CN109344155A - Timber metrical information automatic record method, device, electronic equipment and storage medium - Google Patents
Timber metrical information automatic record method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109344155A CN109344155A CN201810973961.9A CN201810973961A CN109344155A CN 109344155 A CN109344155 A CN 109344155A CN 201810973961 A CN201810973961 A CN 201810973961A CN 109344155 A CN109344155 A CN 109344155A
- Authority
- CN
- China
- Prior art keywords
- information
- wood structure
- structure information
- network model
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The embodiment of the present application discloses a kind of timber metrical information automatic record method, device, electronic equipment and storage medium.Wherein, this method comprises: obtaining the multimedia messages in timber measurement process, text information is converted by the voice messaging in the multimedia messages；The first nerves network model trained by one extracts the first wood structure information from the text information；Store the first wood structure information.The embodiment of the present application passes through artificial intelligence automatically by the speech recognition in multimedia messages at structured message, to provide strong support for the efficient record of dipping data and management.
Description
Technical field
This application involves field of artificial intelligence, and in particular to a kind of timber metrical information automatic record method, device,
Electronic equipment and storage medium.
Background technique
Forestry is the important basic industry in China, as the mankind continually develop forestry, causes environment to occur various
Ecological problem.Trees greening work has obtained the attention of various circles of society, and it is gloomy just to become Science in Future in China to the fine-grained management of forest
The important development direction of woods industry.
One important directions of fine-grained management be the measurement that production, felling, transaction to forest are refined and
Record, one of them regular task are to carry out dipping measurement to the timber of transaction.So-called dipping is i.e. to a certain number of wood
The process that size, material kind, specification of material etc. are measured and recorded.Traditional mode is by manual measurement and manually recorded
Mode is completed.In general, manual operation is measured by one to two people, while completing to record by another recorder.With
The development of artificial intelligence technology, but there are no a kind of automatic technology schemes can replace this artificial mode at present, with
It saves human cost and accelerates efficiency.
Summary of the invention
For above-mentioned technical problem in the prior art, the embodiment of the present application proposes a kind of timber metrical information and remembers automatically
Recording method, device, electronic equipment and computer readable storage medium, to solve log scaling measurement human cost height, low efficiency
The problem of.
The first aspect of the embodiment of the present application provides a kind of timber metrical information automatic record method, comprising:
The multimedia messages in timber measurement process are obtained, convert text for the voice messaging in the multimedia messages
Information；
The first nerves network model trained by one extracts the first wood structureization letter from the text information
Breath；
Store the first wood structure information.
In some embodiments, the method also includes:
When further including image information in determining the multimedia messages, described image information is obtained；
The third nerve network model trained by one extracts the second wood structureization letter from described image information
Breath；
Using the second wood structure information supplement and/or improve the first wood structure information.
In some embodiments, the method also includes:
By the first wood structure information and/or the second wood structure information input to second trained
Neural network model obtains feedback information；
It receives user and supplements the response message of the feedback information and/or improve the first wood structureization letter
Breath.
In some embodiments, the first wood structure information and/or the second wood structure information include
One or more of timber kind, the timber place of production, timber size, lumber quality.
In some embodiments, the feedback information includes that key structure field is reminded, missing information is reminded, mistake letter
One or more of breath prompting.
In some embodiments, the method also includes at least one following retraining steps:
Use the second wood structure information and/or the response message as markup information to the first nerves
Network model carries out retraining；
Use the first wood structure information and/or the second wood structure information as markup information to institute
It states nervus opticus network model and carries out retraining；
And/or
Use the first wood structure information and/or the response message as markup information to the third nerve
Network model carries out retraining.
The second aspect of the embodiment of the present application provides a kind of timber metrical information self-recording unit, comprising:
Speech recognition module will be in the multimedia messages for obtaining the multimedia messages in timber measurement process
Voice messaging is converted into text information；
First nerves network model module, for passing through the first nerves network model trained from the text envelope
The first wood structure information is extracted in breath；
Memory module, for storing the first wood structure information.
In some embodiments, described device further include:
Image collection module when for further including image information in determining the multimedia messages, obtains described image
Information；
Third nerve network model module, for being believed by a third nerve network model trained from described image
The second wood structure information is extracted in breath；
First auxiliary memory module, for using the second wood structure information supplement and/or improving described first
Wood structure information.
In some embodiments, described device further include:
Nervus opticus network model module is used for the first wood structure information and/or the second timber knot
Structure information input obtains feedback information to the nervus opticus network model trained；
Second auxiliary memory module, is supplemented and/or perfect for receiving user to the response message of the feedback information
The first wood structure information.
In some embodiments, the first wood structure information and/or the second wood structure information include
One or more of timber kind, the timber place of production, timber size, lumber quality.
In some embodiments, the feedback information includes that key structure field is reminded, missing information is reminded, mistake letter
One or more of breath prompting.
In some embodiments, described device further includes at least one following retraining module:
First retraining module, for using the second wood structure information and/or the response message as mark
It infuses information and retraining is carried out to the first nerves network model；
Second retraining module, for using the first wood structure information and/or second wood structure
Information carries out retraining to the nervus opticus network model as markup information；
And/or
Third retraining module, for using the first wood structure information and/or the response message as mark
It infuses information and retraining is carried out to the third nerve network model.
The third aspect of the embodiment of the present application provides a kind of electronic equipment, comprising:
Memory and one or more processors；
Wherein, the memory is connect with one or more of processor communications, and being stored in the memory can quilt
The instruction that one or more of processors execute, when described instruction is executed by one or more of processors, the electronics
Equipment is for realizing the method as described in foregoing embodiments.
The fourth aspect of the embodiment of the present application provides a kind of computer readable storage medium, and being stored thereon with computer can
It executes instruction, when the computer executable instructions are executed by a computing apparatus, can be used to realize as described in foregoing embodiments
Method.
5th aspect of the embodiment of the present application provides a kind of computer program product, and the computer program product includes
The computer program being stored on computer readable storage medium, the computer program include program instruction, work as described program
When instruction is computer-executed, it can be used to realize the method as described in foregoing embodiments.
The embodiment of the present application passes through artificial intelligence automatically by the speech recognition in multimedia messages at structured message, thus
Strong support is provided for the efficient record of dipping data and management.
Detailed description of the invention
The feature and advantage of the application can be more clearly understood by reference to attached drawing, attached drawing is schematically without that should manage
Solution is carries out any restrictions to the application, in the accompanying drawings:
Fig. 1 is that a kind of process of timber metrical information automatic record method according to shown in some embodiments of the present application is shown
It is intended to；
Fig. 2 is the schematic diagram for extracting structured message from text information according to shown in some embodiments of the present application；
Fig. 3 is after extracting structured message using first nerves network model according to shown in some embodiments of the present application
Reuse the schematic diagram of a scenario that nervus opticus network model is fed back；
Fig. 4 is a kind of structural frames of timber metrical information self-recording unit according to shown in some embodiments of the present application
Figure；
Fig. 5 is the structural block diagram of a kind of electronic equipment according to shown in some embodiments of the present application.
Specific embodiment
In the following detailed description, many details of the application are elaborated by example, in order to provide to correlation
The thorough understanding of disclosure.However, for those of ordinary skill in the art, the application can obviously not have this
Implement in the case where a little details.It should be understood that " system " used herein, " device ", " unit " and/or " module " art
Language is for distinguishing in the sequence arrangement different components of different stage, element, part or a kind of method of component.However, such as
Identical purpose may be implemented in other expression formulas of fruit, these terms can be replaced by other expression formulas.
It should be understood that when equipment, unit or module be referred to as " ... on ", " being connected to " or " being coupled to " it is another
When equipment, unit or module, can directly in another equipment, unit or module, be connected or coupled to or with other equipment,
Unit or module communication, or may exist intermediate equipment, unit or module, unless context clearly prompts exceptional situation.Example
Such as, term "and/or" used in this application includes any one and all combinations of entry listed by one or more correlations.
Term used herein limits the application range only for describing specific embodiment.Such as present specification
With shown in claims, unless context clearly prompts exceptional situation, " one ", "one", the words such as "an" and/or "the"
Odd number is not refered in particular to, may also comprise plural number.It is, in general, that term " includes " and "comprising" only prompt to include the spy clearly identified
Sign, entirety, step, operation, element and/or component, and such statement do not constitute one it is exclusive enumerate, other features,
Including entirety, step, operation, element and/or component also may include.
Referring to the following description and the annexed drawings, these or other feature and feature, operating method, the phase of structure of the application
Function, the combination of part and the economy of manufacture for closing element can be better understood, and wherein description and accompanying drawings form
Part of specification.It is to be expressly understood, however, that attached drawing is used only as the purpose of illustration and description, it is not intended to limit this
The protection scope of application.It is understood that attached drawing is not necessarily drawn to scale.
Various structures figure used herein is used to illustrate various modifications according to an embodiment of the present application.It should be understood that
, before or following structure be not for limiting the application.The protection scope of the application is subject to claim.
Dipping measurement is the basis of forest tree resource fine-grained management, only sufficiently grasps each timber by dipping measurement
Details just can guarantee that each step of production, felling and transaction to forest all carries out effectively supervision and retrospect.Due to woods
The diversity of the equal-specifications such as timber kind, size, texture, defect, grade, purposes, dipping measurement in the prior art mostly can only be according to
Rely manual operation execute, even if the measuring device slightly higher there are some the degree of automation, can only also mitigate for size, quantity,
The measurement work of this kind of objective information of the volume of timber then can not similar to the information of the heavy dependences subjective experience such as material kind, the place of production, defect
Replace artificial detection completely.During artificial detection, the acquisition of metrical information and record are always to influence dipping efficiency
Key factor, the efficiency being manually entered is extremely low, directly records multimedia messages (including voice, image and/or video etc.) then not
Convenient for statistics, retrieval and verification, thus need acquisition of information and in terms of using automation means improve working efficiency.
In embodiments herein, as shown in Figure 1, providing a kind of timber metrical information automatic record method, pass through
Artificial intelligence is automatically by the speech recognition in multimedia messages at structured message, thus for the efficient record and pipe of dipping data
Reason provides strong support.Specifically, which includes:
S101 obtains the multimedia messages in timber measurement process, and the voice messaging in the multimedia messages is converted
For text information.
It wherein, is most using multimedia messages (especially voice) output measurement result information for artificial detection
Convenient, the highest mode of efficiency, but for information record management end, this unstructured data of multimedia messages is very unfavorable
It in subsequent inquiry, compares and modifies, be the key that influence data-handling efficiency privacy.Therefore, in the reality of the application
It applies in example, multimedia messages is handled first, voice messaging therein is converted to the more tractable text envelope of computer
Breath.Preferably, the process that speech recognition is converted into text can promote effect by the artificial intelligence model after machine learning training
Rate and accuracy；Specifically, identification process generally includes several steps such as pretreatment, feature extraction and pattern match.In view of existing
There is the research for having very more speech recognitions based on artificial intelligence in technology, herein not to the specific implementation side of speech recognition
Formula have ever made more limitations, and any relevant speech recognition technology can be applied in embodiments herein.
S102, the first nerves network model trained by one extract the first wood structure from the text information
Change information.
Wherein, untreated text information is still a kind of unstructured data, although can use its progress
Certain inquiry and modification, but computer equipment can not actually understand the practical significance of non-structured text, be difficult with
Non-structured text information is more effectively managed, such as the comparison of statistics, key message and verification etc..Therefore, in this Shen
In embodiment please, further also by text information processing be structured message, so as to for items fine-grained management project provide
It supports.
The text information processing of the prior art usually all be unable to do without participle, keyword extraction, syntactic analysis and meaning of a word analysis
Etc. processing steps, however semantic analysis and understanding of the mode heavy dependence of the prior art to words sentence, i.e., largely according to
Rely the semantic relation between the vocabulary and word of dictionary, these, which require formerly to put into a large amount of artificial treatment, is just able to satisfy.?
In embodiments herein, it is contemplated that the particularity in log detection field realizes text information by way of artificial intelligence
The processing of structuring.
Specifically, first nerves network is included at least in embodiments herein, first nerves network can be RNN
(Recurrent Neural Network, Recognition with Recurrent Neural Network), LSTM (Long Short-Term Memory, shot and long term note
Recall network) even depth neural network, it is trained by sufficient amount of text information, an ideal neural network model can
To automatically analyze and identify the key content in one section of text information, and it is processed into the structuring suitable for storage management automatically
Field information, thus a large amount of manpower can be saved after scale and avoid the dependence to artificial treatment.In addition, neural network
Model is also equipped with learns perfect ability automatically, can Quick Extended its identifying processing range and promote the accuracy rate of identification, into
One step reduces the demand of manual intervention, and the efficiency of automatic processing is substantially improved.
S103 stores the first wood structure information.
The storage and management of structured message generally realize that database manipulation belongs to the prior art by Database Systems,
Here not reinflated description.Under normal circumstances, the concrete type of structured message and content are obtained according to the text after speech recognition
It arrives, in the preferred embodiment of the application, can also be speculated and be expanded according to historical data and logic.Also that is, it is right
The structured message obtained after the text-processing of speech recognition, can be according to first processing result come comprehensive descision, to fill out
It mends its missing content or corrects its wrong content.
It, can be full-automatic by the text information processing of speech recognition combination artificial intelligence in embodiments herein
Identification, record and the management of log scaling measurement data are completed in ground, relative to traditional scale book means and existing text
The technical solution of processing mode, the embodiment of the present application can effectively replace manual operation, under the premise of guaranteeing that information is accurate and reliable,
Greatly improve working efficiency.
Further, in the preferred embodiment of the application, the training process of text information processing model includes: standard
Standby a large amount of text information, the text information can be acquired the common dipping sample of culler by voice messaging and pass through voice
Identification is transformed.The data sample an of training sample is given in Fig. 2, wherein " Cameroon produces white Chinese parasol tree, length 5
Rice, 70 centimetres of diameter, there is ring crack in an end face " it is one section of word that the culler of text is converted by speech recognition.It is wherein related
The metrical information of trees is marked, for example the information such as the relevant region of current trees, type, attribute and defect are marked in this article
In notebook data；In addition, can also further use the relationship between information to mark, for example the relationship between region and type is used
" place of production " mark.
More training datas are obtained by mode as shown in Figure 2, such as " South Africa produces Korean pine, and 5 meters of length, directly
50 centimetres of diameter, there is slight crack on surface " it is another training data.Certainly, sample shown in Fig. 2 is only simple samples of text,
The text information that may be handled in specific implementation process is longer, and the feature of mark is more, and sample shown in Fig. 2 is not construed as to this Shen
Please technical solution specific embodiment limitation.Using the above-mentioned training data marked, input a neural network (such as
RNN or LSTM network etc.) in be trained, obtain for text identification processing first nerves network model.By described first
Neural network model is connected with front end speech recognition module, is deployed among smart machine；In specific implementation procedure, intelligently set
One section of voice messaging of standby acquisition culler, and text information is converted by speech recognition, and enter this information into first
In RNN the or LSTM network of neural network model, the characteristic attribute of the word of the text is obtained.
Further, according to the characteristic attribute of word, information is stored in the data structure table of a structuring, such as following table
Shown in I.Specific implementation method is instance name, the Property Name, relation name extracted based on neural network, using template
Match or the knowledge base way of recommendation extracts attribute value, infers relationship or instance name.For example, certain described template matching extracts example
For " 5 meters of Cameroon (place of production) white platane wood (instance name) length ", pass through matching instance name (white platane wood) and " length "
Attribute, " 5 meters " of character string before extracting Separate Text With (such as punctuate ", " etc.) are used as attribute value.On the other hand, the knowledge
Library recommended method is to pre-establish knowledge base, and by matching " instance name (white platane wood) ", place of production title " Cameroon " is pushed away
Recommend the relationship " place of production " for providing the two.Wherein, the knowledge base is the triplet sets established based on domain body, the ternary
Group passes through artificial obtained by calibrating for log detection field professional.
Table I structured message sample
Note that simple structure field needed for only listing some dippings in Table I is as an example, in a particular application, it can
The field that can need to use is more, and sample shown in Table I is not construed as the limit to technical scheme specific embodiment
System.
Further, since log scaling information has certain normalization, a part of information field is more important, it is necessary to
Record.Another part is then optional information, can support shortage of data.Meanwhile these fields may it is different with material kind and
There is different importance, such as the material kind for high-end furniture needs to provide more detailed information, such as texture, color, flower
Straight degree of line, circle etc., these details determine the final economic value of the material kind.And the material kind of some more low sides then only needs
It takes dimensions and the simple informations such as volume.Therefore, another nerve can be trained based on obtained structural data
Network, for generating feedback information, to prompt user to input correlation measurement information of high importance.
Fig. 3 gives the example fed back by nervus opticus network model, and user believes in initial multimedia
The voice messaging of " a Malaysian San Cheong wood, 5 meters of length, 50 centimetres of diameter " is had input when ceasing typing.Smart machine is first
Voice messaging is converted to free text, further, free text is converted to structure using first nerves network by smart machine
Change information.The structured fields such as " type ", " place of production ", " length ", " diameter " are contained in the structured message.Since user neglects
" the straight degree of circle " this feature is omited, therefore first network does not find corresponding attribute in free text, the structured field
For default value (Null, as null value).Further, smart machine is by nervus opticus network model, by the attribute of current trees
Information input to the model is handled, and obtains first being fed back to that " Malaysian San Cheong wood circle is directly spent critically important, and whether you want
Record? ".Note that the feedback is obtained not by judgement " the straight degree of circle " field whether default (Null), but
Nervus opticus network model generally should be comprising " circle is straight by the measurement data that training data has obtained Malaysian San Cheong wood
Degree " attribute, when " the straight degree of circle " attribute missing, training data gives corresponding prompt information.In other embodiments, described
Prompt information can provide corresponding suggestion content according to the quantity of the structured message of acquisition, attribute and content.For example, by
When identification mistake occur in the input or speech recognition of recorder's mistake, neural network recognization " place of production " and " kind " are respectively
" Russia " and " rubber ", then provide miscue " Russia do not produce rubber tree, you whether input error? " prompt.
In embodiments herein, further by nervus opticus network model to the wood structure information of extraction into
Row identification and screening generate so that user be helped further to judge key message or the error message of missing for the anti-of prompt
Feedforward information, further to promote the efficiency and accuracy of dipping measurement.
Further, in the preferred embodiment of the application, the multimedia messages of typing can also include when measurement
Image information is assisted in identifying by processing image information and obtains wood structure information.In the prior art, the image of typing
Information is retained typically as aucillary document, and the analysis and identification to picture material are also relied primarily on and be accomplished manually, due to image
Effectiveness of retrieval and accuracy are lower, and the image information in existing scheme can not give dipping other than increasing and storing pressure
Measurement brings apparent benefit.In a preferred embodiment of the present application, further located by a third nerve network model
Image information is managed, therefrom identify and extracts wood structure information when dipping measures.Typically, it is obtained by imaging sensor
Image information, for example, being clapped while user's input voice information " a white platane wood of Cameroon, image are as follows " by camera
Take the photograph the image information of corresponding timber.Optionally, the image information and voice messaging can be the letter in the same multimedia messages
Breath (video flowing etc. with voice and image recorded simultaneously), is also possible to the multimedia messages obtained respectively.Pass through first
Speech recognition and first nerves network model obtain the partial structured information of current timber as first structure information, such as
Type is white platane wood, the place of production is Cameroon etc.；When confirming while getting image information, pass through third nerve network model
Image information is identified, obtains another part structured message of current timber as the second structured message.Wherein,
The field of two structured messages can be identical or different with first structure information, complements one another both when field difference, when
It is mutually authenticated both when field is identical；For example, mentioned by voice obtained type be white platane wood, the place of production be Cameroon etc. letter
Breath has then obtained the content that the information such as size, circle straight degree, defect are used to lack in supplementing structure information by image, into
Third nerve network model of one step based on artificial intelligence can be known by the features such as the texture of timber, color and decorative pattern in image
Not Chu timber kind be mutually authenticated with " white platane wood ", to enrich the structured message of timber using image recognition.
Optionally, can also simultaneously by nervus opticus network model to the first and/or second structured message of timber into
Row feedback, further to supplement or perfect frame information.In another preferred embodiment of the application, multiple neural networks
The information that model obtains is except for being alternatively arranged as the markup information of other neural network models to help in addition to being complementary to one another and improving
The neural network model is helped further to be trained, to obtain more preferably neural network model and knowledge base.For example, logical
Text after crossing the processing speech recognition of first nerves network model has obtained the information such as type, size and defect, passes through third mind
The information such as type, size and defect have also been obtained after network model image recognition, are determining that first nerves network model obtains
Information confidence level it is higher when, use the information as markup information input third nerve network model in, to help third
Neural network model improving performance and accuracy.
Fig. 4 is the timber metrical information self-recording unit schematic diagram according to shown in some embodiments of the present application.Such as Fig. 4
Shown, which includes:
Speech recognition module 410 will be in the multimedia messages for obtaining the multimedia messages in timber measurement process
Voice messaging be converted into text information；
First nerves network model module 420, for passing through the first nerves network model trained from the text
The first wood structure information is extracted in this information；
Memory module 430, for storing the first wood structure information.
In some embodiments, described device further include:
Image collection module when for further including image information in determining the multimedia messages, obtains described image
Information；
Third nerve network model module, for being believed by a third nerve network model trained from described image
The second wood structure information is extracted in breath；
First auxiliary memory module, for using the second wood structure information supplement and/or improving described first
Wood structure information.
In some embodiments, described device further include:
Nervus opticus network model module is used for the first wood structure information and/or the second timber knot
Structure information input obtains feedback information to the nervus opticus network model trained；
Second auxiliary memory module, is supplemented and/or perfect for receiving user to the response message of the feedback information
The first wood structure information.
In some embodiments, the first wood structure information and/or the second wood structure information include
One or more of timber kind, the timber place of production, timber size, lumber quality.
In some embodiments, the feedback information includes that key structure field is reminded, missing information is reminded, mistake letter
One or more of breath prompting.
In some embodiments, described device further includes at least one following retraining module:
First retraining module, for using the second wood structure information and/or the response message as mark
It infuses information and retraining is carried out to the first nerves network model；
Second retraining module, for using the first wood structure information and/or second wood structure
Information carries out retraining to the nervus opticus network model as markup information；
And/or
Third retraining module, for using the first wood structure information and/or the response message as mark
It infuses information and retraining is carried out to the third nerve network model.
With reference to attached drawing 5, the electronic equipment schematic diagram provided for the application one embodiment.As shown in figure 5, the electronic equipment
500 include:
Memory 530 and one or more processors 510；
Wherein, the memory 530 is communicated to connect with one or more of processors 510, is deposited in the memory 530
The instruction 532 that can be executed by one or more of processors is contained, described instruction 532 is by one or more of processors
510 execute, so that one or more of processors 501 execute the method as described in foregoing embodiments.
Certainly, relevant technical staff in the field is appreciated that obtain the multimedia messages such as voice and image, the application
The device and/or electronic equipment of embodiment obviously may also comprise voice acquisition module and image capture module, such as microphone, figure
As sensor, camera etc., modular structure cited by above embodiments is not construed as implementing technical scheme
Limitation.
One embodiment of the application provides a kind of computer readable storage medium, in the computer readable storage medium
Computer executable instructions are stored with, the computer executable instructions execute the side as described in foregoing embodiments after being performed
Method.
It is apparent to those skilled in the art that for convenience and simplicity of description, the equipment of foregoing description
, can be with reference to the corresponding description in aforementioned device embodiment with the specific work process of module, details are not described herein.
Although subject matter described herein is held in the execution on the computer systems of binding operation system and application program
It is provided in capable general context, but it will be appreciated by the appropriately skilled person that may also be combined with other kinds of program module
To execute other realizations.In general, program module include routines performing specific tasks or implementing specific abstract data types,
Program, component, data structure and other kinds of structure.It will be understood by those skilled in the art that subject matter described herein can
It is practiced, including handheld device, multicomputer system, based on microprocessor or can compiled with using other computer system configurations
Journey consumption electronic product, minicomputer, mainframe computer etc., it is possible to use in wherein task by being connected by communication network
In the distributed computing environment that remote processing devices execute.In a distributed computing environment, program module can be located locally and far
In the two of journey memory storage device.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and method and step can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
Scope of the present application.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words
The part of the part or the technical solutions that contribute to original technology can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) execute each embodiment the method for the application all or part of the steps.
And computer-readable storage medium above-mentioned include with store as computer readable instructions, data structure, program module or its
Any mode or technology of the information such as his data are come the physics volatile and non-volatile, removable and can not be situated between because of east realized
Matter.Computer-readable storage medium specifically includes, but is not limited to, USB flash disk, mobile hard disk, read-only memory (ROM, Read-
Only Memory), random access memory (RAM, Random Access Memory), Erasable Programmable Read Only Memory EPROM
(EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory or other solid-state memory technologies, CD-ROM, number are more
Functional disc (DVD), HD-DVD, blue light (Blue-Ray) or other light storage devices, tape, disk storage or other magnetic storages
Equipment or any other medium that can be used to store information needed and can be accessed by computer.
In conclusion present applicant proposes a kind of timber metrical information automatic record method, device, electronic equipment and its meters
Calculation machine readable storage medium storing program for executing.The embodiment of the present application is by artificial intelligence automatically by the speech recognition in multimedia messages at structuring
Information, to provide strong support for the efficient record of dipping data and management.In embodiments herein, pass through voice
Identification combine artificial intelligence text information processing, can fully automatically complete log scaling measurement data identification, record and
Management, relative to traditional scale book means and existing text-processing mode, the technical solution of the embodiment of the present application can have
Effect replaces manual operation, under the premise of guaranteeing that information is accurate and reliable, greatly improves working efficiency.
It should be understood that the above-mentioned specific embodiment of the application is used only for exemplary illustration or explains the application's
Principle, without constituting the limitation to the application.Therefore, that is done without departing from spirit and scope is any
Modification, equivalent replacement, improvement etc., should be included within the scope of protection of this application.In addition, the application appended claims purport
Covering the whole variations fallen into attached claim scope and boundary or this range and the equivalent form on boundary and is repairing
Change example.
Claims (10)
1. a kind of timber metrical information automatic record method characterized by comprising
The multimedia messages in timber measurement process are obtained, convert text envelope for the voice messaging in the multimedia messages
Breath；
The first nerves network model trained by one extracts the first wood structure information from the text information；
Store the first wood structure information.
2. the method according to claim 1, wherein the method also includes:
When further including image information in determining the multimedia messages, described image information is obtained；
The third nerve network model trained by one extracts the second wood structure information from described image information；
Using the second wood structure information supplement and/or improve the first wood structure information.
3. method according to claim 1 or 2, which is characterized in that the method also includes:
By the first wood structure information and/or the second wood structure information input to the nervus opticus trained
Network model obtains feedback information；
It receives user and supplements the response message of the feedback information and/or improve the first wood structure information.
4. according to the method in claim 2 or 3, which is characterized in that the method also includes at least one following retraining
Step:
Use the second wood structure information and/or the response message as markup information to the first nerves network
Model carries out retraining；
Use the first wood structure information and/or the second wood structure information as markup information to described
Two neural network models carry out retraining；
And/or
Use the first wood structure information and/or the response message as markup information to the third nerve network
Model carries out retraining.
5. a kind of timber metrical information self-recording unit characterized by comprising
Speech recognition module, for obtaining the multimedia messages in timber measurement process, by the voice in the multimedia messages
Information is converted into text information；
First nerves network model module, for passing through a first nerves network model trained from the text information
Extract the first wood structure information；
Memory module, for storing the first wood structure information.
6. device according to claim 5, which is characterized in that described device further include:
Image collection module when for further including image information in determining the multimedia messages, obtains described image information；
Third nerve network model module, for passing through a third nerve network model trained from described image information
Extract the second wood structure information；
First auxiliary memory module, for using the second wood structure information supplement and/or improving first timber
Structured message.
7. device according to claim 5 or 6, which is characterized in that described device further include:
Nervus opticus network model module is used for the first wood structure information and/or second wood structure
Information input obtains feedback information to the nervus opticus network model trained；
Second auxiliary memory module, it is described to supplement and/or improve to the response message of the feedback information for receiving user
First wood structure information.
8. device according to claim 6 or 7, which is characterized in that described device further includes at least one following retraining
Module:
First retraining module, for using the second wood structure information and/or the response message to believe as mark
Breath carries out retraining to the first nerves network model；
Second retraining module, for using the first wood structure information and/or the second wood structure information
Retraining is carried out to the nervus opticus network model as markup information；
And/or
Third retraining module, for using the first wood structure information and/or the response message to believe as mark
Breath carries out retraining to the third nerve network model.
9. a kind of electronic equipment characterized by comprising
Memory and one or more processors；
Wherein, the memory is connect with one or more of processor communications, and being stored in the memory can be described
The instruction that one or more processors execute, when described instruction is executed by one or more of processors, the electronic equipment
For realizing method according to any of claims 1-4.
10. a kind of computer readable storage medium, is stored thereon with computer executable instructions, refer to when the computer is executable
When order is executed by a computing apparatus, it can be used to realize method according to any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810973961.9A CN109344155B (en) | 2018-08-24 | 2018-08-24 | Automatic recording method and device for wood measurement information, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810973961.9A CN109344155B (en) | 2018-08-24 | 2018-08-24 | Automatic recording method and device for wood measurement information, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109344155A true CN109344155A (en) | 2019-02-15 |
| CN109344155B CN109344155B (en) | 2021-09-17 |
Family
ID=65291896
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810973961.9A Active CN109344155B (en) | 2018-08-24 | 2018-08-24 | Automatic recording method and device for wood measurement information, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109344155B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111982905A (en) * | 2020-08-26 | 2020-11-24 | 杭州宣迅电子科技有限公司 | Wood quality intelligent detection system based on industrial big data image analysis |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7257531B2 (en) * | 2002-04-19 | 2007-08-14 | Medcom Information Systems, Inc. | Speech to text system using controlled vocabulary indices |
| CN102359938A (en) * | 2011-09-16 | 2012-02-22 | 长沙高新技术产业开发区爱威科技实业有限公司 | Morphological analytical apparatus and method for erythrocytes |
| CN106251865A (en) * | 2016-08-04 | 2016-12-21 | 华东师范大学 | A kind of medical treatment & health record Auto-writing method based on speech recognition |
| CN106649762A (en) * | 2016-12-27 | 2017-05-10 | 竹间智能科技（上海）有限公司 | Intention recognition method and system based on inquiry question and feedback information |
| CN107967491A (en) * | 2017-12-14 | 2018-04-27 | 北京木业邦科技有限公司 | Machine learning method, device, electronic equipment and the storage medium again of plank identification |
| CN108256550A (en) * | 2017-12-14 | 2018-07-06 | 北京木业邦科技有限公司 | A kind of timber classification update method and device |
| CN108320781A (en) * | 2018-03-15 | 2018-07-24 | 安徽科大讯飞医疗信息技术有限公司 | A kind of voice-based medical report generation method and device |
-
2018
- 2018-08-24 CN CN201810973961.9A patent/CN109344155B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7257531B2 (en) * | 2002-04-19 | 2007-08-14 | Medcom Information Systems, Inc. | Speech to text system using controlled vocabulary indices |
| CN102359938A (en) * | 2011-09-16 | 2012-02-22 | 长沙高新技术产业开发区爱威科技实业有限公司 | Morphological analytical apparatus and method for erythrocytes |
| CN106251865A (en) * | 2016-08-04 | 2016-12-21 | 华东师范大学 | A kind of medical treatment & health record Auto-writing method based on speech recognition |
| CN106649762A (en) * | 2016-12-27 | 2017-05-10 | 竹间智能科技（上海）有限公司 | Intention recognition method and system based on inquiry question and feedback information |
| CN107967491A (en) * | 2017-12-14 | 2018-04-27 | 北京木业邦科技有限公司 | Machine learning method, device, electronic equipment and the storage medium again of plank identification |
| CN108256550A (en) * | 2017-12-14 | 2018-07-06 | 北京木业邦科技有限公司 | A kind of timber classification update method and device |
| CN108320781A (en) * | 2018-03-15 | 2018-07-24 | 安徽科大讯飞医疗信息技术有限公司 | A kind of voice-based medical report generation method and device |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111982905A (en) * | 2020-08-26 | 2020-11-24 | 杭州宣迅电子科技有限公司 | Wood quality intelligent detection system based on industrial big data image analysis |
| CN111982905B (en) * | 2020-08-26 | 2021-02-19 | 北新国际木业有限公司 | Wood quality intelligent detection system based on industrial big data image analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109344155B (en) | 2021-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107168945B (en) | Bidirectional cyclic neural network fine-grained opinion mining method integrating multiple features | |
| CN108446540B (en) | Program code plagiarism type detection method and system based on source code multi-label graph neural network | |
| CN107291783B (en) | Semantic matching method and intelligent equipment | |
| CN110489538A (en) | Sentence answer method, device and electronic equipment based on artificial intelligence | |
| CN109165385A (en) | Multi-triple extraction method based on entity relationship joint extraction model | |
| CN106897559B (en) | A kind of symptom and sign class entity recognition method and device towards multi-data source | |
| CN104503998B (en) | For the kind identification method and device of user query sentence | |
| CN106095753B (en) | A kind of financial field term recognition methods based on comentropy and term confidence level | |
| CN107943911A (en) | Data pick-up method, apparatus, computer equipment and readable storage medium storing program for executing | |
| CN106407113B (en) | A kind of bug localization method based on the library Stack Overflow and commit | |
| CN112231472B (en) | Judicial public opinion sensitive information identification method integrated with domain term dictionary | |
| CN109408821B (en) | Corpus generation method and device, computing equipment and storage medium | |
| CN111291570A (en) | Method and device for realizing element identification in judicial documents | |
| CN103995885B (en) | The recognition methods of physical name and device | |
| CN103680493A (en) | Voice data recognition method and device for distinguishing regional accents | |
| WO2020010834A1 (en) | Faq question and answer library generalization method, apparatus, and device | |
| JP2020191075A (en) | Recommendation of web apis and associated endpoints | |
| CN109063713A (en) | A kind of timber discrimination method and system based on the study of construction feature picture depth | |
| Barber et al. | The SALIX method: A semi‐automated workflow for herbarium specimen digitization | |
| CN109857846A (en) | The matching process and device of user's question sentence and knowledge point | |
| CN111339268A (en) | Entity word recognition method and device | |
| CN106933802B (en) | Multi-data-source-oriented social security entity identification method and device | |
| CN111177328B (en) | Question-answer matching system and method, question-answer processing device and medium | |
| CN109408175B (en) | Real-time interaction method and system in general high-performance deep learning calculation engine | |
| CN109344155A (en) | Timber metrical information automatic record method, device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |