CN110263952A - A kind of electronic product intelligence recovery system - Google Patents

Abstract

The present invention relates to field of artificial intelligence, disclose a kind of electronic product intelligence recovery system, the system comprises: database module, computing module, system runs module and terminal installation module, the database module includes external resource mould group, system operation mould group, recycle master die group, data acquisition module group, metadata mould group and data processing mould group, the computing module includes recycling grade mould group and system quotation mould group and deep learning mould group, the system operation module includes system operation mould group, application operating mould group and the safe mould group of system, the terminal installation module includes human-computer interaction module, identify acquisition module, article circulation keeping module, the present invention realizes accurately identifying for electronic product, self-service recycling and arm's length pricing, the intelligent whole recovery operation that electronic product is completed from control, improve recycling Service quality and efficiency of operation.

Description

A kind of electronic product intelligence recovery system

Technical field

The present invention relates to field of artificial intelligence, and in particular to a kind of electronic product intelligence recovery system.

Background technique

With the development of science and technology, the update of electronic product is more and more frequent, substitute the electronic product user choosing abandoned It is idle without selecting to recycle to select sleeping cabinet, forms the electronic product solid waste pollution " barrier lake " of staggering amount, leaves the ecology that can not be estimated Environmental hazard hidden danger.Handle the electronic products such as the idle mobile phone abandoned and tablet computer by way of recycling, for protection environment, Ensure health, maintaining ecological balance and using regenerated resources, develop a circular economy it is particularly important.

The electronic products such as current phone and tablet computer recycling industry situation mainly has online and offline two ways.It is recycled on line It tests that machine operation is complicated, appraisal process is cumbersome, needs for object to be addressed to recycling quotient designated place and manually test after machine and just pay the bill, not only Because unit take-back cost is opaque, cannot complete to complete a business transaction in time, it is easy to breed make profit between two parties, bad fraud phenomenon, Er Qiezhan With fund and time, it is wasted in logistics.It recycles shops address under line to fix, user is inconvenient, recycles still even if trade company visits User is wanted to undertake high cost；Rubbish work as in stop leakage in the roof change face disk, salvage station recycling of trip quotient's recycling, user's value loss.Regional markets Quotient is recycled under line, being limited to its information, technology, resource bottleneck leads to high efficiency at low cost, and formation industry recycling valence low amounts is small, with The status that actual demand seriously disconnects.Conventional industries transition and upgrade, supply side reform and internet+, intelligence+overall situation Under, electronic product recovery technology urgently information hierarchy.Storage huge electronic product in China's is huge environmental pollution hidden danger, must It must be decomposed and be eliminated by effective way；It is huge production factors resource, it is necessary to explore effective mechanism and vitalize utilization；It is huge Industry development chance, it is necessary to innovate effective model value driving.

At present the electronic products such as mobile phone and tablet computer recycling industry technology be concentrated mainly on recycling website, related APP and In terms of terminal installation, the defect grade of Product recycling determines, recycling price, process services are both needed to the either artificial behaviour of manual work Make mechanization, speculative and randomness is extremely serious, and recovery value is reasonable, way of recycling is convenient, ensures and uses there has been no being able to achieve The systemic intellectualized technology scheme of family equity and personal secrets.

Summary of the invention

In view of the deficiencies of the prior art, the present invention provides a kind of electronic product intelligence recovery system, with solution background technique The problems in.

The present invention solves technical problem and adopts the following technical scheme that

The present invention provides a kind of electronic product intelligence recovery systems, including database module, computing module, run module With terminal installation module；

The database module includes recycling category database, data acquisition module group and data processing mould group, for acquiring Processing system recycles the data of required by electronic product, and carries out classification storage；

The computing module includes recycling objective single mode group, identification calculating mould group, recycling price mould group and deep learning mould group, It is objective single for being recycled to electronic product generation, it is that guiding carries out the determination of recycling grade, products quotation finally enriches with the objective list of recycling Objective simple correlation content is recycled, the internal pricing algorithm of defining the level of self feed back optimization is completed at the same time；

The operation module includes system operation mould group, application operating mould group and the safe mould group of system, for providing support The data transmission of each intermodule is called in the control of entire recovery process, establishes safety verification and controlling terminal equipment operation；

The terminal installation module includes human-computer interaction mould group, identification acquisition mould group, industry control mould group, for being mentioned according to instruction The data of electronic product to be recycled are acquired and recycled automatically with operation as a result, realizing for user's control entrance, feedback processing.

Compared with prior art, the present invention have it is following the utility model has the advantages that

The present invention passes through the intelligence of artificial intelligence technology, standardization, automation, intensive and ease for operation, transreplication Property, low-cost advantage, maximally utilize and play the resource value after mobile phone and tablet computer are abandoned and the ecological value, realize recycling Value rationally, way of recycling it is convenient, user's right guarantee, privacy of user safety function, specification recycling grade separation standard with Industry operating regulation and pricing system, and pass through the acquisition analysis system stream of people, logistics, Jin Liu, information flow data, support system Deep learning advanced optimizes and improves recycling service quality and efficiency of operation, improves recycling amount, and it is hidden to decompose electronic waste Suffer from, vitalize production factors resource, drives ecological circulation economy.

The present invention provides two kinds of electronic product recognition methods, the various shapes of electronic product in practical application are covered substantially State identifies type, brand, the name of an article, the model, rule of electronic product to be recycled using artificial intelligence technology by category recognition methods State is guaranteed to keep in good repair in lattice, configuration, color, account number cipher state, string number, electronics three guarantees, is completed by defect recognition methods identification wait return Receive the appearance quality classification and deformation, screen display, component, function, system deficiency classification of electronic product, comprehensively solve Current electronic The category and Defect types and its degree of electronic product to be recycled cannot be quickly identified in Product recycling industry and recycling operation Problem；

The present invention is based on category identification and defect identification define a kind of recycling pricing method, from kind, brand, technology, The various dimensions such as technique, function, hardware, present deal data calculate unit take-back cost, can not only maximize guarantee user's right, and It can maximally utilize and play the resource value after electronic product is abandoned and the ecological value, solving current recycling industry cannot be accurate Determine its recovery value and recycle mode and flow direction, calculate its rationally just and sound unit take-back cost the problem of, solve current Recycling market, arbitrarily quotation, price deception and wheel and deal due tos, form supply and demand not caused by technology and information asymmetry reason Freely, wasting of resources loss, the problem that electronic product solid waste pollutes and ecological environment Hidden hazrads are huge；

The present invention uses customized gridding method and filtration method, breaches tradition based on pixel coordinate, the processing of image two-value, figure As the technical restriction of segmentation and mathematics morphologic detection, it is poor to overcome present image identification technology adaptive performance, once target Image has the problem of larger incompleteness does not often just reach desired result by stronger noise pollution or target image, enhances steady Qualitative and universality, considerably reduces development cost and use cost；

This invention simplifies electronic product recovery process and job steps, reduce intermediate link, considerably reduce money Golden cost of possession, testing cost, removal process affairs cost of labor and logistics cost can help all kinds of main market players to realize benefit Maximum, efficiency highest, least risk, win-win progress；

Specification of the present invention electronic product recycling category and its category classification method and standard have standardized electronic product defect Classification and its classification method and standard have standardized electronic product recycling pricing mechanism, have reduced recycling link manual intervention factor, Electronic product recycling Establishing Market supply and demand can be helped to trust, promote justice of exchange, push supply and demand two smooth；

Compared with the existing technology about the present invention, other substantive distinguishing features outstanding and significant progress are in embodiment part It is described in further detail.

Detailed description of the invention

Upon reading the detailed description of non-limiting embodiments with reference to the following drawings, other feature of the invention, Objects and advantages will become more apparent upon:

Fig. 1 is a kind of structural schematic diagram of electronic product intelligence recovery system of the invention；

Fig. 2 is the single schematic diagram of recycling visitor of the present embodiment；

Fig. 3 is that category database structure schematic diagram is recycled in the present embodiment；

Fig. 4 is category recognition methods flow diagram in the present embodiment；

Fig. 5 is the defect recognition methods schematic diagram of the present embodiment；

Fig. 6,7,8 are respectively the schematic diagram for the extraction characteristic information that the netting analysis of the present embodiment takes the mobile phone as an example；

Fig. 9 is that state user interface schematic diagram is guaranteed to keep in good repair in the electronic product three guarantees of the present embodiment.

Specific embodiment

Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other Embodiment shall fall within the protection scope of the present invention.

It should be noted that certain titles have been used in specification and claims to censure specific components.It answers Work as understanding, those of ordinary skill in the art may censure the same component with different names.Present specification and right In a manner of claim is not using the difference of title as differentiation component, but using the substantial differences of component functionally as area The criterion of subassembly.As used in present specification and claims "comprising" or " comprising " be an open use Language should be interpreted that " including but not limited to " or " including but not limited to ".Specific embodiment part is described to be implemented Example is presently preferred embodiments of the present invention, the range being not intended to limit the invention.

In addition, those skilled in the art will appreciate that, various aspects of the invention can be implemented as system, method or Computer program product.Therefore, various aspects of the invention can be implemented as the form of software and hardware combining, can unite here Referred to as circuit, " module " or " system ".In addition, in some embodiments, various aspects of the invention are also implemented as The form of computer program product in one or more microcontroller readable mediums, comprising micro- in the microcontroller readable medium The readable program code of controller.

Fig. 1-9 is please referred to, the present embodiment provides a kind of electronic product intelligence recovery systems, including database module, calculating Module runs module and terminal installation module；

Wherein database module includes recycling category database, data acquisition module group and data processing mould group, for acquiring Processing system recycles the data of required by electronic product, and carries out classification storage；

Recycling category database is divided according to system operation process, including preset data typing library, category identification member Database, electronic product identification database and the objective one process database of recycling；

Preset data typing library includes external data acquisition database, operation data input database, manual entry data Library, sensitive information monitor database；

Objective one process database is recycled to include interim objective single database, recycle objective single database and recycling terminal database；

Wherein electronic product identification database, data needed for recycling according to the vertical subordinate relation of electronic product and identification are indulged To functional relationship as planning architecture configuration subdata base；

The vertical subordinate relation of electronic product specifically includes the successively electronics categories of subordinate, brand, the name of an article, model, rule Lattice, configuration and color；

Longitudinal functional relationship of data needed for identification is recycled specifically includes the mutually longitudinal associated row of this kind of electronic product Industry data, data incomplete, utilize data and pricing data at category data；

Electronics categories include mobile phone, tablet computer, charger baby, wearable equipment, laptop；

Referring to Fig. 3 in the present embodiment with mobile phone as an example, it is described in detail；

Subdata base is configured according to the vertical subordinate relation of electronic product, specially successively according to the subordinate relation of electronic product Setting recycling category coding and catalogue, framework vertical relation subdata base, classification processing, recording and storage various electronic products Recycle category coding and its category feature, defect feature and basic information data, for system identification and recycling the product provide according to According to vertical relation subdata base specifically includes:

Level-one subdata base, the level-one subdata base include electronics categories database, the electronics categories Database includes mobile phone database, tablet computer database, charger baby database, wearable device database and laptop number According to library；

It is subordinated to the second level subdata base of electronics categories database, second level subdata base includes brand database, such as Shown in Fig. 3, millet mobile phone database, Huawei's mobile phone database can be divided into, iPhone database, Samsung mobile phone database, Oppo mobile phone database etc.；

It is subordinated to the three-level subdata base of brand database, the three-level subdata base includes Brand Operation database, enterprise Industry information database, technical information database, supply chain database and product database；We are herein with Huawei's data in mobile phone Continue to configure for library；Subordinate setting different model product unit under product database, and subordinate configuration ales information table, Product information table, standard picture library, category property data base, defect property data base, wherein the subitem of ales information table include Issue date, listing date, producer price, Sales Channel, sales policy, market price, sales statistics, old for new service；Product Information table subitem includes brand classification, product name, product type, product specification, hardware configuration list and software configuration inventory； Standard picture library subitem includes direct picture, back side image, side image, close up fragmentary, interface image, interface Touch control key feature With path；Category property data base subitem include individualized feature, brand identity, name of an article feature, model features, specification feature, Configuration feature, color characteristic；Defect property data base subitem include appearance quality class, shape-change class, display screen class, component lack Damage class, functional impairment class, system but Sun Lei, vulnerable part invoice price table；

Longitudinal functional relationship of data needed for recycling in the present embodiment according to identification is used as planning architecture configuration subdata base, The functional relationship of data needed for specially realizing identification according to system and recycle, framework recycle electronic product under category coding item and close Join longitudinal functional relationship subdata base of information, electronic product under the various recycling category coding items of classification processing, recording and storage Related information data, be system identification and analysis determine the product recovery value, Land use systems and flow direction, recycling transaction and Sale matching provides foundation；Longitudinal functional relationship subdata base is subordinated to electronics categories database, specifically includes:

It is subordinated to the second level subdata base of electronics categories database, second level subdata includes category database, defect Database, utilizes database and pricing data library at sector database；

Category database includes personalized metadata library, brand recognition metadatabase, name of an article identification metadatabase, model knowledge Other metadatabase, specification identification metadatabase and configuration identification metadatabase；

Data incomplete library includes appearance quality metadatabase, deformation defect metadatabase, screen display defect metadatabase, component Defect metadatabase, functional impairment metadatabase and system deficiency metadatabase；

Sector database includes industry Zone Information database, market information database, technical resource database, professional standard number According to library and industrial chain database；

It include orientation supply database, recycling industry database, recycling market database, aftermarket attachment number using database According to library, regenerated resources database；

Pricing data library includes recycling price metadatabase；

Data acquisition module group collects and records related data according to systemic presupposition and is stored in preset data typing library, including external data Acquisition unit, operation data typing unit, manual entry unit and sensitive information monitoring unit, collect and record correlation according to systemic presupposition Data are stored in corresponding external data acquisition database in preset data typing library, operation data input database, artificial respectively Input database and sensitive information monitor database；

External data acquisition unit presses the functional condition of the external resource mould group, and automatically retrieval collects and records target Internet resources Data, and data are stored in external data acquisition database respectively；

The operation data typing unit presses the system operation mould group functional condition, automatic acquisition and recording system whole process operation Data, and data are stored in operation data input database respectively；

The manual entry unit is default by system function, corresponds to demand by manual type typing or supplement input system Data, and data are stored in manual entry database respectively；

The sensitive information monitoring unit is default by system function, and automatic monitoring or retrieval acquisition target Internet resources are related The instant sensitive information such as resource supply and demand variation, the market price and its fluctuation, industry major event, product discontinuance, version updating, it is real When analysis processing and more new database, and data are stored in sensitive information monitor database respectively；

Data processing mould group is default according to system function, according to the system-computed module and the application operating module and Terminal installation calculates operation demand, is analyzed and processed to the data acquisition module group ephemeral data, generates system metadata branch System operations are held, to simplify system operations and application operating module and terminal installation calling calculation amount, reduce server space Occupancy and data transmission period and bandwidth resources occupy, and improve terminal installation mist computing capability and system fast reaction, stablize fortune Row ability；

The data processing mould group includes type data processing mould group, category data processing mould group, data incomplete processing mould Group, industry data processing mould group handle mould group using data processing mould group, pricing data；

Type data processing mould group is default according to system function, and classification storage recycles the electronic product under category coding item Electronics categories feature and recognition element are extracted in type classification method, standard and rule, analysis, are stored in corresponding category identification member Database, automated tos respond to terminal installation recycling operation and system identification calculates mould group and identifies electronics categories coding to be recycled, It supports system operations and terminal installation mist calculates and application operating；

Category data processing mould group is default according to system function, and classification storage recycles the electronic product under category coding item The category feature and recognition element of electronic product are extracted in category classification method, standard and rule, analysis, are stored in the corresponding category and are compiled Personalized metadatabase, brand recognition metadatabase, the name of an article identify metadatabase, type identifier metadatabase, specification under code item It identifies metadatabase, configuration identification metadatabase, automated tos respond to terminal installation recycling operation and system identification calculates the identification of mould group Electronic product category coding to be recycled, supports system operations and terminal installation mist calculates and application operating；

Data incomplete handles mould group and presets according to system function, and classification storage recycles the electronic product under category coding item The defect feature and recognition element of electronic product are extracted in defect classification method, standard and rule, analysis, are stored in the corresponding category and are compiled Code item under appearance quality metadatabase, deformation defect metadatabase, screen display defect metadatabase, component defect metadatabase, Functional impairment metadatabase, system deficiency metadatabase, automated to respond to terminal installation recycling operation and system identification calculates mould group The defect classification coding for identifying electronic product to be recycled, supports system operations and terminal installation mist calculates and application operating；

Industry data handles mould group and presets according to system function, and classification storage recycles the electronic product under category coding item Related information classification method and rule, analysis extract the related information feature of electronic product and assist in identifying element, are stored in the product Class encodes corresponding category identification metadatabase module and defect under item and identifies metadatabase, and according to systemic presupposition analysis at The related information data that reason system is collected and recorded calculate mould group for system identification, handle mould group using data processing mould group, pricing data The recycling category coding and defect classification for identifying electronic product to be recycled encode, determine recycling mode and flow direction, calculate back It receives price and foundation is provided；

Default according to system function using data processing mould group, classification storage recycles the electronic product under category coding item Using data classification method and rule, analysis extracts the utility value feature of electronic product and assists in identifying element, is stored in the product Class encodes corresponding recycling metadatabase under item, and recycles database according to systemic presupposition analysis process system Data are recycled, determine that recycling mode and flow direction, accounting unit take-back cost provide foundation for system；

Pricing data handles mould group and presets according to system function, and classification storage recycles the electronic product under category coding item Recovery value classification method and rule, analysis extract the pricing information feature of electronic product and assist in identifying element, are stored in the product Class encodes corresponding recycling price metadatabase under item, and believes according to the price in systemic presupposition analysis process system database Data are ceased, unit take-back cost is calculated for system and foundation is provided；

Computing module includes recycling objective single mode group, identification calculating mould group, recycling price mould group and deep learning mould group, is used for Visitor's list is recycled to electronic product generation, the finally substantial recycling of the determination of recycling grade, products quotation is carried out to be oriented to the objective list of recycling Objective simple correlation content is completed at the same time the internal pricing algorithm of defining the level of self feed back optimization；

The single processing mould group of recycling visitor is default according to system function, and management terminal equipment and record storage recycling visitor are single, automatic Terminal installation recycling operation is responded, generates and recycles objective single, the interim visitor's single database of deposit, the objective single database of recycling and recycling respectively Terminal database, and submit to system to complete recycling operation according to the single preset recording content of recycling visitor and sequence；

It is illustrated in figure 2 the single signal of recycling visitor that an identification content of the present embodiment by taking Huawei's mobile phone as an example completes version Figure；

Identification calculates mould group according to system command, and the recycling visitor submitted according to the single processing mould group of recycling visitor is single, and retrieval collects The database purchase associated data passes through the recognition methods of systemic presupposition category and defect classification recognition methods, and identification meter Calculate the category coding and defect classification coding of the procedure identification electronic product to be recycled；

Recycling price mould group according to system command, sentence by the recycling visitor list and system submitted according to the single processing mould group of recycling visitor Surely recycling mode and flow direction, recycling grade determine to collect the database purchase associated data as a result, retrieving, and pass through system Default recycling pricing method and program calculate the unit take-back cost for determining the product；

Deep learning mould group according to system command, to the same category that system completes recycling recycle the image included in objective list and Major parameter is compared, and retrieval collects relevant information, extracts common feature optimization system metadata and work flow, is Realization system precisely identifies, refinement levels, rational quotation, and rapid computations, high-efficient homework, stable operation provide foundation.

Operation module includes system operation mould group, application operating mould group and the safe mould group of system, is supported for offer entire The data transmission of each intermodule is called in the control of recovery process, establishes safety verification and controlling terminal equipment operation；

System runs mould group and runs process according to recycling work flow setting system, sequentially generates various command instruction systems Modules and its functional unit and mist computing terminal equipment be sequentially completed data acquisition storage and analysis processing, identification calculate and Deep learning, information feedback, operation operation, and default, the modification management by system background administration interface progress system function, with And system manager role and permission are preset, modify management and system data typing and parameter setting, modification management；

Application operating mould group system operation process according to as defined in system operation mould group is default, and setting terminal is equipped back It is incorporated as industry to flow and functionality, sequentially generates various task instructions, system command key or document frame, message box, filled by terminal Standby interactive interface is for user's typing, supplement, modification, cancellation, confirmation operation；

The safe mould group of system sets security management processes, plan and security incident prediction scheme according to system demand for security and is System security procedure, establishes firewall, by means such as certification, encryption, monitoring, trackings, provides file encryption to system, identity is recognized Card, secure accessing, using the comprehensive security protection such as protection, access control, by digital certificate to service server, data The independent security deployment of the carry out such as library server, archive server and log server, unified management；

Terminal installation module includes human-computer interaction mould group, identification acquisition mould group, industry control mould group, for providing use according to instruction Family control entrance, feedback processing acquire the data of electronic product to be recycled and recycle operation automatically as a result, realizing:

Human-computer interaction mould group system operation process according to as defined in system operation mould group, the application operating mould group is pre- If setting terminal equips interactive interface and function and function is realized；

Identification acquisition mould group is provided according to the data acquisition module group, system operation mould group, the application operating mould group System operation process it is default, setting terminal equips the image for successively acquiring electronic product to be recycled, reads data；

Industry control mould group, group is according to as defined in the data acquisition module group, system operation mould group, the application operating mould group System runs process and presets, and setting terminal equipment completes successively recycling operation automatically.

Referring to figure 4., a kind of 6,7,8 these electronic products of category recognition methods provided in this embodiment can be mobile phone, Other any electronic equipments for having recovery value such as computer, a kind of electronic product category recognition methods, comprising the following steps:

Setting recycling product classification standard and rule, according to type, brand, the name of an article, model, specification, configuration, color it is vertical Subordinate relation formulates electronic product recovery article class catalogue and category identification coding, presets recycling category database；

Typing by custom built forms crawler software and desk checking typing or manually is acquired, acquires and update the product in real time Category identification encodes the related information of electronic product under item in class database；The related information includes:

Basic information；

Auxiliary information；

Circulation information；

Standard picture；

Wherein basic information includes: that type, brand, the name of an article, model, specification, configuration, color, the system of electronic product are soft Part information and version；

Wherein auxiliary information includes: the brand operator of electronic product, core group parts suppliers, manufacturer, retailer Information, Electronic Product Core skill are disclosed with main Sales Channel and value chain enterprise industrial and commercial registration information and its material items Art, main process technology of production and its processing procedure information；

Wherein circulation information include: issue date of electronic product, producer price, the list marketing date, Sales Channel and Selling price, sales volume, software change version, hardware bill of materials and halt production stop pin date information；

Wherein standard picture include: electronic product direct picture and back side image, side elevational view picture, keypoint part it is micro- Image, one or more designated user's interface images；

It extracts and corresponds to electronic product recognition element and differentiation feature under the category catalogue and coding item, form category and know Other metadata presets category and identifies metadatabase；

Recognition element and differentiation feature in the present embodiment extract electronics by using netting analysis and filtration method and produce Product standard picture obtains, and specifically includes:

Extract the recognition element stored in the standard picture between the electronic product and other variety classes electronic products With differentiation feature；

Extract the knowledge stored in the standard picture between the electronic product and its same type but different brands electronic product Other element and differentiation feature；

Extract store recognition element in the standard picture between the electronic product and its same brand but the different names of an article and Differentiation feature；

Extract the identification member stored in the standard picture between the electronic product and its same brand, the name of an article but different model Element and differentiation feature；

It extracts and stores in the standard picture between the electronic product and its same brand, the name of an article, model but different size Recognition element and differentiation feature；

It extracts and stores the electronic product and its same brand, the name of an article, model and specification in the standard picture, but different configurations Recognition element and differentiation feature between product；

The recognition element and differentiation feature include:

The surface shape and color of electronic product；

Electronic product brand pattern and character；

The layer of structure and color of electronic product enclosure and frame and component；

Functional unit and interactive interface UI design of the electronic product including camera, flash lamp, loudspeaker, key Feature, list style, icon and text position and body coordinate, geometrical relationship；

Netting analysis method in the present embodiment specifically includes:

The transparent four-quadrant coordinate grid of preset calibrations coordinate origin and x-axis, y-axis and its graduated scale is standard detection seat Mark；

Standard image metadata library under category identification coding item is preset, record storage extracts institute using standard detection coordinate State all kinds of standard image datas of the electronic product under category identification coding item；

The extraction standard image data includes:

Standard detection coordinate is overlayed to the surface of electronic product standard picture under the category identification coding item, coordinate is former Point is overlapped with the central point for selecting feature in the standard picture, and x-axis is parallel with top margin, y-axis is vertical, and passes through product material standard Size demarcates the size of coordinate graduated scale in proportion；

Fuselage specification, the display screen specification, casing in the standard picture are extracted by standard detection coordinate With the different function such as the geometry and size, brand picture and text shape and size and camera, flash lamp, loudspeaker of frame The geometry and size and keypoint part microscopic features of component and different key, and demarcate its spatial position coordinate and phase Geometrical relationship between mutually, and the data inputting category is identified into standard image metadata library under coding item；

The designated user interface UI design feature, list style, icon and text are extracted by standard detection coordinate Body characteristics demarcate its spatial position coordinate and its mutual geometrical relationship, and the data inputting category are identified and is encoded Standard image metadata library under；

The multiple designated user interfaces of electronic product under the category identification coding item are extracted by standard detection coordinate to enter The identification feature and coordinate position of key, and access path is recorded according to different user interface set membership, and by data inputting Standard image metadata library under category identification coding item；

The standard surface color of electronic product, mark under the category identification coding item are acquired using filtration method in the present embodiment Quasi- glossiness and standard roughness, and the data inputting category is identified into standard image metadata library under coding item；

In the present embodiment also analysis processing electronic product include type, brand, the name of an article, model, specification, configuration knowledge The characteristic feature of other element and differentiation feature forms the category catalogue and encodes the individualized feature metadata under item, and deposits Enter corresponding category identification metadatabase.

The individualized feature metadata and corresponding recognition element and differentiation feature are also analyzed in the present embodiment, respectively Make category identification template, brand recognition template, name of an article recognition template, the type identifier template, specification of the category electronic product Recognition template and configuration recognition template, and it is stored in corresponding category identification metadatabase.

Also analysis handles the preceding individualized feature metadata and corresponding recognition element in the present embodiment and differentiation is special Sign, corresponding standard picture make the front recognition template, back side recognition template and specified interaction circle of the category electronic product respectively Face recognition template, and it is stored in corresponding category identification metadatabase.

Acquire electronic product characteristic information to be identified, compare category database and category identifies metadatabase, determine described in The category of electronic product to be identified and corresponding coding, specifically include:

Electronic product characteristic information to be identified is acquired, category database is compared and category identifies metadatabase, specifically include:

Acquire direct picture, back side image or the specified interactive interface image of electronic product to be identified；

Respectively by the direct picture of the electronic product to be identified, back side image or specified interactive interface image and corresponding identification Template is compared and is mutually authenticated, and identifies that its category encodes, obtains electronic product all information to be identified.

It wherein compares and is mutually authenticated and specifically include:

If the direct picture of electronic product to be identified, back side image and specified interactive interface image pass sequentially through its personalization Recognition template can accurately identify its category coding, then be automatically performed identification, store information；

Otherwise corresponding identification successively is called according to the relation on attributes of the product category, brand, the name of an article, model, specification, configuration Template is identified: if its recognition template can accurately identify its category coding, being automatically performed identification, is stored information, otherwise mention It takes the identical feature of each step to be mutually authenticated, identifies that its category encodes, store information；

In comparing and being mutually authenticated, through whole contrast contings:

Identification is unable to template and corresponds to configuration, specification, model, successively determines that it is affiliated specification, model, name of an article correspondence Category encodes the minimalist configuration type under item；

Identification is unable to template and corresponds to the name of an article, is determined as that the brand corresponds to brand gradeless and uniformly-priced goods machine under category coding item；

Identification is unable to template and corresponds to brand, and identification decision is gradeless and uniformly-priced goods machine of inferior brand；

Identification is unable to template and corresponds to type, is determined as solid waste machine；

Wherein the direct picture of standard picture and electronic product to be identified, back side image and specified interactive interface image are adopted Interference is reduced with the method for including plus setting coordinate grid and optical filtering color filter plate and image noise reduction or adjusting gray value, improves category Discrimination.

The image that identification is completed under all category catalogues and coding item is compared in the present embodiment, according to identification The recognition template and identified image called in process calculates separately recognition success rate, extracts crucial common feature, Statistical error Template.

Referring to figure 5., 6,7,8, the present embodiment provides a kind of defect recognition methods, by extracting electronic product to be recycled The off-note for encoding the standard picture of electronic product under item in image information with the described category, identifies its various defect class Not, and accordingly determine its recovery value and its recycle mode and flow direction；

The defect recognition methods implementation steps are as follows:

Category catalogue and coding are formulated, defect category database is preset；It can be taken in the present embodiment and category identification side Category catalogue and category identification coding is arranged in the identical mode of method, so that electronic product under the category identification coding item be arranged Defect identification database and defect identify metadatabase；

Defect classification includes appearance quality classification, deformation defect classification, screen under category identification coding item under the present embodiment Aobvious defect classification, component defect classification, functional impairment classification and system deficiency classification；

Formulate the appearance quality classification, deformation defect classification, screen display defect classification, component defect classification, functional impairment The classification standard of classification and system deficiency classification and rule and its defect coding are correspondingly arranged electricity under the category identification coding item The defect categorization metadata library of sub- product；

Appearance quality classification in the present embodiment is divided into 6 grades according to newness degree, wherein the 6th grade is shape breakage classification, is Refer to the electronic product because machine angle caused by the external force such as falling, colliding is damaged, frame is damaged, display screen is damaged, bottom cover top is damaged, bottom The disfigurements such as lid lower part breakage；Its appearance quality classification subclass coding is preset, and outer according to its grade scale and rule mark It sees quality and identifies metadata typing defect categorization metadata library；

Wherein deformation defect classification is divided into beautiful machine, little Hua, great Hua, cracking, outer quick-fried, implosion, overhauls the subclass such as machine, presets Its defect classification subclass coding, and according to its classification standard and rule mark deformation defect identification metadata typing defect classification member Database；

Wherein screen display defect classification is divided into the subclass such as screen film, platen bracket, external screen, touch control layer, liquid crystal layer, backlight, in advance If its screen display defect subclass encodes, and identifies metadata typing defect classification according to its classification standard and rule mark screen display defect Metadatabase；

Wherein component defect classification is divided into the subclass such as charging assembly, camera, mainboard, flash lamp, key, presets its group Part defect subclass coding, and metadata typing defect categorization metadata is identified according to its classification standard and regular annotation component defect Library；

Wherein functional impairment classification is divided into network, WIFI, call, bluetooth, NFC, waves, the subclass such as other function, in advance If its functional impairment subclass encodes, and identifies metadata typing defect classification according to its classification standard and regular marking Function defect Metadatabase；

Wherein system deficiency classification is divided into that version is too low, system Caton, be restarted automatically, the damage of automatic shutdown, function, system The subclass such as crash, cannot be switched on, and presets its system deficiency coding, and according to its classification standard and regular labeling system defect Identify metadata typing defect categorization metadata library；

Analyze that the defect feature and its defect of electronic product under same category identification coding item formed in the present embodiment because Fruit relationship, extracts the logical order of common feature and defect formation, and mark appearance quality, deformation defect, screen display defect, component lack Damage, functional impairment and system deficiency recognition template, typing defect categorization metadata library；

Recognition methods is selected according to electronic product situation to be identified, acquires electronic product characteristic information to be identified, compares and lacks It damages category database and defect identifies metadatabase, determine the category of the electronic product to be identified and correspond to coding；In this reality Apply direct picture, back side image, side elevational view picture and the designated user for specifically including in example and calling the electronic product to be identified Interface image；

Using the standard detection coordinate, direct picture, back side image, the side elevation of the electronic product to be recycled are extracted The geometry and ruler of image and designated user's interface image middle fuselage specification, display screen specification, casing and frame The geometry and size of very little, brand picture and text shape and size and camera, flash lamp, loudspeaker, different key, and mark Its fixed spatial position coordinate and mutual geometrical relationship, and typing defect identification database；

And extract position and the body spy of the designated user interface UI design feature, list style, icon and text Sign, demarcates its spatial position coordinate and its mutual geometrical relationship, and typing defect identification database；

And color, surface gloss, the surface roughness of the acquisition electronic product to be recycled, and typing defect identifies Database；

By the data in defect identification database described in typing and the standard picture number in the standard picture metadatabase According to contrast conting, extracts abnormal data and do not demarcate feature；

If not extracting abnormal data through above-mentioned contrast conting and not demarcating feature, then determine that the electronics to be recycled produces Product are new machine, and by recognition result typing defect identification database；

If extracting abnormal data through above-mentioned contrast conting and not demarcating feature, then successively known using the appearance quality Other template, deformation defect recognition template, screen display defect recognition template, component defect recognition template, functional impairment recognition template and Calculating identification is compared in system deficiency recognition template, and recognition result is distinguished typing defect identification database；

The data that defect identification is completed under coding item, which are compared, to be identified to all categories in the present embodiment, respectively All kinds of defect recognition template recognition success rates are calculated, causality and its logic for extracting crucial common feature and defect formation are suitable Sequence advanced optimizes defect recognition methods and identification step, improves defect recognition accuracy.

Recognition methods in the present embodiment includes: interface method of identification, shape method of identification；

The interface method of identification is suitable for be switched on and the electricity to be identified of interactive interface energy part display device system information Sub- product；Specifically includes the following steps:

The facility information interactive interface standard photographs of the various category electronic products of systemic presupposition；

Brand, model, specification, string number and related text and figure, color, frame shape, camera shooting are extracted using gridding method Key difference feature between head, flash lamp, loudspeaker, key etc. and same brand different size type product；

The facility information interactive interface photo of terminal installation scanning or shooting setting illumination and focal length；

System reads related text content using OCR technique identification；

Pass through pre-set programs contrast conting key difference feature；

Identify brand, model, specification, string number and related text and figure, color, frame shape, camera, flash lamp, Loudspeaker, key etc. and same brand different size type product；

The shape method of identification is suitable for that interactive interface is unable to sesolution apparatus system information and cannot be switched on but shell energy The electronic product to be identified of sesolution apparatus system information, specifically includes the following steps:

The front and back standard photographs of the various category electronic products of systemic presupposition；

Brand, model, specification, string number and related text and figure, color, frame shape, camera shooting are extracted using gridding method Key difference feature between head, flash lamp, loudspeaker, key etc. and same brand different size type product；

The obverse and reverse photo of terminal installation scanning or shooting setting illumination and focal length；

System passes through pre-set programs contrast conting key difference feature；

Identify brand, model, specification, string number and related text and figure, color, frame shape, camera, flash lamp, Loudspeaker, key etc. and same brand different size type product；

Defect recognition methods in the present embodiment further includes recycling grade identification decision step, for identification electronics to be identified The various defect features of product, and determine that it recycles grade accordingly, specifically includes the following steps:

Screen display defect, component defect, functional impairment, shape defect and the appearance of the various category electronic products of systemic presupposition at Feature, type and the grade of color；

Grade is recycled according to logical relation, the syntagmatic setting system between various defect features, type and grade；

Analyze various defect features in the different photos for extracting aforementioned electronic product to be identified respectively using gridding method；

System distinguishes the type and rank of the contrast conting defect by pre-set programs；

System determines recycling grade.

A kind of electronic product recovery method provided in this embodiment, for calculating the recycling etc. for determining electronic product to be recycled Grade and unit take-back cost, specifically includes the following steps:

Referring to figure 4., 5,6,7,8 according to the category catalogue and coding in the category database, presets:

Different brands, model, specification electronic product unit take-back cost be X；

Recycling price radix is W；

Products-hardware cost is K, and hardware allowance for depreciation is k；

Additional value share ratio is P, and brand value preserving rate is p；

Technical value added share ratio is R, and technology value preserving rate is r；

Technique added value share ratio is M, and technique value preserving rate is m；

Function added value share ratio is Q, and function value preserving rate is q；

It is Z that defect, which restores valence/resetting valence,；

Processing cost is Y, wherein tearing machine cost y open₁, environmental protection treatment cost y₂, account number cipher release expense y₃；

It recycles service process cost and expense is L；

Using losing coefficient t；

Reclaiming benefit rate i%；

Wherein:

Recycling price radix W in the present embodiment, encodes the newest producer price of electronic product or market under item using the category Sale price subduction industry is averaged Sales Channel expense as radix；

Products-hardware cost K in the present embodiment encodes electronic product configuration list and zero newest under item using the category The part market price is calculated；

Products-hardware allowance for depreciation k in the present embodiment contrasts general electronic products service life, counts according to yearly depreciation 20% It calculates；

Additional value share P in the present embodiment is set according to trade brand ranking, product market share and related public sentiment It is fixed, a generally default line brand Additional value share 40%, subsidiary brand 25%, three line brands 10%；According to market conditions It is added and subtracted with deep statistical data weighting；

Brand value preserving rate p% in the present embodiment, a generally default line brand year value preserving rate 90%, subsidiary brand 75%, three Line brand 50%；It is added and subtracted according to market conditions and deep statistical data weighting；

Technical value added share R in the present embodiment sets mark, research and development ability and related public sentiment according to similar product technology It is fixed, generally default head enterprise technology added value share 20%, two class enterprises 15%, three classes enterprise 10%；According to market conditions It is added and subtracted with deep statistical data weighting；

Technology value preserving rate r% in the present embodiment, generally default head enterprise technology year value preserving rate 80%, two class enterprises 70%, three classes enterprise 60%；It is added and subtracted according to market conditions and deep statistical data weighting；

Technique added value share M in the present embodiment sets mark, equipment technology and related public sentiment according to similar product technique It is fixed, generally default head enterprise technique added value share 20%, two class enterprises 25%, three classes enterprise 30%；According to market conditions It is added and subtracted with deep statistical data weighting；

Technique value preserving rate m% in the present embodiment, generally default head enterprise technique year value preserving rate 75%, two class enterprises 60%, three classes enterprise 50%；It is added and subtracted according to market conditions and deep statistical data weighting；

Function added value share Q in the present embodiment, compares according to similar product function and associated user comments on public sentiment setting, Generally default head enterprise feature added value value share 20%, two class enterprises 35%, three classes enterprise 50%；According to market conditions and Deep statistical data weighting plus-minus；

Function value preserving rate q% in the present embodiment, generally default head enterprise feature year value preserving rate 80%, two class enterprises 70%, three classes enterprise 60%；It is added and subtracted according to market conditions and deep statistical data weighting；

In the present embodiment defect restore valence/resetting valence Z, according to deformation defect, component defect, functional impairment detection project and Component repair information data analysis result statistics；

Cost of disposal Y includes: y in the present embodiment₁To tear machine cost open；y₂Cost is released for account number cipher, according to recycling product User account and password release complexity setting；y₃For environmental protection treatment cost；Cost of disposal according to recycle mode and Choice of flow direction, y₁And y₂It can add up, y₃With the former not accumulation calculating；

Removal process cost of serving and expense L, equipment artificial according to removal process and logistics Expenses Cost are calculated；

Reclaiming benefit rate i% is set according to the occupied fund time value and industry average rate of profit；

Coefficient t is lost using the time, according to appearance quality classification and grade, is set according to the guarantee period of identification；

According to aforementioned recovery article class catalogue and category identification coding, electronic product under the category identification coding item is set Recycle pricing data library；

Acquisition Product recycling trade information data, recycling quotation information data, component repair information data and regeneration in real time Resource information data, and related public sentiment data, pricing data library is recycled in deposit respectively；

Product in product user interface image is extracted using standard detection coordinate and/or OCR recognition methods as shown in Figure 9 MEID code, electronic product three guarantees guarantee state, user account and cryptographic state, system update version information etc. influence unit take-back cost Information data, respectively deposit recycling pricing data library；

Call under the category identification coding item basic information, auxiliary information, stream communication in the related information of electronic product Breath, calculating its recycling price radix is W_n, products-hardware cost K_n, hardware allowance for depreciation k_n, Additional value share ratio P_n, product Board value preserving rate p_n, technical value added share ratio R_n, technology value preserving rate r_n, technique added value share ratio M_n, technique value preserving rate m_n、 Function added value share ratio Q_n, function value preserving rate q_n, pricing data library is recycled in deposit respectively；

The defect categorical data for calling electronic product under the category identification coding item, according to defect class small pin for the case encoding statistics Its described defect restores valence/resetting valence Z_n, processing cost Y_nAnd its tear machine cost y open_1n, environmental protection treatment cost y_2n, account number cipher solution Except expense y_3n；

It identifies that the related information of electronic product and defect identify data under coding item according to the category, determines its recycling benefit With mode and flow direction, calculates its service process cost and expense is L_n, using losing coefficient t_n, Reclaiming benefit rate i_n%；

Bring formula into:

X_n={ K_n×(1-k_n)+(W_n-K_n)×(P_n×p_n+R_n×r_n+M_n×m_n+Q_n×q_n)×t_n-Z_n-Y_n-L_n}×(1- i_n)

Recycling valence X is estimated up to electronic product to be recycled_n。

It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims Variation is included within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.

In addition, it should be understood that although this specification is described in terms of embodiments, but not each embodiment is only wrapped Containing an independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should It considers the specification as a whole, the technical solutions in the various embodiments may also be suitably combined, forms those skilled in the art The other embodiments being understood that.

Claims (15)

1. a kind of electronic product intelligence recovery system, which is characterized in that including database module, computing module, run module and Terminal installation module；

The database module includes recycling category database, data acquisition module group and data processing mould group, is used for acquisition process System recycles the data of required by electronic product, and carries out classification storage；

The computing module includes recycling objective single mode group, identification calculating mould group, recycling price mould group and deep learning mould group, is used for Visitor's list is recycled to electronic product generation, the finally substantial recycling of the determination of recycling grade, products quotation is carried out to be oriented to the objective list of recycling Objective simple correlation content is completed at the same time the internal pricing algorithm of defining the level of self feed back optimization；

The operation module includes system operation mould group, application operating mould group and the safe mould group of system, is supported entirely for providing The data transmission of each intermodule is called in the control of recovery process, establishes safety verification and controlling terminal equipment operation；

The terminal installation module includes human-computer interaction mould group, identification acquisition mould group, industry control mould group, for providing use according to instruction Family control entrance, feedback processing acquire the data of electronic product and recycle automatically operation as a result, realizing.

2. a kind of electronic product intelligence recovery system according to claim 1, which is characterized in that the recycling category data Library is configured according to system operation process, including preset data typing library, category identification metadatabase, electronic product identification number According to library and recycle objective one process database；

Preset data typing library includes external data acquisition database, operation data input database, manual entry data Library, sensitive information monitor database；

The objective one process database of the recycling includes interim objective single database, recycles objective single database and recycling terminal database.

3. a kind of electronic product intelligence recovery system according to claim 2, which is characterized in that the electronic product identification Longitudinal functional relationship of database, data needed for recycling according to the vertical subordinate relation of electronic product and identification is matched as planning framework Set subdata base；

The vertical subordinate relation of the electronic product, specifically includes the successively electronics categories of subordinate, brand, the name of an article, model, rule Lattice, configuration and color；

Longitudinal functional relationship of data needed for the identification is recycled specifically includes the mutually longitudinal associated row of this kind of electronic product Industry data, data incomplete, utilize data and pricing data at category data；

The electronics categories include mobile phone, tablet computer, charger baby, wearable equipment, laptop.

4. a kind of electronic product intelligence recovery system according to claim 3, which is characterized in that described according to electronic product Vertical subordinate relation configures subdata base, specially sets gradually recycling category coding and mesh according to the subordinate relation of electronic product Record, framework vertical relation subdata base, classification processing, the recycling category of recording and storage various electronic products coding and its category Feature, defect feature and basic information data provide foundation for system identification and recycling, and the vertical relation subdata base is specific Include:

Level-one subdata base, the level-one subdata base include electronics categories database, the electronics categories data Library includes mobile phone database, tablet computer database, charger baby database, wearable device database and laptop data Library；

It is subordinated to the second level subdata base of electronics categories database, the second level subdata base includes brand database；

It is subordinated to the three-level subdata base of brand database, the three-level subdata base includes Brand Operation database, enterprise's letter Cease database, technical information database, supply chain database and product database.

5. a kind of electronic product intelligence recovery system according to claim 4, which is characterized in that described to be recycled according to identification Longitudinal functional relationship of required data specially realizes identification and recycling institute according to system as planning architecture configuration subdata base The functional relationship of data is needed, framework recycles longitudinal functional relationship subdata base of electronic product related information under category coding item, The related information data of electronic product under the various recycling category coding items of classification processing, recording and storage are system identification and are divided Analysis determines that recovery value, Land use systems and the flow direction of the electronic product, recycling transaction and sale matching provide foundation；The longitudinal direction Functional relationship subdata base is subordinated to electronics categories database, specifically includes:

It is subordinated to the second level subdata base of electronics categories database, the second level subdata includes category database, defect Database, utilizes database and pricing data library at sector database；

The category database includes personalized metadata library, brand recognition metadatabase, name of an article identification metadatabase, model knowledge Other metadatabase, specification identification metadatabase and configuration identification metadatabase；

The data incomplete library includes appearance quality metadatabase, deformation defect metadatabase, screen display defect metadatabase, component Defect metadatabase, functional impairment metadatabase and system deficiency metadatabase；

The sector database includes industry Zone Information database, market information database, technical resource database, professional standard number According to library and industrial chain database；

It is described to include orientation supply database, recycling industry database, recycling market database, aftermarket attachment number using database According to library, regenerated resources database；

The pricing data library includes recycling price metadatabase.

6. a kind of electronic product intelligence recovery system according to claim 2, which is characterized in that the data acquisition module group Related data is collected and recorded according to systemic presupposition and is stored in preset data typing library, including the record of external data acquisition unit, operation data Enter unit, manual entry unit and sensitive information monitoring unit, collects and records related data according to systemic presupposition, be stored in present count respectively According to external data acquisition database corresponding in typing library, operation data input database, manual entry database and sensitive letter Cease monitor database；

The external data acquisition unit presses the functional condition of the external resource mould group, and automatically retrieval collects and records target Internet resources Data, and data are stored in external data acquisition database respectively；

The operation data typing unit presses the system operation mould group functional condition, and automatic acquisition and recording system whole process runs number According to, and data are stored in operation data input database respectively；

The manual entry unit is default by system function, and the number of demand is corresponded to by manual type typing or supplement input system According to, and data are stored in manual entry database respectively；

The sensitive information monitoring unit is default by system function, automatic monitoring or retrieval acquisition target Internet resources related resource The instant sensitive information such as supply and demand variation, the market price and its fluctuation, industry major event, product discontinuance, version updating divides in real time Analysis processing and more new database, and data are stored in sensitive information monitor database respectively.

7. a kind of electronic product intelligence recovery system according to claim 5, which is characterized in that the data processing mould group It is default according to system function, operation need are calculated according to the system-computed module and the application operating module and terminal installation It asks, the data acquisition module group ephemeral data is analyzed and processed, generate system metadata and support system operations；

The data processing mould group includes type data processing mould group, category data processing mould group, data incomplete processing mould group, row Industry data processing mould group handles mould group using data processing mould group, pricing data.

8. a kind of electronic product intelligence recovery system according to claim 7, which is characterized in that the type data processing Mould group is default according to system function, classification storage recycle under category coding item the type classification method, standard of the electronic product and Electronics categories feature and recognition element are extracted in rule, analysis, are stored in corresponding category identification metadatabase, automated to respond to terminal Equipment recycling operation and system identification calculate mould group and identify electronics categories coding to be recycled, support system operations and terminal dress Standby mist calculates and application operating；

The category data processing mould group is default according to system function, and classification storage recycles the electronic product under category coding item The category feature and recognition element of electronic product are extracted in category classification method, standard and rule, analysis, are stored in the corresponding category and are compiled Personalized metadatabase, brand recognition metadatabase, the name of an article identify metadatabase, type identifier metadatabase, specification under code item It identifies metadatabase, configuration identification metadatabase, automated tos respond to terminal installation recycling operation and system identification calculates the identification of mould group Electronic product category coding to be recycled, supports system operations and terminal installation mist calculates and application operating；

The data incomplete processing mould group is default according to system function, and classification storage recycles the electronic product under category coding item The defect feature and recognition element of electronic product are extracted in defect classification method, standard and rule, analysis, are stored in the corresponding category and are compiled Code item under appearance quality metadatabase, deformation defect metadatabase, screen display defect metadatabase, component defect metadatabase, Functional impairment metadatabase, system deficiency metadatabase, automated to respond to terminal installation recycling operation and system identification calculates mould group The defect classification coding for identifying electronic product to be recycled, supports system operations and terminal installation mist calculates and application operating；

The industry data processing mould group is default according to system function, and classification storage recycles the electronic product under category coding item Related information classification method and rule, analysis extract the related information feature of electronic product and assist in identifying element, are stored in the product Class encodes corresponding category identification metadatabase module and defect under item and identifies metadatabase, and according to systemic presupposition analysis at The related information data that reason system is collected and recorded calculate mould group for system identification, handle mould group using data processing mould group, pricing data The recycling category coding and defect classification for identifying electronic product to be recycled encode, determine recycling mode and flow direction, calculate back It receives price and foundation is provided；

Described default according to system function using data processing mould group, classification storage recycles the electronic product under category coding item Using data classification method and rule, analysis extracts the utility value feature of electronic product and assists in identifying element, is stored in the product Class encodes corresponding recycling metadatabase under item, and recycles database according to systemic presupposition analysis process system Data are recycled, determine that recycling mode and flow direction, accounting unit take-back cost provide foundation for system；

The pricing data processing mould group is default according to system function, and classification storage recycles the electronic product under category coding item Recovery value classification method and rule, analysis extract the pricing information feature of electronic product and assist in identifying element, are stored in the product Class encodes corresponding recycling price metadatabase under item, and believes according to the price in systemic presupposition analysis process system database Data are ceased, unit take-back cost is calculated for system and foundation is provided.

9. a kind of electronic product intelligence recovery system according to claim 7, which is characterized in that the single processing of recycling visitor Mould group is default according to system function, management terminal equipment and record storage recycling visitor are single, automated to respond to terminal installation recycling operation, It is single to generate recycling visitor, is stored in interim objective single database respectively, recycles objective single database and recycle terminal database, and according to recycling Objective list preset recording content and sequence submit to system to complete recycling operation.

10. a kind of electronic product intelligence recovery system according to claim 9, which is characterized in that the identification calculates mould For group according to system command, the recycling visitor submitted according to the single processing mould group of recycling visitor is single, and retrieval collects the database purchase association Data identify that the electronics produces by the recognition methods of systemic presupposition category and defect classification recognition methods, and identification calculation procedure The category coding and defect classification coding of product；

The recycling price mould group according to system command, sentence by the recycling visitor list and system submitted according to the single processing mould group of recycling visitor Surely recycling mode and flow direction, recycling grade determine to collect the database purchase associated data as a result, retrieving, and pass through system Default recycling pricing method and program calculate the unit take-back cost for determining the electronic product；

The deep learning mould group according to system command, to the same category that system completes recycling recycle the image included in objective list and Major parameter is compared, and retrieval collects relevant information, extracts common feature optimization system metadata and work flow, is Realization system precisely identifies, refinement levels, rational quotation, and rapid computations, high-efficient homework, stable operation provide foundation.

11. a kind of electronic product intelligence recovery system according to claim 10, which is characterized in that the system runs mould Group runs process according to recycling work flow setting system, sequentially generates various command instruction system modules and its function list Member and mist computing terminal equipment be sequentially completed data acquisition storage and analysis processing, identification calculate and deep learning, information feedback, Operation is run, and default system function, modification management and system manager role are carried out by system background administration interface It preset with permission, modify management and system data typing and parameter setting, modification management；

Application operating mould group system operation process according to as defined in system operation mould group is default, and setting terminal is equipped back It is incorporated as industry to flow and functionality, sequentially generates various task instructions, system command key or document frame, message box, filled by terminal Standby interactive interface is for user's typing, supplement, modification, cancellation, confirmation operation；

The safe mould group of system sets security management processes, plan and security incident prediction scheme according to system demand for security and is System security procedure, establishes firewall, by means such as certification, encryption, monitoring, trackings, provides file encryption to system, identity is recognized Card, secure accessing, using the comprehensive security protection such as protection, access control, by digital certificate to service server, data The independent security deployment of the carry out such as library server, archive server and log server, unified management.

12. a kind of electronic product intelligence recovery system according to claim 11, which is characterized in that the human-computer interaction mould Group system operation process according to as defined in system operation mould group, the application operating mould group is default, and setting terminal equipment is handed over Mutual interface and function and function are realized；

The identification acquisition mould group is provided according to the data acquisition module group, system operation mould group, the application operating mould group System operation process it is default, setting terminal equips the image for successively acquiring electronic product to be recycled, reads data；

The industry control mould group, group is according to as defined in the data acquisition module group, system operation mould group, the application operating mould group System runs process and presets, and setting terminal equipment completes successively recycling operation automatically.

13. a kind of electronic product intelligence recovery system according to claim 10, which is characterized in that the category identification side Method specifically includes:

Category catalogue and coding are formulated, category database is preset；

It extracts and corresponds to electronic product recognition element and differentiation feature under the category catalogue and coding item, form category identification member Data preset category and identify metadatabase；

Electronic product characteristic information to be identified is acquired, category database is compared and category identifies metadatabase, is determined described wait know The category of other electronic product and corresponding coding；

To category catalogue in the category database and coding item under correspond to electronic product related information carry out in real time acquisition with more Newly, the related information includes:

Basic information；

Auxiliary information；

Circulation information；

Standard picture information；

The basic information includes: type, brand, the name of an article, model, specification, configuration, color, system software and version information；

The auxiliary information includes: electronic product brand operator, core group parts suppliers, manufacturer, retailer and main Sales Channel and value chain enterprise industrial and commercial registration information and its material items disclose information, Electronic Product Core technology, main Production technology and its procedure information；

The circulation information includes: electronic product issue date, producer price, list marketing date, Sales Channel and sale price Lattice, sales volume, software change version, hardware bill of materials and halt production stop pin date information；

The standard picture information includes: electronic product direct picture and back side image, can show configuration and the friendship of functional information Mutual interface image, side elevational view picture；

The recognition element and differentiation feature are extracted electronic product standard picture by using netting analysis and filtration method and are obtained It arrives, specifically includes:

Extract the recognition element and difference stored in the standard picture between the electronic product and other variety classes electronic products Alienation feature；

Extract the identification member stored in the standard picture between the electronic product and its same type but different brands electronic product Element and differentiation feature；

Extract the recognition element and difference stored in the standard picture between the electronic product and its same brand but the different names of an article Change feature；

Extract store recognition element in the standard picture between the electronic product and its same brand, the name of an article but different model and Differentiation feature；

Extract the identification stored in the standard picture between the electronic product and its same brand, the name of an article, model but different size Element and differentiation feature；

It extracts and stores the electronic product and its same brand, the name of an article, model and specification in the standard picture, but different configuring products Between recognition element and differentiation feature；

The recognition element and differentiation feature include:

The surface shape and color of electronic product；

Electronic product brand pattern and character；

The layer of structure and color of electronic product enclosure and frame and component；

Functional unit and interactive interface UI design of the electronic product including camera, flash lamp, loudspeaker, key are special Position and the body coordinate, geometrical relationship of sign, list style, icon and text；

Analysis processing electronic product includes type, brand, the name of an article, model, specification, the recognition element of configuration and differentiation feature Characteristic feature, form the category catalogue and encode the individualized feature metadata under item, and be stored in corresponding category and identify first number According to library；

The individualized feature metadata and corresponding recognition element and differentiation feature are analyzed, makes the category electronic product respectively Category identification template, brand recognition template, name of an article recognition template, type identifier template, specification recognition template and configuration identification Template, and it is stored in corresponding category identification metadatabase；

The individualized feature metadata and corresponding recognition element and differentiation feature before analysis is handled, corresponding standard picture, point The front recognition template, back side recognition template and specified interactive interface recognition template of the category electronic product are not made, and are stored in Corresponding category identifies metadatabase；

The acquisition electronic product characteristic information to be identified, compares category database and category identifies metadatabase, specifically includes:

If electronic product to be identified can be switched on and interactive interface can sesolution apparatus information, normally execute subsequent identification step Suddenly, the step of identifying in subsequent identification step by interactive interface template is otherwise skipped over；

Acquire direct picture, back side image or the specified interactive interface image of electronic product to be identified；

Respectively by the direct picture of the electronic product to be identified, back side image or specified interactive interface image and corresponding recognition template It is compared and is mutually authenticated, identify that its category encodes, obtain electronic product all information to be identified；

It the comparison and is mutually authenticated and specifically includes:

If the direct picture of electronic product to be identified, back side image and specified interactive interface image pass sequentially through its personalized identification Template can accurately identify its category coding, then be automatically performed identification, store information；

Otherwise corresponding recognition template successively is called according to the relation on attributes of the product category, brand, the name of an article, model, specification, configuration It is identified: if its recognition template can accurately identify its category coding, being automatically performed identification, store information, otherwise extracted each The identical feature of a step is mutually authenticated, and identifies that its category encodes, stores information；

In comparing and being mutually authenticated, through whole contrast contings:

Identification is unable to template and corresponds to configuration, specification, model, successively determines that it is affiliated specification, model, the name of an article and corresponds to category Encode the minimalist configuration type under item；

Identification is unable to template and corresponds to the name of an article, is determined as that the brand corresponds to brand gradeless and uniformly-priced goods machine under category coding item；

Identification is unable to template and corresponds to brand, and identification decision is gradeless and uniformly-priced goods machine of inferior brand；

Identification is unable to template and corresponds to type, is determined as solid waste machine；

The direct picture of the standard picture and electronic product to be identified, back side image and specified interactive interface image are using packet The method for including plus setting coordinate grid and optical filtering color filter plate and image noise reduction or adjust gray value reduces interference, improves category identification Rate；

The image that identification is completed under all category catalogues and coding item is compared, according to the knowledge called in identification process Other template and identified image calculate separately recognition success rate, extract crucial common feature, Statistical error template.

14. a kind of electronic product recovery method according to claim 10, which is characterized in that the defect recognition methods is logical It crosses in the image information for extracting electronic product to be identified and encodes the off-note of the standard picture of electronic product under item with the category, It identifies its various defect classification, and determines its recovery value accordingly and its recycle mode and flow direction；

The defect recognition methods implementation steps are as follows:

Category catalogue and coding are formulated, defect category database is preset；

It presets various category catalogues and encodes appearance quality classification, the deformation defect classification, screen display defect of the electronic product under item Classification standard and rule and its defect coding of classification, component defect classification, functional impairment classification and system deficiency classification, deposit Defect category database；

The classification standard of electronic product and rule and its defect coding under item are encoded according to category, using mesh coordinate method and optical filtering Colour filter method and interactive interface path relation extract appearance quality classification, the deformation defect classification, screen display of the category electronic product The recognition element of defect classification, component defect classification, functional impairment classification and system deficiency classification marks corresponding defect classification and knows Other template；

It is correspondingly arranged defect identification metadatabase；

Recognition methods is selected according to electronic product situation to be identified, acquires electronic product characteristic information to be identified, compares defect class Other database and defect identify metadatabase, determine the category of the electronic product to be identified and correspond to coding；

The recognition methods includes: defect classification interface method of identification, defect classification shape method of identification；

Defect classification interface method of identification is suitable for be switched on and interactive interface can partially show the electricity to be identified of facility information Sub- product；

The defect classification shape method of identification is suitable for that interactive interface is unable to sesolution apparatus system information and cannot be switched on but outside The electronic product to be identified of shell energy sesolution apparatus information；

Defect classification interface method of identification for identification electronic product to be identified hardware configuration defect information, system software lack Damage information and functional impairment information, read device personality user setting, determine its screen display defect classification, functional impairment classification, State, product guarantee state is arranged in system deficiency classification and user account password；

Specifically includes the following steps:

Preset the facility information interactive interface standard picture of various category electronic products；

Using under the functions of the equipments and system health state in mesh coordinate method extract equipment interaction interface standard picture The path relation feature of key feature and interactive interface, mark screen display defect interface recognition template, the identification of functional impairment interface Template and system deficiency interface recognition template；

Terminal installation scans or shoots the facility information interactive interface image of electronic product to be identified；

It is identified using OCR technique and reads related text content, extract hardware configuration defect information, system software defect information, read Take equipment account number cipher setting state, the setting of product guarantee state personalized user；

Facility information interactive interface image and its screen display defect interface by pre-set programs contrast conting electronic product to be identified The difference of recognition template, functional impairment interface recognition template and system deficiency interface recognition template obtains defect information and correlation Content；

The defect classification shape method of identification is suitable for identifying that the appearance quality, deformation defect, screen display of electronic product to be identified lack Damage, component defect, functional impairment and system deficiency feature, damage condition and defect degree,

Specifically includes the following steps:

Preset front and back, the side elevation standard picture of various category electronic products；

Using gridding method extract electronic product front and back, side elevation standard picture, the mode of appearance of nondestructive state, component, Function and system core feature, mark appearance quality, deformation defect, screen display defect, component defect, functional impairment and system deficiency Feature and defect classification recognition template；

Terminal installation scans or shoots front, reverse side and the side elevational view picture of electronic product to be identified；

Mould is identified by the front of pre-set programs contrast conting electronic product to be identified, reverse side and side elevational view picture and defect classification The feature difference of plate；

Identify electronic product to be identified appearance quality classification and deformation, screen display, component, function and system defect classification；

Execute recycling grade identification, for identification the various defect features of electronic product to be identified, and determine its recycling accordingly etc. Grade, specifically includes:

Preset the spy of the screen display defects of various category electronic products, component defect, functional impairment, shape defect and appearance quality Sign, type and rank；

Grade is recycled according to logical relation, the syntagmatic setting system between various defect features, type and grade；

Analyze various defect features in the different photos for extracting aforementioned electronic product to be identified respectively using gridding method；

The type and rank of the contrast conting defect are distinguished by pre-set programs；

Determine the electronic product recycling grade to be identified.

15. a kind of electronic product recovery method according to claim 10, which is characterized in that the recycling pricing method pair Electronic product to be recycled carries out estimating recycling valence, comprising the following steps:

According to the category catalogue and coding in the category database, preset:

Different brands, model, specification electronic product unit take-back cost be X；

Recycling price radix is W；

Products-hardware cost is K, and hardware allowance for depreciation is k；

Additional value share ratio is P, and brand value preserving rate is p；

Technical value added share ratio is R, and technology value preserving rate is r；

Technique added value share ratio is M, and technique value preserving rate is m；

Function added value share ratio is Q, and function value preserving rate is q；

It is Z that defect, which restores valence/resetting valence,；

Processing cost is Y, wherein tearing machine cost y open₁, environmental protection treatment cost y₂, account number cipher release expense y₃；

It recycles service process cost and expense is L；

Using losing coefficient t；

Reclaiming benefit rate i%；

Wherein:

According to the category recognition methods or defect recognition methods, brand, the model, rule of recognition and verification electronic product to be recycled Lattice, transferring its corresponding recycling price radix is W_n；

Products-hardware cost is K_n, hardware allowance for depreciation is k_n；

Additional value share ratio is P_n, brand value preserving rate is p_n；

Technical value added share ratio is R_n, technology value preserving rate is r_n；

Technique added value share ratio is M_n, technique value preserving rate is m_n；

Function added value share ratio is Q_n, function value preserving rate is q_n；

It is Z that defect, which restores valence,_n；

Processing cost is Y_n, wherein tearing machine cost y open_1n, environmental protection treatment cost y_2n, account number cipher release expense y_3n；

It recycles service process cost and expense is L_n；

Using losing coefficient t_n；

Reclaiming benefit rate i_n%；

Bring formula into:

X_n={ K_n×(1-k_n)+(W_n-K_n)×(P_n×p_n+R_n×r_n+M_n×m_n+Q_n×q_n)×t_n-Z_n-Y_n-L_n}×(1-i_n)

Recycling valence X is estimated up to electronic product to be recycled_n。