The content of the invention
A kind of DLP system based on machine learning provided in an embodiment of the present invention, it is one to solve current DLP display wallsThe very huge system of set, the inside uses substantial amounts of chip and sensor component, the operation of whole system is support jointly, itsExploitation and the maintenance difficulties in later stage well imagine, the very difficult technical problem of the management of caused DLP display systems.
A kind of DLP system based on machine learning provided in an embodiment of the present invention, including：
Signal acquisition terminal, network exchange end, display end and machine learning module；
The signal acquisition terminal, the display end and the machine learning module are communicated to connect with the network exchange end；
Wherein, the machine learning module gets the signal acquisition terminal, the display by the network exchange endAll data flows at end, the machine learning module sets and the data flow pair according to the threshold value of preset monitor control index parameterThe data label answered, forms multilayer learning network, and the data flow includes critical performance parameters to be monitored.
Preferably, the signal acquisition terminal is at least 2.
Preferably, the signal acquisition terminal includes：
Signal source video acquisition module, video front processing module, IP coding modules and network sending module；
The signal source video acquisition module, the video front processing module, the IP coding modules and the networkSending module is communicated to connect successively.
Preferably, the display end is at least 2.
Preferably, the display end includes：
Network receiver module, IP decoder modules, video signal processing module and movement drive display module；
The network receiver module, the IP decoder modules, the video signal processing module and the movement drive aobviousShow that module is communicated to connect successively.
The DLP system for being preferably based on machine learning also includes：
Signal dispatching platform, communicates to connect with the network exchange end.
The DLP system for being preferably based on machine learning also includes：
Network storage module, communicates to connect with the network exchange end.
The DLP system for being preferably based on machine learning also includes：
PC transcoding clusters, communicate to connect with the network exchange end.
The DLP system for being preferably based on machine learning also includes：
Ray machine, communicates to connect with the display end.
Preferably, the multilayer learning network is neutral net.
As can be seen from the above technical solutions, the embodiment of the present invention has advantages below：
A kind of DLP system based on machine learning provided in an embodiment of the present invention, including：Signal acquisition terminal, network exchangeEnd, display end and machine learning module；Signal acquisition terminal, display end and machine learning module and network exchange end communicate to connect；Wherein, machine learning module gets signal acquisition terminal, all data flows of display end, machine learning mould by network exchange endRoot tuber sets data label corresponding with data flow according to the threshold value of preset monitor control index parameter, forms multilayer learning network, numberInclude critical performance parameters to be monitored according to stream.In the present embodiment, got by network exchange end by machine learning moduleSignal acquisition terminal, all data flows of display end, machine learning module according to the threshold value of preset monitor control index parameter setting withThe corresponding data label of data flow, forms multilayer learning network, and data flow includes critical performance parameters to be monitored, solves meshPreceding DLP display walls are a set of very huge systems, and the inside uses substantial amounts of chip and sensor component, supports jointlyThe operation of whole system, its exploitation and the maintenance difficulties in later stage well imagine that the management of caused DLP display systems is very difficultTechnical problem.
Specific embodiment
A kind of DLP system based on machine learning provided in an embodiment of the present invention, it is one to solve current DLP display wallsThe very huge system of set, the inside uses substantial amounts of chip and sensor component, the operation of whole system is support jointly, itsExploitation and the maintenance difficulties in later stage well imagine, the very difficult technical problem of the management of caused DLP display systems.
To enable that goal of the invention of the invention, feature, advantage are more obvious and understandable, below in conjunction with the present inventionAccompanying drawing in embodiment, is clearly and completely described, it is clear that disclosed below to the technical scheme in the embodiment of the present inventionEmbodiment be only a part of embodiment of the invention, and not all embodiment.Based on the embodiment in the present invention, this areaAll other embodiment that those of ordinary skill is obtained under the premise of creative work is not made, belongs to protection of the present inventionScope.
Refer to Fig. 1, a kind of one embodiment bag of the DLP system based on machine learning provided in an embodiment of the present inventionInclude：
Signal acquisition terminal 1, network exchange end 2, display end 3 and machine learning module 4；
The signal acquisition terminal 1, the display end 2 and the machine learning module 3 and the communication link of network exchange end 4Connect；
Wherein, the machine learning module 4 gets the signal acquisition terminal 1, described aobvious by the network exchange end 2Show all data flows at end 3, machine learning module 4 sets corresponding with data flow according to the threshold value of preset monitor control index parameterData label, forms multilayer learning network, and data flow includes critical performance parameters to be monitored (such as：The temperature of display end ray machine,The parameter of each important devices of the influence systematic entirety energy such as network speed of exchange network).
Further, the signal acquisition terminal 1 is at least 2.
Further, the signal acquisition terminal 1 includes：
Signal source video acquisition module, video front processing module, IP coding modules and network sending module；
The signal source video acquisition module, the video front processing module, the IP coding modules and the networkSending module is communicated to connect successively.
Further, the display end 3 is at least 2.
Further, the display end 3 includes：
Network receiver module, IP decoder modules, video signal processing module and movement drive display module；
The network receiver module, the IP decoder modules, the video signal processing module and the movement drive aobviousShow that module is communicated to connect successively.
Further, the DLP system based on machine learning also includes：
Signal dispatching platform 5, communicates to connect with the network exchange end.
Further, the DLP system based on machine learning also includes：
Network storage module 6, communicates to connect with the network exchange end.
Further, the DLP system based on machine learning also includes：
PC transcodings cluster 7, communicates to connect with the network exchange end.
Further, the DLP system based on machine learning also includes：
Ray machine 8, communicates to connect with the display end.
Further, the multilayer learning network is neutral net.
It is described with a concrete application scene below, as shown in Fig. 2 application examples includes：
Machine learning functional module, the yellow of the Fig. 2 seen in Figure of description are added on the basis of system overall architectureThe module of background.First, the functional module can be with real-time collecting and the ginseng of each important devices for monitoring influence systematic entirety energyNumber, when exception occur in these parameters, record carries out preliminary mark to these data, records log files, forms ground floor studyNetwork, then proceedes to the mark according to the study of the first layer network as input, the second layer " learning network " is built, with suchPush away, help to deep understanding system and abnormal deeper factor occur.
Assuming that " 8186 chip temperature ", " bulb brightness " and " network data transmission speed " is respectively：Tri- dimensions of T, L, VThe monitoring object for system is spent, respectively, it is allowed to which the threshold range of fluctuation is respectively：Td, Ld, Vd, accordingly label is Tag_T, Tag_l, Tag_v (can set " 1 " mark normal, " 0 " mark is abnormal).
Data flow converges to " machine learning " module by network, and the module can be according to the threshold value of monitor control index parameter beforeSetting enters row label to corresponding data, so " data element " that obtains is this form：[T+-Td,L+-Ld,V+-Vd,Tag_T, Tag_l, Tag_v], and the number of plies of network can with it is corresponding be label number, the final output of the network is normal systemRun or go wrong, when going wrong, it is possible to which backward tracing Primary Component is where to go wrong most possiblyPlace.
In the middle of the easy neutral net, when system goes wrong, we can inquire about the label of currently " data element ",If：[T+-Td, L+-Ld, V+-Vd, 0,0,1], it is possible to which the reason for problem occurs in confirmation system is 8186 chip temperatures(T) and bulb brightness (L) is abnormal caused, can also past last layer review：Check whether Tag_t and Tag_l have each otherThe causality of priority and cause abnormal.Can thus confirm well problem produce it is profound the reason for.
In the present embodiment, the institute of signal acquisition terminal, display end is got by network exchange end by machine learning moduleThere is data flow, machine learning module sets corresponding data label according to the threshold value of preset monitor control index parameter, forms multilayerLearning network, it is a set of very huge system to solve current DLP display walls, and the inside uses substantial amounts of chip and sensingDevice device, supports the operation of whole system jointly, and its exploitation and the maintenance difficulties in later stage well imagine that caused DLP showsShow the very difficult technical problem of the management of system.Development cost can at utmost be reduced；May be such that whole system reaches globalityThe optimization of energy, is favorably improved the adaptive ability of system；System testing is improved profoundly gives the value to be brought of product.
It is apparent to those skilled in the art that, for convenience and simplicity of description, the system of foregoing description,The specific work process of device and unit, may be referred to the corresponding process in preceding method embodiment, will not be repeated here.
In several embodiments provided herein, it should be understood that disclosed system, apparatus and method can be withRealize by another way.For example, device embodiment described above is only schematical, for example, the unitDivide, only a kind of division of logic function there can be other dividing mode when actually realizing, for example multiple units or componentCan combine or be desirably integrated into another system, or some features can be ignored, or do not perform.It is another, it is shown orThe coupling each other for discussing or direct-coupling or communication connection can be the indirect couplings of device or unit by some interfacesClose or communicate to connect, can be electrical, mechanical or other forms.
The unit that is illustrated as separating component can be or may not be it is physically separate, it is aobvious as unitThe part for showing can be or may not be physical location, you can with positioned at a place, or can also be distributed to multipleOn NE.Some or all of unit therein can be according to the actual needs selected to realize the mesh of this embodiment scheme's.
In addition, during each functional unit in each embodiment of the invention can be integrated in a processing unit, it is also possible toIt is that unit is individually physically present, it is also possible to which two or more units are integrated in a unit.Above-mentioned integrated listUnit can both be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.
If the integrated unit is to realize in the form of SFU software functional unit and as independent production marketing or useWhen, can store in a computer read/write memory medium.Based on such understanding, technical scheme is substantiallyThe part for being contributed to prior art in other words or all or part of the technical scheme can be in the form of software productsEmbody, the computer software product is stored in a storage medium, including some instructions are used to so that a computerEquipment (can be personal computer, server, or network equipment etc.) performs the complete of each embodiment methods described of the inventionPortion or part steps.And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM, Read-OnlyMemory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can store journeyThe medium of sequence code.
The above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to precedingEmbodiment is stated to be described in detail the present invention, it will be understood by those within the art that：It still can be to precedingState the technical scheme described in each embodiment to modify, or equivalent is carried out to which part technical characteristic；And theseModification is replaced, and does not make the spirit and scope of the essence disengaging various embodiments of the present invention technical scheme of appropriate technical solution.