CN115009905A

CN115009905A - Film adjusting method based on machine vision

Info

Publication number: CN115009905A
Application number: CN202210597560.4A
Authority: CN
Inventors: 程晓颖; 郑明铭; 石琳; 吴震宇
Original assignee: Zhejiang Sci Tech University ZSTU
Current assignee: Zhejiang Sci Tech University ZSTU
Priority date: 2022-05-30
Filing date: 2022-05-30
Publication date: 2022-09-06

Abstract

The invention discloses a film adjustment method based on machine vision, which comprises the following steps: 1) building a film transport device to place the film and lighting the film above; 2) arranging two cameras and aligning them at different angles The surface of the film; 3), according to the reflection of the film and the tightness of the edge, machine learning determines the degree of deformation of the film, and adjusts the speed of the conveying device, the distance between the rollers and the parallelism, and uses the camera to shoot the reflection and edge of the film surface. The degree of tightness is analyzed. When the speed of the film transmission device, the distance between the rollers and the parallelism adjustment are not suitable, the wrinkle reflection on the surface of the film and the degree of tightness of the edges are very obvious changes, which is very conducive to computer analysis and machine learning judgment. Prevents excessive deformation of the film during transportation, thereby improving the quality of film printing.

Description

Film adjusting method based on machine vision

Technical Field

The invention relates to the technical field of material performance detection, in particular to a film adjusting method based on machine vision.

Background

The film has good mechanical properties and can be widely applied to the field of light industry, such as being used as a printing substrate. However, in the transport conveyance, it is difficult to accurately adjust the speed of the mechanical transfer device, the pitch between the rollers, and the parallelism. The film often can produce fold and deformation in the transportation process, leads to the printing precision to reduce, if can not find in time and adjust conveying speed, interval and the depth of parallelism between the cylinder, probably warp and can be more and more serious, leads to the printed matter qualification rate to descend.

Disclosure of Invention

In order to solve the above problems, an object of the present invention is to provide a film adjusting method based on machine vision, which can detect wrinkles and deformations on the surface of a film without affecting the film conveying conditions, so as to adjust and control the conveying speed, the spacing between rollers and the parallelism, and ensure the stable conveying of the film.

Technical scheme

A film adjusting method based on machine vision comprises the following process steps:

1) illuminating above the film in the film transport device;

2) arranging two cameras and aligning the two cameras at different angles to the surface of the film to prepare for shooting;

3) and the computer performs machine learning according to the reflection condition and the edge tightness condition of the film shot by the camera, so that the deformation degree analysis result of the film is determined according to the image shot by the camera, the speed of the film conveying device and the distance and the parallelism between the rollers are adjusted, and the stable conveying of the film is ensured.

Further comprises the following steps: the film conveying device in the step 1) comprises two rollers, wherein the rollers are provided with films, and the upper sides of the films are provided with illuminating lamps.

Further comprises the following steps: the two cameras in the step 2) are placed above the film transmission device at different angles and shoot the surface of the film, and the installation positions of the two cameras are on the central axis of the wide side of the film.

Further comprises the following steps: determining a deformation degree analysis result of the film according to the image shot by the camera in the step 3) and adjusting the film conveying device:

1. the camera shoots images and transmits the images to the computer;

2. and the computer analyzes the surface reflection and edge tightness of the film and adjusts the speed of the rollers and the distance and the parallelism among the rollers to keep the film stably transported and prevent the film from being deformed and increased.

Advantageous effects

Compared with the prior art, the invention has the following beneficial effects:

1. the camera is adopted to shoot the fold reflection degree and the edge tightness degree of the film surface for analysis, when the speed of the film transmission device and the distance and the parallelism between the rollers are not adjusted properly, the fold reflection condition and the edge tightness degree of the film surface are changed obviously, and the computer is very favorable for analyzing and machine learning judgment, so that the film is prevented from generating overlarge deformation during transmission, and the printing quality of the film is improved;

2. the used camera and the used transportation device are simple in structure, the installation position of the camera does not obstruct the transportation route of the film, and the additional arrangement space of equipment is not required to be increased.

Drawings

FIG. 1 is a schematic view of a thin film transfer device according to the present invention;

FIG. 2 is a schematic diagram of a three-dimensional structure of a placement position of a camera in the present invention;

FIG. 3 is a schematic view of the structure of FIG. 2 from another angle;

FIG. 4 is a top view of FIG. 2;

FIG. 5 is a schematic view of camera-observed overtightening of a film fold;

FIG. 6 is a schematic view of camera-observed film edge bending over slack;

figure 7 is a side view of the film in an over-stretched condition.

Reference numerals

A roller 1, a film 2, a lighting lamp 3, a camera 4, a fold 5, a film edge 6 and a film middle part 7.

Detailed Description

For a better illustration of the invention, reference is made to the following description, taken in conjunction with the accompanying drawings and examples:

with reference to fig. 1-7, the present invention discloses a film adjusting method based on machine vision, which comprises the following process steps:

1) lighting up above the film 2 in the film transport device;

2) arranging two cameras 4 and aligning the surface of the film 2 at different angles to prepare for shooting;

3) and the computer performs machine learning according to the reflection condition and the edge tightness condition of the film 2 shot by the camera 4, so that the deformation degree analysis result of the film 2 is determined according to the image shot by the camera 4, the speed of the film conveying device and the distance and the parallelism between the rollers 1 are adjusted, and the stable conveying of the film is ensured.

Further comprises the following steps: the film conveying device in the step 1) comprises two rollers 1, wherein a film 2 is arranged on each roller 1, and an illuminating lamp 3 is arranged on the upper side of each film 2.

Further comprises the following steps: the two cameras 4 in the step 2) are placed above the film conveying device at different angles and shoot the surface of the film 2, and the installation positions of the two cameras 4 are on the central axis of the wide side of the film 2.

Further comprises the following steps: determining the deformation degree analysis result of the film 2 according to the image shot by the camera 4 in the step 3) and adjusting the film conveying device:

1. the camera 4 shoots images and transmits the images to a computer;

2. and the computer analyzes the surface reflection and edge tightness of the film 2 and adjusts the speed of the rollers 1 and the distance and the parallelism among the rollers 1 to keep the stable transportation of the film and prevent the deformation of the film from increasing.

Specifically, the two rollers 1 rotate at a specific speed, direction, distance and parallelism to drive the film 2, and then the illuminating lamp 3 is placed on the film, so that the camera 4 can observe the light reflection condition of the surface of the film 2;

two cameras 4 are positioned and directed at the surface of the film 2 at different angles; the cameras 4 are respectively arranged on the oblique front upper part and the oblique rear upper part of the film 2, the two cameras 4 form an angle of 180 degrees, namely the mounting positions of the two cameras 4 are on the central axis of the wide edge of the film 2; determining the deformation degree of the film 2 by machine learning according to the reflection of the film 2 and the tightness of the edge; the computer is used for machine learning according to the image shot by the camera 4, if the relative speed of the transportation is too slow, the film 2 loosens, the middle part 7 of the film sags to drive the edge 6 of the film to deform, and the conditions can be captured by the camera 4, then the camera 4 transmits the information to the computer, the computer can increase the relative transmission speed of the two rollers 1 or increase the relative distance of the two rollers 1, if the relative speed of the two rollers 1 is higher, the surface of the film 2 becomes wrinkled 5, the reflection of light changes as shown in figure 5 and is captured by the camera 4, at the moment, the computer can reduce the relative speed of the two rollers 2 or reduce the relative distance between the two rollers 2, if the light on the film 2 is not very uniform, the problem that the parallelism of the two rollers 1 is insufficient is likely to occur, the parallelism of the two rollers 1 can then be adjusted to keep the film 2 transported smoothly.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the technical solutions of the present invention have been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that the technical solutions described in the foregoing embodiments can be modified or some technical features can be replaced equally; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of machine vision based film conditioning comprising the steps of:

1) illuminating above the film (2) in the film transport device;

2) arranging two cameras (4) and aiming at the surface of the film (2) at different angles to prepare for shooting;

3) and the computer performs machine learning according to the reflection and edge tightness of the film (2) shot by the camera (4), so that the deformation degree analysis result of the film (2) is determined according to the image shot by the camera (4), the speed of the film conveying device and the distance and the parallelism between the rollers (1) are adjusted, and the stable conveying of the film is ensured.

2. The known machine vision based film conditioning method according to claim 1, characterized in that: the film conveying device in the step 1) comprises two rollers (1), a film (2) is arranged on each roller (1), and an illuminating lamp (3) is arranged on the upper side of each film (2).

3. The known machine vision based film conditioning method according to claim 2, characterized in that: the two cameras (4) in the step 2) are placed above the film transmission device at different angles and shoot the surface of the film (2), and the installation positions of the two cameras (4) are on the central axis of the wide side of the film (2).

4. The known machine vision based film conditioning method according to claim 3, characterized in that: determining a deformation degree analysis result of the film (2) according to the image shot by the camera (4) in the step 3) and adjusting the film conveying device:

1) the camera (4) shoots images and transmits the images to the computer;

2) the computer analyzes the surface reflection and edge tightness of the film (2) and adjusts the speed of the rollers (1), the distance between the rollers (1) and the parallelism to keep the film stably transported and prevent the film from increasing in deformation.