CN107909107A - Fiber check and measure method, apparatus and electronic equipment - Google Patents

Fiber check and measure method, apparatus and electronic equipment Download PDF

Info

Publication number
CN107909107A
CN107909107A CN201711126617.8A CN201711126617A CN107909107A CN 107909107 A CN107909107 A CN 107909107A CN 201711126617 A CN201711126617 A CN 201711126617A CN 107909107 A CN107909107 A CN 107909107A
Authority
CN
China
Prior art keywords
fiber
deep learning
detected
learning model
fibre
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
Application number
CN201711126617.8A
Other languages
Chinese (zh)
Other versions
CN107909107B (en
Inventor
黄鼎隆
马修·罗伯特·斯科特
傅恺
郭胜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Mailong Intelligent Technology Co ltd
Original Assignee
Shenzhen Malong Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Malong Technologies Co Ltd filed Critical Shenzhen Malong Technologies Co Ltd
Priority to CN201711126617.8A priority Critical patent/CN107909107B/en
Publication of CN107909107A publication Critical patent/CN107909107A/en
Priority to PCT/CN2018/115493 priority patent/WO2019096178A1/en
Application granted granted Critical
Publication of CN107909107B publication Critical patent/CN107909107B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • G06F18/241Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/045Combinations of networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computational Linguistics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Evolutionary Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Image Analysis (AREA)

Abstract

The present invention provides a kind of fiber check and measure method, apparatus and electronic equipment, is related to image identification technical field, wherein, fiber check and measure method includes:Obtain fibre image to be detected;The fiber characteristics of fibre image to be detected are extracted according to predetermined depth learning model, predetermined depth learning model includes a variety of deep learning models based on convolutional neural networks;Fiber characteristics are identified using the grader trained based on predetermined depth learning model, generate the recognition result of fibre image to be detected.The fiber check and measure method can pass through a variety of deep learning models based on convolutional neural networks, independent neutral net is carried out to every kind of fiber and detects identification, so as to when being detected to composite fibre, exclude interference of other fibers to identification, achieve the purpose that to accurately identify fiber.

Description

Fiber check and measure method, apparatus and electronic equipment
Technical field
The present invention relates to image identification technical field, more particularly, to a kind of fiber check and measure method, apparatus and electronic equipment.
Background technology
With the sustained and rapid development of China's economy, booming situation is presented in all trades and professions.It is it follows that each Material of the kind comprising fibre structure is increasingly abundanter, its species, fiber composition, form etc. are multifarious.Current fiber Detection identification method is carried out based on artificial mode, it usually needs collection image, is then observed above-mentioned image, According to existing experience and knowledge, the fiber type in the image is determined, calculate its fibre diameter and quantity etc..This process is time-consuming to take Power, and inefficiency, for the material comprising mixed fiber structure, it is identified difficult.
The content of the invention
In view of this, it is an object of the invention to provide a kind of fiber check and measure method, apparatus and electronic equipment, can pass through A variety of deep learning models based on convolutional neural networks, carry out independent neutral net to every kind of fiber and detect identification, so that When being detected to composite fibre, interference of other fibers to identification is excluded, achievees the purpose that to accurately identify fiber.
In a first aspect, an embodiment of the present invention provides a kind of fiber check and measure method, including:
Obtain fibre image to be detected;
The fiber characteristics of fibre image to be detected are extracted according to predetermined deep learning model, predetermined deep learning model includes A variety of deep learning models based on convolutional neural networks;
Fiber characteristics are identified using the grader trained based on predetermined deep learning model, generate fiber to be detected The recognition result of image.
With reference to first aspect, an embodiment of the present invention provides the first possible embodiment of first aspect, wherein, obtain Fibre image to be detected is taken, is specifically included:
Obtain electron microscope and acquired image of taking pictures is carried out to target detection thing, obtain fibre image to be detected;Mesh Mark detectable substance includes fibre structure.
With reference to first aspect, an embodiment of the present invention provides second of possible embodiment of first aspect, wherein, it is fine Dimensional feature includes:Fiber type, fibre diameter, fiber number.
With reference to first aspect, an embodiment of the present invention provides the third possible embodiment of first aspect, wherein, base It is that the fiber samples data for exceeding certain threshold value by quantity are trained in the deep learning model of convolutional neural networks, it is fine Dimension sample data includes the corresponding picture of different types of fiber.
With reference to first aspect, an embodiment of the present invention provides the 4th kind of possible embodiment of first aspect, wherein, lead to Cross in the following manner and obtain the grader based on the training of predetermined deep learning model:
Utilize the further feature of a variety of deep learning model extraction fiber samples data based on convolutional neural networks;
Based on machine learning algorithm, grader is trained to further feature;
Wherein fiber samples data include fiber characteristics, matching degree and recognition result.
With reference to first aspect, an embodiment of the present invention provides the 5th kind of possible embodiment of first aspect, wherein, profit Fiber characteristics are identified with the grader trained based on predetermined deep learning model, generate the identification of fibre image to be detected As a result, specifically include:
Determine the data format corresponding to the grader based on the training of predetermined deep learning model;
If data format includes binary data format, base is inputted after fiber characteristics are converted to binary data format In the grader of predetermined deep learning model training, with the recognition result of the corresponding fibre image to be detected of generation.
With reference to first aspect, an embodiment of the present invention provides the 6th kind of possible embodiment of first aspect, wherein, Fiber characteristics are identified using the grader trained based on predetermined deep learning model, generate the knowledge of fibre image to be detected After other result, further include:
It is stored in the recognition result of fibre image to be detected as new fiber samples data in fiber samples database.
Second aspect, the embodiment of the present invention provide a kind of fiber detection device, and device includes:
Image collection module, for obtaining fibre image to be detected;
Characteristic extracting module, for extracting the fiber characteristics of fibre image to be detected according to predetermined deep learning model, in advance If deep learning model includes a variety of deep learning models based on convolutional neural networks;
Fiber recognition module, for being known using the grader trained based on predetermined deep learning model to fiber characteristics Not, the recognition result of fibre image to be detected is generated.
The third aspect, the embodiment of the present invention provide a kind of electronic equipment, including memory, processor, are stored on memory There is the computer program that can be run on a processor, processor realizes the method described in first aspect when performing computer program Step.
Fourth aspect, the embodiment of the present invention also provide a kind of meter for the non-volatile program code that can perform with processor Calculation machine computer-readable recording medium, program code make processor perform the method described in first aspect.
The embodiment of the present invention brings following beneficial effect:
In fiber check and measure method provided in an embodiment of the present invention, fibre image to be detected is obtained first, then according to pre- If the fiber characteristics of deep learning model extraction fibre image to be detected, wherein, predetermined deep learning model includes a variety of be based on The deep learning model of convolutional neural networks, finally utilizes the grader based on the training of predetermined deep learning model to fiber characteristics It is identified, generates the recognition result of fibre image to be detected.The fiber check and measure method can be based on convolutional Neural by a variety of The deep learning model of network, carries out independent neutral net to every kind of fiber and detects identification, so as to be detected to composite fibre When, interference of other fibers to identification is excluded, achievees the purpose that to accurately identify fiber.
Other features and advantages of the present invention will illustrate in the following description, also, partly become from specification Obtain it is clear that or being understood by implementing the present invention.The purpose of the present invention and other advantages are in specification, claims And specifically noted structure is realized and obtained in attached drawing.
To enable the above objects, features and advantages of the present invention to become apparent, preferred embodiment cited below particularly, and coordinate Appended attached drawing, is described in detail below.
Brief description of the drawings
, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution of the prior art Embodiment or attached drawing needed to be used in the description of the prior art are briefly described, it should be apparent that, in describing below Attached drawing is some embodiments of the present invention, for those of ordinary skill in the art, before not making the creative labor Put, other attached drawings can also be obtained according to these attached drawings.
Fig. 1 is a kind of flow chart for fiber check and measure method that the embodiment of the present invention one provides;
Fig. 2 is the flow chart for another fiber check and measure method that the embodiment of the present invention one provides;
Fig. 3 is the flow chart for another fiber check and measure method that the embodiment of the present invention one provides;
Fig. 4 is the flow chart for another fiber check and measure method that the embodiment of the present invention one provides;
Fig. 5 is a kind of structure diagram of fiber detection device provided by Embodiment 2 of the present invention;
Fig. 6 is the structure diagram for a kind of electronic equipment that the embodiment of the present invention three provides.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with attached drawing to the present invention Technical solution be clearly and completely described, it is clear that described embodiment is part of the embodiment 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 Lower all other embodiments obtained, belong to the scope of protection of the invention.
Current existing fiber check and measure identification method is carried out based on artificial mode, not only time-consuming and laborious, but also Inefficiency, for the material comprising mixed fiber structure, it is identified difficult.It is of the invention based on this The fiber check and measure method, apparatus and electronic equipment that embodiment provides, can pass through a variety of depth based on convolutional neural networks Model is practised, carrying out independent neutral net to every kind of fiber detects identification, so that when being detected to composite fibre, it is fine to exclude other The interference to identification is tieed up, achievees the purpose that to accurately identify fiber.
For ease of understanding the present embodiment, first to a kind of fiber check and measure method disclosed in the embodiment of the present invention into Row is discussed in detail.
Embodiment one:
An embodiment of the present invention provides a kind of fiber check and measure method, this method can be applied to the fiber check and measure of multiple fields In scene, such as:The fabric fibre in garment production field detects, plant fiber detects etc..Shown in Figure 1, this method includes:
S101:Obtain fibre image to be detected.
Specifically, obtain electron microscope carries out acquired image of taking pictures to target detection thing, fiber to be detected is obtained Image.Wherein, target detection thing is the material comprising fibre structure, such as cloth, high polymer composite material, plant specimen Deng.Electron microscope can be different type, the microscope of different size, and skilled addressee is treated using electron microscope collection The fibre image to be detected of detectable substance.
S102:The fiber characteristics of fibre image to be detected, predetermined depth study mould are extracted according to predetermined deep learning model Type includes a variety of deep learning models based on convolutional neural networks.
Deep learning model based on convolutional neural networks is that the fiber samples data for exceeding certain threshold value by quantity are instructed Get, fiber samples data include the corresponding picture of different types of fiber.
In one preferred embodiment, above-mentioned a variety of deep learning models based on convolutional neural networks can pass through Caffe deep learnings frame is realized.
Specifically, the quantity of the fibre image in fiber samples data is The more the better, and species is The more the better, and data are more, The versatility of the deep learning model based on convolutional neural networks of training generation is better.The embodiment of the present invention using fabric fibre as Example, above-mentioned fibre image includes fiber type, fibre diameter and the different multiple fiber sample data of fiber number, so favourable In subsequently being accurately identified to fibre image to be detected, the identification energy of the deep learning model based on convolutional neural networks is improved Power, even if fiber to be detected is composite fibre, can also identify it exactly.
Step S102 is specifically included:Wrapped using fibre image to be detected as input picture in predetermined deep learning model Carry out feature training successively in the multiple basic units contained, after the completion of training, extract it is multiple it is integrated in full articulamentum or other The feature vector of basic unit's output is specified as corresponding fiber characteristics in the fibre image to be detected.Wherein, fiber characteristics include: Fiber type, fibre diameter, fiber number.
S103:Fiber characteristics are identified using the grader trained based on predetermined deep learning model, generation is to be checked Survey the recognition result of fibre image.
By grader of the fiber characteristics input based on the training of predetermined deep learning model of said extracted, pass through the grader After identification, final recognition result is obtained.Specifically, recognition result is fiber type, fibre diameter and fiber number.
In an optional embodiment, the classification based on the training of predetermined deep learning model is obtained in the following manner Device, it is shown in Figure 2:
S201:It is special using the deep layer of a variety of deep learning model extraction fiber samples data based on convolutional neural networks Sign.
S202:Based on machine learning algorithm, grader is trained to further feature.
Wherein fiber samples data include fiber characteristics, matching degree and recognition result.Above-mentioned machine learning algorithm can be with It is nearest neighbor algorithm, EM algorithm and algorithm of support vector machine etc., specific algorithm can select as the case may be, here not It is construed as limiting.
In an optional embodiment, above-mentioned fiber sample data include triple data;The wherein triple data Including:Source data, belong to the source data same category of forward data and adhered to separately with the source data different classes of anti- To data.
Wherein, source data is the identical sample data of the recognition result got at random from fiber sample data.It is positive Data are the sample data consistent with the recognition result of source data obtained at random from fiber samples data;Of the source data Matching degree with degree higher than forward data.Reverse data is the identification knot with source data obtained at random from fiber samples data The inconsistent sample data of fruit.
In a specific embodiment, triple data are respectively:Fibre image performance is good in fiber samples data The first good picture, the second picture of fibre image poor-performing in fiber samples data, and with the first picture and the second figure The 3rd different picture of piece recognition result.Specific first picture is the highest source data of matching degree, and second picture is clear There are gap, its matching degree be less than the first picture, but still be fibre image for degree, resolution ratio etc. and the first picture, to be positive Data.The reverse data that it is progress reverse contrast in training that 3rd picture, which is then, with secondary by positive and negative contrast, further enhancing The recognition capability of grader, improves neutral net fiber check and measure identification accuracy.
Fiber characteristics are identified using the grader trained based on predetermined deep learning model, generate fiber to be detected The recognition result of image, specifically includes following steps, shown in Figure 3:
S301:Determine the data format corresponding to the grader based on the training of predetermined deep learning model.
S302:If data format includes binary data format, after fiber characteristics are converted to binary data format The grader based on the training of predetermined deep learning model is inputted, with the recognition result of the corresponding fibre image to be detected of generation.
Specifically, if the grader of predetermined deep learning model training supports binary data format, then just by fiber Feature is changed, and is converted to binary data format, then is input in the grader of above-mentioned training, is identified, due to Machine language is binary system, therefore by binary data format, can accelerate the process identified, when being identified, no Need to carry out extra data conversion again, state this can improve the efficiency of identification.
In addition, after the recognition result of fibre image to be detected is generated, it is further comprising the steps of, it is shown in Figure 4:
S401:Fiber samples data are stored in using the recognition result of fibre image to be detected as new fiber samples data In storehouse.
Pass through the data being continuously updated in fiber samples database so that the classification gone out based on deep learning model training The fiber recognition effect of device is more accurate.
In fiber check and measure method provided in an embodiment of the present invention, fibre image to be detected is obtained first, then according to pre- If the fiber characteristics of deep learning model extraction fibre image to be detected, wherein, predetermined deep learning model includes a variety of be based on The deep learning model of convolutional neural networks, finally utilizes the grader based on the training of predetermined deep learning model to fiber characteristics It is identified, generates the recognition result of fibre image to be detected.The fiber check and measure method can be based on convolutional Neural by a variety of The deep learning model of network, carries out independent neutral net to every kind of fiber and detects identification, so as to be detected to composite fibre When, interference of other fibers to identification is excluded, achievees the purpose that to accurately identify fiber.
Embodiment two:
The embodiment of the present invention provides a kind of fiber detection device, and shown in Figure 5, which includes:Image collection module 51st, characteristic extracting module 52, fiber recognition module 53.
Wherein, image collection module 51, for obtaining fibre image to be detected;Characteristic extracting module 52, for according to pre- If the fiber characteristics of deep learning model extraction fibre image to be detected, predetermined deep learning model includes a variety of based on convolution god Deep learning model through network;Fiber recognition module 53, for utilizing the grader based on the training of predetermined deep learning model Fiber characteristics are identified, generate the recognition result of fibre image to be detected.
In the fiber detection device that the embodiment of the present invention is provided, the course of work of modules and foregoing fiber check and measure side Method has identical technical characteristic, therefore, can equally realize above-mentioned function, details are not described herein.
Embodiment three:
The embodiment of the present invention also provides a kind of electronic equipment, shown in Figure 6, which includes:Processor 60, is deposited Reservoir 61, bus 62 and communication interface 63, the processor 60, communication interface 63 and memory 61 are connected by bus 62;Place Reason device 60 is used to perform the executable module stored in memory 61, such as computer program.Processor performs computer program The step of methods of the Shi Shixian as described in embodiment of the method.
Wherein, memory 61 may include high-speed random access memory (RAM, RandomAccessMemory), also may be used Non-labile memory (non-volatile memory), for example, at least a magnetic disk storage can be further included.By at least One communication interface 63 (can be wired or wireless) realizes the communication between the system network element and at least one other network element Connection, can use internet, wide area network, local network, Metropolitan Area Network (MAN) etc..
Bus 62 can be isa bus, pci bus or eisa bus etc..The bus can be divided into address bus, data Bus, controlling bus etc..Only represented for ease of representing, in Fig. 6 with a four-headed arrow, it is not intended that an only bus or A type of bus.
Wherein, memory 61 is used for storage program, and the processor 60 performs the journey after execute instruction is received Sequence, the method performed by device that the stream process that foregoing any embodiment of the embodiment of the present invention discloses defines can be applied to handle In device 60, or realized by processor 60.
Processor 60 is probably a kind of IC chip, has the disposal ability of signal.During realization, above-mentioned side Each step of method can be completed by the integrated logic circuit of the hardware in processor 60 or the instruction of software form.Above-mentioned Processor 60 can be general processor, including central processing unit (Central Processing Unit, abbreviation CPU), network Processor (Network Processor, abbreviation NP) etc.;It can also be digital signal processor (Digital Signal Processing, abbreviation DSP), application-specific integrated circuit (Application Specific Integrated Circuit, referred to as ASIC), ready-made programmable gate array (Field-Programmable Gate Array, abbreviation FPGA) or other are programmable Logical device, discrete gate or transistor logic, discrete hardware components.It can realize or perform in the embodiment of the present invention Disclosed each method, step and logic diagram.General processor can be microprocessor or the processor can also be appointed What conventional processor etc..The step of method with reference to disclosed in the embodiment of the present invention, can be embodied directly in hardware decoding processing Device performs completion, or performs completion with the hardware in decoding processor and software module combination.Software module can be located at Machine memory, flash memory, read-only storage, programmable read only memory or electrically erasable programmable memory, register etc. are originally In the storage medium of field maturation.The storage medium is located at memory 61, and processor 60 reads the information in memory 61, with reference to Its hardware completes the step of above method.
The computer program product of the localization method for the network equipment that the embodiment of the present invention is provided, including store processing The computer-readable recording medium for the non-volatile program code that device can perform, the instruction that said program code includes can be used for holding Method described in row previous methods embodiment, specific implementation can be found in embodiment of the method, and details are not described herein.
It is apparent to those skilled in the art that for convenience and simplicity of description, the device of foregoing description And the specific work process of electronic equipment, the corresponding process in preceding method embodiment is may be referred to, details are not described herein.
Flow chart and block diagram in attached drawing show multiple embodiment method and computer program products according to the present invention Architectural framework in the cards, function and operation.At this point, each square frame in flow chart or block diagram can represent one A part for module, program segment or code, a part for the module, program segment or code are used for realization comprising one or more The executable instruction of defined logic function.It should also be noted that at some as the work(in the realization replaced, marked in square frame Energy can also be with different from the order marked in attached drawing generation.For example, two continuous square frames can essentially be substantially parallel Ground performs, they can also be performed in the opposite order sometimes, this is depending on involved function.It is also noted that block diagram And/or the combination of each square frame and block diagram in flow chart and/or the square frame in flow chart, work(as defined in performing can be used Can or the dedicated hardware based system of action realize, or reality can be carried out with the combination of specialized hardware and computer instruction It is existing.
In the description of the present invention, it is necessary to explanation, term " " center ", " on ", " under ", "left", "right", " vertical ", The orientation or position relationship of the instruction such as " level ", " interior ", " outer " be based on orientation shown in the drawings or position relationship, merely to Easy to describe the present invention and simplify description, rather than instruction or imply signified device or element must have specific orientation, With specific azimuth configuration and operation, therefore it is not considered as limiting the invention.In addition, term " first ", " second ", " the 3rd " is only used for description purpose, and it is not intended that instruction or hint relative importance.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, can be with Realize by another way.Device embodiment described above is only schematical, for example, the division of the unit, Only a kind of division of logic function, can there is other dividing mode when actually realizing, in another example, multiple units or component can To combine or be desirably integrated into another system, or some features can be ignored, or not perform.It is another, it is shown or beg for The mutual coupling, direct-coupling or communication connection of opinion can be by some communication interfaces, device or unit it is indirect Coupling or communication connection, can be electrical, machinery or other forms.
The unit illustrated as separating component may or may not be physically separate, be shown as unit The component shown may or may not be physical location, you can with positioned at a place, or can also be distributed to multiple In network unit.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs 's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, can also That unit is individually physically present, can also two or more units integrate in a unit.
If the function is realized in the form of SFU software functional unit and is used as independent production marketing or in use, can be with It is stored in the non-volatile computer read/write memory medium that a processor can perform.Based on such understanding, the present invention The part that substantially contributes in other words to the prior art of technical solution or the part of the technical solution can be with software The form of product embodies, which is stored in a storage medium, including some instructions use so that One computer equipment (can be personal computer, server, or network equipment etc.) performs each embodiment institute of the present invention State all or part of step of method.And foregoing storage medium includes:USB flash disk, mobile hard disk, read-only storage (ROM, Read- Only Memory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with The medium of store program codes.
Finally it should be noted that:Embodiment described above, is only the embodiment of the present invention, to illustrate the present invention Technical solution, rather than its limitations, protection scope of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair It is bright to be described in detail, it will be understood by those of ordinary skill in the art that:Any one skilled in the art The invention discloses technical scope in, it can still modify the technical solution described in previous embodiment or can be light It is readily conceivable that change, or equivalent substitution is carried out to which part technical characteristic;And these modifications, change or replacement, do not make The essence of appropriate technical solution departs from the spirit and scope of technical solution of the embodiment of the present invention, should all cover the protection in the present invention Within the scope of.Therefore, protection scope of the present invention answers the scope of the claims of being subject to.

Claims (10)

  1. A kind of 1. fiber check and measure method, it is characterised in that including:
    Obtain fibre image to be detected;
    The fiber characteristics of the fibre image to be detected, the predetermined deep learning model are extracted according to predetermined deep learning model Including a variety of deep learning models based on convolutional neural networks;
    The fiber characteristics are identified using the grader trained based on the predetermined deep learning model, are treated described in generation Detect the recognition result of fibre image.
  2. 2. according to the method described in claim 1, it is characterized in that, the acquisition fibre image to be detected, specifically includes:
    Obtain electron microscope and acquired image of taking pictures is carried out to target detection thing, obtain the fibre image to be detected;Institute State target detection thing and include fibre structure.
  3. 3. according to the method described in claim 1, it is characterized in that, the fiber characteristics include:Fiber type, fibre diameter, Fiber number.
  4. 4. according to the method described in claim 1, it is characterized in that, the deep learning model based on convolutional neural networks is Trained by quantity more than the fiber samples data of certain threshold value, the fiber samples data include different types of fibre Tie up corresponding picture.
  5. 5. according to the method described in claim 1, it is characterized in that, it is based on the predetermined depth described in obtaining in the following manner The grader of learning model training:
    Utilize the further feature of the deep learning model extraction fiber samples data based on convolutional neural networks;
    Based on machine learning algorithm, grader is trained to the further feature;
    Wherein described fiber samples data include fiber characteristics, matching degree and recognition result.
  6. 6. according to the method described in claim 1, it is characterized in that, described trained using based on the predetermined deep learning model Grader the fiber characteristics are identified, generate the recognition result of the fibre image to be detected, specifically include:
    Determine the data format corresponding to the grader based on predetermined deep learning model training;
    If the data format includes binary data format, the fiber characteristics are converted to defeated after binary data format Enter the grader based on predetermined deep learning model training, with the identification of the corresponding fibre image to be detected of generation As a result.
  7. 7. according to claim 1-6 any one of them methods, it is characterised in that be based on the predetermined depth in described utilize The fiber characteristics are identified in the grader for practising model training, generate the fibre image to be detected recognition result it Afterwards, further include:
    It is stored in the recognition result of the fibre image to be detected as new fiber samples data in fiber samples database.
  8. 8. a kind of fiber detection device, it is characterised in that described device includes:
    Image collection module, for obtaining fibre image to be detected;
    Characteristic extracting module, for extracting the fiber characteristics of the fibre image to be detected, institute according to predetermined deep learning model Stating predetermined deep learning model includes a variety of deep learning models based on convolutional neural networks;
    Fiber recognition module, for using based on the predetermined deep learning model training grader to the fiber characteristics into Row identification, generates the recognition result of the fibre image to be detected.
  9. 9. a kind of electronic equipment, including memory, processor, it is stored with what can be run on the processor on the memory Computer program, it is characterised in that the processor realizes that the claims 1 to 7 are any when performing the computer program Described in method the step of.
  10. 10. a kind of computer-readable medium for the non-volatile program code that can perform with processor, it is characterised in that described Program code makes the processor perform claim 1 to 7 any one of them method.
CN201711126617.8A 2017-11-14 2017-11-14 Fiber detection method and device and electronic equipment Active CN107909107B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201711126617.8A CN107909107B (en) 2017-11-14 2017-11-14 Fiber detection method and device and electronic equipment
PCT/CN2018/115493 WO2019096178A1 (en) 2017-11-14 2018-11-14 Fiber detection method and apparatus, and electronic device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711126617.8A CN107909107B (en) 2017-11-14 2017-11-14 Fiber detection method and device and electronic equipment

Publications (2)

Publication Number Publication Date
CN107909107A true CN107909107A (en) 2018-04-13
CN107909107B CN107909107B (en) 2020-09-15

Family

ID=61845585

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711126617.8A Active CN107909107B (en) 2017-11-14 2017-11-14 Fiber detection method and device and electronic equipment

Country Status (2)

Country Link
CN (1) CN107909107B (en)
WO (1) WO2019096178A1 (en)

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108563906A (en) * 2018-05-02 2018-09-21 北京航空航天大学 A kind of short fiber reinforced composite macro property prediction technique based on deep learning
CN109460471A (en) * 2018-11-01 2019-03-12 信融源大数据科技(北京)有限公司 A method of the mode based on self study establishes kinds of fibers spectrum library
CN109492544A (en) * 2018-10-19 2019-03-19 兰波(苏州)智能科技有限公司 A method of classified by enhancing optical microscopy to animal origin
CN109543532A (en) * 2018-10-19 2019-03-29 兰波(苏州)智能科技有限公司 A method of the Fibre sorting based on classification standard and human behavior
CN109583564A (en) * 2018-10-31 2019-04-05 东华大学 Extremely similar animal origin automatic identifying method based on VGG convolutional neural networks
WO2019096178A1 (en) * 2017-11-14 2019-05-23 深圳码隆科技有限公司 Fiber detection method and apparatus, and electronic device
CN110749554A (en) * 2018-07-23 2020-02-04 波音公司 Characterization of the ratio of melted tissue strands in a layer of fibrous material
CN110924052A (en) * 2018-09-20 2020-03-27 格力电器(武汉)有限公司 Washing machine control method and device
WO2020113694A1 (en) * 2018-12-04 2020-06-11 深圳码隆科技有限公司 Spinning machine fault monitoring method and apparatus, and server
CN111665244A (en) * 2019-03-05 2020-09-15 温力力 Method based on textile fiber identification and component detection system
CN111665243A (en) * 2019-03-05 2020-09-15 温力力 Textile fiber identification and component detection system
CN111860510A (en) * 2020-07-29 2020-10-30 浙江大华技术股份有限公司 X-ray image target detection method and device
WO2020236446A1 (en) * 2019-05-17 2020-11-26 Corning Incorporated Predicting optical fiber manufacturing performance using neural network
CN112241713A (en) * 2020-10-22 2021-01-19 江苏美克医学技术有限公司 Vaginal microorganism identification method and device based on pattern identification and deep learning
CN112489019A (en) * 2020-12-01 2021-03-12 合肥工业大学 Method for rapidly identifying chopped fibers in GFRC image based on deep learning
CN112529829A (en) * 2019-08-28 2021-03-19 银河水滴科技(北京)有限公司 Training method and device for burr positioning and burr detection model
CN112669271A (en) * 2020-12-22 2021-04-16 深圳酷派技术有限公司 Object surface defect detection method, related device and computer storage medium
CN113255790A (en) * 2021-05-31 2021-08-13 创新奇智(北京)科技有限公司 Textile fiber component identification method, electronic device and storage medium
CN113640326A (en) * 2021-08-18 2021-11-12 华东理工大学 Multistage mapping reconstruction method for nano-pore resin-based composite material micro-nano structure

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112446392A (en) * 2019-08-28 2021-03-05 银河水滴科技(北京)有限公司 Training of sealing ring classification model and classification method and device of sealing ring
CN111696064B (en) * 2020-06-15 2023-08-18 北京金山云网络技术有限公司 Image processing method, device, electronic equipment and computer readable medium
CN113210264B (en) * 2021-05-19 2023-09-05 江苏鑫源烟草薄片有限公司 Tobacco sundry removing method and device
CN113936277A (en) * 2021-09-14 2022-01-14 北京焦耳科技有限公司 Double-mode identification method, system and medium for handwritten name and school number
CN116796159B (en) * 2023-08-17 2023-11-17 浙江恒逸石化有限公司 Dyeing effect prediction method, training method and device of dyeing effect prediction model

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05116967A (en) * 1991-10-23 1993-05-14 Furukawa Electric Co Ltd:The Production of optical fiber base material
CN101096819A (en) * 2006-06-27 2008-01-02 中国纺织科学研究院 Organization discrimination method of fabrics
CN101178396A (en) * 2007-11-29 2008-05-14 东华大学 Cashmere fibre, fleece fibre blended spinning quantitative determination instrument and method
CN101424680A (en) * 2008-12-11 2009-05-06 东华大学 Computer automatic recognition apparatus and method for profile fiber
US20130045877A1 (en) * 2011-08-19 2013-02-21 Agency For Science, Technology And Research Methods to form substrates for optical sensing by surface enhanced raman spectroscopy (sers) and substrates formed by the methods
CN103149210A (en) * 2013-02-25 2013-06-12 东华大学 System and method for detecting fabric cashmere content based on scale graphic features
CN105654123A (en) * 2015-12-26 2016-06-08 河北工业大学 Identification method of woven fabric tissue chart
CN105760877A (en) * 2016-02-19 2016-07-13 天纺标检测科技有限公司 Wool and cashmere identification algorithm based on gray level co-occurrence matrix model

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9436895B1 (en) * 2015-04-03 2016-09-06 Mitsubishi Electric Research Laboratories, Inc. Method for determining similarity of objects represented in images
CN106295629B (en) * 2016-07-15 2018-06-15 北京市商汤科技开发有限公司 structured text detection method and system
CN106778685A (en) * 2017-01-12 2017-05-31 司马大大(北京)智能系统有限公司 Electrocardiogram image-recognizing method, device and service terminal
CN106960219B (en) * 2017-03-10 2021-04-16 百度在线网络技术(北京)有限公司 Picture identification method and device, computer equipment and computer readable medium
CN107909107B (en) * 2017-11-14 2020-09-15 深圳码隆科技有限公司 Fiber detection method and device and electronic equipment

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05116967A (en) * 1991-10-23 1993-05-14 Furukawa Electric Co Ltd:The Production of optical fiber base material
CN101096819A (en) * 2006-06-27 2008-01-02 中国纺织科学研究院 Organization discrimination method of fabrics
CN101178396A (en) * 2007-11-29 2008-05-14 东华大学 Cashmere fibre, fleece fibre blended spinning quantitative determination instrument and method
CN101424680A (en) * 2008-12-11 2009-05-06 东华大学 Computer automatic recognition apparatus and method for profile fiber
US20130045877A1 (en) * 2011-08-19 2013-02-21 Agency For Science, Technology And Research Methods to form substrates for optical sensing by surface enhanced raman spectroscopy (sers) and substrates formed by the methods
CN103149210A (en) * 2013-02-25 2013-06-12 东华大学 System and method for detecting fabric cashmere content based on scale graphic features
CN105654123A (en) * 2015-12-26 2016-06-08 河北工业大学 Identification method of woven fabric tissue chart
CN105760877A (en) * 2016-02-19 2016-07-13 天纺标检测科技有限公司 Wool and cashmere identification algorithm based on gray level co-occurrence matrix model

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
F.H. SHE: "Identification and classification of animal fibres using artificial neural networks", 《JOURNAL OF TEXILE ENGINEERING》 *
何鹏程: "改进的卷积神经网络模型及其应用研究", 《中国优秀硕士学位论文全文数据库-信息科技辑》 *
孙硕磊: "卷积神经网络在异型纤维识别中的研究和应用", 《中国优秀硕士学位论文全文数据库-信息科技辑》 *

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019096178A1 (en) * 2017-11-14 2019-05-23 深圳码隆科技有限公司 Fiber detection method and apparatus, and electronic device
CN108563906B (en) * 2018-05-02 2022-03-22 北京航空航天大学 Short fiber reinforced composite material macroscopic performance prediction method based on deep learning
CN108563906A (en) * 2018-05-02 2018-09-21 北京航空航天大学 A kind of short fiber reinforced composite macro property prediction technique based on deep learning
CN110749554A (en) * 2018-07-23 2020-02-04 波音公司 Characterization of the ratio of melted tissue strands in a layer of fibrous material
CN110924052A (en) * 2018-09-20 2020-03-27 格力电器(武汉)有限公司 Washing machine control method and device
CN109543532B (en) * 2018-10-19 2023-07-28 兰波(苏州)智能科技有限公司 Fiber classification method based on classification standard and human behavior
CN109543532A (en) * 2018-10-19 2019-03-29 兰波(苏州)智能科技有限公司 A method of the Fibre sorting based on classification standard and human behavior
CN109492544A (en) * 2018-10-19 2019-03-19 兰波(苏州)智能科技有限公司 A method of classified by enhancing optical microscopy to animal origin
CN109492544B (en) * 2018-10-19 2023-01-03 兰波(苏州)智能科技有限公司 Method for classifying animal fibers through enhanced optical microscope
CN109583564A (en) * 2018-10-31 2019-04-05 东华大学 Extremely similar animal origin automatic identifying method based on VGG convolutional neural networks
CN109460471A (en) * 2018-11-01 2019-03-12 信融源大数据科技(北京)有限公司 A method of the mode based on self study establishes kinds of fibers spectrum library
CN109460471B (en) * 2018-11-01 2021-09-24 信融源大数据科技(北京)有限公司 Method for establishing fiber category map library based on self-learning mode
WO2020113694A1 (en) * 2018-12-04 2020-06-11 深圳码隆科技有限公司 Spinning machine fault monitoring method and apparatus, and server
CN111665244B (en) * 2019-03-05 2024-03-19 温力力 Method based on textile fiber identification and component detection system
CN111665243B (en) * 2019-03-05 2024-03-19 温力力 Textile fiber identification and component detection system
CN111665243A (en) * 2019-03-05 2020-09-15 温力力 Textile fiber identification and component detection system
CN111665244A (en) * 2019-03-05 2020-09-15 温力力 Method based on textile fiber identification and component detection system
WO2020236446A1 (en) * 2019-05-17 2020-11-26 Corning Incorporated Predicting optical fiber manufacturing performance using neural network
CN112529829B (en) * 2019-08-28 2024-04-19 银河水滴科技(北京)有限公司 Training method and device for burr positioning and burr detection model
CN112529829A (en) * 2019-08-28 2021-03-19 银河水滴科技(北京)有限公司 Training method and device for burr positioning and burr detection model
CN111860510A (en) * 2020-07-29 2020-10-30 浙江大华技术股份有限公司 X-ray image target detection method and device
CN111860510B (en) * 2020-07-29 2021-06-18 浙江大华技术股份有限公司 X-ray image target detection method and device
CN112241713B (en) * 2020-10-22 2023-12-29 江苏美克医学技术有限公司 Method and device for identifying vaginal microorganisms based on pattern recognition and deep learning
CN112241713A (en) * 2020-10-22 2021-01-19 江苏美克医学技术有限公司 Vaginal microorganism identification method and device based on pattern identification and deep learning
CN112489019A (en) * 2020-12-01 2021-03-12 合肥工业大学 Method for rapidly identifying chopped fibers in GFRC image based on deep learning
CN112669271A (en) * 2020-12-22 2021-04-16 深圳酷派技术有限公司 Object surface defect detection method, related device and computer storage medium
CN113255790A (en) * 2021-05-31 2021-08-13 创新奇智(北京)科技有限公司 Textile fiber component identification method, electronic device and storage medium
CN113640326A (en) * 2021-08-18 2021-11-12 华东理工大学 Multistage mapping reconstruction method for nano-pore resin-based composite material micro-nano structure
CN113640326B (en) * 2021-08-18 2023-10-10 华东理工大学 Multistage mapping reconstruction method for micro-nano structure of nano-porous resin matrix composite material

Also Published As

Publication number Publication date
CN107909107B (en) 2020-09-15
WO2019096178A1 (en) 2019-05-23

Similar Documents

Publication Publication Date Title
CN107909107A (en) Fiber check and measure method, apparatus and electronic equipment
Li et al. An effective data augmentation strategy for CNN-based pest localization and recognition in the field
CN107845092A (en) Clothes LOGO effect detection methods, device and electronic equipment
CN106951925A (en) Data processing method, device, server and system
CN109446889A (en) Object tracking method and device based on twin matching network
CN110084374A (en) Construct method, apparatus and prediction technique, device based on the PU model learnt
CN107220277A (en) Image retrieval algorithm based on cartographical sketching
CN105513066B (en) It is a kind of that the generic object detection method merged with super-pixel is chosen based on seed point
CN111353580B (en) Training method of target detection network, electronic equipment and storage medium
CN105608459A (en) Commodity image segmentation method and commodity image segmentation device
CN110246579B (en) Pathological diagnosis method and device
CN109740752A (en) Depth model training method and device, electronic equipment and storage medium
CN102316352A (en) Stereo video depth image manufacturing method based on area communication image and apparatus thereof
CN110796135A (en) Target positioning method and device, computer equipment and computer storage medium
CN109410290A (en) The method and apparatus for determining operating area boundary
CN112819796A (en) Tobacco shred foreign matter identification method and equipment
CN109857878A (en) Article mask method and device, electronic equipment and storage medium
CN113435456A (en) Rock slice component identification method and device based on machine learning and medium
CN114419087A (en) Focus image generation method and device, electronic equipment and storage medium
Lin et al. Integrated circuit board object detection and image augmentation fusion model based on YOLO
CN112883926B (en) Identification method and device for form medical images
CN103871089A (en) Image superpixel meshing method based on fusion
CN105719261A (en) Point cloud data combination system and method
CN108876776A (en) A kind of method of generating classification model, eye fundus image classification method and device
CN111539390A (en) Small target image identification method, equipment and system based on Yolov3

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
PP01 Preservation of patent right

Effective date of registration: 20211123

Granted publication date: 20200915

PP01 Preservation of patent right
PD01 Discharge of preservation of patent

Date of cancellation: 20220415

Granted publication date: 20200915

PD01 Discharge of preservation of patent
TR01 Transfer of patent right

Effective date of registration: 20231226

Address after: 518000 floor 33, Yantian modern industry service center, 3018 Shayan Road, Shatoujiao street, Yantian District, Shenzhen City, Guangdong Province

Patentee after: Shenzhen mailong Intelligent Technology Co.,Ltd.

Address before: 518000 B, 5 / F, building 2, international creative port, industrial east street, Yantian District, Shenzhen City, Guangdong Province

Patentee before: SHENZHEN MALONG TECHNOLOGY Co.,Ltd.

TR01 Transfer of patent right