Broken yarn and head finding device based on machine vision
Technical Field
The invention relates to a broken yarn and end-finding device based on machine vision and a using method thereof, belonging to the technical field of automatic joint of ring spun yarn.
Background
The spinning process is an important link of the spinning process, and the problem of yarn breakage (the phenomenon that continuous spun yarn strips are broken between the spun yarn output from a front roller and a bobbin) in the spinning production process directly causes interruption of the spinning production process and seriously affects the spinning production efficiency. The broken yarn joint work of the ring spinning frame which accounts for the largest percentage in the spinning frame still needs to be manually completed by a skilled vehicle stopping worker, the labor intensity of the worker is increased, the production intelligence and informatization of textile enterprises are reduced, the automatic joint device of the ring spinning frame which can be put into use is not produced autonomously at home at present, and although the automatic joint device is shown in international spinning machine exhibition abroad, the technical development of the domestic automatic joint technology is lagged due to technical blockade.
To realize the automatic piecing of the spinning frame, the automatic end-finding of broken yarn is firstly completed, including the searching and capturing of the broken yarn, and then the broken yarn is fed into a front roller after passing through a steel wire ring and a yarn guide hook to complete piecing. The main methods for finding and capturing broken yarns in the traditional automatic piecing device can be divided into two types, one is to suck the broken yarns through negative pressure airflow, and the other is to use felt to rub the surface of a bobbin to glue the broken yarns. The first method has the defects that after yarn breakage, because the spindle still rotates at high speed, the broken yarn is tightly attached to the surface of the bobbin, and the success rate of direct suction is low; the second type of method has the disadvantage of large operating area and requires spindle counter-rotation. The common disadvantage of both types of methods is that there is no post-broken yarn capture detection, i.e. it is not known whether the capture has been successful. Therefore, the searching and capturing of the broken yarn of the traditional automatic piecing device still stays at the stage of blindman image, the success rate of finding the yarn is low, and the time consumption is long.
Disclosure of Invention
The invention aims to solve the technical problem of automatic broken yarn searching and capturing in the ring spinning spun yarn production process.
In order to solve the problems, the technical scheme adopted by the invention is to provide a broken yarn finding device based on machine vision, which comprises a robot, a rotary spindle, an air pump device, a suction nozzle, a hairbrush, a machine vision system and a control cabinet; a robot is correspondingly arranged on one side of the rotating spindle for mounting the yarn breaking bobbin, and a brush and a suction nozzle are arranged on a tail end executing mechanism arranged on the robot; the suction nozzle is connected with an air pump device arranged on one side of the robot; a machine vision system is correspondingly arranged on the yarn breaking bobbin arranged on the rotating spindle; the robot, the rotary spindle, the air pump device and the machine vision system are connected with the control cabinet.
Preferably, the robot is set as a six-degree-of-freedom industrial robot.
Preferably, the rotating spindle is provided with a device which can rotate around the central shaft of the rotating spindle in an infinite angle and is used for completing the forward and reverse rotation and angle control of the bobbin.
Preferably, the air pump device is a device capable of continuously generating negative pressure air flow, and comprises a negative pressure air pump and an air pressure conduit; the negative pressure air pump is a pump or a compressor capable of conveying an air medium.
Preferably, the brush is a brush or a felt tool that separates the yarn breakage attached to the surface of the bobbin from the bobbin by rubbing the surface of the bobbin.
Preferably, the machine vision system includes an industrial camera, a light source and a background board for acquiring and processing images of broken yarns.
Preferably, the control cabinet comprises an industrial control system, a rotary spindle drive, a drive for an air pump device and a drive for a robot.
Preferably, the suction nozzle is in a horn shape, one end of the suction nozzle is connected to the negative pressure air pump through an air pressure conduit, and the other end of the suction nozzle is provided with a nozzle opening for sucking broken yarns.
Preferably, the robot is provided with an end actuating mechanism which is provided with an end suction station for taking and placing the broken yarn bobbin from the spinning frame to the rotating spindle, a brushing-off station for moving the brush to the surface of the yarn bobbin, and a suction station for moving the suction nozzle to the broken yarn suction point; and the switching of different stations is completed through the rotation of the robot tail end executing mechanism.
The invention also provides a working method of the broken yarn and head-finding device based on machine vision, which comprises the following steps:
step 1: the robot takes and places the broken yarn bobbin from the spinning frame to the rotating spindle;
step 2: after the mechanical arm puts the broken yarn bobbin on the rotating spindle, the rotating spindle starts to drive the yarn bobbin to rotate around a certain direction; when the rotating spindle starts to rotate, the industrial camera collects the broken yarn pictures in real time and uploads the broken yarn pictures to an embedded system of the control cabinet for processing, so that the preprocessing of the broken yarn pictures and the recognition and positioning of the broken yarn contours are completed; if the broken yarn contour is not detected, namely the broken yarn is attached to the surface of the bobbin, the tail end of the robot is switched to a brushing station, a hairbrush is moved to the surface of the bobbin, the broken yarn is exposed outside the bobbin through the rotation of a rotating spindle in the unwinding direction, and then the detection and the positioning of the broken yarn are completed; if the broken yarn is not attached to the surface of the bobbin, the detection and the positioning of the broken yarn can be directly carried out;
and step 3: after the broken yarn is positioned, converting the pixel coordinate of the broken yarn into a robot coordinate and calculating a suction position, converting the tail end of the robot into a suction station and moving a suction nozzle to the suction position;
and 4, step 4: starting an air pump device to complete the capture of broken yarns;
and 5: the bobbin is unwound, in order to conveniently detect whether the yarn is successfully captured or not, the rotating spindle needs to rotate for a fixed angle in the unwinding direction of the yarn again, meanwhile, the suction nozzle moves for a fixed distance in the direction away from the bobbin, and the picture after the suction is collected again and uploaded to an embedded system for processing;
step 6: if the yarn breakage is detected to be captured, the whole automatic yarn breakage end-finding process is finished, and if not, the step 2 is returned to find the ends again.
The invention introduces a machine vision method to complete the identification and positioning of broken yarns, finishes the capture of the broken yarns by using a brush separation and negative pressure suction mode, and uses the machine vision to detect the capture result after the capture is finished, thereby finally finishing the automatic end-finding of the broken yarns. The invention solves the problems of long yarn searching and capturing time and low efficiency of the automatic piecing device of the existing ring spinning frame.
Compared with the prior art, the invention has the following beneficial effects:
1. a broken yarn detection method based on machine vision is introduced, the identification and the positioning aiming at the broken yarn are realized, and the defects of large broken yarn absorption operation area and long consumed time in the prior art are overcome;
2. the capture mode of combining the brush and the adsorption ensures that broken yarns attached to the bobbin can be successfully captured, and solves the problem that negative pressure adsorption in the prior art has poor capture effect on broken yarns attached to the surface of the bobbin.
3. And the detection after suction based on machine vision is introduced to ensure the successful capture of broken yarns, thereby greatly improving the successful power of the capture of broken yarns.
4. The hairbrush and the suction nozzle are assembled at the tail end of the robot, bobbin taking and placing, broken yarn brushing and broken yarn capturing stations are switched through the change of the pose of the tail end of the robot, broken yarn searching and capturing can be completed by only one set of robot, the mechanical structure of the automatic broken yarn finding device is greatly simplified, and the defect that the mechanical mechanism of the automatic broken yarn finding device in the prior art is complex is overcome.
Drawings
FIG. 1 is a schematic structural diagram of a broken yarn automatic head-finding device based on machine vision according to an embodiment of the invention;
FIG. 2 is a schematic view of an end effector mechanism of a robot in accordance with an embodiment of the present invention;
FIG. 3 is a general flow chart of automatic broken yarn end-changing according to the embodiment of the invention;
FIG. 3 is a flow chart of automatic broken yarn end-changing; the figure b is a broken yarn detection algorithm flow chart;
reference numerals: 1. the bobbin picking and placing device comprises a control cabinet 2, a robot 3, an air pump device 4, a robot tail end actuating mechanism 5, a brush 6, a suction nozzle 7, a background plate 8, a bobbin 9, a machine vision system 10, a rotating spindle 11, a suction nozzle opening 12, a suction nozzle opening 13 and a bobbin picking and placing device;
Detailed Description
In order to make the invention more comprehensible, preferred embodiments are described in detail below with reference to the accompanying drawings:
as shown in fig. 1-3, the invention provides a machine vision-based broken yarn and head-finding device, which comprises a robot 2, a rotating spindle 10, an air pump device 3, a suction nozzle 6, a brush 5, a machine vision system 9 and a control cabinet 1; a robot 2 is correspondingly arranged on one side of a rotating spindle 10 for mounting a yarn-breaking bobbin 8, and a brush 5 and a suction nozzle 6 are arranged on a tail end actuating mechanism 4 arranged on the robot 2; the suction nozzle 6 is connected with an air pump device 3 arranged on one side of the robot 2; a machine vision system 9 is correspondingly arranged on a yarn breaking bobbin 8 arranged on the rotating spindle 10; the robot 2, the rotary spindle 10, the air pump device 3 and the machine vision system 9 are connected with the control cabinet 1. The robot 2 is a six-degree-of-freedom industrial robot, and the rotating spindle 10 is a device which can rotate around the central axis of the rotating spindle in an infinite angle and is used for finishing the forward and reverse rotation and angle control of the bobbin 8. The air pump device 3 is a device capable of continuously generating negative pressure airflow and comprises a negative pressure air pump and an air pressure guide pipe; the negative pressure air pump is a pump or a compressor capable of conveying an air medium. The brush 5 is a brush or a felt-like tool that can separate the yarn breakage attached to the surface of the bobbin from the bobbin by rubbing on the surface of the bobbin. The machine vision system 9 includes an industrial camera, light source and background board 7 for capturing and processing images of broken yarn. The control cabinet 1 comprises an industrial control system, a rotary spindle driving device, a driving device of an air pump device and a driving device of a robot. The suction nozzle 6 is in a horn shape, one end of the suction nozzle is connected to the negative pressure air pump through an air pressure conduit, and the other end of the suction nozzle is provided with a nozzle opening which is opened and used for sucking broken yarns. The tail end executing mechanism 4 arranged on the robot 2 is provided with a tail end sucking station for taking and placing the broken yarn bobbin 8 from the spinning frame to the rotating spindle 10, a brushing-off station for moving a brush to the surface of the bobbin and a sucking station for moving a suction nozzle to a broken yarn sucking point; and the switching of different stations is completed through the rotation of the robot tail end executing mechanism.
A working method of a yarn breaking and end changing device based on machine vision is characterized in that: the method comprises the following steps:
step 1: the robot 2 takes and places the broken yarn bobbin 8 from the spinning frame to the rotating spindle 10;
step 2: the rotating spindle 10 starts to drive the bobbin 8 to rotate around a certain direction after the mechanical arm puts the bobbin with yarn broken on the rotating spindle; when the rotating spindle 10 starts to rotate, the industrial camera collects the broken yarn pictures in real time and uploads the broken yarn pictures to the embedded system of the control cabinet 1 for processing, so that the preprocessing of the broken yarn pictures and the recognition and positioning of the broken yarn contours are completed; if the broken yarn contour is not detected, namely the broken yarn is attached to the surface of the bobbin, the tail end of the robot is switched to a brushing station, the hairbrush 5 is moved to the surface of the bobbin, the broken yarn is exposed outside the bobbin through the rotation of the rotating spindle in the unwinding direction, and then the detection and the positioning of the broken yarn are completed; if the broken yarn is not attached to the surface of the bobbin, the detection and the positioning of the broken yarn can be directly carried out;
and step 3: after the broken yarn is positioned, converting the pixel coordinate of the broken yarn into a robot coordinate and calculating a suction position, converting the tail end of the robot into a suction station and moving the suction nozzle 6 to the suction position;
and 4, step 4: starting the air pump device 3 to complete the capture of broken yarns;
and 5: the bobbin is unwound, in order to conveniently detect whether the yarn is successfully captured or not, the rotating spindle 10 needs to rotate for a fixed angle in the unwinding direction of the yarn again, meanwhile, the suction nozzle 6 moves for a fixed distance in the direction away from the bobbin 8, and the picture after the suction is collected again and uploaded to an embedded system for processing;
step 6: if the yarn breakage is detected to be captured, the whole automatic yarn breakage end-finding process is finished, and if not, the step 2 is returned to find the ends again.
As shown in fig. 1: a broken yarn and head-finding device based on machine vision comprises the following parts: the robot comprises a robot 2, a rotary spindle 10, an air pump device 3, a hairbrush 5, an industrial camera, a light source, a background plate 7 and a control cabinet 1. The parts of the device are described as follows: the robot 2 adopts a common six-degree-of-freedom industrial robot and is used for taking and placing bobbins and moving a brush and a suction nozzle; the rotating spindle 10 is driven by a stepping motor and is used for driving the bobbin 8 to complete forward and reverse rotation and angle control; the air pump device 3 is a device capable of continuously generating negative pressure airflow, comprises a negative pressure air pump, an air pressure conduit and a suction nozzle and is used for completing the adsorption of broken yarns; the brush 5 is made of nylon yarn and is used for rubbing the surface of the bobbin to separate broken yarn attached to the surface of the bobbin from the bobbin when the broken yarn is attached to the surface of the bobbin; the industrial camera is a camera commonly used in industrial production and is used for acquiring images of broken yarns. The light source is an LED light source and is used for providing a light source for an industrial camera to obtain photos. The background plate 7 is a pure black acrylic plate, and serves as a background in a bobbin photograph captured by a camera to improve the contrast between broken yarn and the background. The core component in the control cabinet 1 is an embedded system raspberry pie, and is used for processing broken yarn images, controlling the motion of a robot, controlling a rotary spindle and controlling an air pump.
The brush 5 and the suction nozzle 6 are both assembled on the end executing mechanism 4 of the robot 2, the switching of the end sucking station, the brushing station and the sucking station of the robot is completed through the rotation of the end of the robot, and the bobbin taking and placing, the brush moving and the suction nozzle moving are respectively completed. The rotating spindle 10 starts to rotate around a certain direction after the mechanical arm takes and places the broken yarn bobbin 8 from the spinning frame onto the rotating spindle 10, and a camera acquires a broken yarn photo; after the completion of the suction of the broken yarn, the rotating spindle needs to be rotated again by a fixed angle in the direction of the yarn unwinding for the convenience of the detection of whether the catching was successful.
After recognizing and completing the positioning of the broken yarn, the suction nozzle 6 is moved to a broken yarn suction position; the air pump is started after the suction nozzle 6 moves to the broken yarn suction position, and the broken yarn is captured; when the rotating spindle 10 starts to rotate, the industrial camera acquires a real-time picture of a bobbin and uploads the real-time picture to a raspberry group to finish the processing of a broken yarn picture; after the broken yarn is sucked, the bobbin is unwound, the suction nozzle moves backwards, the picture of the bobbin and the suction nozzle area is obtained again and is uploaded to a raspberry pie, and whether the broken yarn is captured successfully or not is detected. An embedded system raspberry group carries a Linux system, a master control program is compiled by using a Python programming language, and an OpenCV (computer vision library) library is called to complete preprocessing of a broken yarn image, identification and positioning of a broken yarn contour and detection of whether the broken yarn is successfully absorbed or not after being absorbed; and the raspberry-shaped expansion board is used for completing the cooperative control of the stepping motor in the rotating spindle 10, the robot 2 and the air pump in the air pump device 3.
The rotating spindle 10 is positioned at one side of the robot, and the forward and reverse rotation and angle control of the rotating spindle are completed by using a stepping motor. In order to improve the image processing effect of broken yarn, a black background plate 7 different from white yarn is placed behind the rotating spindle 10, and a camera and a light source are placed in front of the rotating spindle 10 to complete the collection of the broken yarn image. The negative pressure air pump in the air pump device 3 adopts a double-pump-head diaphragm pump driven by a direct current motor, is arranged on the other side of the robot 2, and when the two pump heads are connected in series, the negative pressure can reach-90 Kpa, and the flow is about 16L/min.
An industrial control system in the control cabinet 1 uses an embedded system raspberry pi, a master control program is compiled by using a Python programming language, and an OpenCV (computer vision library) library is called to complete preprocessing of a broken yarn image, identification and positioning of a broken yarn contour and detection of whether the broken yarn is successfully absorbed or not after being absorbed; and the raspberry pi expansion board is used for completing the cooperative control of the stepping motor, the robot 2 and the air pump 3 in the rotary spindle 10.
The robot end actuator 4 is shown in fig. 2, wherein a brush 5 and a suction nozzle 6 are both assembled on the robot end actuator 4, and the robot end actuator 4 is provided with a bobbin taking and placing device 13, a suction nozzle opening 11 and a suction nozzle interface 12; the suction nozzle interface 12 is connected to the negative pressure air pump through a conduit, when the tail end of the robot is positioned at a bobbin taking and placing station, the brush 5 and the suction nozzle 6 are both positioned at a horizontal position, and the robot can complete the bobbin taking and placing. When broken yarns adsorbed on the bobbin are separated from the bobbin 8 or the broken yarns are required to be sucked by the hairbrush 5, the tail end of the robot rotates by 90 degrees to a corresponding station, the hairbrush 5 and the suction nozzle 6 are both located at vertical positions, and the broken yarns can be sucked by the hairbrush. Wherein the caliber of the nozzle opening 11 is 35mm x 7mm, the hairbrush 5 is made of nylon yarn, and the length of the hairbrush 5 is equal to the height of the bobbin 8.
The process of automatic yarn-breaking end-changing based on machine vision is shown in fig. 3, wherein the process of automatic yarn-breaking end-changing in fig. 3 is shown in a diagram a; can be described specifically as: 1) the robot takes and places the broken yarn bobbin from the spinning frame to the rotating spindle; 2) the spindle starts to rotate, and meanwhile, the industrial camera collects broken yarn pictures in real time and uploads the broken yarn pictures to the embedded system for processing; 3) if the broken yarn is attached to the surface of the bobbin, the hairbrush is adopted to rub the surface of the bobbin so that the broken yarn head is exposed outside the bobbin, and then the detection and the positioning of the broken yarn are completed; if the broken yarn is not attached to the surface of the bobbin, the detection and the positioning of the broken yarn are directly carried out; 4) after the broken yarn is positioned, converting the pixel coordinate of the broken yarn into a robot coordinate, and driving the tail end of the robot to move the suction nozzle to a broken yarn suction position; 5) starting an air pump to capture broken yarns; 6) unwinding the bobbin, and acquiring the picture after absorption again and uploading the picture to an embedded system for processing; 7) if the yarn breakage is detected to be captured, the whole automatic broken yarn end-finding process is finished, otherwise, the process returns to the step 2) to find the ends again.
The algorithm flow chart of the broken yarn detection is shown as a b chart in fig. 3, and the principle of the broken yarn detection method is as follows: 1) preprocessing a broken yarn image, including image smoothing processing, converting an image gray format and binarizing the image through threshold processing; 2) extracting a spool side area close to the mechanical arm as an interested area; 3) searching a broken yarn contour by using an edge detection method, and immediately stopping the rotation of the rotary spindle after finding the broken yarn contour; 4) returning to the average value of the coordinates of the pixel points of the broken yarn profile; 5) and converting the pixel coordinate into a robot coordinate by using a coordinate conversion algorithm, and returning the coordinate point to the robot as a broken yarn suction point.
While the invention has been described with respect to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention. Those skilled in the art can make various changes, modifications and equivalent arrangements, which are equivalent to the embodiments of the present invention, without departing from the spirit and scope of the present invention, and which may be made by utilizing the techniques disclosed above; meanwhile, any changes, modifications and variations of the above-described embodiments, which are equivalent to those of the technical spirit of the present invention, are within the scope of the technical solution of the present invention.