each component type, the composition of the electronic / electrical equipment component scraps contained in the captured image can be analyzed. The analysis result is output in step S108.
  • the weight ratio of the electronic / electrical equipment component waste existing in the captured image for each component type is determined by the image analysis. You can ask. As a result, the raw material composition can be quickly estimated regardless of the individual's experience and skill. As a result, the operator can determine the purchase conditions of the raw materials and the selection method of the raw materials at an earlier stage, so that the entire factory can be operated more efficiently.
  • the background image around the electronic / electrical equipment component scraps may differ depending on the conditions of the imaging destination, and in some cases, the machine learning system may not be able to properly extract the electronic / electrical equipment component scraps.
  • the machine learning system is used for extracting electronic / electrical equipment component waste, which includes information on a new recognition frame obtained by synthesizing an image obtained by combining background information and a border image of electronic / electrical equipment component waste. By newly learning the learning data for the purpose, it is possible to obtain a more flexible composition analysis device corresponding to various conditions in the captured image.
  • the outline, color, area ratio to the recognition frame, and background of the electronic / electrical equipment parts scraps contained in the raw materials that are not reflected in the learning data.
  • FIG. 4 shows an example of comparison between the result of estimating the area and the measured value by extracting the substrate and the plastic as electronic / electrical devices from the captured image according to the analysis method according to the embodiment of the present invention. It is a table to represent. According to the method for analyzing the composition of electronic / electrical equipment component scraps according to the embodiment of the present invention, an appropriate evaluation can be performed with a measurement error of less than 5.0% for the substrate and a measurement error of less than 10% for the plastic. When two types of parts were included, a recognition rate of 90% or more could be achieved. Therefore, it can be said that this method has a sufficient effect in roughly grasping the composition of the raw material.
  • FIG. 5 is a table showing an example of comparison of recognition results with respect to actual measurement values when electronic / electrical equipment component scraps are extracted from captured images using the machine learning system according to the embodiment of the present invention.
  • Example 10 shows the actual measurement value of the number of extracts and the number of recognitions (extractions), the number of false recognitions, the recognition rate, and the false recognition rate when the characteristics are reflected in the learning data used for learning of the machine learning system. The certainty of the case was set to 0.3.
  • Example 2 and Comparative Example 2 show an example in which the raw material information is not reflected in the learning data. The certainty of Example 2 was 0.07, and the certainty of Comparative Example 2 was 0.3.
  • Example 2 As can be seen from the results of Example 2 and Comparative Example 1, the number of false recognitions by the machine learning system increases slightly by lowering the certainty, but the lower the certainty, the better the recognition rate is obtained. Has been done. It can be said that this method has a sufficient effect in roughly grasping the composition of the raw material.
  • the processing device 100 digitizes and analyzes the average area, number, average particle size, weight ratio, etc. of each of a plurality of component types from the captured image, so that the electrons can be manually sorted as in the conventional case.
  • the raw material composition can be quantified and grasped remarkably more quickly than the raw material composition of electrical equipment parts waste is evaluated.
  • the processing apparatus 100 based on the raw material analysis result analyzed by the processing apparatus 100, information on the operating conditions of the sorter that sorts a specific part type from a plurality of part types, for example, selection of a sorter for sorting raw materials. Then, the operating conditions such as the sorting conditions and the sorting order can be determined, and the sorting process can be performed based on the operating conditions.
  • the air volume of the wind power sorter can be adjusted according to the average particle size of a plurality of component types based on the raw material analysis results analyzed by the machine learning system.
  • the air volume can be, for example, 5 to 20 m / s, more preferably 5 to 12 m / s, and further 5 to 10 m / s.
  • Wind power sorting can be repeated twice or more depending on the raw material analysis result.
  • a picking process for removing lumpy copper wire debris can be performed.
  • the position information of the copper wire scraps stored in the storage device 110 is output to the picking device as the sorter 13, and the picking device can remove the copper wire scraps according to the output result.
  • This copper wire scrap can be sent to, for example, a valuable metal recovery process.
  • the sorting process using a sieving machine can be performed between the first wind power sorting and the second wind power sorting.
  • a sieving machine is adopted as the sorting machine 13, and the size of the sieve mesh of the sieving machine for sorting characteristic part types based on the average particle size of a plurality of part types obtained in the raw material analysis result. Can be changed.
  • the present invention relates to the sorter 13 used in the magnetic force sorting step, the eddy current sorting step, and the optical sorting step of optically sorting metal objects and non-metallic objects.
  • the image data is analyzed in real time while continuously photographing the electronic / electrical equipment parts scraps being transported, and the raw material composition is analyzed. Can be done.
  • the raw material composition of electronic / electrical equipment parts waste is evaluated by hand sorting, and the result is reflected in the operation management of the sorting process and the setting of operating conditions.
  • the raw material composition is manually sorted. It was not possible to perform rapid processing with the method of grasping.
  • the composition of the component scraps in the electronic / electrical equipment component scraps whose composition changes from moment to moment is instantly determined by image analysis and separation based on predetermined classification data. Since it can be quantified, a large amount of electronic / electrical equipment parts waste can be quickly sorted under more appropriate conditions.
  • the sorting efficiency (results) of the parts scraps 13 and 14 can be evaluated based on the amount of change in the parts scraps.
  • sorting machines 13 and 14 such as picking devices, color sorters, and metal sorters, individual separation of parts types becomes easy. ..
  • the display device 130 displays the analysis result with a frame having a different color for each raw material type, the operator can easily recognize it, so that the false detection of the composition analysis device can be easily recognized.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Mathematical Physics (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Health & Medical Sciences (AREA)
  • Environmental & Geological Engineering (AREA)
  • Chemical & Material Sciences (AREA)
  • Geometry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Image Analysis (AREA)
  • Sorting Of Articles (AREA)
  • Processing Of Solid Wastes (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

Provided are a method for analyzing the composition of an electronic/electrical apparatus component layer, a method for processing an electronic/electrical apparatus component layer, a device for analyzing the composition of an electronic/electrical apparatus component layer, and a device for processing an electronic/electrical apparatus component layer that make it possible to improve the precision of image recognition, and to efficiently analyze the component layer composition of an electronic/electrical apparatus component layer irrespective of the experience or skill of a person. A method for analyzing the composition of an electronic/electrical apparatus component layer, the method including: extracting an electronic/electrical apparatus component layer from within a captured image in which is captured a raw material including a plurality of electronic/electrical apparatus component layers including a plurality of types of components, through use of an image recognition process in which is used a machine learning system, and then performing compositional analysis; and setting the reliability of the machine learning system to a first threshold value when information about the raw material constituting the subject of compositional analysis is reflected in learning data used in learning by the machine learning system, or setting said reliability to a second threshold value lower than the first threshold value and extracting an electronic/electrical apparatus component layer when information about the raw material constituting the subject of compositional analysis is not reflected in the learning data.

Description

電子・電気機器部品屑の組成解析方法、電子・電気機器部品屑の処理方法、電子・電気機器部品屑の組成解析装置及び電子・電気機器部品屑の処理装置Electronic / Electrical Equipment Parts Waste Composition Analysis Method, Electronic / Electrical Equipment Parts Waste Disposal Method, Electronic / Electrical Equipment Parts Waste Composition Analysis Device and Electronic / Electrical Equipment Parts Waste Processing Equipment
　本発明は、電子・電気機器部品屑の組成解析方法、電子・電気機器部品屑の処理方法、電子・電気機器部品屑の組成解析装置及び電子・電気機器部品屑の処理装置に関する。 The present invention relates to a method for analyzing the composition of electronic / electrical equipment parts waste, a method for processing electronic / electrical equipment parts waste, a composition analysis device for electronic / electrical equipment parts waste, and an electronic / electrical equipment parts waste processing apparatus.
　近年、資源保護の観点から、廃家電製品・ＰＣや携帯電話等の電子・電気機器部品屑から、有価金属を回収することがますます盛んになってきている。また、電子・電気機器部品屑の処理量は近年増加する傾向にあり、その効率的な回収方法が検討され、提案されている。 In recent years, from the viewpoint of resource protection, it has become more and more popular to recover valuable metals from waste home appliances, scraps of electronic and electrical equipment parts such as PCs and mobile phones. In addition, the amount of waste of electronic / electrical equipment parts processed has tended to increase in recent years, and efficient collection methods have been studied and proposed.
　例えば、特開２０１５－１２３４１８号公報（特許文献１）では、銅を含む電気・電子機器部品屑を焼却後、所定のサイズ以下に粉砕し、粉砕した電気・電子機器部品屑を銅の溶錬炉で処理することが記載されている。 For example, in Japanese Patent Application Laid-Open No. 2015-123418 (Patent Document 1), electrical / electronic equipment component scraps containing copper are incinerated, crushed to a predetermined size or less, and the crushed electrical / electronic equipment component scraps are smelted with copper. It is described that it is processed in a furnace.
　しかしながら、電子・電気機器部品屑の処理量が増加することにより、電子・電気機器部品屑に含まれる物質の種類によってはその後の銅製錬工程での処理に好ましくない物質（製錬阻害物質）が従来よりも多量に投入されることとなる。このような銅製錬工程に装入される製錬阻害物質の量が多くなると、電子・電気機器部品屑の投入量を制限せざるを得なくなる状況が生じる。 However, due to the increase in the amount of electronic / electrical equipment component waste processed, some substances (smelting inhibitor) that are not preferable for processing in the subsequent copper smelting process may be produced depending on the type of substance contained in the electronic / electrical equipment component waste. It will be input in a larger amount than before. If the amount of smelting inhibitor charged into such a copper smelting process increases, there will be a situation in which the amount of waste of electronic / electrical equipment parts must be limited.
　例えば、電子・電気機器部品屑には、様々な形状及び種類の部品屑が含まれており、供給元の違い等によりその原料組成が変化する。銅製錬工程に投入される原料を適切に選別するために、現在は、電子・電気機器部品屑の原料組成を予め目視判定や化学分析によって評価し、その結果を選別処理の操業管理、運転条件の設定に反映させることが行われている。 For example, electronic / electrical equipment parts scraps include parts scraps of various shapes and types, and the raw material composition changes depending on the difference in supply source and the like. In order to properly select the raw materials to be input to the copper smelting process, the raw material composition of electronic and electrical equipment parts scraps is currently evaluated in advance by visual judgment and chemical analysis, and the results are evaluated for the operation management and operating conditions of the sorting process. It is reflected in the setting of.
　しかしながら、目視により原料組成を判定する手法では、個人の経験や技能によって評価結果にバラつきがあり、定量的な評価もできていない。原料組成特定のための化学分析や手選別も時間を要する。 However, with the method of visually determining the raw material composition, the evaluation results vary depending on individual experience and skills, and quantitative evaluation is not possible. Chemical analysis and manual selection to identify the raw material composition also take time.
特開２０１５－１２３４１８号公報Japanese Unexamined Patent Publication No. 2015-123418
　本発明者らは目視以外の原料組成把握のための方法として、画像認識技術の利用を検討し、種々の検討を行った。しかしながら、対象とする電子・電気機器部品屑は、種々の部品が混在する上、供給元の状況に応じて組成も大きく異なり、形状も様々であるため、画像認識による電子・電気機器部品屑の個体識別を精度良く行えない場合がある。 The present inventors examined the use of image recognition technology as a method for grasping the raw material composition other than visual inspection, and conducted various studies. However, since various parts are mixed in the target electronic / electrical equipment parts waste, the composition is greatly different depending on the situation of the supplier, and the shape is also various. Therefore, the electronic / electrical equipment parts waste by image recognition is used. Individual identification may not be performed accurately.
　上記課題を鑑み、本開示は、画像認識精度を向上でき、個人の経験や技能に関係なく、電子・電気機器部品屑中の部品屑の組成を短時間で効率良く解析することが可能な電子・電気機器部品屑の組成解析方法、電子・電気機器部品屑の処理方法、電子・電気機器部品屑の組成解析装置及び電子・電気機器部品屑の処理装置を提供する。 In view of the above problems, the present disclosure can improve the image recognition accuracy, and can efficiently analyze the composition of the component waste in the electronic / electrical equipment component waste in a short time regardless of the individual experience and skill. -Provides a method for analyzing the composition of electrical equipment parts waste, a method for processing electronic / electrical equipment parts waste, a composition analysis device for electronic / electrical equipment parts waste, and a processing device for electronic / electrical equipment parts waste.
　本発明の実施の形態に係る電子・電気機器部品屑の組成解析方法は一実施態様において、複数の部品種を含む複数の電子・電気機器部品屑を含む原料を撮像した撮像画像の中から、機械学習システムを利用した画像認識処理を用いて、電子・電気機器部品屑を抽出し、組成解析を行うことを含み、機械学習システムの学習に用いられる学習データに、組成解析対象とする原料の情報が反映されている場合は、機械学習システムの確信度を第１の閾値に設定し、組成解析対象とする原料の情報が学習データに反映されていない場合は、第１の閾値よりも低い第２の閾値に設定して電子・電気機器部品屑を抽出することを含む電子・電気機器部品屑の組成解析方法である。 The method for analyzing the composition of electronic / electrical equipment component scraps according to the embodiment of the present invention is, in one embodiment, from an image taken from an image of a raw material containing a plurality of electronic / electrical equipment component scraps including a plurality of component types. Using image recognition processing using a machine learning system, including extracting scraps of electronic and electrical equipment parts and performing composition analysis, the learning data used for learning the machine learning system includes raw materials for composition analysis. If the information is reflected, the certainty of the machine learning system is set to the first threshold, and if the information of the raw material to be analyzed for composition is not reflected in the learning data, it is lower than the first threshold. This is a method for analyzing the composition of electronic / electrical equipment component waste, which includes extracting electronic / electrical equipment component waste by setting a second threshold value.
　本発明の実施の形態に係る電子・電気機器部品屑の処理方法は一実施態様において、上記組成解析結果に基づいて、複数の部品種の中から特定の部品種を選別する選別工程を含む電子・電気機器部品屑の処理方法である。 In one embodiment, the method for treating electronic / electrical equipment component waste according to the embodiment of the present invention includes a sorting step of selecting a specific component type from a plurality of component types based on the composition analysis result. -It is a method of disposing of scraps of electrical equipment parts.
　本発明の実施の形態に係る電子・電気機器部品屑の組成解析装置は一実施態様において、複数の部品種を含む複数の電子・電気機器部品屑を含む原料を撮像した撮像画像の中から、機械学習システムを利用した画像認識処理を用いて、電子・電気機器部品屑を抽出し、組成解析を行う電子・電気機器部品屑の組成解析装置であって、機械学習システムの学習に用いられる学習データに、組成解析対象とする原料の情報が反映されている場合は、機械学習システムの確信度を第１の閾値に設定し、組成解析対象とする原料の情報が学習データに反映されていない場合は、第１の閾値よりも低い第２の閾値に設定して電子・電気機器部品屑を抽出する処理装置を備える電子・電気機器部品屑の組成解析装置である。 In one embodiment, the electronic / electrical equipment component scrap composition analyzer according to the embodiment of the present invention is obtained from an image taken from an image of a raw material containing a plurality of electronic / electrical equipment component scraps including a plurality of component types. A learning device used to learn machine learning systems, which is a composition analysis device for electronic and electrical equipment component scraps that extracts electronic and electrical equipment component scraps and analyzes the composition using image recognition processing using a machine learning system. When the information of the raw material to be the composition analysis is reflected in the data, the certainty of the machine learning system is set to the first threshold value, and the information of the raw material to be the composition analysis is not reflected in the learning data. In this case, it is a composition analysis device for electronic / electrical equipment component waste including a processing device for extracting electronic / electrical equipment component waste by setting a second threshold value lower than the first threshold value.
　本発明の実施の形態に係る電子・電気機器部品屑の処理装置は一実施態様において、複数の部品種を含む複数の電子・電気機器部品屑を撮像する撮像装置と、撮像画像の中から、機械学習システムを利用した画像認識処理を用いて、電子・電気機器部品屑を抽出し、組成解析を行う電子・電気機器部品屑の組成解析装置であって、機械学習システムの学習に用いられる学習データに、組成解析対象とする原料の情報が反映されている場合は、機械学習システムの確信度を第１の閾値に設定し、組成解析対象とする原料の情報が学習データに反映されていない場合は、第１の閾値よりも低い第２の閾値に設定して電子・電気機器部品屑を抽出する処理装置を備える組成解析装置と、組成解析装置によって解析された組成解析結果に基づいて電子・電気機器部品屑から特定の部品屑を選別する選別機とを備える電子・電気機器部品屑の処理装置である。 In one embodiment, the electronic / electrical equipment component waste processing apparatus according to the embodiment of the present invention includes an imaging device that images a plurality of electronic / electrical equipment component waste including a plurality of component types, and an image captured image. A learning device used to learn machine learning systems, which is a composition analysis device for electronic and electrical equipment component scraps that extracts electronic and electrical equipment component scraps and analyzes the composition using image recognition processing using a machine learning system. When the information of the raw material to be the composition analysis is reflected in the data, the certainty of the machine learning system is set to the first threshold value, and the information of the raw material to be the composition analysis is not reflected in the learning data. In this case, a composition analysis device provided with a processing device for extracting electronic / electrical equipment component scraps by setting a second threshold value lower than the first threshold, and electrons based on the composition analysis result analyzed by the composition analysis device. -It is a processing device for electronic / electrical equipment parts waste equipped with a sorter for sorting specific parts waste from electrical equipment parts waste.
　本開示によれば、画像認識精度を向上でき、個人の経験や技能に関係なく、電子・電気機器部品屑中の部品屑の組成を短時間で効率良く解析することが可能な電子・電気機器部品屑の組成解析方法、電子・電気機器部品屑の処理方法、電子・電気機器部品屑の組成解析装置及び電子・電気機器部品屑の処理装置が提供できる。 According to the present disclosure, the image recognition accuracy can be improved, and the composition of component scraps in electronic / electrical device scraps can be efficiently analyzed in a short time regardless of individual experience and skill. It is possible to provide a method for analyzing the composition of parts scraps, a method for treating electronic / electrical equipment parts waste, an electronic / electrical equipment parts waste composition analysis device, and an electronic / electrical equipment parts waste treatment device.
本発明の実施の形態に係る電子・電気機器部品屑の処理装置を示すブロック図である。It is a block diagram which shows the electronic / electrical equipment component waste processing apparatus which concerns on embodiment of this invention. 図２（ａ）は、撮像画像中に存在する電子・電気機器部品屑（基板）に認識枠を付与した画像の例を表す写真であり、図２（ｂ）は、認識枠を付与した電子・電気機器部品屑を部品種（基板・プラスチック）毎に並べた例を示す写真である。FIG. 2A is a photograph showing an example of an image in which a recognition frame is attached to electronic / electrical equipment component scraps (boards) existing in the captured image, and FIG. 2B is an electron having the recognition frame attached. -It is a photograph showing an example in which electrical equipment component scraps are arranged by component type (board / plastic). 電子・電気機器部品屑の画像解析処理の一例を示すフローチャートである。It is a flowchart which shows an example of the image analysis processing of the electronic / electrical equipment component waste. 撮像画像の中から電子・電気機器として基板とプラスチックとをそれぞれ抽出し、その面積を推測した結果と面積の実測値との比較の例を表す表である。It is a table showing an example of comparison between the result of estimating the area and the measured value of the area by extracting each of the substrate and the plastic as electronic / electrical equipment from the captured image. 本発明の実施の形態に係る機械学習システムを用いて撮像画像の中から電子・電気機器部品屑を抽出した場合の実測値に対する認識結果の比較の例を表す表である。It is a table which shows the example of the comparison of the recognition result with respect to the measured value at the time of extracting the electronic / electrical equipment component scrap from the captured image by using the machine learning system which concerns on embodiment of this invention.
　以下、図面を参照しながら本発明の実施の形態を説明する。以下に示す実施の形態は、この発明の技術的思想を具体化するための装置や方法を例示するものであってこの発明の技術的思想は構成部品の構造、配置等を下記のものに特定するものではない。 Hereinafter, embodiments of the present invention will be described with reference to the drawings. The embodiments shown below exemplify devices and methods for embodying the technical idea of the present invention, and the technical idea of the present invention specifies the structure, arrangement, etc. of components as follows. It is not something to do.
　本発明の実施の形態に係る電子・電気機器部品屑の処理装置は、図１に示すように、複数の部品種を含む複数の電子・電気機器部品屑を含む原料を撮像した撮像画像を取得可能な撮像装置１２と、機械学習システムを利用した画像認識処理を用いて、電子・電気機器部品屑を抽出し、組成解析を行う電子・電気機器部品屑の組成解析装置１０と、組成解析装置１０によって解析された組成解析結果に基づいて電子・電気機器部品屑から特定の部品屑を選別する選別機１３とを備える。 As shown in FIG. 1, the electronic / electrical equipment component waste processing apparatus according to the embodiment of the present invention acquires an image of a raw material containing a plurality of electronic / electrical equipment component waste including a plurality of component types. A composition analysis device 10 and a composition analysis device for electronic / electrical device component scraps that extract electronic / electrical device component scraps and perform composition analysis using a possible image pickup device 12 and image recognition processing using a machine learning system. It is provided with a sorting machine 13 that sorts specific parts scraps from electronic / electrical equipment parts scraps based on the composition analysis result analyzed by 10.
　本実施形態における「電子・電気機器部品屑」とは、廃家電製品・ＰＣや携帯電話等の電子・電気機器を破砕した屑であり、回収された後、適当な大きさに破砕されたものを指す。本実施形態では、電子・電気機器部品屑とするための破砕は、処理者自身が行ってもよいが、市中で破砕されたものを購入等したものでもよい。 The "electronic / electrical equipment component waste" in the present embodiment is waste crushed electronic / electrical equipment such as waste home appliances / PCs and mobile phones, and is crushed to an appropriate size after being collected. Point to. In the present embodiment, the crushing for making electronic / electrical equipment parts waste may be performed by the processor himself, or may be crushed in the city and purchased.
　破砕方法として、特定の装置には限定されず、せん断方式でも衝撃方式でもよいが、できる限り、部品の形状を損なわない破砕が望ましい。従って、細かく粉砕することを目的とする粉砕機のカテゴリーに属する装置は含まれない。 The crushing method is not limited to a specific device, and may be a shearing method or an impact method, but crushing that does not impair the shape of parts is desirable as much as possible. Therefore, equipment belonging to the category of crushers intended for fine crushing is not included.
　電子・電気機器部品屑は、基板、筐体などに使われるプラスチック（合成樹脂類）、金属片、銅線屑、コンデンサー、ＩＣチップ、その他、等の複数の部品種からなり、処理目的に応じて更に細かく分類することができる。以下に限定されるものではないが、本実施形態では、粒度５０ｍｍ以下に破砕されている電子・電気機器部品屑を好適に処理することができる。粒度の下限は特に限定されないが、５ｍｍ以上、より典型的には１０ｍｍ以上、更には１５ｍｍ以上である。 Electronic / electrical equipment component scraps consist of multiple component types such as plastics (synthetic resins) used for substrates and housings, metal pieces, copper wire scraps, capacitors, IC chips, etc., depending on the processing purpose. Can be further classified. Although not limited to the following, in the present embodiment, electronic / electrical equipment component scraps crushed to a particle size of 50 mm or less can be suitably treated. The lower limit of the particle size is not particularly limited, but is 5 mm or more, more typically 10 mm or more, and further 15 mm or more.
　組成解析装置１０は、本実施形態に係る組成解析アルゴリズムに従って撮像画像の画像解析処理を行う処理装置１００と、各種処理に必要な情報を記憶する記憶装置１１０と、入力装置１２０と、表示装置１３０とを備える。組成解析装置１０は、ネットワーク１１を介して、処理装置１００による処理結果を、ネットワーク１１を介して接続されたサーバ１５又は選別機１４へ送信可能に構成されている。 The composition analysis device 10 includes a processing device 100 that performs image analysis processing of captured images according to the composition analysis algorithm according to the present embodiment, a storage device 110 that stores information necessary for various processes, an input device 120, and a display device 130. And. The composition analysis device 10 is configured to be able to transmit the processing result of the processing device 100 to the server 15 or the sorting machine 14 connected via the network 11 via the network 11.
　処理装置１００は、学習データを用いた機械学習システムを利用することにより、撮像画像に対して画像認識処理を行い、撮像画像中の電子・電気機器部品屑の組成解析を行う。機械学習には、ディープラーニング等を利用した画像認識のための種々の解析ソフトが利用可能である。記憶装置１１０には、処理装置１００による処理に必要な情報が格納される。 The processing device 100 performs image recognition processing on the captured image by using a machine learning system using the learning data, and analyzes the composition of electronic / electrical equipment component scraps in the captured image. For machine learning, various analysis software for image recognition using deep learning or the like can be used. The storage device 110 stores information necessary for processing by the processing device 100.
　例えば、記憶装置１１０には、電子・電気機器部品屑の画像情報から、複数の部品種、即ち、基板、プラスチック、金属片、銅線屑、コンデンサー、ＩＣチップ、その他（コネクタ、フィルム状部品屑、被覆線屑等）の少なくとも２種類以上、好ましくは７種類以上、更には１０種類以上に分類するための抽出データ、抽出された電子・電気機器部品屑に対して認識枠を付与するための認識枠データ、認識枠中の電子・電気機器部品屑の面積を計測又は推定するための部品種面積率データ、認識枠で囲まれた電子・電気機器部品屑の組成解析に必要な解析データ等を含む種々の組成解析情報等が格納される。 For example, in the storage device 110, a plurality of component types, that is, a substrate, a plastic, a metal piece, a copper wire scrap, a condenser, an IC chip, and others (connector, film-shaped component scrap), etc. , Covered wire scraps, etc.), preferably 7 or more types, and further 10 or more types of extracted data, and for assigning a recognition frame to the extracted electronic / electrical equipment parts scraps. Recognition frame data, part type area ratio data for measuring or estimating the area of electronic / electrical equipment parts waste in the recognition frame, analysis data necessary for composition analysis of electronic / electrical equipment parts waste surrounded by the recognition frame, etc. Various composition analysis information including the above are stored.
　抽出処理に利用される抽出情報としては、例えば、電子・電気機器部品屑の輪郭等の幾何形状を抽出するための形状情報、電子・電気機器部品屑の色彩を抽出するための色彩情報、電子・電気機器部品屑の周囲の背景画像の色彩、凹凸による影等の背景情報を抽出するための背景情報等、電子・電気機器部品屑の回転画像等を含む。 The extraction information used in the extraction process includes, for example, shape information for extracting geometric shapes such as contours of electronic / electrical equipment component scraps, color information for extracting the color of electronic / electrical equipment component scraps, and electronics. -Includes rotating images of electronic / electrical equipment parts scraps, etc., such as background information for extracting background information such as colors of background images around electrical equipment parts scraps and shadows due to unevenness.
　処理装置１００は、記憶装置１１０に格納された学習データを用いた機械学習により、撮像画像の中から、複数の電子・電気機器部品屑を抽出し、これを複数の部品種毎に区別して分類する。学習データには過去の画像解析結果に基づく電子・電気機器部品屑の特徴を検出するための種々の情報が含まれており、入力装置１２０を介して新規データが入力され、機械学習システムの学習に用いられる学習データを更新（反映）することにより、電子・電気機器部品屑の認識精度を向上することが行われる。 The processing device 100 extracts a plurality of electronic / electrical equipment component scraps from the captured image by machine learning using the learning data stored in the storage device 110, and classifies them by classifying them according to the plurality of component types. do. The training data includes various information for detecting the characteristics of electronic / electrical equipment component scraps based on the past image analysis results, and new data is input via the input device 120 to learn the machine learning system. By updating (reflecting) the learning data used in the above, the recognition accuracy of electronic / electrical equipment component scraps is improved.
　しかしながら、電子・電気機器部品屑は、原料として搬入される前に破砕処理が行われている場合が多く、定まった形を有さず、種々の部品を含み得るため、これらの全てを学習データに反映させることが困難な場合がある。例えば、新しい入手ルートからの電子・電気機器部品屑は、学習データに反映させることが困難であることが多い。機械学習システムの学習に用いられる学習データに組成解析対象とする原料の情報が反映されていない場合には、抽出処理を行ってもその精度が高くなくなり、誤認識率が急激に高くなる。具体的には、誤認識率が一定値を超える場合、例えば本実施形態では誤認識率が１５％、更には１０％を超える場合は、機械学習システムの学習に用いられる学習データに組成解析対象とする原料の情報が反映されていない場合として判断できる。 However, electronic / electrical equipment parts waste is often crushed before being carried in as a raw material, does not have a fixed shape, and may contain various parts. Therefore, all of these are learned data. It may be difficult to reflect in. For example, it is often difficult to reflect the scraps of electronic / electrical equipment parts from a new acquisition route in the learning data. If the learning data used for learning the machine learning system does not reflect the information of the raw material to be the composition analysis target, the accuracy will not be high even if the extraction process is performed, and the erroneous recognition rate will increase sharply. Specifically, when the false recognition rate exceeds a certain value, for example, in the present embodiment, when the false recognition rate exceeds 15% and further exceeds 10%, the learning data used for learning the machine learning system is subject to composition analysis. It can be judged as a case where the information of the raw material to be used is not reflected.
　選別処理対象とする新規の原料が搬送された場合に、処理装置１００がその新規の原料の撮像画像の中から、電子・電気機器部品屑の抽出処理を行うに際し、機械学習システムの学習に用いられる学習データに、組成解析対象とする原料の情報が反映されている場合には、機械学習システムの確信度を第１の閾値に設定し、組成解析対象とする原料の情報が学習データに反映されていない場合は、第１の閾値よりも低い第２の閾値に設定する。 When a new raw material to be sorted is transported, the processing device 100 uses it for learning a machine learning system when performing an extraction process of electronic / electrical equipment component waste from the captured image of the new raw material. When the information of the raw material to be the composition analysis target is reflected in the learning data to be obtained, the certainty of the machine learning system is set to the first threshold value, and the information of the raw material to be the composition analysis target is reflected in the learning data. If not, the second threshold is set to be lower than the first threshold.
　組成解析対象とする原料の情報が学習データに反映されていない場合は、第１の閾値よりも低い第２の閾値に設定することにより、電子・電気機器部品屑の認識対象を広げることができるため、認識対象範囲を緩和することができる。その結果、新規の原料に対しても電子・電気機器部品屑の抽出率を向上させることができ、原料の組成解析の精度を向上させることができる。なお、本実施形態の確信度とは、図１の処理装置１００が備える機械学習システムによる電子・電気機器部品屑の抽出結果の確からしさの度合いを示す。 When the information of the raw material to be the composition analysis target is not reflected in the learning data, the recognition target of the electronic / electrical equipment component waste can be expanded by setting the second threshold value lower than the first threshold value. Therefore, the recognition target range can be relaxed. As a result, the extraction rate of electronic / electrical equipment parts waste can be improved even for a new raw material, and the accuracy of the composition analysis of the raw material can be improved. The certainty of the present embodiment indicates the degree of certainty of the extraction result of the electronic / electrical equipment component waste by the machine learning system included in the processing device 100 of FIG.
　電子・電気機器部品屑の抽出処理の場合、確信度は高く設定すると電子・電気機器部品屑の抽出のための認識率を高めることができる一方で、抽出個数が少なくなるため、原料の組成判断のための抽出処理としては適切とはいえない。一方、認識率を低く設定しすぎることで誤認識が多くなることから、原料の組成判断のための抽出処理としては適切とはいえない。本実施形態では、第１の閾値を０．２～０．５とすることが好ましく、より好ましくは、０．２５～０．４とする。第２の閾値は０．０１～０．１とすることが好ましく、より好ましくは０．０２～０．０９とする。その結果、新規原料に対しても撮像画像から、目視による実測値に比べて７０％以上の電子・電気機器部品屑を抽出することができ、これにより組成解析の精度を高めることができる。 In the case of extraction processing of electronic / electrical equipment parts waste, if the certainty is set high, the recognition rate for extracting electronic / electrical equipment parts waste can be increased, but the number of extractions is small, so the composition of the raw material is judged. It cannot be said that it is appropriate as an extraction process for. On the other hand, if the recognition rate is set too low, erroneous recognition increases, so that it is not appropriate as an extraction process for determining the composition of raw materials. In the present embodiment, the first threshold value is preferably 0.2 to 0.5, more preferably 0.25 to 0.4. The second threshold value is preferably 0.01 to 0.1, more preferably 0.02 to 0.09. As a result, 70% or more of the electronic / electrical equipment component scraps can be extracted from the captured image even for the new raw material as compared with the measured value by visual inspection, and thus the accuracy of the composition analysis can be improved.
　処理装置１００は、確信度の第１及び第２の閾値の設定値に従って抽出された電子・電気機器部品屑に対し、認識枠を付与する。認識枠は、電子・電気機器部品屑の輪郭を縁取った最小枠としてもよいし、電子・電気機器部品屑及び電子・電気機器部品屑の周囲の背景の画像を含む認識枠を付与してもよい。 The processing device 100 assigns a recognition frame to the electronic / electrical equipment component scraps extracted according to the set values of the first and second threshold values of the certainty. The recognition frame may be a minimum frame that borders the outline of the electronic / electrical equipment component waste, or a recognition frame including an image of the background around the electronic / electrical equipment component waste and the electronic / electrical equipment component waste is added. May be good.
　電子・電気機器部品屑の周囲の背景の画像を含む認識枠を付与する場合には、電子・電気機器部品屑の輪郭と外接する外接図形を認識枠として付与することが好ましい。外接図形の形状は、外接矩形、外接円形、外接多角形など任意の形状を有していてもよい。 When a recognition frame including an image of the background around the electronic / electrical equipment component scraps is given, it is preferable to give a circumscribed figure circumscribing the outline of the electronic / electrical equipment parts scraps as the recognition frame. The shape of the inscribed figure may have any shape such as a circumscribed rectangle, a circumscribed circle, and a circumscribed polygon.
　認識枠は、操作者が複数の部品屑を見分けやすいように、部品種毎に異なる特性（色彩、線の濃淡、線の種類（太さ、点線／実線など））を有して付与されることが望ましい。図２（ａ）は撮像画像から電子・電気機器部品屑として基板を抽出した場合の例を表す写真である。図２（ｂ）の紙面上部が基板、紙面下部がプラスチックの抽出物の例を示す。 The recognition frame is given with different characteristics (color, line shading, line type (thickness, dotted line / solid line, etc.)) for each part type so that the operator can easily distinguish a plurality of part scraps. Is desirable. FIG. 2A is a photograph showing an example in which a substrate is extracted as scraps of electronic / electrical equipment parts from a captured image. The upper part of the paper surface of FIG. 2B shows an example of a substrate, and the lower part of the paper surface shows an example of a plastic extract.
　処理装置１００は、更に、付与された認識枠内に存在する電子・電気機器部品屑の面積を計測又は推測する。電子・電気機器部品屑の面積の計測は、撮像画像に基づいて面積を計測することが可能な面積計測ツールを用いることができる。電子・電気機器部品屑の面積を機械学習システムを用いて推定する場合には、処理装置１００が、記憶装置１１０に記憶された、認識枠に対する電子・電気機器部品屑の面積率の情報を少なくとも有する部品種面積率データを用いて、複数の部品種毎に、認識枠が付された電子・電気機器部品屑の合計面積を推測する。部品種面積率データとしては、電子・電気機器部品屑の輪郭の形状情報、認識枠の面積、認識枠内に占める電子・電気機器部品屑の面積及び背景の面積、認識枠に対する電子・電気機器部品屑の面積率の情報、位置情報等の種々の情報を含むことができる。 The processing device 100 further measures or estimates the area of the electronic / electrical equipment component waste existing in the given recognition frame. For the measurement of the area of the scraps of electronic / electrical equipment parts, an area measurement tool capable of measuring the area based on the captured image can be used. When estimating the area of electronic / electrical equipment component waste using a machine learning system, the processing device 100 at least obtains information on the area ratio of electronic / electrical equipment component waste with respect to the recognition frame stored in the storage device 110. Using the component type area ratio data, the total area of electronic / electrical equipment component scraps with recognition frames is estimated for each of a plurality of component types. The part type area ratio data includes the shape information of the outline of the electronic / electrical equipment component scrap, the area of the recognition frame, the area of the electronic / electrical equipment component scrap and the background area occupied in the recognition frame, and the electronic / electrical equipment with respect to the recognition frame. Various information such as information on the area ratio of parts scraps and position information can be included.
　処理装置１００は、更に、部品種毎の電子・電気機器部品屑の合計面積の推測結果と予め定められた複数の部品種毎の単位面積当たりの想定重量とを乗算し、複数の部品種に含まれる電子・電気機器部品屑の重量を解析することで、複数の部品種の重量比率をそれぞれ解析し、これにより、撮像画像内に存在する電子・電気機器部品屑の組成を解析する。 The processing device 100 further multiplies the estimation result of the total area of electronic / electrical equipment component scraps for each component type by the estimated weight per unit area for each of a plurality of predetermined component types to obtain a plurality of component types. By analyzing the weight of the electronic / electrical equipment component scraps contained, the weight ratios of the plurality of component types are analyzed, respectively, and thereby the composition of the electronic / electrical equipment component scraps existing in the captured image is analyzed.
　複数の部品種の単位面積当たりの想定重量は、操業結果に応じて予め操作者により入力装置１２０等を介して設定しておくことができる。以下に限定されるものではないが、例えば、電子・電気機器部品屑を基板、プラスチック、その他部品の３種類に分類する場合、基板屑の想定重量を例えば２．０ｇ／ｃｍ2、プラスチックの想定重量を１．５ｇ／ｃｍ2、その他の部品を１．０ｇ／ｃｍ2と設定することができる。 The estimated weight per unit area of the plurality of component types can be set in advance by the operator via the input device 120 or the like according to the operation result. Although not limited to the following, for example, when classifying electronic / electrical equipment parts scraps into three types of substrates, plastics, and other parts, the estimated weight of the substrate scraps is, for example, 2.0 g / cm 2 , and the assumed plastics. The weight can be set to 1.5 g / cm 2 and the other parts can be set to 1.0 g / cm 2.
　なお、処理装置１００は、上記で説明した面積の情報の他に、抽出した部品種毎にその部品種を構成する部品の個数（個）、上記の面積の推測結果と個数とから算出される平均粒径、夫々の部品種を構成する元素の重量比などのその他物理的特性についても解析し、表示装置１３０等に出力することもできる。 In addition to the area information described above, the processing device 100 is calculated from the number (pieces) of the parts constituting the extracted part type for each extracted part type, and the estimation result and the number of the above area. Other physical properties such as the average particle size and the weight ratio of the elements constituting each component type can also be analyzed and output to the display device 130 or the like.
　処理装置１００は、更に、複数の部品種の重量比率の解析結果に基づいて、複数の部品種を選別するための選別機の運転条件の情報を生成することが好ましい。選別機としては、ピッキング、カラーソーター、メタルソーター、渦電流選別機、風力選別機、篩別機などの種々の選別機がある。例えば、複数の部品種の解析結果から、処理装置１００が、例えば基板とプラスチックとを選別するカラーソーターの運転条件を生成し、生成した運転条件を記憶装置１１０へ格納する。記憶装置１１０へ格納された運転条件は、選別機１３、１４へ出力されて、選別機１３、１４が、出力された運転条件に応じて選別処理を行うことができる。 It is preferable that the processing device 100 further generates information on the operating conditions of the sorter for sorting the plurality of part types based on the analysis result of the weight ratio of the plurality of part types. As the sorter, there are various sorters such as a picking machine, a color sorter, a metal sorter, an eddy current sorter, a wind power sorter, and a sieving machine. For example, from the analysis results of a plurality of component types, the processing device 100 generates operating conditions for a color sorter that selects, for example, a substrate and plastic, and stores the generated operating conditions in the storage device 110. The operating conditions stored in the storage device 110 are output to the sorters 13 and 14, and the sorters 13 and 14 can perform the sorting process according to the output operating conditions.
　処理装置１００は、電子・機器部品屑の撮像画像から抽出された電子・電気機器部品屑のそれぞれの位置情報を取得してもよい。そして、特定の電子・電気機器部品屑の位置を抽出してこれを選別するための特定の選別機１３、１４に対し、位置情報を出力するように構成されてもよい。例えば、基板と金属片はメタルソーター等の特定の選別機１３、１４では分離できないが、画像情報で個別に位置情報が得られれば、ピッキング機能を備える選別機１３、１４によってこれらを選別することができるようになる。 The processing device 100 may acquire the position information of each of the electronic / electrical equipment component scraps extracted from the captured image of the electronic / equipment component scraps. Then, the position information may be output to the specific sorters 13 and 14 for extracting the position of the specific scraps of electronic / electrical equipment parts and sorting them. For example, the substrate and the metal piece cannot be separated by specific sorters 13 and 14 such as a metal sorter, but if the position information can be obtained individually from the image information, they can be sorted by the sorters 13 and 14 having a picking function. Will be able to.
　処理装置１００は、抽出処理及び推測処理に必要な学習データを機械学習により更に学習することができる。例えば、処理装置１００は、基板、プラスチック、金属片、銅線屑などの複数の部品種毎の特徴、例えば本実施形態では、各部品種毎の色彩、形状、認識枠と部品屑の面積率の関係等の情報に関し、数百枚～数万枚のデータの入力に基づいてその特徴を学習し、抽出処理及び推測処理の精度を向上させるように学習することができる。 The processing device 100 can further learn the learning data required for the extraction process and the guess process by machine learning. For example, the processing device 100 has characteristics for each of a plurality of component types such as a substrate, plastic, metal pieces, and copper wire scraps, for example, in the present embodiment, the color, shape, recognition frame, and area ratio of the component scraps for each component type. It is possible to learn the characteristics of information such as relationships based on the input of hundreds to tens of thousands of sheets of data, and to improve the accuracy of extraction processing and estimation processing.
　また、電子・電気機器部品屑の誤認識等が生じた場合に、或いは電子・電気機器部品屑の抽出漏れが生じた場合には、誤認識又は抽出漏れが生じた電子・電気機器部品屑の特性を処理装置１００の機械学習システムが更に学習するように構成されることが好ましい。例えば、処理装置１００は、電子・電気機器部品屑の抽出処理によって抽出されなかった電子・電気機器部品屑の輪郭の情報、抽出処理で抽出されなかった電子・電気機器部品屑と背景の画像とを含む新たな認識枠の情報、その認識枠に対する電子・電気機器部品屑の面積率の情報等の入力に応じた機械学習等により新たな学習モデルを作製することができる。 In addition, when misrecognition of electronic / electrical equipment parts waste occurs, or when extraction omission of electronic / electrical equipment parts waste occurs, misrecognition or extraction omission occurs in the electronic / electrical equipment parts waste. It is preferable that the machine learning system of the processing device 100 further learns the characteristics. For example, the processing device 100 includes information on the outline of the electronic / electrical equipment component waste not extracted by the extraction process, the electronic / electrical equipment component waste not extracted by the extraction process, and an image of the background. A new learning model can be created by machine learning or the like in response to input of information on a new recognition frame including the above, information on the area ratio of electronic / electrical equipment component scraps with respect to the recognition frame, and the like.
　本発明の実施の形態に係る電子・電気機器部品屑の組成解析装置によれば、撮像画像の画像処理によって、電子・電気機器部品屑の部品種毎の重量比率を数値化して評価することができる。これにより、原料の組成にとって、原料に対して物理選別を行うべきか、キルン炉等による焼却処理を行うか等の選別条件の選択をより効率的に行うことができる。撮像画像の認識結果に基づいて、例えば、原料中のプラスチック比率が多い場合には、プラスチックの重量比率に基づいて購入条件を見直すことや、キルン炉の熱負荷等を調整することも可能となる。 According to the composition analyzer for electronic / electrical equipment component waste according to the embodiment of the present invention, the weight ratio of each component type of electronic / electrical equipment component waste can be quantified and evaluated by image processing of the captured image. can. Thereby, for the composition of the raw material, it is possible to more efficiently select the sorting conditions such as whether the raw material should be physically sorted or incinerated by a kiln furnace or the like. Based on the recognition result of the captured image, for example, when the ratio of plastic in the raw material is large, it is possible to review the purchase conditions based on the weight ratio of plastic and adjust the heat load of the kiln furnace. ..
　図１に示す電子・電気機器部品屑の処理装置を用いた電子・電気機器部品屑の処理方法の一例について、図３のフローチャートを用いて説明する。まず、ステップＳ１００において、撮像画像が取得される。撮像画像は入力装置１２０或いはネットワーク１１を介して入力された撮像画像でもよいし、撮像装置１２による撮像結果を用いてもよい。ステップＳ１０１において、撮像画像に含まれる、組成解析対象とする原料の情報が、機械学習システムの学習に用いられる学習データに既に反映されているか否かが判断される。判断は入力装置１２０を介して操作者が手動で行っても良いし、処理装置１００が記憶装置１１０のデータを参照して自動的に行っても良い。 An example of a method for processing electronic / electrical equipment component waste using the electronic / electrical equipment component waste processing apparatus shown in FIG. 1 will be described with reference to the flowchart of FIG. First, in step S100, the captured image is acquired. The captured image may be an captured image input via the input device 120 or the network 11, or the captured image obtained by the imaging device 12 may be used. In step S101, it is determined whether or not the information of the raw material to be the composition analysis target included in the captured image is already reflected in the learning data used for learning the machine learning system. The determination may be made manually by the operator via the input device 120, or may be made automatically by the processing device 100 with reference to the data of the storage device 110.
　撮像画像に撮像された組成解析対象とする原料の情報が、機械学習システムの学習に用いられる学習データに反映されている場合は、ステップＳ１０２に進み、機械学習システムの確信度が第１の閾値に設定される。組成解析対象とする原料の情報が学習データに反映されていない場合は、ステップＳ１０３へ進み、第１の閾値よりも低い第２の閾値に設定され、ステップＳ１０４へ進む。 If the information of the raw material to be analyzed for composition captured in the captured image is reflected in the learning data used for learning the machine learning system, the process proceeds to step S102, and the certainty of the machine learning system is the first threshold value. Is set to. If the information of the raw material to be the composition analysis target is not reflected in the learning data, the process proceeds to step S103, is set to a second threshold value lower than the first threshold value, and proceeds to step S104.
　ステップＳ１０４～ステップＳ１０７において、機械学習システムを利用した画像認識処理が行われる。ステップＳ１０４において、処理装置１００は、設定された機械学習システムの確信度に基づいて、撮像画像内に存在する電子・電気機器部品屑を、部品種毎（例えば、基板、プラスチック、金属片、銅線屑、コンデンサー、ＩＣチップ、その他の部品種の７分類）に分類して、抽出する。分類結果は、表示装置１３０等によって表示されることができる。 In steps S104 to S107, image recognition processing using a machine learning system is performed. In step S104, the processing apparatus 100 removes the electronic / electrical equipment component scraps existing in the captured image for each component type (for example, substrate, plastic, metal piece, copper) based on the set certainty of the machine learning system. Classify into 7 categories of wire scraps, capacitors, IC chips, and other parts) and extract. The classification result can be displayed by the display device 130 or the like.
　ステップＳ１０５において、処理装置１００は、撮像画像から抽出され複数の電子・電気機器部品屑に対し、複数の部品種毎に、電子・電気機器部品屑の周囲の背景の画像を含む認識枠を付与する。ステップＳ１０６において、処理装置１００は、認識枠に対する電子・電気機器部品屑の面積率の情報を有する部品種面積率データに基づいて、複数の部品種毎に、認識枠が付された前記電子・電気機器部品屑の面積（合計面積）を推測する。ステップＳ１０７において、処理装置１００は、面積の推測結果と、複数の部品種の単位面積当たりの想定重量を乗算して複数の部品種の重量比率をそれぞれ解析することにより、撮像画像内の電子・電気機器部品屑の組成を解析する。例えば、画像内の部品種毎の合計面積に、単位面積当たりの重量を乗算すると、部品屑の総重量を概ね算出することができる。各部品種毎の部品屑の総重量をそれぞれ対比することで、撮像画像内に含まれる電子・電気機器部品屑の組成が解析できる。解析結果は、ステップＳ１０８において出力される。 In step S105, the processing apparatus 100 assigns a recognition frame including an image of the background around the electronic / electrical equipment component scraps to the plurality of electronic / electrical equipment component scraps extracted from the captured image for each of the plurality of component types. do. In step S106, the processing device 100 attaches a recognition frame to each of a plurality of parts types based on the part type area ratio data having information on the area ratio of the electronic / electrical equipment parts waste with respect to the recognition frame. Estimate the area (total area) of electrical equipment parts scraps. In step S107, the processing apparatus 100 multiplies the area estimation result and the assumed weight per unit area of the plurality of component types to analyze the weight ratios of the plurality of component types, thereby performing the electrons in the captured image. Analyze the composition of electrical equipment component scraps. For example, by multiplying the total area of each part type in the image by the weight per unit area, the total weight of the part scraps can be roughly calculated. By comparing the total weight of the component scraps for each component type, the composition of the electronic / electrical equipment component scraps contained in the captured image can be analyzed. The analysis result is output in step S108.
　本発明の実施の形態に係る電子・電気機器部品屑の組成解析方法及び組成解析装置によれば、画像解析によって、撮像画像中に存在する電子・電気機器部品屑の部品種毎の重量比率を求めることができる。これにより、原料組成を、個人の経験や技能に関係なく迅速に推測することができる。その結果、操作者が、原料の購入条件や原料の選別方法の判断をより早期に行うことができるようになるため、工場全体をより効率的に運用できる。 According to the composition analysis method and the composition analysis apparatus of the electronic / electrical equipment component waste according to the embodiment of the present invention, the weight ratio of the electronic / electrical equipment component waste existing in the captured image for each component type is determined by the image analysis. You can ask. As a result, the raw material composition can be quickly estimated regardless of the individual's experience and skill. As a result, the operator can determine the purchase conditions of the raw materials and the selection method of the raw materials at an earlier stage, so that the entire factory can be operated more efficiently.
　電子・電気機器部品屑の周囲の背景画像は、撮像先の条件により異なる場合があり、場合によっては、機械学習システムが電子・電気機器部品屑を適切に抽出できない場合がある。本実施形態では、機械学習システムが、背景の情報と、電子・電気機器部品屑の縁取り画像を組み合わせた画像を合成した新たな認識枠の情報を含む、電子・電気機器部品屑の抽出処理のための学習データを新たに学習することにより、撮像画像中の種々の条件に対応したより柔軟な組成解析装置を得ることができる。また、本システムを用いた場合にも誤認識率が高い場合には、学習データに反映されていない原料に含まれる電子・電気機器部品屑の輪郭、色彩、認識枠に対する面積率、及び背景の少なくともいずれかの情報を、機械学習システムに学習させることで電子・電気機器部品屑の認識精度を高めることができる。 The background image around the electronic / electrical equipment component scraps may differ depending on the conditions of the imaging destination, and in some cases, the machine learning system may not be able to properly extract the electronic / electrical equipment component scraps. In the present embodiment, the machine learning system is used for extracting electronic / electrical equipment component waste, which includes information on a new recognition frame obtained by synthesizing an image obtained by combining background information and a border image of electronic / electrical equipment component waste. By newly learning the learning data for the purpose, it is possible to obtain a more flexible composition analysis device corresponding to various conditions in the captured image. If the false recognition rate is high even when this system is used, the outline, color, area ratio to the recognition frame, and background of the electronic / electrical equipment parts scraps contained in the raw materials that are not reflected in the learning data. By having the machine learning system learn at least one of the information, it is possible to improve the recognition accuracy of the scraps of electronic / electrical equipment parts.
　図４は、本発明の実施の形態に係る解析方法に従って、撮像画像の中から電子・電気機器として基板とプラスチックとをそれぞれ抽出し、その面積を推測した結果と実測値との比較の例を表す表である。本発明の実施の形態に係る電子・電気機器部品屑の組成解析方法によれば、基板については５．０％未満の測定誤差で、プラスチックについては１０％未満の測定誤差で適切な評価ができており、二種類の部品を含む場合には、認識率９０％以上を達成することができた。そのため、原料の組成を数値的に大まかに把握する上では、本手法により十分な効果が得られているといえる。 FIG. 4 shows an example of comparison between the result of estimating the area and the measured value by extracting the substrate and the plastic as electronic / electrical devices from the captured image according to the analysis method according to the embodiment of the present invention. It is a table to represent. According to the method for analyzing the composition of electronic / electrical equipment component scraps according to the embodiment of the present invention, an appropriate evaluation can be performed with a measurement error of less than 5.0% for the substrate and a measurement error of less than 10% for the plastic. When two types of parts were included, a recognition rate of 90% or more could be achieved. Therefore, it can be said that this method has a sufficient effect in roughly grasping the composition of the raw material.
　図５は、本発明の実施の形態に係る機械学習システムを用いて撮像画像の中から電子・電気機器部品屑を抽出した場合の実測値に対する認識結果の比較の例を表す表である。実施例１０は、機械学習システムの学習に用いられる学習データにその特徴を反映させた場合の抽出個数の実測値と認識（抽出）個数、誤認識個数、認識率及び誤認識率を示し、この場合の確信度を０．３とした。実施例２及び比較例２は、学習データにその原料情報が反映されていない場合の例を示す。実施例２の確信度を０．０７とし、比較例２の確信度を０．３とした。実施例２と比較例１の結果からわかるように、確信度を低くすることにより機械学習システムによる誤認識個数は若干増加するが、確信度を低くした方が、認識率は良好な結果が得られている。原料の組成を数値的に大まかに把握する上では、本手法により十分な効果が得られているといえる。 FIG. 5 is a table showing an example of comparison of recognition results with respect to actual measurement values when electronic / electrical equipment component scraps are extracted from captured images using the machine learning system according to the embodiment of the present invention. Example 10 shows the actual measurement value of the number of extracts and the number of recognitions (extractions), the number of false recognitions, the recognition rate, and the false recognition rate when the characteristics are reflected in the learning data used for learning of the machine learning system. The certainty of the case was set to 0.3. Example 2 and Comparative Example 2 show an example in which the raw material information is not reflected in the learning data. The certainty of Example 2 was 0.07, and the certainty of Comparative Example 2 was 0.3. As can be seen from the results of Example 2 and Comparative Example 1, the number of false recognitions by the machine learning system increases slightly by lowering the certainty, but the lower the certainty, the better the recognition rate is obtained. Has been done. It can be said that this method has a sufficient effect in roughly grasping the composition of the raw material.
　本発明は上記の実施の形態によって記載したが、この開示の一部をなす論述及び図面はこの発明を限定するものであると理解すべきではない。即ち、本発明は各実施形態に限定されるものではなく、その要旨を逸脱しない範囲で構成要素を変形して具体化できる。また、各実施形態に開示されている複数の構成要素の適宜な組み合わせにより、種々の発明を形成できる。例えば、実施形態に示される全構成要素からいくつかの構成要素を削除してもよい。更に、異なる実施形態の構成要素を適宜組み合わせてもよい。 Although the present invention has been described in accordance with the above embodiments, the statements and drawings that form part of this disclosure should not be understood to limit the invention. That is, the present invention is not limited to each embodiment, and the components can be modified and embodied within a range that does not deviate from the gist thereof. In addition, various inventions can be formed by appropriately combining the plurality of components disclosed in each embodiment. For example, some components may be removed from all the components shown in the embodiments. Further, the components of different embodiments may be combined as appropriate.
　例えば、処理装置１００が、撮像画像の中から複数の部品種毎の平均の面積、個数、平均粒径、重量比などを数値化して解析することにより、従来のように、手選別で電子・電気機器部品屑の原料組成を評価するよりも著しく迅速にその原料組成を数値化して把握することができる。 For example, the processing device 100 digitizes and analyzes the average area, number, average particle size, weight ratio, etc. of each of a plurality of component types from the captured image, so that the electrons can be manually sorted as in the conventional case. The raw material composition can be quantified and grasped remarkably more quickly than the raw material composition of electrical equipment parts waste is evaluated.
　更に、処理装置１００が解析した原料解析結果に基づいて、複数の部品種の中から特定の部品種を選別する選別機の運転条件の情報、例えば、原料を選別処理するための選別機の選択と、選別条件、選別順序等の操業条件を決定し、その操業条件に基づいて選別処理を行うことができる。 Further, based on the raw material analysis result analyzed by the processing apparatus 100, information on the operating conditions of the sorter that sorts a specific part type from a plurality of part types, for example, selection of a sorter for sorting raw materials. Then, the operating conditions such as the sorting conditions and the sorting order can be determined, and the sorting process can be performed based on the operating conditions.
　例えば、電子・電気機器部品屑に対して風力選別機を用いて風力選別を行って軽量物と重量物とに選別することにより、選別後の処理物中の基板とプラスチックの重量比率を上げるための処理を行うことができる。この場合、選別機１３、１４による処理においては、機械学習システムが解析した原料解析結果に基づいて、複数の部品種の平均粒径に応じて、風力選別機の風量を調整することができる。風量は例えば５～２０ｍ／ｓ、より好ましくは５～１２ｍ／ｓ、更には５～１０ｍ／ｓ程度とすることができる。風力選別は原料解析結果に応じて２回以上繰り返して行うことができる。 For example, in order to increase the weight ratio of the substrate and plastic in the processed material after sorting by performing wind power sorting using a wind power sorter on the scraps of electronic and electrical equipment parts to sort them into lightweight and heavy goods. Can be processed. In this case, in the processing by the sorters 13 and 14, the air volume of the wind power sorter can be adjusted according to the average particle size of a plurality of component types based on the raw material analysis results analyzed by the machine learning system. The air volume can be, for example, 5 to 20 m / s, more preferably 5 to 12 m / s, and further 5 to 10 m / s. Wind power sorting can be repeated twice or more depending on the raw material analysis result.
　或いは、上記の風力選別を実施する前に、ピッキング装置を用いたピッキング処理を行うことにより、塊状の銅線屑を取り除くピッキング処理を行うことができる。このピッキング処理に際しては、記憶装置１１０に記憶された銅線屑の位置情報を選別機１３としてのピッキング装置に出力し、ピッキング装置がその出力結果に応じて銅線屑を取り除くことができる。この銅線屑は、例えば有価金属回収工程へ送ることができる。 Alternatively, by performing a picking process using a picking device before carrying out the above wind power sorting, a picking process for removing lumpy copper wire debris can be performed. In this picking process, the position information of the copper wire scraps stored in the storage device 110 is output to the picking device as the sorter 13, and the picking device can remove the copper wire scraps according to the output result. This copper wire scrap can be sent to, for example, a valuable metal recovery process.
　風力選別を二回以上繰り返す場合は、第１回目の風力選別と第２回目の風力選別との間に篩別機を用いた選別処理を行うことができる。この場合、選別機１３としては篩別機が採用され、原料解析結果で得られる複数の部品種の平均粒径に基づいて、特性の部品種を選別するための篩別機の篩目の寸法を変更することができる。 When the wind power sorting is repeated twice or more, the sorting process using a sieving machine can be performed between the first wind power sorting and the second wind power sorting. In this case, a sieving machine is adopted as the sorting machine 13, and the size of the sieve mesh of the sieving machine for sorting characteristic part types based on the average particle size of a plurality of part types obtained in the raw material analysis result. Can be changed.
　上記で説明した手法の他にも、磁力選別工程、渦電流選別工程、及び金属物と非金属物とを光学的に選別する光学式選別工程に用いられる選別機１３に対してそれぞれ本発明の実施の形態に係る組成解析装置による組成解析結果を活用することで、搬送中の電子・電気機器部品屑を連続的に撮影しながら、その画像データをリアルタイムに解析し、原料組成を解析することができる。 In addition to the methods described above, the present invention relates to the sorter 13 used in the magnetic force sorting step, the eddy current sorting step, and the optical sorting step of optically sorting metal objects and non-metallic objects. By utilizing the composition analysis result by the composition analyzer according to the embodiment, the image data is analyzed in real time while continuously photographing the electronic / electrical equipment parts scraps being transported, and the raw material composition is analyzed. Can be done.
　従来、電子・電気機器部品屑の原料組成は、手選別によって評価し、その結果を選別処理の操業管理、運転条件の設定に反映させることが行われていたが、しかしながら、手選別により原料組成を把握する手法では、迅速な処理を行うことができなかった。 Conventionally, the raw material composition of electronic / electrical equipment parts waste is evaluated by hand sorting, and the result is reflected in the operation management of the sorting process and the setting of operating conditions. However, the raw material composition is manually sorted. It was not possible to perform rapid processing with the method of grasping.
　本発明の実施の形態によれば、時々刻々とその組成が変化する電子・電気機器部品屑の中からその中の部品屑の組成を画像解析と所定の分類データに基づく分離によって、瞬時に判別し数値化することができるため、大量の電子・電気機器部品屑をより適切な条件で迅速に選別を行うことができる。 According to the embodiment of the present invention, the composition of the component scraps in the electronic / electrical equipment component scraps whose composition changes from moment to moment is instantly determined by image analysis and separation based on predetermined classification data. Since it can be quantified, a large amount of electronic / electrical equipment parts waste can be quickly sorted under more appropriate conditions.
　更に、選別機１３、１４による処理前後の部品屑の原料組成を画像解析することで、部品屑の変化量に基づいて、選別機１３、１４の選別効率（成績）を評価することができる。電子・電気機器部品屑の原料組成を判別するとともにその位置情報を抽出し、ピッキング装置やカラーソーター、メタルソーターなどの選別機１３、１４と連動させることで、部品種の個別分離が容易になる。また、表示装置１３０に解析結果として各原料種毎に色の異なる枠を付けて表示させることで操作者が認識しやすくなるため、組成解析装置の誤検知も認識しやすくなる。 Further, by image-analyzing the raw material composition of the parts scraps before and after the processing by the sorters 13 and 14, the sorting efficiency (results) of the parts scraps 13 and 14 can be evaluated based on the amount of change in the parts scraps. By discriminating the raw material composition of electronic / electrical equipment parts waste, extracting its position information, and linking it with sorting machines 13 and 14 such as picking devices, color sorters, and metal sorters, individual separation of parts types becomes easy. .. Further, since the display device 130 displays the analysis result with a frame having a different color for each raw material type, the operator can easily recognize it, so that the false detection of the composition analysis device can be easily recognized.
　１０…組成解析装置
　１１…ネットワーク
　１２…撮像装置
　１３、１４…選別機
　１５…サーバ
　１００…処理装置
　１１０…記憶装置
　１２０…入力装置
　１３０…表示装置 10 ... Composition analysis device 11 ... Network 12 ... Imaging device 13, 14 ... Sorting machine 15 ... Server 100 ... Processing device 110 ... Storage device 120 ... Input device 130 ... Display device

Claims (8)

  1. 　複数の部品種を含む複数の電子・電気機器部品屑を含む原料を撮像した撮像画像の中から、機械学習システムを利用した画像認識処理を用いて、前記電子・電気機器部品屑を抽出し、組成解析を行うことを含み、
    　前記機械学習システムの学習に用いられる学習データに、組成解析対象とする原料の情報が反映されている場合は、前記機械学習システムの確信度を第１の閾値に設定し、組成解析対象とする原料の情報が前記学習データに反映されていない場合は、第１の閾値よりも低い第２の閾値に設定して前記電子・電気機器部品屑を抽出すること
    　を含むことを特徴とする電子・電気機器部品屑の組成解析方法。     The electronic / electrical equipment component scraps are extracted from captured images of raw materials containing a plurality of electronic / electrical equipment component scraps including a plurality of component types by using image recognition processing using a machine learning system. Including performing composition analysis
    When the learning data used for learning the machine learning system reflects the information of the raw material to be the composition analysis target, the certainty of the machine learning system is set to the first threshold value and the composition analysis target. When the information of the raw material is not reflected in the learning data, the electronic / electrical equipment component waste is extracted by setting the second threshold value lower than the first threshold value. A method for analyzing the composition of electrical equipment parts waste.
  2. 　前記第１の閾値を０．２～０．５とし、前記第２の閾値を０．０１～０．１とすることを含む請求項１に記載の電子・電気機器部品屑の組成解析方法。 The method for analyzing the composition of electronic / electrical equipment component scraps according to claim 1, wherein the first threshold value is 0.2 to 0.5 and the second threshold value is 0.01 to 0.1.
  3. 　前記電子・電気機器部品屑を抽出し、組成解析を行うことが、
    　設定された前記確信度に基づいて抽出された前記電子・電気機器部品屑に対し、前記電子・電気機器部品屑及び前記電子・電気機器部品屑の周囲の背景の画像を含む認識枠を付与し、
    　前記認識枠に対する前記電子・電気機器部品屑の面積率の情報を少なくとも有する部品種面積率データに基づいて、前記複数の部品種毎に、前記認識枠が付された前記電子・電気機器部品屑の合計面積を推測し、
    　前記合計面積の推測結果と前記複数の部品種毎の単位面積当たりの想定重量とを乗算し、前記複数の部品種毎の前記電子・電気機器部品屑の重量比率をそれぞれ解析することにより、前記撮像画像内の前記電子・電気機器部品屑の組成解析を行うこと
    　を含むことを特徴とする請求項１又は２に記載の電子・電気機器部品屑の組成解析方法。     Extracting the scraps of electronic and electrical equipment parts and performing composition analysis can be performed.
    A recognition frame including an image of the background around the electronic / electrical equipment component waste and the electronic / electrical equipment component waste is added to the electronic / electrical equipment component waste extracted based on the set certainty. ,
    Based on the part type area ratio data having at least information on the area ratio of the electronic / electrical equipment component waste with respect to the recognition frame, the electronic / electrical equipment component waste with the recognition frame attached to each of the plurality of component types. Estimate the total area of
    By multiplying the estimation result of the total area by the assumed weight per unit area for each of the plurality of component types and analyzing the weight ratio of the electronic / electrical equipment component scraps for each of the plurality of component types, the said The method for analyzing the composition of electronic / electrical equipment component scraps according to claim 1 or 2, wherein the composition analysis of the electronic / electrical equipment component scraps in the captured image is performed.
  4. 　前記学習データに反映されていない原料に含まれる前記電子・電気機器部品屑の輪郭、色彩、前記認識枠に対する前記面積率、及び前記背景の少なくともいずれかの情報を、前記機械学習システムに学習させることを含む請求項３に記載の電子・電気機器部品屑の組成解析方法。 The machine learning system is made to learn at least one of the contours, colors, the area ratio with respect to the recognition frame, and the background of the electronic / electrical equipment component scraps contained in the raw material that are not reflected in the learning data. The method for analyzing the composition of electronic / electrical equipment component scraps according to claim 3, which includes the above.
  5. 　前記複数の部品種が、基板及びプラスチックを少なくとも含む請求項１に記載の電子・電気機器部品屑の組成解析方法。 The method for analyzing the composition of electronic / electrical equipment component scraps according to claim 1, wherein the plurality of component types include at least a substrate and a plastic.
  6. 　請求項１～５のいずれか１項に記載の組成解析結果に基づいて、前記複数の部品種の中から特定の部品種を選別する選別工程を含むことを特徴とする電子・電気機器部品屑の処理方法。 Electronic / electrical equipment component scraps comprising a sorting step of selecting a specific component type from the plurality of component types based on the composition analysis result according to any one of claims 1 to 5. Processing method.
  7. 　複数の部品種を含む複数の電子・電気機器部品屑を含む原料を撮像した撮像画像の中から、機械学習システムを利用した画像認識処理を用いて、前記電子・電気機器部品屑を抽出し、組成解析を行う電子・電気機器部品屑の組成解析装置であって、
    　前記機械学習システムの学習に用いられる学習データに、組成解析対象とする原料の情報が反映されている場合は、前記機械学習システムの確信度を第１の閾値に設定し、組成解析対象とする原料の情報が前記学習データに反映されていない場合は、第１の閾値よりも低い第２の閾値に設定して前記電子・電気機器部品屑を抽出する処理装置を備える
    ことを特徴とする電子・電気機器部品屑の組成解析装置。     The electronic / electrical equipment component scraps are extracted from captured images of raw materials containing a plurality of electronic / electrical equipment component scraps including a plurality of component types by using image recognition processing using a machine learning system. It is a composition analysis device for electronic / electrical equipment parts waste that performs composition analysis.
    When the learning data used for learning the machine learning system reflects the information of the raw material to be the composition analysis target, the certainty of the machine learning system is set to the first threshold value and the composition analysis target. When the information of the raw material is not reflected in the learning data, the electron is provided with a processing device for extracting the electronic / electrical equipment component waste by setting the second threshold value lower than the first threshold value. -Composition analyzer for scraps of electrical equipment parts.
  8. 　複数の部品種を含む複数の電子・電気機器部品屑を撮像する撮像装置と、
    　撮像画像の中から、機械学習システムを利用した画像認識処理を用いて、前記電子・電気機器部品屑を抽出し、組成解析を行う電子・電気機器部品屑の組成解析装置であって、前記機械学習システムの学習に用いられる学習データに、組成解析対象とする原料の情報が反映されている場合は、前記機械学習システムの確信度を第１の閾値に設定し、組成解析対象とする原料の情報が前記学習データに反映されていない場合は、第１の閾値よりも低い第２の閾値に設定して前記電子・電気機器部品屑を抽出する処理装置を備える組成解析装置と、
    　前記組成解析装置によって解析された組成解析結果に基づいて前記電子・電気機器部品屑から特定の部品屑を選別する選別機と
    　を備える電子・電気機器部品屑の処理装置。     An imaging device that captures images of multiple electronic and electrical equipment component scraps, including multiple component types,
    A composition analysis device for electronic / electrical equipment component scraps that extracts the electronic / electrical equipment component scraps from captured images using image recognition processing using a machine learning system and analyzes the composition of the machine. When the learning data used for learning the learning system reflects the information of the raw material to be the composition analysis target, the certainty of the machine learning system is set to the first threshold value, and the raw material to be the composition analysis target is set. When the information is not reflected in the learning data, a composition analysis device including a processing device for extracting the electronic / electrical equipment component waste by setting a second threshold value lower than the first threshold value, and
    An electronic / electrical equipment component waste processing device including a sorter that sorts specific component waste from the electronic / electrical equipment component waste based on the composition analysis result analyzed by the composition analyzer.
PCT/JP2021/014225 2020-04-01 2021-04-01 Method for analyzing composition of electronic/electrical apparatus component layer, method for processing electronic/electrical apparatus component layer, device for analyzing composition of electronic/electrical apparatus component layer, and device for processing electronic/electrical apparatus component layer WO2021201251A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2020-066206 2020-04-01
JP2020066206A JP7301783B2 (en) 2020-04-01 2020-04-01 Composition analysis method for electronic/electrical equipment parts scrap, electronic/electrical equipment parts scrap processing method, electronic/electrical equipment parts scrap composition analysis device, and electronic/electrical equipment parts scrap processing equipment

Publications (1)

Publication Number Publication Date
WO2021201251A1 true WO2021201251A1 (en) 2021-10-07

Family

ID=77929260

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/014225 WO2021201251A1 (en) 2020-04-01 2021-04-01 Method for analyzing composition of electronic/electrical apparatus component layer, method for processing electronic/electrical apparatus component layer, device for analyzing composition of electronic/electrical apparatus component layer, and device for processing electronic/electrical apparatus component layer

Country Status (3)

Country Link
JP (1) JP7301783B2 (en)
TW (1) TW202203150A (en)
WO (1) WO2021201251A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022224478A1 (en) * 2021-04-21 2022-10-27 Ｊｘ金属株式会社 Electrical and electronic component scrap processing method, and electrical and electronic component scrap processing device

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7351043B1 (en) 2023-01-24 2023-09-26 三菱電機株式会社 Control method, program, and electrostatic separation device for electrostatic separation device

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007505733A (en) * 2003-09-20 2007-03-15 キネテイツク・リミテツド Apparatus and method for classifying targets in a waste stream
JP2014081943A (en) * 2013-11-14 2014-05-08 Omron Corp Data structure, library preparation device, electronic equipment analyzer, library providing system
JP2015232449A (en) * 2014-06-09 2015-12-24 アサヒプリテック株式会社 Device for evaluating the amount of valuables included in waste electronic circuit board in non-destructive way
WO2019026551A1 (en) * 2017-07-31 2019-02-07 荏原環境プラント株式会社 Waste composition estimation device, system, program, method, and data structure
WO2020090941A1 (en) * 2018-10-31 2020-05-07 Ｊｘ金属株式会社 Composition analyzer of electronic/electrical equipment parts waste, treatment device of electronic/electrical equipment parts waste, and treatment method of electronic/electrical equipment parts waste
JP2020197954A (en) * 2019-06-03 2020-12-10 Ｊｘ金属株式会社 Image extraction processing method of object, composition analysis method of electronic/electric instrument component scraps, composition analysis apparatus of electronic/electric instrument component scraps, and processing method of electronic/electric instrument component scraps
JP2020197953A (en) * 2019-06-03 2020-12-10 Ｊｘ金属株式会社 Image extraction processing method of object, composition analysis method of electronic/electric instrument component scraps, composition analysis apparatus of electronic/electric instrument component scraps, and processing method of electronic/electric instrument component scraps

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108932510A (en) * 2018-08-20 2018-12-04 贵州宜行智通科技有限公司 A kind of rubbish detection method and device
CN110717426A (en) * 2019-09-27 2020-01-21 卓尔智联(武汉)研究院有限公司 Garbage classification method based on domain adaptive learning, electronic equipment and storage medium

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007505733A (en) * 2003-09-20 2007-03-15 キネテイツク・リミテツド Apparatus and method for classifying targets in a waste stream
JP2014081943A (en) * 2013-11-14 2014-05-08 Omron Corp Data structure, library preparation device, electronic equipment analyzer, library providing system
JP2015232449A (en) * 2014-06-09 2015-12-24 アサヒプリテック株式会社 Device for evaluating the amount of valuables included in waste electronic circuit board in non-destructive way
WO2019026551A1 (en) * 2017-07-31 2019-02-07 荏原環境プラント株式会社 Waste composition estimation device, system, program, method, and data structure
WO2020090941A1 (en) * 2018-10-31 2020-05-07 Ｊｘ金属株式会社 Composition analyzer of electronic/electrical equipment parts waste, treatment device of electronic/electrical equipment parts waste, and treatment method of electronic/electrical equipment parts waste
JP2020197954A (en) * 2019-06-03 2020-12-10 Ｊｘ金属株式会社 Image extraction processing method of object, composition analysis method of electronic/electric instrument component scraps, composition analysis apparatus of electronic/electric instrument component scraps, and processing method of electronic/electric instrument component scraps
JP2020197953A (en) * 2019-06-03 2020-12-10 Ｊｘ金属株式会社 Image extraction processing method of object, composition analysis method of electronic/electric instrument component scraps, composition analysis apparatus of electronic/electric instrument component scraps, and processing method of electronic/electric instrument component scraps

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022224478A1 (en) * 2021-04-21 2022-10-27 Ｊｘ金属株式会社 Electrical and electronic component scrap processing method, and electrical and electronic component scrap processing device
JP2022166727A (en) * 2021-04-21 2022-11-02 Ｊｘ金属株式会社 Electrical and electronic component scrap processing method, and electrical and electronic component scrap processing device
JP7264936B2 (en) 2021-04-21 2023-04-25 Ｊｘ金属株式会社 Electric/electronic component waste processing method and electric/electronic component waste processing apparatus

Also Published As

Publication number Publication date
JP7301783B2 (en) 2023-07-03
JP2021159881A (en) 2021-10-11
TW202203150A (en) 2022-01-16

Similar Documents

Publication Publication Date Title
JP7092889B2 (en) Composition analyzer for electronic / electrical equipment parts waste, electronic / electrical equipment parts waste processing equipment, and electronic / electrical equipment parts waste processing method
JP7328012B2 (en) Electronic/electrical device parts scrap composition analysis method, electronic/electrical device parts scrap composition analysis device, and electronic/electrical equipment parts scrap processing method
WO2021201251A1 (en) Method for analyzing composition of electronic/electrical apparatus component layer, method for processing electronic/electrical apparatus component layer, device for analyzing composition of electronic/electrical apparatus component layer, and device for processing electronic/electrical apparatus component layer
JP7328011B2 (en) Electronic/electrical device parts scrap composition analysis method, electronic/electrical device parts scrap composition analysis device, and electronic/electrical equipment parts scrap processing method
JP2021522070A5 (en)
WO2021201250A1 (en) Composition analysis method for electronic/electrical equipment component waste, processing method for electronic/electrical equipment component waste, composition analyzer for electronic/electrical equipment component waste, and processing device for electronic/electrical equipment component waste
TWI779947B (en) Disposal method of electrical and electronic parts scraps and disposal device of electrical and electronic parts scraps
CN113369155B (en) Renewable waste product identification detection and automatic recovery system and method
JP3681316B2 (en) Image processing apparatus, waste processing apparatus using the same, image processing method, and medium on which image processing program is recorded
Paulraj et al. Classification of recyclables from E-waste stream using thermal imaging-based technique
US12026867B2 (en) Apparatus for analyzing composition of electronic and electrical device part scraps, device for processing electronic and electrical device part scraps, and method for processing electronic and electrical device part scraps
WO2022102176A1 (en) Sorting method for electronic component scraps and processing method for electronic component scraps
CN112215149A (en) Accessory sorting system and method based on visual detection
JP2022078833A (en) Processing method of electronic component waste, image analysis system of electronic component waste, and sorting system of electronic component waste
Kofler et al. Detecting Star Cracks in Topography Images of Specular Back Surfaces of Structured Wafers
JP2023057552A (en) Apparatus and method for assigning material value score to printed circuit board waste material or portion thereof, and system for sorting printed circuit board waste material
Gundupalli et al. Thermal imaging-based classification of the E-waste stream
CN115063735A (en) Worker card identification method and device and electronic equipment
CN113963220A (en) Security check image classification model training method, security check image classification method and device
CN113791090A (en) Rapid verification system and method for welding defects of recovered circuit board
JP2019171343A (en) Processing method of electronic-electrical equipment component scrap

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21780131

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21780131

Country of ref document: EP

Kind code of ref document: A1