CN211402180U

CN211402180U - System for identifying cloth cover defects

Info

Publication number: CN211402180U
Application number: CN201921765302.2U
Authority: CN
Inventors: 金玲玲
Original assignee: Shenzhen Lintsense Technology Co ltd
Current assignee: Shenzhen Lintsense Technology Co ltd
Priority date: 2019-10-18
Filing date: 2019-10-18
Publication date: 2020-09-01
Anticipated expiration: 2029-10-18

Abstract

The application discloses a system for cloth cover defect discernment includes: the intelligent cloth inspecting equipment is deployed on each cloth inspecting field, and at least comprises a cloth surface image acquisition device which is used for acquiring a cloth surface image to be inspected; the network transmission equipment is deployed at each cloth inspection site and used for uploading cloth surface images acquired by the intelligent cloth inspection equipment at the corresponding cloth inspection site to the cloud server; the cloud server comprises a cache distributor and an image recognizer, wherein the cache distributor is used for receiving the cloth cover images uploaded by the network transmission equipment, the image recognizer is provided with an input connected to the output of the cache distributor and an output providing an identification, and the identification indicates whether the identified cloth cover images contain cloth cover defects. According to the scheme of the embodiment of the application, the leakage of the core data in the database can be prevented, and the condition that the GPU is easily damaged in the transportation process in the prior art is also avoided.

Description

System for identifying cloth cover defects

Technical Field

The application relates to the technical field of fabric detection, in particular to a system for identifying cloth cover defects.

Background

In the production line of fabrics such as woven fabric, knitted fabric, non-woven fabric, etc., it is necessary to detect whether the produced fabric has a surface defect, for example, whether the fabric has stains, holes, fuzz, etc. The traditional manual detection often has the problems of low detection precision and low detection speed, and in recent years, the requirements of customers on the product quality are higher and higher, so that the problems of improving the detection accuracy and saving manpower are concerned.

Due to the adoption of manual detection means, the error rate is high and the efficiency is low, in recent years, a cloth surface defect detection technology based on machine vision is gradually developed to replace manual detection, namely intelligent cloth inspection equipment, wherein a camera is used for collecting cloth images, and a detection algorithm (such as an image recognition model) operated by an industrial personal computer is used for detecting the cloth images so as to recognize cloth surface defects. The detection algorithm is usually deployed on a GPU (Graphics Processing Unit) of the industrial personal computer, a database for identifying cloth cover defects is also deployed on the industrial personal computer, so that core data in the database is easy to leak, meanwhile, the industrial personal computer is easy to damage in the transportation process, the after-sale cost is high, and in addition, for a manufacturer, the industrial personal computer can be debugged only when a user checks cloth on the site, and the maintenance cost is increased.

Disclosure of Invention

In view of the above problems, embodiments of the present application provide a system for cloth cover defect identification, which can solve the technical problems mentioned in the above background section.

A system for cloth cover defect identification according to an embodiment of the application includes: the intelligent cloth inspecting equipment is deployed on each cloth inspecting field, and at least comprises a cloth surface image acquisition device which is used for acquiring a cloth surface image to be inspected; the network transmission equipment is deployed at each cloth inspection site and used for uploading cloth surface images acquired by the intelligent cloth inspection equipment at the corresponding cloth inspection site to the cloud server; the cloud server comprises a cache distributor and an image recognizer, wherein the cache distributor is used for receiving the cloth cover images uploaded by the network transmission equipment, the image recognizer is provided with an input connected to the output of the cache distributor and an output providing an identification, and the identification indicates whether the identified cloth cover images contain cloth cover defects.

Further, the cloud server further comprises a memory coupled to the image recognizer for storing image recognition models, and an image model trainer having an input coupled to the output of the cache distributor and an output coupled to the memory for storing the trained image models in the memory after the image models are trained using the cloth cover image feature vectors.

Further, the network transmission equipment is also used for feeding the identification back to a control device of the intelligent cloth inspecting equipment of the corresponding cloth inspecting field.

Further, the control device comprises a PLC control cabinet or a signal input and output device.

Further, the system also comprises a user terminal device which is arranged at each perching field and is connected to the network transmission device.

Further, the user terminal equipment comprises an industrial personal computer or a remote terminal controller.

Further, the cloth cover image acquisition device comprises a CCD industrial camera or a CMOS industrial camera.

Further, the network transmission device comprises a hundred mega switch or a gigabit switch.

Further, the network transmission device is connected with the cloud server through a mobile wireless access network.

Further, the mobile radio access network is a 5G mobile radio access network.

From the above description, it can be seen that the detection algorithm is deployed on the cloud server in the scheme of the embodiment of the application, the corresponding database for detecting cloth cover defects is also deployed on the cloud server, and during detection, cloth cover images acquired by the intelligent cloth inspecting equipment are transmitted to the cloud server through the network transmission device for detection, so that core data in the database can be prevented from leaking, meanwhile, the situation that a GPU (graphics processing Unit) is easily damaged in the transportation process of the industrial personal computer in the prior art is also avoided, in addition, the cloud server can be placed at a detection algorithm provider, and the algorithm provider does not need to debug the detection algorithm on the cloth inspecting site of a user.

Drawings

FIG. 1 is a schematic diagram of a system for cloth cover defect identification according to one embodiment of the present application;

fig. 2 is a schematic diagram of a cloud server according to an embodiment of the present application;

fig. 3 is a schematic diagram of a system for cloth cover defect identification according to another embodiment of the present application.

Detailed Description

The embodiments of the present application will be described in detail below, and the embodiments described by referring to the drawings are exemplary only for the purpose of explaining the present application and are not to be construed as limiting the present application.

In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.

In the description of the present application, it is to be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; may be mechanically connected, may be electrically connected or may be in communication with each other; either directly or indirectly through intervening media, either internally or in any other relationship. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art as appropriate.

The following disclosure provides many different embodiments or examples for implementing different features of the application. To simplify the disclosure of the present application, the components and settings of a specific example are described below. Of course, they are merely examples and are not intended to limit the present application. In addition, examples of various specific processes and materials are provided herein, but one of ordinary skill in the art may recognize applications of other processes and/or use of other materials.

Referring to fig. 1, a system for identifying a cloth cover defect in an embodiment of the present application may specifically include: the intelligent perching device 100, the network transmission device 200 and the cloud server 300. The intelligent perching device 100 is connected with the cloud server 300 through the network transmission device 200.

In this embodiment, a plurality of intelligent perching devices 100 may be deployed at the same perching site, or may be deployed at different perching sites, and when a plurality of intelligent perching devices 100 are deployed at different perching sites, they may be connected to the cloud server 300 through the corresponding network transmission device 200, respectively. The intelligent cloth inspecting equipment 100 at least comprises a cloth surface image acquisition device 102, and the cloth surface image acquisition device 102 is used for acquiring cloth surface images to be inspected. The cloth cover image capturing Device 102 may include a CCD (Charge Coupled Device) or a CMOS (Complementary Metal oxide semiconductor) industrial camera, wherein the CCD is a semiconductor component for recording light changes in a digital camera, and the CMOS is a sensor with a sensitivity generally 10 times lower than that of the CCD sensor.

In this embodiment, the network transmission device 200 is deployed at a cloth inspection site, and is configured to upload a cloth surface image acquired by the intelligent cloth inspection device 100 at the corresponding cloth inspection site to the cloud server 300. Specifically, the cloth cover image collecting device 102 may be directly connected to the network transmission device 200 through an ethernet interface, and transmit the cloth cover image to the network transmission device 200 through an ethernet transmission protocol. Network transport device 200 may comprise, for example, a hundred megabyte switch or a gigabit switch.

Referring to fig. 2, in this embodiment, the cloud server 300 may include a cache distributor 302 and an image recognizer 304, the image recognizer 304 has an input connected to the output of the cache distributor 302 and an output providing an identifier, the cache distributor 302 is configured to cache the cloth cover image uploaded by the network transmission device 200 and distribute the cached cloth cover image to the image recognizer 304, and the image recognizer 304 performs recognition detection on the cloth cover image and provides an identifier corresponding to a recognition result, where the identifier indicates whether the recognized cloth cover image includes a cloth cover defect. Specifically, the image recognizer 304 may detect the cloth cover image by using an image model, for example, the image model may be an image recognition neural network model, which is trained by using feature vectors of the cloth cover image. The image recognizer 304 may be, for example, a GPU (graphics processing Unit) deployed in the cloud server 300, and the GPU (graphics processing Unit) is also called a display core, a visual processor, a display chip, or an accelerator card, and is a microprocessor specially used for performing image operation on a computer, a workstation, and a part of mobile devices (such as a tablet computer, a smart phone, and the like).

With continued reference to fig. 2, in a preferred embodiment, the cloud server 300 further may include a memory 306 coupled to the image recognizer 304, the memory 306 being configured to store an image recognition model, and an image model trainer 308, the image model trainer 308 having an input coupled to the output of the buffer distributor 302 and an output coupled to the memory 306, the image model trainer 308 is used for training the image model by using the feature vectors of the cloth cover image distributed by the buffer distributor 302, and, after training, stores the trained image model in the memory 306 to update the image model in the memory 306, alternatively, a plurality of image models may be stored in the memory 306, and the image recognizer 304 may recognize the cloth image by using the updated image model in the memory 306, or may select one of the plurality of image models in the memory 306 according to a predetermined rule to recognize the cloth image.

In this embodiment, the network transmission device 200 may be connected to the cloud server 300 through a mobile wireless access network, specifically, the mobile wireless access network may be a 5G mobile wireless access network. Of course, the above connection method is only for illustration, and in the specific implementation, other connection methods other than the above connection method may be adopted according to specific situations.

With continued reference to fig. 1, in a preferred embodiment, the network transmission device 200 is further configured to feed back the identification output by the image recognizer 304 to the control apparatus 104 of the intelligent perching device 100 at the corresponding perching site. The control device 104 may be configured to control one or more of start-up and shut-down of the intelligent cloth inspecting apparatus 100, conveying of cloth, labeling, alarming, and light source adjustment, and may also obtain related operation data of the intelligent cloth inspecting apparatus 100. The network transmission device 200 feeds back the identification to the control apparatus 104 to cause the control apparatus 104 to perform an operation corresponding to the identification. The control device 104 may be, for example, a PLC (Programmable logic controller) control cabinet or a signal input/output device (I/O). The control device 104 may be connected to the network transmission apparatus 200 through a serial port. Further, the network transmission device 200 may also upload the operation/operation parameters and data of the control apparatus 104 to the cloud server 300. The control device 104 may be connected to sensors, drive motors, etc. on the intelligent perching apparatus 100.

With continued reference to fig. 1, in a preferred embodiment, the system may further include a user terminal device 400 disposed at each of the perching sites, the user terminal device 400 being connected to the network transmission device 200. The user terminal device 400 may include an industrial personal computer or a remote terminal controller. In one embodiment, as shown in fig. 1, the user terminal device 400 is a remote terminal controller, which can be connected with the network transmission device 200 by wireless routing or wired connection. The remote Terminal Controller may be, for example, a TC (Terminal Controller) or a smart Terminal, and the smart Terminal may include, for example, one or more of a tablet and a smart phone. The remote terminal controller may access the cloud server 300 for viewing or remote control. In another embodiment, as shown in fig. 3, the user terminal device 400 is an industrial personal computer, the cloth cover image acquisition device 102 and the control device 104 may be connected to the network transmission device 200 through the user terminal device 400, the cloth cover image acquired by the cloth cover image acquisition device 102 is uploaded to the cloud server 300 through the user terminal device 400, and the control device 104 receives the identifier provided by the cloud server 300 through the user terminal device 400 or uploads the relevant parameters and data of the intelligent cloth inspection device 100 to the cloud server 300.

As can be seen from the above description, in the solution of the embodiment of the present application, the detection algorithm is deployed on the cloud server 300, and the corresponding database for detecting the cloth cover defect is also deployed on the cloud server, during the detection, the cloth cover image collected by the intelligent cloth inspecting equipment 100 is uploaded to the cloud server 300 through the network transmission equipment 200 for detection, the cloud server 300 can feed back the identification of the reaction detection result to the intelligent cloth inspecting equipment 100, so as to control the intelligent perching device 100 to perform corresponding operations, the user can remotely control or check the detection of the cloud server 300 through the user terminal device 400, so as to prevent the core data in the database from leaking, meanwhile, the condition that the GPU arranged in the industrial personal computer in the transportation process of the industrial personal computer in the prior art is easy to damage is avoided, in addition, the cloud server 300 can be placed at a detection algorithm provider, and the algorithm provider does not need to go to the site of a user to debug the algorithm.

The detailed description set forth above in connection with the appended drawings describes exemplary embodiments but does not represent all embodiments that may be practiced or fall within the scope of the claims. The term "exemplary" used throughout this specification means "serving as an example, instance, or illustration," and does not mean "preferred" or "advantageous" over other embodiments. The detailed description includes specific details for the purpose of providing an understanding of the described technology. However, the techniques may be practiced without these specific details. In some instances, well-known structures and devices are shown in block diagram form in order to avoid obscuring the concepts of the described embodiments.

The previous description of the disclosure is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other variations without departing from the scope of the disclosure. Thus, the disclosure is not intended to be limited to the examples and designs described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A system for cloth cover defect identification, characterized by, includes:

the intelligent cloth inspecting equipment is deployed on each cloth inspecting field, and at least comprises a cloth surface image acquisition device which is used for acquiring a cloth surface image to be inspected;

the network transmission equipment is deployed at each cloth inspection site and used for uploading cloth surface images acquired by the intelligent cloth inspection equipment at the corresponding cloth inspection site to the cloud server;

the cloud server comprises a cache distributor and an image recognizer, wherein the cache distributor is used for receiving the cloth cover images uploaded by the network transmission equipment, the image recognizer is provided with an input connected to the output of the cache distributor and an output providing an identification, and the identification indicates whether the identified cloth cover images contain cloth cover defects.

2. The system of claim 1, wherein the cloud server further comprises a memory coupled to the image recognizer for storing image recognition models, and an image model trainer having an input coupled to the output of the cache distributor and an output coupled to the memory for storing trained image models in the memory after the image models are trained using cloth image feature vectors.

3. The system for cloth cover defect identification as claimed in claim 1 or 2, wherein the network transmission device is further used for feeding back the identification to the control device of the intelligent cloth inspection device of the corresponding cloth inspection field.

4. The system for cloth cover defect identification as defined in claim 3, wherein said control device comprises a PLC control cabinet or a signal input and output device.

5. The system for cloth cover defect identification as defined in claim 1 or 2, wherein the system further comprises a user terminal device disposed at each perching field, the user terminal device being connected to the network transmission device.

6. The system for cloth cover defect identification as claimed in claim 5, wherein the user terminal device comprises an industrial personal computer or a remote terminal controller.

7. The system for cloth cover defect identification as defined in claim 1 or 2, wherein said cloth cover image acquisition device comprises a CCD industrial camera or a CMOS industrial camera.

8. The system for cloth cover defect identification as set forth in claim 1 or 2, wherein said network transmission device comprises a hundred mega switch or a giga switch.

9. The system for cloth cover defect identification as claimed in claim 1 or 2, wherein the network transmission device is connected with the cloud server through a mobile wireless access network.

10. The system for cloth cover defect identification as defined in claim 9, wherein the mobile radio access network is a 5G mobile radio access network.