WO2019096178A1

WO2019096178A1 - Fiber detection method and apparatus, and electronic device

Info

Publication number: WO2019096178A1
Application number: PCT/CN2018/115493
Authority: WO
Inventors: 斯科特·马修·罗伯特; 黄鼎隆; 傅恺; 郭胜
Original assignee: 深圳码隆科技有限公司
Priority date: 2017-11-14
Filing date: 2018-11-14
Publication date: 2019-05-23
Also published as: CN107909107A; CN107909107B

Abstract

A fiber detection method and apparatus, and an electronic equipment, relating to the technical field of image recognition. The fiber detection method comprises: obtaining a fiber image to be detected (S101); extracting fiber features of the fiber image to be detected according to a preset deep learning model, the preset deep learning model comprising multiple deep learning models based on a convolutional neural network (S102); and utilizing a classifier trained based on the preset deep learning model to recognize the fiber features, and generating a recognition result of the fiber image to be detected (S103). According to the fiber detection method, independent neural network detection recognition can be performed on each type of fibers by means of the deep learning models based on the convolutional neural network, and therefore, interference on recognition from other fibers during blend fiber detection is eliminated, and the purpose of accurate fiber recognition is achieved.

Description

Fiber testing method, device and electronic device

Cross-reference to related applications

The present application claims the priority of the Chinese Patent Application No. 201711126617.8, entitled "Fiber Detection Method, Apparatus, and Electronic Apparatus", filed on November 14, 2017, the entire contents of which is incorporated herein by reference. .

Technical field

The present application relates to the field of image recognition technologies, and in particular, to a fiber detecting method, device, and electronic device.

Background technique

With the sustained and rapid development of China's economy, all walks of life have shown a booming trend. Along with this, various materials containing fibrous structures are becoming more and more abundant, and their types, fiber compositions and forms vary widely. The current fiber detection and identification method is based on manual methods. It is usually necessary to collect images, and then observe the above images. According to the existing experience and knowledge, the fiber types in the images are determined, and the fiber diameter and number are calculated. This process is time consuming and labor inefficient, especially for materials containing hybrid fiber structures, which are difficult to identify.

Summary of the invention

In view of this, the purpose of the present application is to provide a fiber detecting method, device and electronic device capable of performing independent neural network detection and identification on each fiber through various deep learning models based on convolutional neural networks, thereby When the mixed fiber is detected, the interference of other fibers on the identification is excluded, and the purpose of accurately identifying the fiber is achieved.

In a first aspect, an embodiment of the present application provides a fiber detecting method, including:

Obtaining an image of the fiber to be tested;

Extracting fiber characteristics of the fiber image to be detected according to a preset depth learning model, and the preset depth learning model includes a plurality of deep learning models based on a convolutional neural network;

The fiber feature is identified by a classifier trained based on the preset depth learning model to generate a recognition result of the fiber image to be detected.

With reference to the first aspect, the embodiment of the present application provides a first possible implementation manner of the first aspect, wherein acquiring the image of the fiber to be detected includes:

An image obtained by taking an image of a target object by an electron microscope is obtained, and a fiber image to be detected is obtained; the target object includes a fiber structure.

In connection with the first aspect, embodiments of the present application provide a second possible embodiment of the first aspect, wherein the fiber characteristics comprise: fiber type, fiber diameter, and fiber count.

With reference to the first aspect, the embodiment of the present application provides a third possible implementation manner of the first aspect, wherein the deep learning model based on the convolutional neural network is trained by using fiber sample data exceeding a certain threshold, the fiber sample The data includes pictures corresponding to different types of fibers.

With reference to the first aspect, the embodiment of the present application provides a fourth possible implementation manner of the first aspect, wherein the classifier based on the preset deep learning model training is obtained by:

Extracting deep features of fiber sample data using a variety of deep learning models based on convolutional neural networks;

Training a classifier for deep features based on a machine learning algorithm;

Among them, the fiber sample data includes fiber characteristics, matching degree and recognition result.

With reference to the first aspect, the embodiment of the present application provides a fifth possible implementation manner of the first aspect, wherein the fiber feature is identified by using a classifier trained based on the preset depth learning model to generate a recognition result of the fiber image to be detected. Specifically, including:

Determining a data format corresponding to the classifier trained based on the preset deep learning model;

If the data format includes a binary data format, the fiber feature is converted into a binary data format and then input into a classifier based on the preset depth learning model training to generate a recognition result corresponding to the fiber image to be detected.

With reference to the first aspect, the embodiment of the present application provides a sixth possible implementation manner of the first aspect, wherein the fiber feature is identified by using a classifier trained based on the preset depth learning model to generate an image of the fiber to be detected. After the results, it also includes:

The recognition result of the fiber image to be detected is stored as a new fiber sample data in the fiber sample database.

In a second aspect, an embodiment of the present application provides a fiber detecting device, where the device includes:

An image acquisition module configured to acquire a fiber image to be detected;

a feature extraction module configured to extract a fiber feature of the fiber image to be detected according to a preset depth learning model, where the preset depth learning model includes a plurality of deep learning models based on a convolutional neural network;

The fiber identification module is configured to identify the fiber feature by using a classifier trained based on the preset depth learning model to generate a recognition result of the fiber image to be detected.

In a third aspect, an embodiment of the present application provides an electronic device, including a memory and a processor, where the computer stores a computer program executable on a processor, and when the processor executes the computer program, the steps of the method described in the first aspect are implemented. .

In a fourth aspect, the embodiment of the present application further provides a computer readable medium having a processor-executable non-volatile program code, the program code causing a processor to perform the method described in the first aspect.

The embodiments of the present application bring the following beneficial effects:

In the fiber detecting method provided by the embodiment of the present application, the fiber image to be detected is first obtained, and then the fiber feature of the fiber image to be detected is extracted according to a preset depth learning model, wherein the preset depth learning model includes a plurality of convolutional neural networks. The deep learning model finally uses the classifier based on the preset deep learning model to identify the fiber features and generate the recognition result of the fiber image to be detected. The fiber detection method can perform independent neural network detection and identification on each fiber through a variety of deep learning models based on convolutional neural networks, thereby eliminating the interference of other fibers on the identification of the mixed fibers, and accurately identifying the fibers. the goal of.

Other features and advantages of the present application will be set forth in the description which follows and become apparent from the description. The objectives and other advantages of the present invention are realized and attained by the structure of the invention.

The above described objects, features, and advantages of the present invention will become more apparent from the following description.

DRAWINGS

In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings to be used in the specific embodiments or the description of the prior art will be briefly described below, and obviously, the attached in the following description The drawings are some embodiments of the present application, and those skilled in the art can obtain other drawings based on these drawings without any creative work.

1 is a flow chart of a fiber detecting method according to Embodiment 1 of the present application;

2 is a flow chart of another fiber detecting method according to Embodiment 1 of the present application;

3 is a flow chart of another fiber detecting method according to Embodiment 1 of the present application;

4 is a flow chart of another fiber detecting method according to Embodiment 1 of the present application;

FIG. 5 is a schematic structural diagram of a fiber detecting device according to Embodiment 2 of the present application; FIG.

FIG. 6 is a schematic structural diagram of an electronic device according to Embodiment 3 of the present application.

Detailed ways

The technical solutions of the present application will be clearly and completely described in the following with reference to the accompanying drawings. It is obvious that the described embodiments are a part of the embodiments of the present application, and not all of them. An embodiment. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present application without departing from the inventive scope are the scope of the present application.

At present, the existing fiber detection and identification method is performed on an artificial basis, which is not only time-consuming and laborious, but also inefficient, especially for materials containing a hybrid fiber structure, which is difficult to identify. Based on this, the fiber detecting method, device and electronic device provided by the embodiments of the present application can perform independent neural network detection and identification on each fiber through various deep learning models based on convolutional neural networks, thereby detecting mixed fibers. When the interference of other fibers is recognized, the purpose of accurately identifying the fiber is achieved.

In order to facilitate the understanding of the present embodiment, a fiber detecting method disclosed in the embodiment of the present application is first introduced in detail.

Embodiment 1:

The embodiment of the present application provides a fiber detecting method, which can be applied to fiber detecting scenarios in various fields, such as cloth fiber detection and plant fiber testing in the field of garment production. Referring to Figure 1, the method includes:

S101: Acquire an image of the fiber to be detected.

Specifically, an image acquired by photographing the target object by an electron microscope is acquired, and an image of the fiber to be detected is obtained. Among them, the target test substance is a material containing a fibrous structure such as cloth, a high polymer composite material, and a plant specimen. The electron microscope can be a microscope of different types and different specifications, and the relevant professional uses an electron microscope to collect an image of the fiber to be detected of the object to be detected.

S102: Extracting fiber characteristics of the fiber image to be detected according to the preset depth learning model, and the preset depth learning model includes a plurality of deep learning models based on a convolutional neural network.

The deep learning model based on convolutional neural network is trained by fiber sample data exceeding a certain threshold, and the fiber sample data includes pictures corresponding to different types of fibers.

In a preferred embodiment, the above various deep learning models based on convolutional neural networks can be implemented by the Caffe deep learning framework.

Specifically, the more the number of fiber images in the fiber sample data, the better, the more the types, the more the data, the better the versatility of the training-generated deep learning model based on the convolutional neural network. In the embodiment of the present application, taking the cloth fiber as an example, the fiber image includes a plurality of fiber sample data having different fiber types, fiber diameters, and fiber numbers, which is advantageous for accurate recognition of the fiber image to be detected later, and the convolutional neural network is improved. The ability of the deep learning model to recognize, even if the fiber to be tested is a hybrid fiber, can be accurately identified.

The step S102 specifically includes: performing the feature training in the plurality of base layers included in the preset depth learning model as the input image as the input image, and extracting the fully connected layer or other designated base layers in the multiple integrations after the training is completed. The output feature vector is used as the corresponding fiber feature in the image of the fiber to be detected. Among them, fiber characteristics include: fiber type, fiber diameter and fiber number.

S103: Identify a fiber feature by using a classifier trained based on a preset depth learning model to generate a recognition result of the fiber image to be detected.

The extracted fiber features are input into a classifier trained based on a preset depth learning model, and after the classifier is identified, the final recognition result is obtained. Specifically, the recognition results are fiber type, fiber diameter, and fiber number.

In an optional implementation manner, the classifier based on the preset deep learning model training is obtained in the following manner, as shown in FIG. 2:

S201: Extracting deep features of fiber sample data using a plurality of deep learning models based on convolutional neural networks.

S202: Train the classifier for the deep feature based on the machine learning algorithm.

Among them, the fiber sample data includes fiber characteristics, matching degree and recognition result. The above machine learning algorithm may be a neighboring algorithm, a maximum expectation algorithm, and a support vector machine algorithm. The specific algorithm may be selected according to a specific situation, which is not limited herein.

In an optional embodiment, the fiber sample data includes triple data; wherein the triple data includes: source data, forward data belonging to the same category as the source data, and subordinate to the source data Reverse data for different categories.

The source data is sample data with the same recognition result randomly obtained from the fiber sample data. The forward data is sample data that is randomly obtained from the fiber sample data and is consistent with the recognition result of the source data; the matching degree of the source data is higher than the matching degree of the forward data. The reverse data is sample data that is randomly obtained from the fiber sample data and is inconsistent with the recognition result of the source data.

In a specific embodiment, the triplet data is: a first picture with good performance of the fiber image in the fiber sample data, a second picture with poor performance of the fiber image in the fiber sample data, and the first picture and the second picture The picture identifies a third picture with a different result. The specific first picture is the source data with the highest matching degree, and the second picture has a gap with the first picture in terms of definition and resolution, and the matching degree is lower than the first picture, but is still a fiber image, and is a forward data. . The third picture is the reverse data of the reverse contrast in the training, and the positive opposition ratio is used to further enhance the recognition ability of the classifier and improve the accuracy of the neural network fiber detection and recognition.

The classifier for training based on the preset depth learning model is used to identify the fiber features, and the recognition result of the fiber image to be detected is generated, which specifically includes the following steps, as shown in FIG. 3:

S301: Determine a data format corresponding to the classifier trained based on the preset depth learning model.

S302: If the data format includes a binary data format, converting the fiber feature into a binary data format, and inputting a classifier based on the preset depth learning model training to generate a recognition result corresponding to the fiber image to be detected.

Specifically, if the classifier trained by the preset deep learning model supports the binary data format, the fiber feature is converted, converted into a binary data format, and then input into the trained classifier for recognition, because the machine language is Binary, so the binary data format can speed up the recognition process, and no additional data conversion is required when performing the recognition, thereby improving the efficiency of recognition.

In addition, after generating the recognition result of the fiber image to be detected, the following steps are further included, as shown in FIG. 4:

S401: The recognition result of the fiber image to be detected is saved as a new fiber sample data in the fiber sample database.

By continuously updating the data in the fiber sample database, the fiber recognition effect of the classifier trained based on the deep learning model is more accurate.

Embodiment 2:

The embodiment of the present application provides a fiber detecting device. As shown in FIG. 5, the device includes an image acquiring module 51, a feature extracting module 52, and a fiber identifying module 53.

The image obtaining module 51 is configured to acquire a fiber image to be detected. The feature extraction module 52 is configured to extract fiber characteristics of the fiber image to be detected according to the preset depth learning model, and the preset depth learning model includes multiple convolutional neural networks. The deep learning model; the fiber identification module 53 is configured to identify the fiber features by using a classifier trained based on the preset depth learning model to generate a recognition result of the fiber image to be detected.

In the fiber detecting device provided by the embodiment of the present application, the working process of each module has the same technical features as the fiber detecting method described above, and therefore, the above functions can also be implemented, and details are not described herein again.

Embodiment 3:

The embodiment of the present application further provides an electronic device. As shown in FIG. 6, the electronic device includes: a processor 60, a memory 61, a bus 62, and a communication interface 63. The processor 60, the communication interface 63, and the memory 61 pass through the bus. 62 is connected; the processor 60 is configured to execute an executable module, such as a computer program, stored in the memory 61. The steps of the method as described in the method embodiments are implemented when the processor executes a computer program.

The memory 61 may include a high speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is implemented by at least one communication interface 63 (which may be wired or wireless), and may use an Internet, a wide area network, a local network, a metropolitan area network, or the like.

The bus 62 can be an ISA bus, a PCI bus, or an EISA bus. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one double-headed arrow is shown in Figure 6, but it does not mean that there is only one bus or one type of bus.

The memory 61 is configured to store a program, and the processor 60 executes the program after receiving the execution instruction, and the method executed by the device defined by the flow process disclosed in any embodiment of the present application may be applied to the processing. In processor 60, or implemented by processor 60.

Processor 60 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method may be completed by an integrated logic circuit of hardware in the processor 60 or an instruction in the form of software. The processor 60 may be a general-purpose processor, including a central processing unit (CPU) and a network processor (NP), and may also be a digital signal processor (DSP). ), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component. The methods, steps, and logical block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general purpose processor may be a microprocessor or the processor or any conventional processor or the like. The steps of the method disclosed in the embodiments of the present application may be directly implemented by the hardware decoding processor, or may be performed by a combination of hardware and software modules in the decoding processor. The software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like. The storage medium is located in the memory 61, and the processor 60 reads the information in the memory 61 and performs the steps of the above method in combination with its hardware.

The computer program product of the method for locating a network device provided by the embodiment of the present application includes a computer readable storage medium storing non-volatile program code executable by a processor, the program code including instructions configurable to execute the front For the specific implementation of the method, refer to the method embodiment, and details are not described herein again.

A person skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the device and the electronic device described above can refer to the corresponding process in the foregoing method embodiments, and details are not described herein again.

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present application. In this regard, each block of the flowchart or block diagram can represent a module, a program segment, or a portion of code that comprises one or more of the Executable instructions. It should also be noted that in some alternative implementations, the functions noted in the blocks may also occur in a different order than that illustrated in the drawings. For example, two consecutive blocks may be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts, can be implemented in a dedicated hardware-based system that performs the specified function or function. Or it can be implemented by a combination of dedicated hardware and computer instructions.

In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inside" and "outside" etc. The orientation or positional relationship of the indications is based on the orientation or positional relationship shown in the drawings, for the convenience of the description and the simplified description, and does not indicate or imply that the device or component referred to has a specific orientation or a specific orientation. Construction and operation are therefore not to be construed as limiting the application. Moreover, the terms "first", "second" and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The device embodiments described above are merely illustrative. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner. For example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some communication interface, device or unit, and may be electrical, mechanical or otherwise.

The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as separate products, may be stored in a non-transitory computer readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application, which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including The instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .

Finally, it should be noted that the above-mentioned embodiments are only specific embodiments of the present application, and are used to explain the technical solutions of the present application, and are not limited thereto. The scope of protection of the present application is not limited thereto, although reference is made to the foregoing. The present invention has been described in detail with reference to the embodiments of the present invention. It will be understood by those skilled in the art that the technical solutions described in the foregoing embodiments can still be modified within the technical scope of the present disclosure. The changes may be easily conceived, or equivalently substituted for some of the technical features; and the modifications, variations, or substitutions of the present invention are not intended to depart from the spirit and scope of the technical solutions of the embodiments of the present application. Within the scope of protection. Therefore, the scope of protection of this application should be determined by the scope of protection of the claims.

Claims

A fiber detecting method, comprising:

Obtaining an image of the fiber to be tested;

Extracting fiber features of the image to be detected according to a preset depth learning model, the preset depth learning model comprising a plurality of deep learning models based on a convolutional neural network;

The fiber feature is identified by a classifier trained based on the preset depth learning model to generate a recognition result of the fiber image to be detected.
The method according to claim 1, wherein the obtaining the image of the fiber to be detected comprises:

An image obtained by photographing a target object by an electron microscope is obtained to obtain an image of the fiber to be detected; the target object includes a fiber structure.
The method of claim 1 wherein said fiber characteristics comprise: fiber type, fiber diameter, and fiber count.
The method of claim 1 wherein said convolutional neural network based deep learning model is trained by fiber sample data having a number exceeding a certain threshold, said fiber sample data comprising different types of fibers corresponding to image.
The method according to claim 1, wherein the classifier based on the preset deep learning model training is obtained by:

Extracting deep features of fiber sample data using a deep learning model based on convolutional neural networks;

Training a classifier for the deep feature based on a machine learning algorithm;

The fiber sample data includes fiber characteristics, matching degree, and recognition result.
The method according to claim 1, wherein the identifying the fiber feature by using a classifier trained based on the preset depth learning model, and generating the recognition result of the to-be-detected fiber image, specifically includes:

Determining, according to the data format corresponding to the classifier trained by the preset depth learning model;

If the data format includes a binary data format, the fiber feature is converted into a binary data format, and the classifier trained based on the preset depth learning model is input to generate a recognition result corresponding to the fiber image to be detected.
The method according to any one of claims 1 to 6, wherein the fiber feature is identified by using a classifier trained based on the preset depth learning model to generate the image of the fiber to be detected. After identifying the results, it also includes:

The recognition result of the image of the fiber to be detected is stored as new fiber sample data in the fiber sample database.
A fiber detecting device, characterized in that the device comprises:

An image acquisition module configured to acquire a fiber image to be detected;

a feature extraction module, configured to extract a fiber feature of the to-be-detected fiber image according to a preset depth learning model, where the preset depth learning model includes a plurality of deep learning models based on a convolutional neural network;

And a fiber identification module configured to identify the fiber feature by using a classifier trained based on the preset depth learning model to generate a recognition result of the fiber image to be detected.
An electronic device comprising a memory and a processor having stored thereon a computer program executable on the processor, wherein the processor executes the computer program to implement the above claims 1 to 7 The steps of any of the methods described.
A computer readable medium having a processor-executable non-volatile program code, the program code causing the processor to perform the method of any one of claims 1 to 7.