A visual recognition system and method for cigarette stem packages

Abstract

The present invention relates to a visual recognition system and method for tobacco stem packages, including a moving crossbeam assembly, a first visual camera, a second visual camera, a dust-proof mechanism, and a controller. This technical solution completely replaces manual operation with an intelligent system, realizing all processes of transferring tobacco stem packages from a transport vehicle to a designated area, from the designated area to a tobacco stem conveyor, unpacking and discharging the tobacco stem packages, and recycling empty tobacco stem packages, truly achieving the automation and intelligence of tobacco stem feeding.

Description

Technical Field

[0001] The present invention relates to the technical field of cigarette production automation and intelligence, and particularly relates to a visual recognition system and method for tobacco stalk packages. Background Art

[0002] At present, in the tobacco primary processing workshop, most of the feeding of tobacco stalks is completed manually: the tobacco stalk packages are transported by a transport vehicle to a designated area and stacked irregularly. A remote control is used to move the hook (currently three hooks) in the east-west and up-down directions. The operator uses the remote control to move the hook to the place where the tobacco stalk package needs to be opened, and manually hooks three bags of tobacco stalk packages with the three hooks respectively (the tobacco stalk packages in the north-south direction need to be pulled by the operator with the hook to this place for hooking). After the hooking is completed, when the remote control is used to move the tobacco stalk package to a position level with the dust removal hood, the operator fixes the tobacco stalk package with one hand and operates the cutting knife with the other hand to cut open the bottom of the tobacco stalk package, and then manually shakes it to make the tobacco stalks completely fall onto the conveyor belt (the same operation is performed for the other two tobacco stalk packages, but there is a process of rotating the hook to find the unopened tobacco stalk package in the middle). After the unpacking is completed, the empty tobacco stalk package is manually placed at the specified place. In order to improve the overall automation level and achieve staff reduction and efficiency increase, on the premise of meeting the production process requirements, it is planned to use automated equipment and robots to realize the automation of production. Summary of the Invention

[0003] The purpose of the present invention is to provide a visual recognition system and method for tobacco stalk packages, which can realize functions such as the identification and positioning of tobacco stalk packages on the transport vehicle and in the designated area, and the detection of sparks during the unpacking process of the tobacco stalk packages.

[0004] In order to achieve the above purpose, the present invention adopts the following technical solutions:

[0005] The intelligent tobacco stalk feeding system based on mobile vision technology can not only complete the grasping of packages from the transport vehicle and placing them in the designated area, but also complete the grasping, unpacking and conveying to the conveyor belt from the designated area, realizing overall automation; the original hoist can be reserved for backup after transformation to prevent equipment failure, and manual operation can be continued to ensure production; a visual recognition system and method for tobacco stalk packages have completed the identification of characters on the transport vehicle to determine the type of tobacco stalks, the positioning and identification of tobacco stalk packages on the transport vehicle, the positioning and identification of tobacco stalk packages in the designated placement area, the detection of sparks during the unpacking process of the unpacking and discharging mechanism, and the detection of leakage and scattering during the transfer process of the tobacco stalk packages.

[0006] The beneficial effects of the present invention are:

[0007] This technical solution completely replaces manual operation with an intelligent system, realizing all processes of the transfer of tobacco stalk packages from the transport vehicle to the designated area, from the designated area to the tobacco stalk conveyor, the unpacking and discharging of tobacco stalk packages, and the recycling of empty tobacco stalk packages, truly realizing the automation and intelligence of tobacco stalk feeding. Description of the Drawings

[0008] Figure 1 This is a schematic diagram of the system for visual recognition of tobacco stem packages in the present invention;

[0009] Figure 2 This is a schematic diagram of the moving vision structure above the dual-axis truss robot;

[0010] Figure 3 This is a schematic diagram of the vision structure of the tobacco stem package breaking and discharging mechanism. Detailed Embodiments

[0011] The technical solutions of the present invention will be described in detail below through embodiments. The following embodiments are merely exemplary and can only be used to explain and illustrate the technical solutions of the present invention, rather than being construed as a limitation to the technical solutions of the present invention.

[0012] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "front", "rear", "vertical", "horizontal", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. This is 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. Therefore, it cannot be understood as a limitation to the present invention. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0013] In the description of the present invention, it should be noted that unless otherwise clearly specified and defined, the terms "installed", "connected", and "connected" 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 directly connected or indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0014] The technical solutions of the present invention will be further described in detail through the following specific embodiments in combination with the drawings.

[0015] As Figures 1 to 3As shown in the figure, a visual recognition system for tobacco stem packages has an overall structure where a first vision camera 7 is installed on a moving crossbeam assembly 1 above an intelligent tobacco stem feeding system based on mobile vision technology. The first vision camera 7 is entirely within a dust-proof mechanism 8 to prevent dust from adhering to the camera lens and affecting the visual effect. This first vision camera is used to determine the type of tobacco stems based on the characters on the transport vehicle 3, locate and identify the tobacco stem packages on the transport vehicle 3, locate and identify the tobacco stem packages in the designated placement area, and detect leakage and spillage during the transfer process of the tobacco stem packages. A second vision camera 9 is installed in the feeding dust-proof mechanism 2 above the tobacco stem conveyor 4 for detecting sparks during the bag-opening process of the bag-breaking and discharging mechanism 5. The two vision cameras cooperate with each other to form the vision system of the intelligent tobacco stem feeding system based on mobile vision technology, ensuring the normal production of the intelligent tobacco stem feeding system based on mobile vision technology.

[0016] For the moving crossbeam assembly 1 above the intelligent tobacco stem feeding system based on mobile vision technology, a support beam (section steel) 6 is erected above the workshop of the tobacco stem feeding station. A guide rail 11 is fixedly installed on the support beam 6, and a rack 10 is fixedly installed on the support beam 6. A first connecting block 13 is installed on the slider of the guide rail 11. A motor mounting plate 14 is fixedly installed on the first connecting block 13. A motor 15 is installed on the motor mounting plate 14. A gear 12 is installed on the shaft of the motor 15, and the gear 12 cooperates with the rack 10. With the rotation of the motor 15, the overall movement is achieved. A dust-proof mechanism 8 is installed on the first connecting block 13, and the entire lens of the first vision camera 7 is installed inside the dust-proof mechanism 8 to prevent dust from entering and contaminating the lens.

[0017] A second vision camera 9 is installed in the feeding dust-proof mechanism 2. A funnel 16 is fixedly installed above the tobacco stem conveyor. Cylinders 18 are installed on the two opposite sides of the funnel. Two sealing flaps 17 are connected to the two side seams of the funnel 16 through hinges. A second connecting block 19 is fixedly installed on the side of the sealing flap 17. A rotating shaft 20 is connected to the second connecting block 19 and is also connected to the movable shaft of the cylinder 18. With the telescoping of the cylinder, the flipping of the sealing flap 17 is achieved.

[0018] The second vision camera is used to identify the type of tobacco stems on the transport vehicle 3. Single characters "A", "B", "C", and "D" are set to represent different types of tobacco stems respectively, with one-to-one correspondence. When the transport vehicle 3 stops at a fixed position to trigger a signal, the mobile vision moves to the upper part, and the second vision camera takes a visual picture. Since the visual picture will be distorted, visual graphic correction and straightening processing are required. Threshold segmentation, dynamic threshold segmentation are performed using fixed thresholds or automatically selected thresholds, and the connected regions in the image are extracted for character segmentation, so that each single character corresponds to a region. The morphological method is used to connect the separated parts of the same character, and a separate region of interest (ROI) is defined for each expected character in the image. The region of interest (ROI) of the character must include the single character to be separated. Character features are extracted, and the halcon preset dot matrix classifier is used for character recognition. The type of the tobacco stem is judged according to the recognized character.

[0019] For the positioning and recognition of the tobacco stem packages on the transport vehicle 3, an absolute coordinate system is set at the stop position of the transport vehicle. According to the visual picture, the coordinate points of each tobacco stem package and their stacking relationship are judged. The tobacco stem package to be grabbed first is at the top layer. For multiple tobacco stem packages at the top layer, their outer dimensions, tilt angles, and distances from the gripper are detected. Second, the ones with larger outer dimensions are grabbed first; third, the ones with shorter distances are grabbed first; fourth, the ones with smaller tilt angles are grabbed first.

[0020] For the positioning and recognition of the tobacco stem packages in the specified placement area, an absolute coordinate system is established in the specified placement area. Four placement area coordinates are defined according to the four different types of tobacco stems "A", "B", "C", and "D". The area coordinates are divided according to the maximum size of the tobacco stem packages within each type of tobacco stem placement area (while considering the clamping and picking space size of the dual-axis truss robot, one area is reserved for turnover within the same type of tobacco stem area), and then the tobacco stem packages are placed row by row, column by column, and layer by layer. Each time a tobacco stem package is placed, when the weight is higher than the weight of the previous tobacco stem package, another area is placed; when it is lower than the weight of the previous tobacco stem package, it is placed layer by layer. In this way, it can be ensured that the lighter and smaller tobacco stem packages are placed on the upper layer. When the same type of tobacco stem area is full, it is placed according to the lowest layer, so that the tobacco stem packages can be placed not too high. Since quality ratio is required, the system records the quality and placement position of each grabbed tobacco stem package for quality ratio to pick up packages. When the tobacco stem package with the required quality is at the lower layer, the turnover area is used for clamping, and the quality and new placement position of the tobacco stem package are recorded.

[0021] During the detection of leakage and scattering during the transfer of tobacco stem packages, when the gripper of the truss robot picks up a tobacco stem package from the transport vehicle and places it in the designated placement area, there may be damaged tobacco stem packages, resulting in the phenomenon of tobacco stem leakage and scattering. Therefore, during the transfer of tobacco stem packages, the first vision camera needs to follow and detect whether there is leakage and scattering of tobacco stem packages; when the gripper of the truss robot picks up a tobacco stem package from the designated area and places it on the feeding dust-proof mechanism for bag-breaking and feeding, the picked-up tobacco stem packages are all intact, so there will be no leakage and scattering phenomenon; during the transfer of tobacco stem packages from the transport vehicle to the designated placement area, the mobile vision follows the movement, takes continuous photos, extracts the visual image features, and judges whether there are new features in the continuous group of visual images; if there are new features, it is judged that there is leakage and scattering of tobacco stem, and the leaking tobacco stem package is placed in the area of leaking tobacco stem packages, and then processed manually later.

[0022] During the detection of sparks during the bag-breaking and discharging process of the bag-breaking discharging mechanism 5, the second vision camera is installed in the feeding dust-proof mechanism 2, and its installation position is directly opposite to the position of the blade during the bag-breaking process of the bag-breaking discharging mechanism 5; taking the flame image as the research object, through image acquisition, image processing, and image recognition, methods such as grayscale conversion and binarization are used to separate the flame from the background, extract features such as the area and centroid of the flame, and save them in the system. During the process of the blade cutting the tobacco stem package, continuous photos are taken, the features are extracted and compared with the image features in the system to judge whether there is a flame.

[0023] The present invention includes a first vision camera 7 installed on the moving crossbeam 1 above the intelligent tobacco stem feeding system based on mobile vision technology and a second vision camera 9 installed inside the feeding dust-proof mechanism 2. The two vision cameras form the vision system of the intelligent tobacco stem feeding system based on mobile vision technology.

[0024] Overall operation process of the scheme:

[0025] The transport vehicle 3 stops at a fixed position, the sensor detects the signal and transmits it to the first vision camera 7;

[0026] The first vision camera 7 moves to the designated position to collect the tobacco stem type number information on the transport vehicle 3 and the information of the tobacco stem packages on the vehicle, and determines the bag-taking order through the vision algorithm;

[0027] The dual-axis truss robot axis 1 takes the bag, weighs it according to the given order and places it in the designated area. During the transfer process, the first vision camera 7 follows and takes continuous photos to detect whether there is leakage and scattering of tobacco stem;

[0028] The dual-axis truss robot axis 2 moves to the corresponding position to take the bag according to the tobacco stem package placement area information given by the first vision camera 7;

[0029] The dual-axis truss robot axis 2 then moves to the upper edge of the feeding dust-proof mechanism 2 of the tobacco stem conveyor 4 for bag-breaking and jitter discharging;

[0030] The second axis of the dual-axis truss robot finally performs the action of emptying the tobacco stem package.

[0031] Repeat the above action process to carry out production.

[0032] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A visual recognition method for tobacco stem packages, using a visual recognition system for tobacco stem packages, characterized in that, It includes the following steps: S1. The transport vehicle stops at a fixed position, and the sensor detects the signal of the transport vehicle and transmits it to the first vision camera; S2. The first vision camera moves to a specified position to collect the information of the tobacco stem variety number on the transport vehicle and the information of the tobacco stem bales on the transport vehicle. The controller determines the grasping order of the tobacco stem bales through a vision algorithm and transmits the information of the grasping order of the tobacco stem bales to the dual-axis truss robot; S3. The first motion axis of the dual-axis truss robot places the tobacco stem bales at a specified area after weighing them according to the given grasping order of the tobacco stem bales; S4. The second motion axis of the dual-axis truss robot moves to the corresponding position to grasp the tobacco stem bales according to the information of the placement area of the tobacco stem bales; S5. After the second motion axis of the dual-axis truss robot grasps the tobacco stem bales in step S4, it moves above the feeding dust-proof mechanism of the tobacco stem conveyor to break the bales and shake and unload the materials; S6. After shaking and unloading the materials, the dual-axis truss robot performs the action of processing the empty tobacco stem bales; S7. Repeat steps S1 to S6 to carry out production; The method for the first vision camera to realize the information of the tobacco stem variety number on the transport vehicle is as follows: different identifiers are set to represent different types of tobacco stems, and the identifiers are in one-to-one correspondence with a certain type of tobacco stem. The transport vehicle relies on the signal triggered by the fixed position, and the first vision camera moves above the transport vehicle. The first vision camera takes a visual picture and judges the type of tobacco stem by identifying the identifier according to the visual picture; The method for determining the grasping order of the tobacco stem bales includes the following steps: 21) Set an absolute coordinate system at the parking position of the transport vehicle; 22) Judge the coordinate points of each tobacco stem bale and their mutual stacking relationship according to the visual picture. The tobacco stem bale to be grasped first is at the top layer; 23) Detect the outer dimension, inclination angle and the distance from the gripper of the multiple tobacco stem bales at the top layer. The tobacco stem bale with a larger outer dimension is grasped second, the tobacco stem bale closer to the gripper is grasped third, and the tobacco stem bale with a smaller inclination angle is grasped fourth.

2. The visual recognition method of tobacco stem packages according to claim 1, characterized in that The method for identifying the identifier according to the visual picture includes the following steps: 1) If the visual picture is deformed, visual graphic correction and rotation processing are required; 2) For the visual picture without deformation or the visual picture after correction and rotation processing, use a fixed threshold or an automatically selected threshold for threshold segmentation, dynamic threshold segmentation, and extract the connected regions in the visual picture for identifier segmentation, so that a single identifier corresponds to one region; 3) Use the morphological method to connect the separated parts of the same identifier, and define a separate region of interest for each expected identifier in the visual picture. The region of interest of the identifier must include the single identifier to be separated; 4) Extract the identifier features and use the halcon preset dot matrix classifier for identifier recognition; judge the type of tobacco stem according to the recognized identifier.

3. The visual recognition method for tobacco stem packages according to claim 1, wherein The method for placing in the specified area in step S3 includes the following steps: 31) Set an absolute coordinate system in the specified placement area; 32) Delimit the corresponding placement area coordinates according to the type of tobacco stem with the set identifier. Divide the area coordinates according to the maximum size of the tobacco stem bales in the placement area of each type of tobacco stem, and then place the tobacco stem bales row by row, column by column, and layer by layer. 33) Each time when placing the tobacco stem bale, if the weight is higher than that of the previous tobacco stem bale, place it in another area; if it is lower than the weight of the previous tobacco stem bale, place it layer by layer.

4. The visual recognition method for tobacco stem packages according to claim 1, wherein In step S3, during the process that the gripper of the first moving axis of the dual-axis truss robot picks up the tobacco stem bale from the transport vehicle to the designated area, the first vision camera follows the movement, takes continuous photos, extracts the image features, and judges whether there are new features in the continuous group of pictures; if there are new features, it is judged that tobacco stem leakage occurs, and the leaking tobacco stem bale is placed in the area for leaking tobacco stem bales, and then processed manually.

5. The visual recognition method of tobacco stem packages according to claim 1, characterized in that In step S5, during the bag-breaking process, Mars detection is carried out. Specifically: Taking the flame image as the research object, through image acquisition, image processing, and image recognition, the graying and binarization methods are used to separate the flame from the background, extract the area and centroid features of the flame, and save them in the system; The second vision camera takes continuous photos during the process that the blade cuts the tobacco stem bale, extracts the features and compares them with the image features in the system to judge whether there is a flame.

6. The visual recognition method of tobacco stem packages according to claim 1, characterized in that The tobacco stem bale vision recognition system includes a moving crossbeam assembly, a first vision camera, a second vision camera, a dust-proof mechanism and a controller; The moving crossbeam assembly is arranged above the intelligent tobacco stem feeding system, the dust-proof mechanism is installed on the moving crossbeam assembly, and the first vision camera is installed on the moving crossbeam assembly within the range of the dust-proof mechanism; The second vision camera is installed in the feeding dust-proof mechanism above the tobacco stem conveyor; Both the first vision camera and the second vision camera are electrically connected to the controller.

7. The visual recognition method for tobacco stem packages according to claim 6, characterized in that, The moving crossbeam assembly includes a support beam, a guide rail, a guide rail slider, a rack, a first connecting block, a motor mounting plate, a motor and a gear; The support beam is erected above the tobacco stem feeding station, the guide rail and the rack are both fixedly installed on the support beam, the guide rail slider is slidably arranged on the guide rail, the first connecting block is installed on the guide rail slider, the motor mounting plate is fixedly installed on the first connecting block, the motor is installed on the motor mounting plate, the gear is installed on the output shaft of the motor, the gear cooperates with the rack, and the dust-proof mechanism is installed on the first connecting block.

8. The visual recognition method of tobacco stem packages according to claim 6, characterized in that, The feeding dust-proof mechanism includes a funnel, a cylinder, two sealing flaps, a second connecting block and a rotating shaft; The funnel is installed above the tobacco stem conveyor, the cylinder is installed on the side of the two opposite sides of the funnel, the two sealing flaps are connected to the two side seams of the funnel through hinges, the second connecting block is fixedly installed on the side of the sealing flap, and the second connecting block is connected to the movable shaft of the cylinder through the rotating shaft.

Images