CN219153837U

CN219153837U - Mount paper location laminating device

Info

Publication number: CN219153837U
Application number: CN202320210548.3U
Authority: CN
Inventors: 何春红
Original assignee: Dongguan City College
Current assignee: Dongguan City College
Priority date: 2023-02-14
Filing date: 2023-02-14
Publication date: 2023-06-09
Anticipated expiration: 2033-02-14

Abstract

The utility model relates to the technical field of positioning and laminating, and discloses a mounting paper positioning and laminating device, which comprises: the first camera is arranged on the bracket; the attaching station is positioned below the camera; a conveyor belt for transporting articles; the second camera is positioned above one conveying belt; the station to be grabbed is positioned on the conveying belt; the box body is placed on the conveying belt; and the manipulator is used for grabbing articles. The utility model provides a mounting paper positioning and laminating device, which determines an image acquisition mode suitable for a mounting paper positioning and laminating imaging system, and can avoid the phenomenon of optical deformation during sampling to a certain extent; the positioning accuracy is high, promotes the laminating speed.

Description

Mount paper location laminating device

Technical Field

The utility model relates to the field of positioning and laminating, in particular to a paper mounting positioning and laminating device.

Background

At present, the lamination of the paper mounting and box body of the top and bottom cover box making machine adopts a photoelectric tracking hydraulic servo deviation rectifying and positioning technology, and the positioning and lamination procedures of the box body are realized by gradually searching the paper mounting edge through a photoelectric sensor arranged at the front end of the clamp. The method has the problems of low positioning precision, low attaching speed, poor correction capability, complex mechanical structure, weak equipment expansibility and the like

For solving above-mentioned problem, propose a mount paper location laminating device in this application.

Disclosure of Invention

Object of the utility model

In order to solve the technical problems in the background art, the utility model provides a mounting paper positioning and laminating device, which determines an image acquisition mode suitable for a mounting paper positioning and laminating imaging system and can avoid the phenomenon of optical deformation in sampling to a certain extent; the positioning accuracy is high, promotes the laminating speed.

(II) technical scheme

In order to solve the above problems, the present utility model provides a mounting paper positioning and attaching device, comprising:

the first camera is arranged on the bracket;

the attaching station is positioned below the camera;

a conveyor belt for transporting articles;

the second camera is positioned above one conveying belt;

the station to be grabbed is positioned on the conveying belt;

the box body is placed on the conveying belt;

and the manipulator is used for grabbing articles.

Preferably, the manipulator is a four-axis manipulator.

Preferably, a mount paper is placed on the conveyor belt.

Preferably, the first camera (2) and the second camera (6) are high-definition industrial cameras.

Preferably, the camera I (2) and the attaching station (3) are vertically arranged.

The technical scheme of the utility model has the following beneficial technical effects:

the rapid detection of the deflection angle position information of the image edge and the detected object is realized; the phenomenon of optical deformation during sampling can be avoided to a certain extent; the positioning accuracy is high, promotes the laminating speed.

Drawings

Fig. 1 is a schematic structural diagram of a mounting paper positioning and laminating device provided by the utility model.

Fig. 2 is a diagram of a camera calibration method in a mounting paper positioning and laminating device according to the present utility model.

Fig. 3 is a diagram of a servo control system in the paper mounting positioning and laminating device provided by the utility model.

Reference numerals: 1. mounting paper; 2. a first camera; 3. a laminating station; 5. a conveyor belt; 6. a second camera; 7. a station to be grabbed; 8. a case body; 9. and a manipulator.

Detailed Description

The objects, technical solutions and advantages of the present utility model will become more apparent by the following detailed description of the present utility model with reference to the accompanying drawings. It should be understood that the description is only illustrative and is not intended to limit the scope of the utility model. In addition, in the following description, descriptions of well-known structures and techniques are omitted so as not to unnecessarily obscure the present utility model.

As shown in fig. 1-3, the present utility model provides a mounting paper positioning and attaching device, which includes:

a first camera 2 mounted on the bracket;

the attaching station 3 is positioned below the camera;

a conveyor belt 5 for conveying articles; a paper mounting 1 is arranged on the conveying belt 5;

a second camera 6, which is located above one of the conveyor belts 5;

a station 7 to be grabbed, which is positioned on the conveying belt 5;

the box body 8 is placed on the conveying belt 5;

a manipulator 9 for gripping the article; specifically, the manipulator 9 is a four-axis manipulator.

The first camera (2) and the second camera (6) are high-definition industrial cameras.

The first camera (2) and the attaching station (3) are vertically arranged.

The number of the conveyer belts 5 is two, and the conveyer belts are vertically arranged. The first camera 2 and the second camera 6 are industrial cameras.

Aiming at the requirements of higher success rate of image edge extraction and higher positioning precision, the advantages and disadvantages of each traditional edge detection operator are analyzed through result comparison, and an adaptive threshold Canny image edge detection method based on a discipline algorithm is provided on the basis, so that automatic detection of image edge contours is realized; based on the concept of combining thickness with segmentation of image areas, the Hough transform operator is used for realizing rapid detection of the image edge space position information.

Feature extraction of images: in order to improve the image edge extraction precision, after the traditional edge detection operator is subjected to preliminary analysis, the image edge extraction is determined to be carried out by adopting a Canny operator with stronger adaptability. On the premise of obtaining a clearer image contour, the Hough transformation operator is used for realizing the rapid detection of the image edge and the deflection angle position information of the measured object;

the method is characterized in that the accurate extraction of the spatial position information of the box body and the paper mounting is realized, firstly, the images are processed, the contour edges of the images are extracted by using a Canny operator, the spatial position information extraction is completed by a Hough transformation algorithm on the basis, and the process realized by the self-adaptive Canny edge detection algorithm is decomposed into the following steps:

s1, reading and calling an rgb2gray ('lena') function to convert a true color image into a gray image;

s2, smoothing the image by using a Gaussian filter to eliminate noise;

s3, calculating the gradient amplitude and direction of the pixel points at the edge of the image;

s4, suppressing non-maximum value to eliminate false response to the edge;

s5, self-adaptive threshold calculation and direction and threshold judgment;

s6, displaying the target image.

It should be noted that, the first camera 2 and the second camera 6 form a visual detection unit.

Further, in order to avoid the phenomenon that the pictures sampled by the camera are blurred and distorted due to vibration generated in the movement of the manipulator:

referring to fig. 1, the flying vision imaging system is a mode of indirectly collecting image information of an element through the principle of optical reflection of a plane mirror, an image collecting unit is generally composed of two cameras, the two cameras are respectively and fixedly installed on a machine frame and the tail end of a pick-up head and synchronously move along with the pick-up head, the camera installation mode of the paper mounting positioning and laminating station vision imaging system is improved on the basis of flying vision, the camera used for collecting position information of a box body is fixedly installed on a bracket, the other camera is fixedly installed above a paper mounting and laminating station, the mode cancels the plane mirror indirect image collecting mode, and the phenomenon of optical deformation during sampling can be avoided to a certain extent.

In an alternative embodiment, the machine vision system is further comprised of:

the system consists of a motion and control part of an image acquisition and processing and box making platform;

the flow is as follows:

placing a workpiece on a workbench, simply fixing the workpiece by a clamp, controlling a camera to move to a fixed position of the workbench, photographing films on the workpiece to acquire images, and automatically processing and fitting the images to obtain the outline of a positioning area after each image acquisition;

the initial coordinate of the box making, which is extracted through image processing and analysis in the process of program running, is a pixel coordinate taking an image coordinate system as a reference, and cannot be directly used as a physical coordinate for driving a motion platform, and a visual coordinate system is required to be calibrated, so that the extracted pixel coordinate is converted into a physical coordinate which can be identified by the motion platform;

since there is a small amount of noise on the image captured by the camera that affects the subsequent processing of the image, image enhancement processing is required on the resulting grayscale image. Methods of image enhancement fall into two categories: the enhancement method based on the spatial domain and the frequency domain, wherein the low-pass filtering method in the enhancement method of the frequency domain can well reduce noise in the image and avoid the noise from affecting the processing of the image. In the subsequent image processing process, certain grabbing errors are also caused due to the problem of contour transition, so that further binarization processing is needed for the enhanced image, and a binary image with more vivid contour contrast is obtained by adjusting the gray image threshold value, so that the efficiency of image processing is improved, and the recognition capability of a system is improved. Since the binarized image cannot completely exclude the interference of some fine areas on the image capturing, a morphological processing method can be adopted at this time. And (3) adopting a certain structural unit to perform corrosion operation on the binary image to remove parts with smaller areas, and then recovering the size of the original identification area through expansion operation. Therefore, the identification area is not reduced, and the tiny area interfering with contour fitting is removed at the same time, so that an accurate coordinate value is obtained.

Referring to fig. 2, imaging system camera calibration and optimization:

in order to improve the accuracy and precision of the monocular vision system for extracting the spatial position information of the object, the camera needs to be calibrated. The key step of camera calibration is to build a reasonable geometric imaging model of the camera, wherein the model is a mathematical description of the physical imaging process of an actual camera, and the determination of internal and external parameters of the camera, particularly the distortion coefficient and the effective focal length of a lens, are key to camera calibration. The imaging process of things is essentially the mapping conversion relation of physical space points, and the mapping relation is determined by camera geometric imaging model parameters and is also the key for establishing the geometric relation between the space points and the pixel points. And (3) solving the internal and external parameters of the camera and the mapping relation between the space points and the pixel points by establishing a geometric imaging model of the camera, namely calibrating the camera.

The calibration of the camera is divided into two parts, internal and external parameters, respectively. The optical characteristics of the camera are represented by internal parameters including focal length, distortion coefficient of the lens and scale factor; the external parameters of the camera reflect the mutual mapping and position conversion relationship between the camera coordinate system and the world coordinate system.

At present, a perfect theoretical basis and a perfect method system are formed by researching a camera calibration method. The camera calibration method can be roughly classified into a conventional camera calibration method, an active vision camera calibration method, and a camera self-calibration method. Combining with the actual research background of the project, under the condition of higher requirements on positioning and attaching precision of the mounting paper, the calibration method improved based on the traditional camera calibration principle is selected in the aspect of comprehensively considering the calibration precision and difficulty in realization.

After years of development, the latter optimizes and improves the traditional camera calibration method, wherein the RAC method of Tsai and the plane calibration method of Zhang Zhengyou are more prominent and widely adopted. The RAC method of Tsai is also called a two-step method, and is a calibration method based on radial constraints. Firstly, obtaining external parameters of a camera, and secondly, obtaining internal parameters; according to the existence of distortion of the camera lens, the camera parameters are solved by adopting an over-determined equation summation nonlinear optimization method, and the method is moderate in solving operation difficulty and high in accuracy. The calibration method based on the plane target is proposed by Zhang Zhengyou and the like, is simple and feasible, and becomes a common camera calibration method in the field of machine vision. The calibration method based on the plane target needs to shoot at least 3 pieces of chessboard paper pictures from different angles, the camera and the target have no fixed constraint, and the calibration method can move in the visual field of the camera without specific motion parameters. In order to improve the success rate of camera calibration and reduce the working strength, a plane calibration method proposed by Zhang Zhengyou and the like is adopted. The specific calibration method principle is shown in figure 2. The calibration of the camera at least needs to sample the target pictures from three different angles, so that homography matrixes corresponding to each picture are different. When solving the homography matrix, the extracted image feature points at least cannot be smaller than four, and the more the number of the feature points is, the higher the calibration precision is, which is why the Zhang's calibration method adopts chessboard paper for calibration. The calibration experiment of the camera is carried out in Matlab software, and the calibration work of the camera parameters is completed by calling a camera calibration kit Camera Calibration Toolbox built in Matlab and adopting a plane target-based calibration method proposed by Zhang Zhengyou et al.

Further, referring to fig. 3, the visual servo positioning and attaching system:

and a hardware control module of an executing mechanism of the paper mounting positioning and laminating device is built by adopting an upper computer control strategy based on a PC and a motion control card, and integration with a visual detection system is completed in LabVIEW software. The mixed editing and developing of LabVIEW and Matlab software are realized through a node calling technology (CLF), and an operation interface of a visual servo control system is designed by means of strong graphic program editing capability of LabVIEW software.

The machine vision servo positioning and attaching system consists of a vision detection unit and a manipulator servo motion control system, and the integrated design of system software and hardware is completed in LabVIEW software. The LabVIEW software has strong software and hardware integration and system fusion capabilities, and provides a special software development tool for machine vision and motion control, so that the LabVIEW software is adopted to design upper computer control software and a system operation interface for the construction of the machine vision servo control system. The hardware architecture of the paper mounting positioning and laminating device consists of a high-definition industrial camera, an upper computer, a stepping driver and a four-axis planetary reduction manipulator. In order to improve the flexibility and the later expansion capability of the system, the upper computer adopts a mode of PC plus four-axis operation control card, and fully exerts the powerful operation capability and abundant hardware interface advantages of the PC.

An automatic vision calibration recognition system and an image rapid processing system:

machine vision systems include both machine (mechanical, motion, control) and vision (hardware and software). The system is composed of a box making platform, a box control platform and a box control platform. The flow is as follows: the workpiece is placed on a workbench, the workpiece is simply fixed by a clamp, then a camera is controlled to move to a fixed position of the workbench, a film on the workpiece is photographed and images are collected, and after the images are collected each time, the images are automatically processed and fitted to form the outline of a positioning area.

The initial coordinate of the box making, which is extracted through image processing and analysis in the process of program running, is a pixel coordinate taking an image coordinate system as a reference, and cannot be directly used as a physical coordinate for driving a motion platform, and a visual coordinate system is required to be calibrated. The extracted pixel coordinates are converted into physical coordinates recognizable by the motion platform.

Since there is a small amount of noise on the image captured by the camera that affects the subsequent processing of the image, image enhancement processing is required on the resulting grayscale image. Methods of image enhancement fall into two categories: the enhancement method based on the spatial domain and the frequency domain, wherein the low-pass filtering method in the enhancement method of the frequency domain can well reduce noise in the image and avoid the noise from affecting the processing of the image. In the subsequent image processing process, a certain grabbing error is brought due to the problem of contour transition, so that further binarization processing is needed for the enhanced image. The binary image with more vivid contour contrast is obtained by adjusting the gray image threshold value, so that the image processing efficiency is improved, and the recognition capability of the system is improved. Since the binarized image cannot completely exclude the interference of some fine areas on the image capturing, a morphological processing method can be adopted at this time. And (3) adopting a certain structural unit to perform corrosion operation on the binary image to remove parts with smaller areas, and then recovering the size of the original identification area through expansion operation. Therefore, the identification area is not reduced, and the tiny area interfering with contour fitting is removed at the same time, so that an accurate coordinate value is obtained.

According to the utility model, aiming at the requirements of image acquisition and mounting paper positioning and laminating processes, a flight visual imaging concept is introduced and a typical flight visual imaging process and principle are analyzed. In order to improve the imaging quality and stability of a camera, the phenomena of blurring and distortion of pictures caused by vibration in the imaging process of the camera are avoided, a flight vision imaging hardware architecture is improved, and the camera arranged at a picking-up position at the tail end of a manipulator is fixedly arranged on a bracket. Through an actual image sampling effect experiment, the reasonable and reliable arrangement mode of the camera is verified.

2. The self-adaptive threshold Canny image edge detection method based on the discipline algorithm realizes automatic detection of the image edge contour. Aiming at the requirements of higher success rate of image edge extraction and higher positioning precision, the advantages and disadvantages of each traditional edge detection operator are compared and analyzed through experimental results, an adaptive threshold Canny image edge detection method based on a discipline algorithm is provided on the basis, and automatic detection of image edge contours is realized; based on the idea of combining thickness with segmentation of image areas, the Hough transform operator is used for realizing the rapid detection of the image edge space position information; and the feasibility of the operator is verified through experiments, and the experimental result meets the requirements of paper mounting, positioning and laminating.

3. A full-automatic high-speed image recognition correction system is designed. And accurately positioning filling positions of various filling product types through full-automatic height correction and marker identification correction and automatically adjusting the filling height.

It is to be understood that the above-described embodiments of the present utility model are merely illustrative of or explanation of the principles of the present utility model and are in no way limiting of the utility model. Accordingly, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present utility model should be included in the scope of the present utility model. Furthermore, the appended claims are intended to cover all such changes and modifications that fall within the scope and boundary of the appended claims, or equivalents of such scope and boundary.

Claims

1. Mount paper location laminating device, its characterized in that includes:

a camera I (2) mounted on the bracket;

a fitting station (3) positioned below the camera;

a conveyor belt (5) for transporting the articles;

a second camera (6) positioned above one of the conveyor belts (5);

the station (7) to be grabbed is positioned on the conveying belt (5);

the box body (8) is placed on the conveying belt (5);

and the manipulator (9) is used for grabbing articles.

2. A mounting positioning and laminating device according to claim 1, characterized in that the manipulator (9) is a four-axis manipulator.

3. A mounting positioning and laminating device according to claim 2, characterized in that the mounting (1) is placed on the conveyor belt (5).

4. The laminating device for positioning paper mounting according to claim 1, wherein the first camera (2) and the second camera (6) are high-definition industrial cameras.

5. A mounting positioning and laminating device according to claim 1, characterized in that the camera one (2) and the laminating station (3) are arranged vertically.