CN110321850A - Garment material automatic identifying method, device, system, equipment and storage medium - Google Patents
Garment material automatic identifying method, device, system, equipment and storage medium Download PDFInfo
- Publication number
- CN110321850A CN110321850A CN201910603902.7A CN201910603902A CN110321850A CN 110321850 A CN110321850 A CN 110321850A CN 201910603902 A CN201910603902 A CN 201910603902A CN 110321850 A CN110321850 A CN 110321850A
- Authority
- CN
- China
- Prior art keywords
- image
- garment material
- identification
- clothes
- clothing
- 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.)
- Pending
Links
- 239000000463 material Substances 0.000 title claims abstract description 142
- 238000000034 method Methods 0.000 title claims abstract description 80
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 48
- 239000004744 fabric Substances 0.000 claims abstract description 39
- 238000003062 neural network model Methods 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 11
- 238000004891 communication Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 5
- 230000007787 long-term memory Effects 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 238000005520 cutting process Methods 0.000 claims description 4
- 238000011478 gradient descent method Methods 0.000 claims description 4
- 230000015654 memory Effects 0.000 claims description 4
- 230000006870 function Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 4
- 230000003321 amplification Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 239000012141 concentrate Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 230000016776 visual perception Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000009394 selective breeding Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/698—Matching; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses a kind of garment material automatic identifying method, device, system, equipment and computer readable storage mediums.Wherein, method includes the triggering image taking instruction when the positional relationship of microimaging head and clothes to be captured meets default shooting condition, so that microimaging head carries out image taking to clothes to be captured automatically according to pre-set image enlargement ratio；When the image of clothing for receiving the acquisition of microimaging head, the garment material identification model automatic identification constructed in advance and the fabric classification for exporting image of clothing are utilized；Garment material identification model is based on element study method, using obtained by image pattern collection training neural network model；Image pattern collection includes multiple style book images for carrying garment material class label.The application realizes garment material identification automation, not only increases garment material recognition efficiency, also reduces garment material identification cost.
Description
Technical field
The present embodiments relate to image identification technical field, more particularly to a kind of garment material automatic identifying method,
Device, system, equipment and computer readable storage medium.
Background technique
In garment marketing industry, especially on-line selling, the image of clothing and corresponding clothes of high quality are often required to use
Dress information is described to present customers clothes.
The relevant technologies usually rely on artificial shooting image of clothing, then by manual identified garment material and in clothing information
Increase fabric information in description.But artificial shooting is not only very high to the competency profiling of shooting personnel, but also shooting efficiency ratio
It is lower, in particular for high-volume clothes camera application scene, shoot higher cost.In addition, showing image by conventional garment
The affiliated type of garment material and quality good or not can not accurately be differentiated.And with the fast development of garment production processing technology and
User to clothes require it is higher and higher, the type of garment material is also more and more abundant, for example, one major class of silk fabrics just include spin,
Thin silk fabric, sieve, silk fabric, satin, thin,tough silk, crape, yarn, brocade, raw silk, suede, form, the fabric of Ge Shisi seed type, and it is different from different dimensions
Mode classification, such as from raw material building form can be divided into uni-material fabric, blended fabric, union, hand over and four major class of fabric.
Therefore, the identification of garment material is a very crucial and cumbersome job, and the identification work of garment material is still with people at present
Based on the mode of work, not only efficiency is relatively low, but also higher cost.
Summary of the invention
The embodiment of the present disclosure provides a kind of garment material automatic identifying method, device, equipment and computer-readable storage
Medium realizes garment material identification automation, improves garment material recognition efficiency, reduces garment material identification cost.
In order to solve the above technical problems, the embodiment of the present invention the following technical schemes are provided:
On the one hand the embodiment of the present invention provides a kind of garment material automatic identifying method, comprising:
When the positional relationship of detection microimaging head and clothes to be captured meets default shooting condition, transmission image taking refers to
It enables, image taking is carried out to the clothes to be captured automatically to trigger the microimaging head according to pre-set image enlargement ratio；
When the image of clothing for receiving the microimaging head acquisition, certainly using the garment material identification model constructed in advance
The dynamic fabric classification identified and export the image of clothing；
Wherein, the garment material identification model is to utilize image pattern collection training neural network based on element study method
Model obtains；Described image sample set includes multiple style book images for carrying garment material class label.
Optionally, the training process of the garment material identification model includes:
N number of garment material classification is randomly selected from described image sample set, every class garment material classification includes m clothes
Sample image, to generate batch training sample set；
Image preprocessing is carried out to the style book image in the batch processing sample set, and using gradient descent method to pre-
The shot and long term memory network model parameter first constructed is updated, to complete a model training；
If model training number reaches preset times threshold value, the training process of the garment material identification model is terminated.
Optionally, the style book image progress image preprocessing in the batch processing sample set includes:
Style book image is concentrated to carry out random image scaling described batch of training sample；
The size of style book image after scaling is cut to preset standard size；
Selective erasing processing is carried out to the image block of the style book image after cutting, the size of described image block is not more than
The 1/2 of corresponding style book picture size.
Optionally, described to include: to described batch of training sample concentration style book image progress random image scaling
Style book image is concentrated to carry out Random Graph described batch of training sample according to the scale value within the scope of pre-set zoom
As scaling；The zoom ranges are [0.3,3.0].
Optionally, described image shooting instruction includes image enlargement ratio, and described image enlargement ratio is 20,50 or 100.
The embodiment of the invention also provides a kind of garment material automatic identification equipments, comprising:
Model prebuild module is obtained for being based on element study method using image pattern collection training neural network model
Garment material identification model；Described image sample set includes multiple style book images for carrying garment material class label；
Image taking control module meets default clap for the positional relationship when detection microimaging head and clothes to be captured
Take the photograph condition, send image taking instruction, with trigger the microimaging head according to pre-set image enlargement ratio automatically to it is described to
It shoots clothes and carries out image taking；
Fabric identification module, for utilizing what is constructed in advance when the image of clothing for receiving the microimaging head acquisition
Garment material identification model automatic identification and the fabric classification for exporting the image of clothing.
On the other hand the embodiment of the present invention provides a kind of garment material automatic recognition system, comprising:
Microimaging head, wireless communication components and data processor；The microimaging head and the data processor are logical
It crosses the wireless communication components and carries out data communication；
The data processor is used to meet default shooting when the positional relationship of the microimaging head and clothes to be captured
Image taking instruction is sent when condition, to trigger the microimaging head according to pre-set image enlargement ratio automatically to described wait clap
It takes the photograph clothes and carries out image taking；When the image of clothing for receiving the microimaging head acquisition, the garment surface constructed in advance is utilized
Material identification model automatic identification and the fabric classification for exporting the image of clothing；
Wherein, the garment material identification model is to utilize image pattern collection training neural network based on element study method
Model obtains；Described image sample set includes multiple image of clothing samples for carrying garment material class label.
Optionally, the data processor is arranged in intelligent mobile terminal；The garment material identification model is deployed to cloud
Hold server；
The intelligent mobile terminal sends image recognition instruction to the cloud server, so that the cloud server is sharp
Recognition result is simultaneously fed back to the intelligence by the fabric classification of the image of clothing described in the garment material identification model automatic identification
It can mobile terminal.
The embodiment of the invention also provides a kind of garment material automatic identification equipment, including processor, the processor is used
The step of realizing the garment material automatic identifying method as described in preceding any one when executing the computer program stored in memory.
The embodiment of the present invention finally additionally provides a kind of computer readable storage medium, the computer readable storage medium
On be stored with garment material automatic identification procedure, realize when the garment material automatic identification procedure is executed by processor as former
The step of one garment material automatic identifying method.
The advantages of technical solution provided by the present application, is, using garment material identification model to camera in different amplifications
The clothes micro-image shot under multiplying power carries out the automatic identification of garment material, the corresponding fabric classification of output fabric micro-image
Information realizes garment material identification automation, not using only the technique of family visual perception garment material and different garment face
Otherness between material also greatly improves the efficiency of garment material identification work, reduces garment material identification cost, has
Conducive to quickly and accurately detail pictures and classification information of sale displaying fabric on line, the displaying information of clothes is enriched.
In addition, the embodiment of the present invention also directed to garment material automatic identifying method provide corresponding realization device, system,
Equipment and computer readable storage medium, further such that the method have more practicability, described device, system, equipment and
Computer readable storage medium has the advantages that corresponding.
It should be understood that the above general description and the following detailed description are merely exemplary, this can not be limited
It is open.
Detailed description of the invention
It, below will be to embodiment or correlation for the clearer technical solution for illustrating the embodiment of the present invention or the relevant technologies
Attached drawing needed in technical description is briefly described, it should be apparent that, the accompanying drawings in the following description is only this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of flow diagram of garment material automatic identifying method provided in an embodiment of the present invention；
Fig. 2 is a kind of training flow diagram of garment material identification model provided in an embodiment of the present invention；
Fig. 3 is a kind of specific embodiment structure chart of garment material automatic identification equipment provided in an embodiment of the present invention；
Fig. 4 is a kind of specific embodiment structure chart of garment material automatic recognition system provided in an embodiment of the present invention.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.Obviously, described embodiments are only a part of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall within the protection scope of the present invention.
The description and claims of this application and term " first ", " second ", " third " " in above-mentioned attached drawing
Four " etc. be for distinguishing different objects, rather than for describing specific sequence.Furthermore term " includes " and " having " and
Their any deformations, it is intended that cover and non-exclusive include.Such as contain a series of steps or units process, method,
System, product or equipment are not limited to listed step or unit, but may include the step of not listing or unit.
After describing the technical solution of the embodiment of the present invention, the various non-limiting realities of detailed description below the application
Apply mode.
Referring first to Fig. 1, Fig. 1 is a kind of process signal of garment material automatic identifying method provided in an embodiment of the present invention
Figure, the embodiment of the present invention may include the following contents:
S101: judging whether microimaging head and the positional relationship of clothes to be captured meet default shooting condition, if so,
Execute S102.
In embodiments of the present invention, shooting condition can for default setting, execute in different application scenario triggered camera
The condition of image taking, different application scene are for example related with user demand to clothing cloth to be captured.Microimaging head can be
It is moved under the control of drive system, until being moved to target position, then executes image capture operations.In the target position
Place, microimaging head is at a distance from clothes to be captured and locating spatial position meets default shooting condition.Wherein, can exist in advance
It hangs and multiple sensors is set on the workbench of clothes, sensor is believed at a distance from clothes to be captured for calculating microimaging head
The relative position of breath and microimaging head and clothes to be captured, that is, spatial position locating for microimaging head.System root
It is pre-set and is stored to the data information of shooting condition and sensor acquisition according to photographed scene according to user and is compared, come
Execute S101 operation.
S102: image taking instruction is sent.
It is understood that system send image taking to instruct to microimaging hair, microimaging head receives the instruction
Execute image capture operations automatically afterwards.In addition, can also artificially judge when the positional relationship of microimaging head and clothes to be captured is full
Sufficient shooting condition, to system input picture photographing information, such as take the photograph instruction triggering can by operation mobile computing device on
The application program of prepackage is completed.After system receives the information, triggering image taking instruction transmission so that microimaging head according to
Pre-set image enlargement ratio carries out image taking to clothes to be captured automatically.
Optionally, image taking instruction may include image enlargement ratio, and image enlargement ratio may be, for example, but be not intended to limit
In 20,50 or 100.Certainly, image enlargement ratio also directly can pass through the knob or key being arranged thereon in microimaging head end
Carry out artificial selection.It can include image being sent to microimaging hair to promote the degree of automation of entire fabric identification
The image taking of enlargement ratio instructs.
S103: judge whether to receive the image of clothing of microimaging head acquisition, if so, executing S104.
S104: the garment material identification model automatic identification constructed in advance and the fabric classification for exporting image of clothing are utilized.
In this application, garment material identification model may be based on element study method, utilize image pattern collection training nerve
Obtained by network model；Image pattern collection includes multiple style book images for carrying garment material class label.Each clothes sample
The label of this image can be manually labeled in advance.In order to further enhance the accuracy of model identification, reduce due to hard
The influence for the noise that part equipment is introduced in image acquisition process, the clothes of the image pattern collection of training garment material identification model
Sample image is and the image of clothing of clothes to be captured is acquired by the same camera.
In technical solution provided in an embodiment of the present invention, using garment material identification model to camera in different amplifications
The clothes micro-image shot under multiplying power carries out the automatic identification of garment material, the corresponding fabric classification of output fabric micro-image
Information realizes garment material identification automation, not using only the technique of family visual perception garment material and different garment face
Otherness between material also greatly improves the efficiency of garment material identification work, reduces garment material identification cost, has
Conducive to quickly and accurately detail pictures and classification information of sale displaying fabric on line, the displaying information of clothes is enriched.
In a specific embodiment, it please refers to shown in Fig. 2, the training of garment material identification model, which can be used, to be based on
LSTM (shot and long term memory network), garment material can be used in the training method of meta learning (meta learning), model framework
The training process of identification model can include:
N number of garment material classification is randomly selected from image pattern concentration, every class garment material classification includes m style book
Image, to generate batch training sample set.Such as can from image pattern concentrate randomly select 5 garment material classifications, then for
Each garment material classification of above-mentioned 5 garment material classifications randomly chooses 5 style books for being subordinate to the garment material classification
Image.
Image preprocessing is carried out to the style book image in batch processing sample set, and using gradient descent method to preparatory structure
The shot and long term memory network model parameter built is updated, to complete a model training.
Optionally, first style book image can be concentrated to carry out random image scaling batch training sample, it specifically can be according to pre-
If any one scale value in zoom ranges concentrates style book image to carry out random image scaling, scaling batch training sample
Range can be [0.3,3.0].Then the size of the style book image after scaling is cut to preset standard size, such as can incited somebody to action
It is 256*256 that style book image, which is reduced,.The image block of the style book image after cutting can finally be carried out at selective erasing
Reason, i.e., be set to 0 or other constant values for the pixel value of selection area a certain in style book image, the size of selection area is big
What small and coordinate position was all randomly generated, still, the size of image block is not more than the 1/2 of corresponding style book picture size.
If style book image length is S, width S, then the length value of selection area is no more than S/2, and width value is no more than S/2, choosing
The coordinate position numberical range for determining region can be 0~255.
Judge whether the model training number of garment material identification model reaches preset times threshold value, if so, terminating clothes
Fill the training process of fabric identification model；If it is not, then repeating above-mentioned steps carries out model training, until reaching preset time
Training process is terminated after number threshold value, and saves trained model.
From the foregoing, it will be observed that the embodiment of the present invention is by the training method based on meta learning, and random scaling, selective erasing etc.
Data preprocessing method completes the training of garment material identification model, realizes garment material identification automation, greatly improves clothes
Fill the efficiency of fabric identification work.
The embodiment of the present invention provides corresponding realization device also directed to garment material automatic identifying method, further such that
The method has more practicability.Garment material automatic identification equipment provided in an embodiment of the present invention is introduced below, under
The garment material automatic identification equipment of text description can correspond to each other reference with above-described garment material automatic identifying method.
Referring to Fig. 3, Fig. 3 is garment material automatic identification equipment provided in an embodiment of the present invention in a kind of specific embodiment
Under structure chart, the device can include:
Model prebuild module 301 is obtained for being based on element study method using image pattern collection training neural network model
To garment material identification model；Image pattern collection includes multiple style book images for carrying garment material class label.
Image taking control module 302 meets pre- for the positional relationship when detection microimaging head and clothes to be captured
If shooting condition, image taking instruction is sent, to trigger microimaging head according to pre-set image enlargement ratio automatically to be captured
Clothes carry out image taking.
Fabric identification module 303, for utilizing the clothes constructed in advance when the image of clothing for receiving the acquisition of microimaging head
Dress fabric identification model automatic identification and the fabric classification for exporting image of clothing.
Optionally, in some embodiments of the present embodiment, the model prebuild module 301 can also include:
It criticizes training sample set and generates submodule, for randomly selecting N number of garment material classification, every class from image pattern concentration
Garment material classification includes m style book image, to generate batch training sample set；
Image preprocessing submodule, for carrying out image preprocessing to the style book image in batch processing sample set, and
The shot and long term memory network model parameter constructed in advance is updated using gradient descent method, to complete a model training；
Model training terminates submodule and then terminates garment material for reaching preset times threshold value in model training number
The training process of identification model.
In other embodiments of the present embodiment, described image pretreatment submodule for example can also include:
Image scaling unit, for concentrating style book image to carry out random image scaling batch training sample；
Image cropping unit, for the size of the style book image after scaling to be cut to preset standard size；
Processing unit is wiped, carries out selective erasing processing, image for the image block to the style book image after cutting
The size of block is not more than the 1/2 of corresponding style book picture size.
The function of each functional module of garment material automatic identification equipment described in the embodiment of the present invention can be according to the above method
Method specific implementation in embodiment, specific implementation process are referred to the associated description of above method embodiment, herein not
It repeats again.
From the foregoing, it will be observed that the embodiment of the present invention realizes garment material identification automation, garment material recognition efficiency is improved,
Reduce garment material identification cost.
The embodiment of the invention also provides a kind of garment material automatic recognition systems, refer to Fig. 4, it may include microimaging
First 41, wireless communication components 42 and data processor 43, microimaging head 41 and data processor 43 can groups by wireless communication
Part 42 carries out data communication.
Wherein, microimaging head 41 can carry out the camera of Image Acquisition for any one under different enlargement ratios,
This does not influence the realization of the application.Wireless communication components 42 may be, for example, wifi etc., and those skilled in the art can be according to reality
Application scenarios are selected, and the application does not do any restriction to this.
Data processor 43 can be used for meeting default shooting item when the positional relationship of microimaging head 41 and clothes to be captured
When part send image taking instruction, with trigger microimaging head 41 according to pre-set image enlargement ratio automatically to clothes to be captured into
Row image taking；When the image of clothing for receiving the acquisition of microimaging head 41, the garment material identification model constructed in advance is utilized
Automatic identification and the fabric classification for exporting image of clothing；Garment material identification model may be based on element study method, utilize image
Obtained by sample set training neural network model；Image pattern collection includes multiple image of clothing samples for carrying garment material class label
This.
Optionally, data processor 43 can be deployed in the intelligent mobile terminals such as intelligent terminal, such as mobile phone, plate, may be used also
It is deployed in cloud server etc., this does not influence the realization of the application.Garment material identification model can be deployed in intelligent terminal, example
Such as mobile phone, plate intelligent mobile terminal, can also be deployed in cloud server etc., for the calling of data processor 43, realize clothes
Fill the automatic identification of fabric.In one embodiment, data processor 42 may be provided at intelligent mobile terminal, garment material
Identification model is deployed to cloud server；Intelligent mobile terminal, can when using garment material identification model identification image of clothing
Image recognition instruction is sent to cloud server, so that cloud server utilizes garment material identification model automatic identification clothes figure
Recognition result is simultaneously fed back to intelligent mobile terminal by the fabric classification of picture.
It is realized in a kind of specific embodiment, on mobile intelligent terminal and sends image taking instruction and calling garment surface
Expect to realize when identification model automatic identification function to be mounted in the application program of the mobile intelligent terminal in advance, that is to say, that
Pre- may include image taking module, fabric identification module, user on the displayed page of the application program of mobile intelligent terminal
Corresponding function can be realized by clicking, certainly, which may also include image enlargement ratio selecting unit, automatic right
Burnt unit etc., certainly, those skilled in the art can also increase or decrease functional module according to practical application scene, and the application is to this
Any restriction is not done.
The function of each functional module of garment material automatic recognition system described in the embodiment of the present invention can be according to the above method
Method specific implementation in embodiment, specific implementation process are referred to the associated description of above method embodiment, herein not
It repeats again.
From the foregoing, it will be observed that the embodiment of the present invention realizes garment material identification automation, garment material recognition efficiency is improved,
Reduce garment material identification cost.
The embodiment of the invention also provides a kind of garment material automatic identification equipments, specifically can include:
Memory, for storing computer program；
Processor realizes garment material automatic identification side described in any one embodiment as above for executing computer program
The step of method.
The function of each functional module of garment material automatic identification equipment described in the embodiment of the present invention can be according to the above method
Method specific implementation in embodiment, specific implementation process are referred to the associated description of above method embodiment, herein not
It repeats again.
From the foregoing, it will be observed that the embodiment of the present invention realizes garment material identification automation, garment material recognition efficiency is improved,
Reduce garment material identification cost.
The embodiment of the invention also provides a kind of computer readable storage mediums, are stored with garment material automatic identification journey
Sequence, garment material automatic identification described in any one embodiment as above when the garment material automatic identification procedure is executed by processor
The step of method.
The function of each functional module of computer readable storage medium described in the embodiment of the present invention can be according to above method reality
The method specific implementation in example is applied, specific implementation process is referred to the associated description of above method embodiment, herein no longer
It repeats.
From the foregoing, it will be observed that the embodiment of the present invention realizes garment material identification automation, garment material recognition efficiency is improved,
Reduce garment material identification cost.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
For setting, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part
Explanation.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond the scope of this invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
It above can to a kind of garment material automatic identifying method provided by the present invention, device, system, equipment and computer
Storage medium is read to be described in detail.Specific case used herein explains the principle of the present invention and embodiment
It states, the above description of the embodiment is only used to help understand the method for the present invention and its core ideas.It should be pointed out that for this skill
For the those of ordinary skill in art field, without departing from the principle of the present invention, several change can also be carried out to the present invention
Into and modification, these improvements and modifications also fall within the scope of protection of the claims of the present invention.
Claims (10)
1. a kind of garment material automatic identifying method characterized by comprising
Shooting condition is preset when the positional relationship of detection microimaging head and clothes to be captured meets, sends image taking instruction,
Image taking is carried out to the clothes to be captured automatically to trigger the microimaging head according to pre-set image enlargement ratio；
When the image of clothing for receiving the microimaging head acquisition, known automatically using the garment material identification model constructed in advance
Fabric classification that is other and exporting the image of clothing；
Wherein, the garment material identification model is to utilize image pattern collection training neural network model based on element study method
Gained；Described image sample set includes multiple style book images for carrying garment material class label.
2. garment material automatic identifying method according to claim 1, which is characterized in that the garment material identification model
Training process include:
N number of garment material classification is randomly selected from described image sample set, every class garment material classification includes m style book
Image, to generate batch training sample set；
Image preprocessing is carried out to the style book image in the batch processing sample set, and using gradient descent method to preparatory structure
The shot and long term memory network model parameter built is updated, to complete a model training；
If model training number reaches preset times threshold value, the training process of the garment material identification model is terminated.
3. garment material automatic identifying method according to claim 2, which is characterized in that described to the batch processing sample
The style book image of concentration carries out image preprocessing
Style book image is concentrated to carry out random image scaling described batch of training sample；
The size of style book image after scaling is cut to preset standard size；
Selective erasing processing is carried out to the image block of the style book image after cutting, the size of described image block is no more than corresponding
The 1/2 of style book picture size.
4. garment material automatic identifying method according to claim 3, which is characterized in that described to described batch of training sample
Concentration style book image carries out random image scaling
Style book image is concentrated to carry out random image contracting described batch of training sample according to the scale value within the scope of pre-set zoom
It puts；The zoom ranges are [0.3,3.0].
5. garment material automatic identifying method according to any one of claims 1 to 4, which is characterized in that described image
Shooting instruction includes image enlargement ratio, and described image enlargement ratio is 20,50 or 100.
6. a kind of garment material automatic identification equipment characterized by comprising
Model prebuild module obtains clothes using image pattern collection training neural network model for being based on element study method
Fabric identification model；Described image sample set includes multiple style book images for carrying garment material class label；
Image taking control module meets default shooting item for the positional relationship when detection microimaging head and clothes to be captured
Part sends image taking instruction, to trigger the microimaging head according to pre-set image enlargement ratio automatically to described to be captured
Clothes carry out image taking；
Fabric identification module, for utilizing the clothes constructed in advance when the image of clothing for receiving the microimaging head acquisition
Fabric identification model automatic identification and the fabric classification for exporting the image of clothing.
7. a kind of garment material automatic recognition system characterized by comprising
Microimaging head, wireless communication components and data processor；The microimaging head and the data processor pass through institute
It states wireless communication components and carries out data communication；
The data processor is used to meet default shooting condition when the positional relationship of the microimaging head and clothes to be captured
When send image taking instruction, to trigger the microimaging head according to pre-set image enlargement ratio automatically to the clothes to be captured
Put into row image taking；When the image of clothing for receiving the microimaging head acquisition, known using the garment material constructed in advance
Other model automatic identification and the fabric classification for exporting the image of clothing；
Wherein, the garment material identification model is to utilize image pattern collection training neural network model based on element study method
Gained；Described image sample set includes multiple image of clothing samples for carrying garment material class label.
8. garment material automatic recognition system according to claim 7, which is characterized in that the data processor setting exists
Intelligent mobile terminal；The garment material identification model is deployed to cloud server；
The intelligent mobile terminal sends image recognition instruction to the cloud server, so that the cloud server utilizes institute
It states the fabric classification of image of clothing described in garment material identification model automatic identification and recognition result is fed back into the intelligent sliding
Dynamic terminal.
9. a kind of garment material automatic identification equipment, which is characterized in that including processor, the processor is for executing memory
It is realized when the computer program of middle storage as described in any one of claim 1 to 5 the step of garment material automatic identifying method.
10. a kind of computer readable storage medium, which is characterized in that be stored with garment surface on the computer readable storage medium
Expect automatic identification procedure, realizes when the garment material automatic identification procedure is executed by processor such as any one of claim 1 to 5
The step of garment material automatic identifying method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910603902.7A CN110321850A (en) | 2019-07-05 | 2019-07-05 | Garment material automatic identifying method, device, system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910603902.7A CN110321850A (en) | 2019-07-05 | 2019-07-05 | Garment material automatic identifying method, device, system, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110321850A true CN110321850A (en) | 2019-10-11 |
Family
ID=68122786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910603902.7A Pending CN110321850A (en) | 2019-07-05 | 2019-07-05 | Garment material automatic identifying method, device, system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110321850A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112734383A (en) * | 2021-01-12 | 2021-04-30 | 浙江理工大学 | Method for automatically identifying clothing process |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1914917A (en) * | 2004-01-30 | 2007-02-14 | 株式会社茉莉特斯 | Imaging device, imaging lens, and method for writing data to imaging lens |
| CN101718522A (en) * | 2009-12-25 | 2010-06-02 | 华南农业大学 | Non-contact type nondestructive detecting device for plant growth and detecting method thereof |
| CN102096827A (en) * | 2011-01-18 | 2011-06-15 | 东华大学 | Profiled fiber automatic recognition method based on scale invariability and support vector machine classification |
| CN102741881A (en) * | 2009-12-15 | 2012-10-17 | 七梦科技有限公司 | Textile article identification device and textile article gripping system |
| CN203824920U (en) * | 2014-02-26 | 2014-09-10 | 上海工程技术大学 | Simple and easy shell fabric identification instrument |
| CN104239584A (en) * | 2014-10-15 | 2014-12-24 | 江南大学 | Lace fabric retrieval system and method based on Android |
| CN105092590A (en) * | 2015-08-17 | 2015-11-25 | 大连理工大学 | Test device and test method for observing hole machining outlet |
| CN105785933A (en) * | 2014-12-24 | 2016-07-20 | 佛山市顺德区美的电热电器制造有限公司 | Garment nursing machine, control method and control system thereof |
| CN107358243A (en) * | 2017-07-14 | 2017-11-17 | 深圳码隆科技有限公司 | A kind of method and apparatus of cloth identification |
| CN208121445U (en) * | 2018-03-30 | 2018-11-20 | 昆山保扬新型材料科技有限公司 | A kind of cloth inspecting machine that can be calibrated automatically |
| CN108875777A (en) * | 2018-05-03 | 2018-11-23 | 浙江大学 | Kinds of fibers and blending rate recognition methods in textile fabric based on two-way neural network |
| CN109143560A (en) * | 2018-09-03 | 2019-01-04 | 中国海洋大学 | A kind of automatic zoom device of multiple dimensioned planktonic organism detector lens and its application method |
| CN109344924A (en) * | 2018-09-10 | 2019-02-15 | 湖州易有科技有限公司 | Identification commodity method based on deep learning and the commodity storage based on machine vision and identifying system |
| CN109948577A (en) * | 2019-03-27 | 2019-06-28 | 无锡雪浪数制科技有限公司 | A kind of cloth recognition methods, device and storage medium |
| CN109961022A (en) * | 2019-03-07 | 2019-07-02 | 首都师范大学 | A kind of clothes sketch input fabric material identification analogy method based on deep learning |
-
2019
- 2019-07-05 CN CN201910603902.7A patent/CN110321850A/en active Pending
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1914917A (en) * | 2004-01-30 | 2007-02-14 | 株式会社茉莉特斯 | Imaging device, imaging lens, and method for writing data to imaging lens |
| CN102741881A (en) * | 2009-12-15 | 2012-10-17 | 七梦科技有限公司 | Textile article identification device and textile article gripping system |
| CN101718522A (en) * | 2009-12-25 | 2010-06-02 | 华南农业大学 | Non-contact type nondestructive detecting device for plant growth and detecting method thereof |
| CN102096827A (en) * | 2011-01-18 | 2011-06-15 | 东华大学 | Profiled fiber automatic recognition method based on scale invariability and support vector machine classification |
| CN203824920U (en) * | 2014-02-26 | 2014-09-10 | 上海工程技术大学 | Simple and easy shell fabric identification instrument |
| CN104239584A (en) * | 2014-10-15 | 2014-12-24 | 江南大学 | Lace fabric retrieval system and method based on Android |
| CN105785933A (en) * | 2014-12-24 | 2016-07-20 | 佛山市顺德区美的电热电器制造有限公司 | Garment nursing machine, control method and control system thereof |
| CN105092590A (en) * | 2015-08-17 | 2015-11-25 | 大连理工大学 | Test device and test method for observing hole machining outlet |
| CN107358243A (en) * | 2017-07-14 | 2017-11-17 | 深圳码隆科技有限公司 | A kind of method and apparatus of cloth identification |
| CN208121445U (en) * | 2018-03-30 | 2018-11-20 | 昆山保扬新型材料科技有限公司 | A kind of cloth inspecting machine that can be calibrated automatically |
| CN108875777A (en) * | 2018-05-03 | 2018-11-23 | 浙江大学 | Kinds of fibers and blending rate recognition methods in textile fabric based on two-way neural network |
| CN109143560A (en) * | 2018-09-03 | 2019-01-04 | 中国海洋大学 | A kind of automatic zoom device of multiple dimensioned planktonic organism detector lens and its application method |
| CN109344924A (en) * | 2018-09-10 | 2019-02-15 | 湖州易有科技有限公司 | Identification commodity method based on deep learning and the commodity storage based on machine vision and identifying system |
| CN109961022A (en) * | 2019-03-07 | 2019-07-02 | 首都师范大学 | A kind of clothes sketch input fabric material identification analogy method based on deep learning |
| CN109948577A (en) * | 2019-03-27 | 2019-06-28 | 无锡雪浪数制科技有限公司 | A kind of cloth recognition methods, device and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| 包青平: "基于深度学习的服装图像分类与检索", 《中国优秀硕士学位论文全文数据库 工程科技I辑》 * |
| 张玮: "基于深度学习的织物面料成分分类算法研究", 《中国优秀硕士学位论文全文数据库 工程科技I辑》 * |
| 曾磐: "基于深度学习的智能洗衣机系统构建", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112734383A (en) * | 2021-01-12 | 2021-04-30 | 浙江理工大学 | Method for automatically identifying clothing process |
| CN112734383B (en) * | 2021-01-12 | 2024-04-12 | 浙江理工大学 | Method for automatically identifying clothing technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106251338B (en) | Target integrity detection method and device | |
| CN110378943A (en) | Image processing method, device, electronic equipment and storage medium | |
| CN109325954A (en) | Image partition method, device and electronic equipment | |
| CN109325933A (en) | A kind of reproduction image-recognizing method and device | |
| CN108121957A (en) | The method for pushing and device of U.S. face material | |
| CN108765278A (en) | A kind of image processing method, mobile terminal and computer readable storage medium | |
| CN109191414A (en) | A kind of image processing method, device, electronic equipment and storage medium | |
| CN110163076A (en) | A kind of image processing method and relevant apparatus | |
| CN106663322B (en) | The method and apparatus of feature in image for identification | |
| CN105426455A (en) | Method and device for carrying out classified management on clothes on the basis of picture processing | |
| KR20180066863A (en) | Control method and storage medium | |
| CN107257438A (en) | Take pictures based reminding method, device, terminal and computer-readable storage medium | |
| CN106203286A (en) | The content acquisition method of a kind of augmented reality, device and mobile terminal | |
| CN107172354A (en) | Method for processing video frequency, device, electronic equipment and storage medium | |
| CN108829250A (en) | A kind of object interaction display method based on augmented reality AR | |
| CN111507246A (en) | Method, device, system and storage medium for selecting marked object through gesture | |
| CN110163211A (en) | A kind of image-recognizing method, device and storage medium | |
| CN101533520A (en) | Portrait marking method and device | |
| WO2017125975A1 (en) | Makeup trend analysis device, makeup trend analysis method, and makeup trend analysis program | |
| CN109190617A (en) | A kind of rectangle detection method of image, device and storage medium | |
| CN110136166A (en) | A kind of automatic tracking method of multichannel picture | |
| CN109474785A (en) | The focus of electronic device and electronic device tracks photographic method | |
| CN108985443A (en) | Action identification method and its neural network generation method, device and electronic equipment | |
| CN110084765A (en) | A kind of image processing method, image processing apparatus and terminal device | |
| CN109191585A (en) | Information processing method, device, equipment and computer readable 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191011 |