Yarn vehicle code identification method

Abstract

The invention provides a yarn vehicle code identification method, relates to the technical field of glass fiber production and manufacturing, and solves the technical problems that in the prior art, the efficiency of manual vehicle code identification is relatively low, errors are easy to occur after long-time identification, logistics conveying equipment frequently stops, and the production efficiency is influenced. The device comprises the following operation steps that when a yarn vehicle passes through a vehicle code recognition device, the vehicle code recognition device can obtain an image of a yarn vehicle code plate on the yarn vehicle; image information of the yarn vehicle code plate is uploaded to an ROI area detection network, and the ROI area detection network obtains an information section of the yarn vehicle code plate from the image of the yarn vehicle code plate; the information section is uploaded to a vehicle code identification network, and the vehicle code identification network obtains the vehicle code information contained in the information section and verifies the vehicle code information; if the verification is passed, vehicle code information is uploaded to an upper computer system for distributing the direction of the yarn vehicle; and if the verification is not passed, giving an alarm.

Description

Technical Field

[0001] The invention relates to the technical field of glass fiber production and manufacturing, in particular to a yarn carrier code identification method. Background Art

[0002] The manufacturing process of glass fiber includes several different production areas. Different types of glass fiber products have different production processes, production steps and required processing time. In the glass fiber production workshop, yarn balls are transported between various production areas using yarn carts as carriers. In the same workshop, different batches and types of yarn balls share the same yarn cart. In order to distinguish the batches and types of yarn balls on the yarn cart, a code plate containing the car code is installed on the top crossbeam of the yarn cart. When the yarn ball is loaded onto the yarn cart, the material information of the yarn ball is bound to the car code information contained in the code plate and uploaded to the host computer system. A car code information collection area is set at the junction of every two production areas of the glass fiber logistics production line. When the yarn cart passes the collection point, the host computer system determines the batch and type of the yarn ball on the yarn cart based on the acquired car code information, thereby sending a destination instruction to the logistics transportation system to control the yarn cart to drive to different production areas.

[0003] The existing yarn car code disk manually identifies the car code and inputs it into the upper computer system to allocate the yarn car's destination. However, the efficiency of manual identification of car codes is relatively low, and long-term identification is prone to errors, causing the logistics conveying equipment to stop frequently, affecting production efficiency, and is not suitable for large-scale automated glass fiber production lines. Summary of the invention

[0004] The purpose of the present invention is to provide a yarn car code identification method to solve the technical problem that the efficiency of manual code identification in the prior art is relatively low, long-term identification is prone to errors, and logistics conveying equipment often stops and affects production efficiency. The preferred technical solution among the many technical solutions provided by the present invention can produce many technical effects as described below.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] The present invention provides a yarn train code identification method, comprising the following steps:

[0007] Step S1: a yarn carriage identification device is installed on an automated glass fiber production line. A yarn carriage moves on the automated glass fiber production line. When the yarn carriage passes through the yarn carriage identification device, the yarn carriage identification device acquires an image of a yarn carriage identification plate on the yarn carriage.

[0008] Step S2: the image information of the yarn machine number plate is uploaded to the ROI area detection network, and the ROI area detection network obtains the information segment of the yarn machine number plate from the image of the yarn machine number plate;

[0009] Step S3: the information segment is uploaded to the vehicle code recognition network, and the vehicle code recognition network obtains the vehicle code information contained in the information segment and verifies the vehicle code information;

[0010] Step S4: if the verification is passed, the vehicle code information is uploaded to the upper computer system for allocating the destination of the yarn vehicle; if the verification is not passed, an alarm is issued.

[0011] Optionally, in step S2, the information section of the yarn machine code plate includes a mark section, a digital section and a check section, and the digital section is located between the mark section and the check section.

[0012] Optionally, in step S3, the vehicle code recognition network may obtain characters in the flag section to distinguish the type of the yarn ball on the yarn vehicle or the direction of the yarn vehicle.

[0013] Optionally, in step S3, the vehicle code recognition network obtains a plurality of digital characters in the digital segment and adds all the digital characters to obtain a digital result;

[0014] The vehicle code recognition network obtains the verification digit of the verification section and verifies the digital result with the verification digit.

[0015] Optionally, in step S4, if the digital result is consistent with the verification digit, the verification passes; if the digital result is inconsistent with the verification digit, the verification fails.

[0016] Optionally, the vehicle code recognition network includes a forward vehicle code recognition network and a reverse vehicle code recognition network;

[0017] First, the forward vehicle code recognition network is used to verify the information section of the yarn vehicle code plate. If the verification is passed, the vehicle code information is uploaded to the upper computer system for allocating the destination of the yarn vehicle.

[0018] If the verification fails, the reverse vehicle code recognition network is used to verify the information section of the yarn vehicle code card. If the verification passes, the vehicle code information is uploaded to the upper computer system for allocating the destination of the yarn vehicle.

[0019] If the verification fails, an alarm is issued.

[0020] Optionally, in step S1, the vehicle code recognition device includes a support frame, an industrial camera, a camera light source and a photoelectric switch, the support frame and the photoelectric switch are both installed on the automated glass fiber production line, and the industrial camera and the camera light source are installed on one end of the support frame;

[0021] The industrial camera is used to obtain an image of the yarn car number plate on the yarn car;

[0022] The camera light source is used to illuminate the front side of the yarn machine number plate;

[0023] The photoelectric switch is used to collect information on whether the yarn carrier is in place.

[0024] Optionally, the vehicle code identification device further comprises a back panel lighting source, wherein the back panel lighting source is mounted on the other end of the support frame, and the back panel lighting source is used to illuminate the back side of the yarn vehicle code plate.

[0025] The present invention provides a yarn car code recognition method, which obtains the image of the yarn car code plate through a code recognition device, and verifies it by using an ROI area detection network and a code recognition network, so that the direction of the yarn car can be quickly allocated. The ROI area detection network and the code recognition network are based on deep learning technology, and have powerful learning and image processing capabilities. Therefore, they can cope with problems such as code plate bending, image weak light, reflection, etc., reduce the requirements of the code recognition device on the on-site environment, facilitate the installation and deployment of the equipment, and at the same time improve the code recognition rate, reduce the number of stops and the stop time of the glass fiber production line, and improve production efficiency. It solves the technical problem that the efficiency of manual code recognition in the prior art is relatively low, and long-term recognition is prone to errors, resulting in frequent stops of logistics transportation equipment and affecting production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 It is a structural schematic diagram of a yarn train code identification device in a yarn train code identification method provided by an embodiment of the present invention;

[0028] Figure 2 It is an operation step diagram of a yarn train code identification method provided by an embodiment of the present invention.

[0029] Figure 3It is an image of a yarn train number plate in a yarn train number recognition method provided by an embodiment of the present invention;

[0030] Figure 4 It is a forward arrangement diagram of a yarn car number plate in a yarn car number identification method provided by an embodiment of the present invention;

[0031] Figure 5 The invention discloses a reverse arrangement diagram of a yarn train number plate in a yarn train number recognition method provided by an embodiment of the present invention.

[0032] In the figure, 1. Automated glass fiber production line; 2. Yarn car; 3. Yarn car code plate; 4. Support frame; 5. Industrial camera; 6. Camera light source; 7. Photoelectric switch; 8. Back panel lighting source. DETAILED DESCRIPTION

[0033] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0034] In the description of the present invention, it should be noted that, unless otherwise specified, the meaning of "plurality" is two or more; the orientations or positional relationships indicated by the terms "upper", "lower", "left", "right", "inner", "outer", "front end", "rear end", "head", "tail", etc. are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0035] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0036] The present invention provides a yarn train code identification method, comprising the following steps:

[0037] Step S1: The vehicle code recognition device is installed on the automated glass fiber production line 1, and the yarn car 2 moves on the automated glass fiber production line 1. When the yarn car 2 passes through the vehicle code recognition device, the vehicle code recognition device will obtain an image of the yarn car code plate 3 on the yarn car 2;

[0038] Step S2: the image information of the yarn machine number plate 3 is uploaded to the ROI area detection network, and the ROI area detection network obtains the information segment of the yarn machine number plate 3 from the image of the yarn machine number plate 3;

[0039] Step S3: the information segment is uploaded to the vehicle code recognition network, and the vehicle code recognition network obtains the vehicle code information contained in the information segment and verifies the vehicle code information;

[0040] Step S4: If the verification is passed, the vehicle code information is uploaded to the upper computer system, which can be a logistics scheduling system for allocating the destination of the yarn car 2; if the verification fails, an alarm is issued. The present invention provides a yarn car vehicle code recognition method, which obtains the image of the yarn car code card 3 through the vehicle code recognition device, and verifies it by using the ROI area detection network and the vehicle code recognition network, so that the yarn car destination can be quickly allocated. The ROI area detection network and the vehicle code recognition network are based on deep learning technology and have strong learning ability and image processing ability. Therefore, they can deal with problems such as code card bending, image weak light, reflection, etc., reduce the requirements of the vehicle code recognition device for the on-site environment, facilitate the installation and deployment of the equipment, and at the same time improve the vehicle code recognition rate, reduce the number of stops and parking time of the glass fiber production line, and improve production efficiency. It solves the technical problem that the efficiency of manual vehicle code recognition in the prior art is relatively low, and long-term recognition is prone to errors, which causes the logistics transportation equipment to stop frequently and affects production efficiency.

[0041] Before using the present invention, images of all situations of the yarn car code plate 3 should be collected in advance, such as code disk images under weak light conditions, code disk images under reflective conditions, code disk images with bending conditions, and normal code disk images. The information segments and car code information are labeled to form a code disk image data set for network training. When annotating the image, use a rectangular frame to select the information segment. It should be ensured that the rectangular frame contains all the car code information. The vertex coordinates of the rectangular frame are output for training the ROI area detection network. The collected image is cut using the vertex coordinates of the rectangular frame and the cut image is saved to obtain a car code data set. The car code information in the image is labeled for training the car code recognition network. Use the trained network to perform information segment detection and car code recognition.

[0042] As an optional implementation, in step S2, the information section of the yarn car code card 3 includes a mark section, a digital section and a check section, and the digital section is located between the mark section and the check section. Figure 4, the left side is the mark section, the middle is the number section, and the right side is the check section; the reverse arrangement is, see Figure 5 , the left side is the check section, the middle is the number section, the right side is the mark section, and the characters in it are also mirrored.

[0043] As an optional implementation, in step S3, the vehicle code recognition network will obtain the character in the mark segment, which can be an English character, such as "A", etc., or it can be a special character, such as "→", etc., used to distinguish the type of yarn ball on the yarn car 2 or the direction of the yarn car 2.

[0044] In step S3, the vehicle code recognition network obtains multiple digital characters in the digital section and adds all the digital characters to obtain a digital result; the vehicle code recognition network obtains the verification digit in the verification section and verifies the digital result and the verification digit.

[0045] In step S4, if the digital result is consistent with the check digit, the check is passed; if the digital result is inconsistent with the check digit, the check is failed.

[0046] As an optional implementation, the vehicle code recognition network includes a forward vehicle code recognition network and a reverse vehicle code recognition network;

[0047] First, the forward vehicle code recognition network is used to verify the information section of the yarn vehicle code plate 3. If the verification passes, the vehicle code information is uploaded to the upper computer system for allocating the destination of the yarn vehicle 2.

[0048] If the verification fails, the reverse vehicle code recognition network is used to verify the information section of the yarn vehicle code card 3. If the verification passes, the vehicle code information is uploaded to the upper computer system for allocating the destination of the yarn vehicle 2.

[0049] If the verification fails, an alarm is issued.

[0050] After the information segment of the yarn car code card 3 is obtained from the image of the yarn car code card 3 by using the ROI area detection network, the forward car code recognition network is first used to recognize the car code information, and the result is verified after recognition. If the first digit is a character in the preset flag segment, and the sum of the characters of the car code information in the digital segment is equal to the check digit in the check segment, the verification is passed, and the recognition result is uploaded to the host computer system to assign the direction of the yarn car 2; if the verification fails, the reverse car code recognition network is used to recognize the car code information, and the result is verified after recognition. If the last digit is a reverse character in the preset flag segment, and the sum of the characters of the car code information in the digital segment is equal to the check digit after the characters in the check segment are exchanged, the verification is passed, and the recognition result is uploaded to the host computer system to assign the direction of the yarn car, otherwise the verification fails.

[0051] For example, the recognition result of the forward vehicle code recognition network is '←1234515'. Figure 4 As shown, the first character is the positive character in the pre-set flag section, and the sum of the vehicle code characters in the digital section is 15, which is the same as the number 15 in the last two check sections, so the check passes. If the recognition result of the reverse vehicle code recognition network is '5154321→', Figure 5 As shown, the last character is the reverse character in the preset flag section, and the sum of the vehicle code characters in the digital section is 15, which is the same as the check digit 15 after the characters in the first two check sections are exchanged, and the check is passed.

[0052] As an optional implementation, a vehicle code information collection area is set at the junction of every two production areas in the automated glass fiber production line 1, and the vehicle code recognition device is deployed at the vehicle code information collection area. In step S1, the vehicle code recognition device includes a support frame 4, an industrial camera 5, a camera light source 6 and a photoelectric switch 7. The support frame 4 and the photoelectric switch 7 are both installed on the automated glass fiber production line 1. The industrial camera 5 and the camera light source 6 are installed on one end of the support frame 4. The support frame 4 can be made of aluminum alloy; a slide is provided on the support frame 4 to facilitate the lifting and lowering movement of the industrial camera 5, the camera light source 6 and the back panel lighting source 8, so that the height of the industrial camera 5, the camera light source 6 and the back panel lighting source 8 can be adjusted to correspond to the yarn car code sign 3 on the top of the yarn car 2; the support frame 4 is a gantry structure, and the columns on both sides of the support frame 4 are vertically fixed on both sides of the automated glass fiber production line 1, and the support frame 4 spans the support frame 4 to ensure that it can remain stable when the yarn car 2 passes.

[0053] The industrial camera 5 is used to obtain the image of the yarn car code sign 3 on the yarn car 2, and the industrial camera 5 has a focus function, and also includes an Ethernet port with a PoE function for image transmission and camera power supply. The installation position of the industrial camera 5 can ensure that when the yarn car 2 passes through the car code information collection area, the industrial camera 5 can capture the full view and details of the yarn car code sign 3 of the yarn car 2;

[0054] The camera light source 6 is used to illuminate the front of the yarn car number sign 3. The camera light source 6 can emit uniform and stable light. By adjusting the installation position of the camera light source 6, the information section of the yarn car number sign 3 can be illuminated;

[0055] The photoelectric switch 7 is used to collect information about whether the yarn car 2 is in place. When the photoelectric switch 7 detects a signal, it means that the yarn car 2 has arrived at the car code information collection area. The photoelectric switch 7 will send the information that the yarn car 2 is in place to the host computer system, and then the host computer system controls the industrial camera 5 to take pictures and obtain the image of the yarn car code sign 3. The photoelectric switch 7 can ensure that the yarn car 2 is accurately stopped in the car code information collection area, thereby ensuring that the industrial camera 5 can take a complete image of the yarn car code sign 3.

[0056] As an optional embodiment, the vehicle code recognition device further includes a back panel lighting source 8, which is mounted on the other end of the support frame 4. When the industrial camera 5 acquires the image of the yarn vehicle code sign 3, the back panel lighting source 8 is used to illuminate the back side of the yarn vehicle code sign 3, and the exposure parameter of the industrial camera 5 is reduced. Figure 3 , so that a code plate image containing white characters in a black background photo can be obtained, which can reduce the impact of vehicle code contamination on recognition accuracy.

[0057] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A yarn machine code identification method, characterized in that: The steps include: Step S1: a yarn car (2) is installed on the automated glass fiber production line (1), and when the yarn car (2) passes the yarn car identification device, the yarn car identification device will obtain an image of the yarn car identification plate (3) on the yarn car (2); Step S2: the image information of the yarn machine number plate (3) is uploaded to the ROI area detection network, and the ROI area detection network obtains the information segment of the yarn machine number plate (3) from the image of the yarn machine number plate (3); Step S3: the information segment is uploaded to the vehicle code recognition network, and the vehicle code recognition network obtains the vehicle code information contained in the information segment and verifies the vehicle code information; Step S4: if the verification is passed, the vehicle code information is uploaded to the upper computer system for allocating the destination of the yarn vehicle (2); if the verification is not passed, an alarm is issued.

2. A yarn train code identification method according to claim 1, characterized in that: In step S2, the information section of the yarn machine code plate (3) comprises a mark section, a digital section and a check section, and the digital section is located between the mark section and the check section.

3. A yarn train code identification method according to claim 2, characterized in that: In step S3, the vehicle code recognition network obtains the characters in the marking section to distinguish the type of yarn ball on the yarn vehicle (2) or the direction of the yarn vehicle (2).

4. A yarn train code identification method according to claim 2, characterized in that: In step S3, the vehicle code recognition network obtains multiple digital characters in the digital segment and adds all the digital characters to obtain a digital result; The vehicle code recognition network obtains the verification digit of the verification section and verifies the digital result with the verification digit.

5. A yarn train code identification method according to claim 4, characterized in that: In step S4, if the digital result is consistent with the verification digit, the verification is passed; if the digital result is inconsistent with the verification digit, the verification is failed.

6. A yarn train code identification method according to claim 1, characterized in that: The vehicle code recognition network includes a forward vehicle code recognition network and a reverse vehicle code recognition network; First, the forward vehicle code recognition network is used to verify the information section of the yarn vehicle code plate (3). If the verification is passed, the vehicle code information is uploaded to the upper computer system for allocating the destination of the yarn vehicle (2); If the verification fails, the reverse vehicle code recognition network is used to verify the information section of the yarn vehicle code card (3). If the verification passes, the vehicle code information is uploaded to the upper computer system for allocating the destination of the yarn vehicle (2); If the verification fails, an alarm is issued.

7. A yarn train code identification method according to claim 1, characterized in that: In step S1, the vehicle code recognition device comprises a support frame (4), an industrial camera (5), a camera light source (6) and a photoelectric switch (7), wherein the support frame (4) and the photoelectric switch (7) are both installed on the automated glass fiber production line (1), and the industrial camera (5) and the camera light source (6) are installed on one end of the support frame (4); The industrial camera (5) is used to obtain an image of the yarn wheel number plate (3) on the yarn wheel (2); The camera light source (6) is used to illuminate the front side of the yarn machine number plate (3); The photoelectric switch (7) is used to collect information on whether the yarn car (2) is in place.

8. A yarn train code identification method according to claim 7, characterized in that: The vehicle code identification device also includes a back panel lighting light source (8), which is installed on the other end of the support frame (4) and is used to illuminate the back side of the yarn vehicle code plate (3).

Images