CN110328157B - Intelligent sorting device based on machine vision and sorting method thereof - Google Patents
Intelligent sorting device based on machine vision and sorting method thereof Download PDFInfo
- Publication number
- CN110328157B CN110328157B CN201910645494.1A CN201910645494A CN110328157B CN 110328157 B CN110328157 B CN 110328157B CN 201910645494 A CN201910645494 A CN 201910645494A CN 110328157 B CN110328157 B CN 110328157B
- Authority
- CN
- China
- Prior art keywords
- station
- sorting
- camera
- advection
- articles
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/342—Sorting according to other particular properties according to optical properties, e.g. colour
- B07C5/3422—Sorting according to other particular properties according to optical properties, e.g. colour using video scanning devices, e.g. TV-cameras
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
- B07C5/362—Separating or distributor mechanisms
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C2501/00—Sorting according to a characteristic or feature of the articles or material to be sorted
- B07C2501/0063—Using robots
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Sorting Of Articles (AREA)
- Specific Conveyance Elements (AREA)
Abstract
The invention discloses an intelligent sorting device based on machine vision and a sorting method thereof, wherein the device comprises a sorting transmission mechanism, a manipulator sorting mechanism, a support frame, a vision sorting mechanism and a PLC (programmable logic controller); the sorting transmission mechanism comprises a feeding station, an advection detection station, a transportation station, a transfer discharging station and an advection station; the vision letter sorting mechanism includes packing state system, differentiates the shape system, distinguishes look system and camera, packing state system, differentiates the shape system and distinguishes the station of look system sharing camera, and the camera is installed on the bottom outer wall of every support frame, and every packing state system judges the packing state of article and classifies types, combines colour and shape, and the locking target article shunts, and the camera main function is location, discernment and controls manipulator letter sorting mechanism through the PLC controller and carries out work. The problem of current commodity circulation parcel letter sorting technique's manual operation complicacy, inefficiency and difficult letter sorting is solved.
Description
Technical Field
The invention belongs to cargo sorting equipment for intelligent logistics, and particularly relates to an intelligent sorting device based on machine vision and a sorting method thereof.
Background
With the increasing cost of labor, replacing manpower with machines to do repetitive and high-intensity labor is an important direction for modern machine research. Meanwhile, with the improvement of living standard and the acceleration of life rhythm of people, more and more people buy daily necessities in an online shopping mode, the prevalence of online shopping drives the development of express delivery, and the express deliverers need to distribute thousands of express deliveries every day, which often causes the problems of delay of the express deliveries, untimely sorting, untimely delivery and the like, in addition, the goods to be delivered are complex, the goods have different packaging forms such as box body packaging, clothes packaging bags, envelope bags and the like, the packaging states are different, most of the packaging is complete, and some articles with damaged packages or without packages can not be sorted quickly, the articles with different specifications and shapes are not easy to sort, the method has the defects of great problems, insufficient refinement, unreasonable and incomplete classification, low sorting efficiency, complex work, high error rate and the like during manual sorting. Therefore, a need exists in the art for an intelligent sorting apparatus based on machine vision to solve this problem.
Disclosure of Invention
The invention aims to provide an intelligent sorting device based on machine vision and a sorting method thereof, which are used for solving the defects of complex manual operation, low efficiency and difficulty in sorting of the conventional logistics package sorting technology.
In order to achieve the technical purpose, the technical scheme adopted by the invention is as follows:
an intelligent sorting device based on machine vision comprises a sorting transmission mechanism, a manipulator sorting mechanism, a support frame, a vision sorting mechanism and a PLC (programmable logic controller);
the sorting transmission mechanism comprises a feeding station, an advection detection station, a transportation station, a transfer discharging station and an advection station;
the visual sorting mechanism comprises a packaging state system, a shape distinguishing system, a color distinguishing system and a camera, wherein the packaging state system, the shape distinguishing system and the color distinguishing system share the station of the camera, the camera is installed on the outer wall of the bottom of each supporting frame, the camera adopts a 2D special color distinguishing and shape distinguishing camera, and the judging speed of the camera is set to be 15 Hz; each packaging state system judges the packaging state of the articles according to the images captured by the cameras and classifies the articles, the target articles are locked to be shunted by combining colors and shapes, and the cameras mainly have the functions of positioning, identifying and controlling the mechanical arm sorting mechanism to work through the PLC.
Preferably, the model of the PLC is GuardLogix Controller 1756-L73, the PLC is set as a safety PLC, a human-computer interface of the PLC is set as 2711P-T10C4D8, and an Ethernet module of the PLC is set as 1756-EN 2T.
Preferably, a conveying driving module is arranged at a feeding station of the sorting transmission mechanism, the feeding station is set to be conveyed by a differential tension belt, a gravity sensing detection platform is mounted on the advection detection station, and the gravity sensing detection platform consists of a gravity sensor and a shunting turnout; the end fixed mounting of advection detection station has first reposition of redundant personnel station, the end at first reposition of redundant personnel station is installed to the feed end of advection station, and the discharge end of advection station installs the second reposition of redundant personnel station, categorised transport area is installed to the end of second reposition of redundant personnel station, categorised transport area connects the transportation station, and the top fixed mounting of categorised transport area has the support frame, the transportation station adopts the linear mode of lining up of multistage gradient differential distance to transmit article, handover ejection of compact station is installed in the ejection of compact department of transportation station, install manipulator letter sorting mechanism on the support frame, manipulator letter sorting mechanism sets up the top in categorised transport area.
Preferably, the gravity sensor and the diversion fork are correspondingly provided with three sections, the gravity sensor is set to be in three stages of 1-5 kg, 5-15 kg and more than 15kg, the diversion fork is communicated with the feeding end of the first diversion station, the articles can be subjected to weight sorting work, when the articles move to the advection detection station, advection distance pulling is carried out on the advection detection station, and the weight of the articles on the gravity sensing detection platform can be sorted into three weight sections of 1-5 kg, 5-15 kg and more than 15kg by combining the gravity sensor and the diversion fork in the gravity sensing detection platform, then the articles enter the first diversion station and are transported to the advection station.
Preferably, the number of the first shunting stations and the number of the second shunting stations are three, the first shunting stations and the second shunting stations are arranged in a stepped mode, and the height difference between the first shunting stations and the second shunting stations is set to be 30-40 CM.
Preferably, above-mentioned manipulator letter sorting mechanism passes through the ethernet with the PLC controller and realizes being connected, manipulator letter sorting mechanism includes two hand claw armed levers and vertical guide rail, hand claw armed lever and vertical guide rail all set up on the support frame, there is the rack board through slider and spout swing joint on the support frame, the inner wall fixed mounting of vertical guide rail has vertical motor, the output shaft outer wall fixed mounting of vertical motor has the gear, intermeshing between the rack board on gear and the support frame, horizontal wrist rotary cylinder body is installed on two hand claw armed lever tops, horizontal wrist rotary cylinder body switches on the hydraulic pump, the inner wall at vertical guide rail is installed to the hydraulic pump, hand claw armed lever top is provided with horizontal motor, swing joint has vertical lead screw on the vertical guide rail, vertical lead screw is connected in the output shaft transmission of horizontal motor.
Preferably, the outer wall of the output shaft of the transverse motor is provided with a screw rod guide sleeve matched with the vertical screw rod, the inner wall of the screw rod guide sleeve is meshed with the outer wall of the vertical screw rod, the sorting device is stable in structure through the arrangement of the transverse motor and the vertical screw rod, and the sorting device can correspondingly move left, right, up and down on three sorting transport belts through the control of a PLC (programmable logic controller) so as to facilitate sorting work;
preferably, the inner walls of the paw arm rods are provided with paw rotating cylinders, the output ends of the paw rotating cylinders are fixedly connected with corresponding fingers, a paw clamping cylinder is arranged between the two paw arm rods, the output ends of the paw clamping cylinders are connected with the paw arm rods to control the two paw arm rods to be clamped, horizontal rotating work can be carried out at the clamping arms of the sorting device through the arrangement of a transverse wrist rotating cylinder body, and the manipulator sorting mechanism is provided with the paw rotating cylinders and can correspondingly turn over in the vertical direction;
the vertical motor and the horizontal motor are both set as servo motors, and the vertical motor, the horizontal motor, the hydraulic pump, the gripper clamping cylinder and the gripper rotating cylinder are all connected with the output end of the PLC through electric signals.
A sorting method of an intelligent sorting device based on machine vision comprises the following steps:
step 1: feeding articles; the materials are intensively placed at a feeding port on a feeding station to wait for a conveying roller to carry out feeding work;
step 2: sorting by weight; when the articles move to the advection detection station, advection distance pulling is carried out on the advection detection station, and the weight of the articles on the gravity sensing detection platform can be sorted into three sections by combining a gravity sensor and a diversion fork in the gravity sensing detection platform, and then the three sections enter a first diversion station and are transported to the advection station;
step 3: processing the information state of the article; the tail end of each advection station locks a target article for shunting, a camera on a support frame performs image acquisition, preprocessing, camera calibration, target positioning and three-dimensional calculation, a packaging state system, a shape distinguishing system and a color distinguishing system in a visual sorting mechanism shoots the article at the discharge end of the advection station, and acquires and processes information of the shape, the color distinguishing and the packaging state of the article;
step 4: intelligently sorting; the camera in the visual sorting mechanism is used for processing the shot information and sending the processed information to the PLC controller, the PLC controller feeds the processed information back to the manipulator sorting mechanism, and the manipulator sorting mechanism can clamp and sort objects with different shapes, different packaging states or colors on a classified conveying belt;
step 5: transporting the goods; after sorting, transporting the articles to a transport station for transport;
step 6: discharging the articles; distributing to each corresponding handover discharging station through a plurality of gradients, driving the sorted articles by the handover discharging stations, and translating the articles to the corresponding sorting discharging ports for collection.
Preferably, the camera operation in Step3 includes image acquisition, preprocessing, camera calibration, target positioning and three-dimensional calculation;
the image acquisition is to acquire an image by selecting a proper industrial camera for acquiring the image;
the pretreatment process comprises the following steps: in order to obtain high-quality images and improve the accuracy of a system, noise is eliminated and other interference factors are removed by using image preprocessing, wherein the image preprocessing comprises threshold segmentation, smoothing processing and HDR (high-resolution) methods;
the camera calibration process comprises the following steps: internal and external parameters of a left camera and a right camera are obtained through calibration of a binocular camera so as to improve the system precision;
the target positioning process comprises the following steps: creating a template for the target standard image based on the gray scale or the shape for executing template search to obtain an affine matrix;
the three-dimensional calculation process comprises the following steps: obtaining the angle of the space on the plane through three points on the plane; the coordinates in the three-dimensional space are obtained through two central points of the left image and the right image in combination with a parallax principle, and three-dimensional data of the surface of the express parcel are obtained. .
The invention has the following beneficial effects:
according to the intelligent sorting device based on machine vision and the sorting method thereof, due to the fact that the gravity sensing detection platform is arranged in the sorting transmission mechanism, pre-sorting work can be conducted on articles according to different weights, then camera information in the vision sorting mechanism is collected, ultra-high-speed article sorting can be achieved, accuracy is high, through the fact that three refining systems in the vision sorting mechanism are sequentially arranged, data of each article can be collected, feedback summarizing and sorting work can be conducted, sorting efficiency is improved, and sorting types are refined; machine vision-based mechanical arm sorting mechanism of intelligent sorting device has that the letter sorting angle is many, the scope covers extensively and easily sorts work, has avoided the manual work to sort, has improved the production efficiency of enterprise to this mechanical arm sorting mechanism stable in structure and cost of maintenance are low. The intelligent sorting device and the system replace manpower, can intelligently and automatically sort objects, reduce the workload of express delivery personnel, save the enterprise cost, shorten the delivery time of express delivery objects and improve the working efficiency.
Drawings
FIG. 1 is a flow chart of the operation of an intelligent sorting device based on machine vision and a sorting method thereof according to the present invention;
FIG. 2 is a top view of a sorting transmission mechanism of the intelligent sorting device based on machine vision and the sorting method thereof according to the present invention;
FIG. 3 is a side view of a sorting transmission mechanism of the intelligent sorting device based on machine vision and the sorting method thereof according to the invention;
FIG. 4 is a cross-sectional view A-A of the sorting actuator of FIG. 2 illustrating an intelligent sorting apparatus and a sorting method thereof based on machine vision according to the present invention;
FIG. 5 is an enlarged view of B in FIG. 4 of the intelligent sorting apparatus and the sorting method based on machine vision according to the present invention;
fig. 6 is a material visual sorting flow chart of the intelligent sorting device based on machine vision and the sorting method thereof.
Wherein the reference numerals are: 1. a sorting transmission mechanism; 2. a vision sorting mechanism; 3. a manipulator sorting mechanism; 4. a support frame; 5. a PLC controller; 6. a gravity sensing detection platform; 11. a feeding station; 12. a advection detection station; 13. a transport station; 14. transferring a discharging station; 15. a first splitting station; 16. a sorting conveyor belt; 17. a advection station; 18. a second diversion station; 21. a packaging status system; 22. identifying a shape system; 23. a color discrimination system; 24. a camera; 30. a gripper arm lever; 31. a vertical motor; 32. a transverse motor; 33. a hydraulic pump; 34. a transverse wrist rotating cylinder body; 35. a gripper clamping cylinder; 36. a paw rotating cylinder; 37. a gear; 38. a screw rod guide sleeve; 39. a rack plate; 301. a vertical guide rail; 302. and a vertical screw rod.
Detailed Description
Embodiments of the present invention are described in further detail below with reference to the accompanying drawings.
The utility model provides an intelligence sorting device based on machine vision which characterized in that: the sorting device comprises a sorting transmission mechanism 1, a manipulator sorting mechanism 3, a support frame 4, a visual sorting mechanism 2 and a PLC (programmable logic controller) 5;
the sorting transmission mechanism 1 comprises a feeding station 11, a advection detection station 12, a transportation station 13, a transfer discharging station 14 and a advection station 17;
the visual sorting mechanism 2 comprises a packaging state system 21, a shape distinguishing system 22, a color distinguishing system 23 and a camera 24, the packaging state system 21, the shape distinguishing system 22 and the color distinguishing system 23 share a station of the camera 24, the camera 24 is installed on the outer wall of the bottom of each support frame 4, the camera 24 adopts a 2D special color distinguishing and shape distinguishing camera, and the judging speed of the camera is set to be 15 Hz; each packaging state system 21 judges the packaging state of the articles according to the images captured by the cameras 24, classifies the articles, locks the target articles for distribution by combining colors and shapes, and the cameras 24 mainly have the functions of positioning, identifying and controlling the mechanical arm sorting mechanism 3 to work through the PLC 5.
In specific implementation, the model of the PLC 5 is GuardLogix Controller 1756-L73 and is set as a safety PLC, the human-computer interface of the PLC is set as 2711P-T10C4D8, and the Ethernet module of the PLC is set as 1756-EN 2T.
Referring to fig. 2 and 3, in specific implementation, a feeding station 11 of the sorting transmission mechanism 1 is provided with a transportation driving module, the feeding station 11 is set to be in differential tension belt transmission, a gravity sensing detection platform 6 is installed on a advection detection station 12, and the gravity sensing detection platform 6 is composed of a gravity sensor and a diversion turnout; the end of the advection detection station 12 is fixedly provided with a first shunting station 15, the feed end of the advection station 17 is arranged at the end of the first shunting station 15, the discharge end of the advection station 17 is provided with a second shunting station 18, the end of the second shunting station 18 is provided with a classification conveyer belt 16, the classification conveyer belt 16 is connected with the conveying station 13, a support frame 4 is fixedly arranged above the classification conveyer belt 16, the conveying station 13 adopts a multi-level gradient constant-distance linear queuing mode to transmit articles, a handover discharge station 14 is arranged at the discharge position of the conveying station 13, a manipulator sorting mechanism 3 is arranged on the support frame 4, and the manipulator sorting mechanism 3 is arranged above the classification conveyer belt 16.
In specific implementation, the gravity sensor and the diversion fork are correspondingly provided with three sections, the gravity sensor is set to have three stages of 1-5 kg, 5-15 kg and more than 15kg, the diversion fork is communicated with the feeding end of the first diversion station 15, the weight sorting work can be carried out on the articles, when the articles move to the advection detection station 12, the advection distance pulling is carried out on the advection detection station 12, and the weight of the articles on the gravity sensing detection platform 6 can be sorted into three sections of 1-5 kg, 5-15 kg and more than 15kg by combining the gravity sensor and the diversion fork in the gravity sensing detection platform 6, then the articles enter the first diversion station 15 and are transported to the advection station 17.
In practical implementation, three corresponding first splitting stations 15 and three corresponding second splitting stations 18 are installed, the first splitting stations 15 and the second splitting stations 18 are arranged in a stepped manner, the height difference is set to be 30-40 CM, and 9 types of articles with different specifications can be correspondingly sorted out by the aid of the three first shunting stations 15 and the three second shunting stations 18 and the sorting working line, so that the sorting efficiency of the articles can be improved, and the sorting of the article types becomes more detailed, so that the subsequent operation can be conveniently carried out by the working personnel, the sorting work can be simultaneously carried out, the ultra-high-speed article sorting can be realized, the accuracy is high, and through being equipped with three kinds of system that refine in the vision letter sorting mechanism 2 in proper order, can collect the data of every article, feed back and gather and letter sorting work, and then improved letter sorting efficiency and refined the letter sorting kind.
Referring to fig. 4 and 5, in specific implementation, the manipulator sorting mechanism 3 is connected to the PLC controller 5 through an ethernet, the manipulator sorting mechanism 3 includes two gripper arms 30 and a vertical guide rail 301, the gripper arms 30 and the vertical guide rail 301 are both disposed on the support frame 4, a rack plate 39 is movably connected to the support frame 4 through a slider and a chute, a vertical motor 31 is fixedly mounted on an inner wall of the vertical guide rail 301, a gear 37 is fixedly mounted on an outer wall of an output shaft of the vertical motor 31, the gear 37 is engaged with the rack plate 39 on the support frame 4, a horizontal wrist rotary cylinder 34 is mounted at top ends of the two gripper arms 30, the horizontal wrist rotary cylinder 34 is connected to a hydraulic pump 33, the hydraulic pump 33 is mounted on an inner wall of the vertical guide rail 301, a horizontal motor 32 is disposed at top end of the gripper arm 30, and a vertical lead screw 302 is movably connected to the vertical, the output shaft of the transverse motor 32 is in transmission connection with a vertical screw rod 302;
in specific implementation, the outer wall of the output shaft of the transverse motor 32 is provided with the screw guide sleeve 38 matched with the vertical screw 302, the inner wall of the screw guide sleeve 38 is meshed with the outer wall of the vertical screw 302, and the sorting device is stable in structure and can correspondingly move left, right, up and down on the three classified conveying belts 16 under the control of the PLC 5 so as to facilitate sorting;
in specific implementation, the inner walls of the paw arm rods 30 are all provided with paw rotating cylinders 36, the output ends of the paw rotating cylinders 36 are fixedly connected with corresponding fingers, a paw clamping cylinder 35 is arranged between the two paw arm rods 30, the output ends of the paw clamping cylinders 35 are connected with the paw arm rods 30 to control the clamping of the two paw arm rods 30, the clamping arms of the sorting device can horizontally rotate through the arrangement of a transverse wrist rotating cylinder body 34, the manipulator sorting mechanism 3 is provided with the paw rotating cylinders 36 and can correspondingly turn over in the vertical direction, the functionality and the applicability of the manipulator sorting mechanism 3 are improved,
during concrete implementation, above-mentioned vertical motor 31 and horizontal motor 32 are servo motor, vertical motor 31, horizontal motor 32, hydraulic pump 33, the tight jar 35 of hand claw clamp and the rotatory jar 36 output of all through signal of telecommunication connection PLC controller 5 of hand claw, this manipulator letter sorting mechanism 3 has that the letter sorting angle is many, the scope covers extensively and easily sorts the work, avoided the manual work to sort, the production efficiency of enterprise has been improved, and manipulator letter sorting mechanism 3's stable in structure and cost of maintenance are low.
In specific implementation, the visual sorting mechanism 2 can classify the packing states of the articles, that is, three conditions of complete packing, no packing or damaged packing, and perform corresponding classification one by one, and distinguish three types of "cuboid", "cylinder", "flat" and "other shapes", and can classify the articles with different colors, for example, a brown cuboid box body and a black flat bag body, and further can further distinguish common articles, decorative bag bodies and other fragile articles with red marks, and one of the three classification conveyor belts 16 is an article with incomplete or no packing; the other cuboid cylindrical box body; and the last black flat bag body or the fragile product of the red mark of other shapes are arranged, and are sorted to the corresponding classification conveying belt 16 through the mechanical arm sorting mechanism 3 to be conveyed, and the corresponding classification operation is carried out, so that the sorting work is more refined.
Referring to fig. 1, a sorting method of an intelligent sorting apparatus based on machine vision includes the following steps:
step 1: feeding articles; the materials are intensively placed at a feeding port on a feeding station 11 to wait for a conveying roller to carry out feeding work;
step 2: sorting by weight; when the article moves to the advection detection station 12, the advection distance is carried out on the advection detection station 12, and the weight of the article on the gravity sensing detection platform 6 can be sorted into three sections by combining the gravity sensor and the shunting turnout in the gravity sensing detection platform 6, and then the article enters the first shunting station 15 and is transported to the advection station 17;
step 3: processing the information state of the article; the tail end of each advection station 17 locks a target article for shunting, a camera 24 on a support frame 4 carries out image acquisition, preprocessing, camera 24 calibration, target positioning and three-dimensional calculation, a packaging state system 21, a shape distinguishing system 22 and a color distinguishing system 23 in a visual sorting mechanism 2 shoot the article at the discharge end of the advection station 17, and acquire and process information of the shape, color distinguishing and packaging state of the article;
step 4: intelligently sorting; the camera 24 in the visual sorting mechanism 2 processes the shot information and sends the processed information to the PLC controller 5, the PLC controller 5 feeds the information back to the mechanical arm sorting mechanism 3, and the mechanical arm sorting mechanism 3 can clamp and sort objects with different shapes, different packaging states or different colors on the classification conveying belt 16;
step 5: transporting the goods; after sorting, the articles are transported to a transport station 13 for transport;
step 6: discharging the articles; distributed to each corresponding handover discharging station 14 through a plurality of gradients, the handover discharging station 14 drives the sorted articles to move horizontally to the corresponding sorting discharging ports for collection.
Referring to fig. 6, in a specific implementation, the operation of the camera 24 in Step3 includes image acquisition, preprocessing, camera 24 calibration, target positioning, and three-dimensional calculation;
the image acquisition is to acquire an image by selecting a proper industrial camera for acquiring the image;
the pretreatment process comprises the following steps: in order to obtain high-quality images and improve the accuracy of a system, noise is eliminated and other interference factors are removed by using image preprocessing, wherein the image preprocessing comprises threshold segmentation, smoothing processing and HDR (high-resolution) methods;
the camera calibration process comprises the following steps: internal and external parameters of a left camera and a right camera are obtained through calibration of a binocular camera so as to improve the system precision;
the target positioning process comprises the following steps: creating a template for the target standard image based on the gray scale or the shape for executing template search to obtain an affine matrix;
the three-dimensional calculation process comprises the following steps: obtaining the angle of the space on the plane through three points on the plane; the coordinates in the three-dimensional space are obtained through two central points of the left image and the right image in combination with a parallax principle, and three-dimensional data of the surface of the express parcel are obtained.
After the intelligent sorting device based on machine vision and the sorting method thereof are adopted, the gravity sensing detection platform arranged in the sorting transmission mechanism can enable the articles to be pre-sorted according to different weights, and then the camera information in the vision sorting mechanism is acquired, so that the ultra-high-speed article sorting can be realized, the accuracy is high, and through sequentially arranging three refining systems in the vision sorting mechanism, the data of each article can be collected, the feedback summarizing and the sorting work can be carried out, so that the sorting efficiency is improved, and the sorting types are refined; machine vision-based mechanical arm sorting mechanism of intelligent sorting device has that the letter sorting angle is many, the scope covers extensively and easily sorts work, has avoided the manual work to sort, has improved the production efficiency of enterprise to this mechanical arm sorting mechanism stable in structure and cost of maintenance are low. The intelligent sorting device and the system replace manpower, can intelligently and automatically sort objects, reduce the workload of express delivery personnel, save the enterprise cost, shorten the delivery time of express delivery objects and improve the working efficiency.
Claims (9)
1. The utility model provides an intelligence sorting device based on machine vision which characterized in that: the sorting device comprises a sorting transmission mechanism (1), a mechanical arm sorting mechanism (3), a support frame (4), a visual sorting mechanism (2) and a PLC (programmable logic controller) (5);
the sorting transmission mechanism (1) comprises a feeding station (11), a advection detection station (12), a transportation station (13), a transfer discharging station (14) and an advection station (17);
the visual sorting mechanism (2) comprises a packaging state system (21), a shape distinguishing system (22), a color distinguishing system (23) and a camera (24), the packaging state system (21), the shape distinguishing system (22) and the color distinguishing system (23) share stations of the camera (24), the camera (24) is installed on the outer wall of the bottom of each support frame (4), the camera (24) adopts a 2D special color distinguishing and shape distinguishing camera, and the judging speed of the camera is set to be 15 Hz; each packaging state system (21) judges the packaging state of the articles according to the images captured by the cameras (24), classifies the articles, locks the target articles by combining colors and shapes, and divides the flow, and the cameras (24) mainly have the functions of positioning, identifying and controlling the mechanical arm sorting mechanism (3) to work through the PLC (5);
a conveying driving module is arranged on a feeding station (11) of the sorting transmission mechanism (1), the feeding station (11) is arranged for differential tension belt conveying, a gravity sensing detection platform (6) is arranged on a advection detection station (12), and the gravity sensing detection platform (6) consists of a gravity sensor and a shunting turnout; the end fixed mounting of advection detection station (12) has first reposition of redundant personnel station (15), the end at first reposition of redundant personnel station (15) is installed to the feed end of advection station (17), and the discharge end of advection station (17) installs second reposition of redundant personnel station (18), categorised transport area (16) is installed to the end of second reposition of redundant personnel station (18), categorised transport area (16) are connected transport station (13), and the top fixed mounting of categorised transport area (16) has support frame (4), transport station (13) adopt the linear queuing mode of multistage gradient differential distance to transmit article, the ejection of compact department at transport station (13) is installed in handover ejection of compact station (14), install manipulator letter sorting mechanism (3) on support frame (4), manipulator letter sorting mechanism (3) set up the top in categorised transport area (16).
2. The machine vision based intelligent sorting device according to claim 1, characterized in that: the model of the PLC Controller (5) is GuardLogix Controller 1756-L73, the PLC Controller is set to be a safety PLC, a human-computer interface of the PLC Controller is set to be 2711P-T10C4D8, and an Ethernet module of the PLC Controller is set to be 1756-EN 2T.
3. The machine vision based intelligent sorting device according to claim 1, characterized in that: the gravity sensor and the shunting fork are correspondingly provided with three intervals, the gravity sensor is set to be 1-5 kg, 5-15 kg and three gears larger than 15kg, the feeding end of the shunting fork and the feeding end of the first shunting station (15) are communicated with each other, the weight sorting work can be carried out on articles, when the articles move to the advection detection station (12), advection distance drawing is carried out on the advection detection station (12), and the weight of the articles on the gravity sensing detection platform (6) can be sorted into three weight intervals of 1-5 kg, 5-15 kg and more than 15kg by combining the gravity sensor and the shunting fork in the gravity sensing detection platform (6), then the articles enter the first shunting station (15) and are transported to the advection station (17).
4. The machine vision based intelligent sorting device according to claim 1, characterized in that: the three shunting stations are respectively and correspondingly arranged on the first shunting station (15) and the second shunting station (18), the first shunting station (15) and the second shunting station (18) are arranged in a stepped mode, and the height difference between the first shunting station and the second shunting station is set to be 30-40 CM.
5. The machine vision based intelligent sorting device according to claim 1, characterized in that: the manipulator sorting mechanism (3) is connected with the PLC (5) through the Ethernet, the manipulator sorting mechanism (3) comprises two gripper arm rods (30) and a vertical guide rail (301), the gripper arm rods (30) and the vertical guide rail (301) are arranged on a support frame (4), a rack plate (39) is movably connected onto the support frame (4) through a sliding block and a sliding groove, a vertical motor (31) is fixedly arranged on the inner wall of the vertical guide rail (301), a gear (37) is fixedly arranged on the outer wall of an output shaft of the vertical motor (31), the gear (37) is meshed with the rack plate (39) on the support frame (4), a transverse wrist rotating cylinder body (34) is arranged on the top ends of the two gripper arm rods (30), the transverse wrist rotating cylinder body (34) is communicated with a hydraulic pump (33), and the hydraulic pump (33) is arranged on the inner wall of the vertical guide rail (301), a transverse motor (32) is arranged at the top end of the paw arm lever (30), a vertical screw rod (302) is movably connected to the vertical guide rail (301), and an output shaft of the transverse motor (32) is in transmission connection with the vertical screw rod (302).
6. The intelligent sorting device based on machine vision according to claim 5, characterized in that:
lead screw guide pin bushing (38) that match with vertical lead screw (302) are installed to the output shaft outer wall of horizontal motor (32), the inner wall of lead screw guide pin bushing (38) and the outer wall intermeshing of vertical lead screw (302), through the setting of horizontal motor (32) and vertical lead screw (302), this sorting device stable in structure to can be on three categorised transportation area (16), through the control of PLC controller (5), the removal work about going on that corresponds, so that sort the work.
7. The intelligent sorting device based on machine vision according to claim 5, characterized in that:
the inner walls of the paw arm rods (30) are provided with paw rotating cylinders (36), the output ends of the paw rotating cylinders (36) are fixedly connected with corresponding fingers, a paw clamping cylinder (35) is arranged between the two paw arm rods (30), the output end of the paw clamping cylinder (35) is connected with the paw arm rods (30) to control the clamping of the two paw arm rods (30), through the arrangement of a transverse wrist rotating cylinder body (34), the clamping arms of the sorting device can horizontally rotate, and a manipulator sorting mechanism (3) is provided with the paw rotating cylinders (36) and can correspondingly turn over in the vertical direction;
the vertical motor (31) and the horizontal motor (32) are both set to be servo motors, and the vertical motor (31), the horizontal motor (32), the hydraulic pump (33), the paw clamping cylinder (35) and the paw rotating cylinder (36) are all connected with the output end of the PLC (5) through electric signals.
8. A sorting method of an intelligent sorting device based on machine vision is characterized in that: the method comprises the following steps:
step 1: feeding articles; the materials are intensively placed at a feeding port on a feeding station (11) to wait for a conveying roller to carry out feeding work;
step 2: sorting by weight; when the articles move to the advection detection station (12), advection distance pulling is carried out on the advection detection station (12), the weight of the articles on the gravity sensing detection platform (6) can be sorted into three sections by combining a gravity sensor and a diversion fork in the gravity sensing detection platform (6), and then the three sections enter a first diversion station (15) and are transported to the advection station (17);
step 3: processing the information state of the article; the tail end of each advection station (17) locks a target article for shunting, a camera (24) on a support frame (4) performs image acquisition, preprocessing, camera (24) calibration, target positioning and three-dimensional calculation, a packaging state system (21), a shape distinguishing system (22) and a color distinguishing system (23) in a visual sorting mechanism (2) shoots the article at the discharge end of the advection station (17), and acquires and processes information of the shape, the color distinguishing and the packaging state of the article;
step 4: intelligently sorting; the shot information is processed by a camera (24) in the visual sorting mechanism (2) and is sent to a PLC (programmable logic controller) controller (5), the PLC controller (5) feeds back the information to the manipulator sorting mechanism (3), and the manipulator sorting mechanism (3) can clamp and sort objects with different shapes, different packaging states or colors on a classification conveyor belt (16);
step 5: transporting the goods; after being sorted, the articles are transported to a transport station (13) for transport;
step 6: discharging the articles; the goods are distributed to each corresponding transfer discharging station (14) through a plurality of gradients, the transfer discharging stations (14) drive the sorted goods to move to the corresponding sorting discharging ports for collection.
9. The sorting method of the intelligent sorting device based on the machine vision according to the claim 8, characterized in that: the camera (24) operation in the Step3 comprises image acquisition, preprocessing, camera (24) calibration, target positioning and three-dimensional calculation;
the image acquisition is to acquire an image by selecting a proper industrial camera for acquiring the image;
the pretreatment process comprises the following steps: in order to obtain high-quality images and improve the accuracy of a system, noise is eliminated and other interference factors are removed by using image preprocessing, wherein the image preprocessing comprises threshold segmentation, smoothing processing and HDR (high-resolution) methods;
the camera calibration process comprises the following steps: internal and external parameters of a left camera and a right camera are obtained through calibration of a binocular camera so as to improve the system precision;
the target positioning process comprises the following steps: creating a template for the target standard image based on the gray scale or the shape for executing template search to obtain an affine matrix;
the three-dimensional calculation process comprises the following steps: obtaining the angle of the space on the plane through three points on the plane; the coordinates in the three-dimensional space are obtained through two central points of the left image and the right image in combination with a parallax principle, and three-dimensional data of the surface of the express parcel are obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910645494.1A CN110328157B (en) | 2019-07-17 | 2019-07-17 | Intelligent sorting device based on machine vision and sorting method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910645494.1A CN110328157B (en) | 2019-07-17 | 2019-07-17 | Intelligent sorting device based on machine vision and sorting method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110328157A CN110328157A (en) | 2019-10-15 |
| CN110328157B true CN110328157B (en) | 2021-06-29 |
Family
ID=68145632
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910645494.1A Active CN110328157B (en) | 2019-07-17 | 2019-07-17 | Intelligent sorting device based on machine vision and sorting method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110328157B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110694926A (en) * | 2019-11-05 | 2020-01-17 | 中邮科技有限责任公司 | Special-shaped part detection and distribution system and method |
| CN113814182A (en) * | 2020-07-24 | 2021-12-21 | 北京京东振世信息技术有限公司 | Sorting system, method and storage medium |
| CN112354877A (en) * | 2020-10-17 | 2021-02-12 | 南通贝源机械制造有限公司 | Automatic sorting assembly line |
| CN112886363A (en) * | 2021-03-17 | 2021-06-01 | 珠海瑞捷电气股份有限公司 | Assembly system of current terminal |
| CN113457992A (en) * | 2021-06-10 | 2021-10-01 | 南京轩世琪源软件科技有限公司 | Sorting robot based on visual identification and sorting method |
| CN113333326B (en) * | 2021-06-30 | 2022-09-09 | 上海首坤智能科技有限公司 | Full-automatic defect detecting and sorting equipment for bowls and dishes |
| CN113714125B (en) * | 2021-08-19 | 2024-04-09 | 河南麒展实业有限公司 | Cylinder sorting device and sorting method |
| CN113843182A (en) * | 2021-09-27 | 2021-12-28 | 深圳技术大学 | Automated sorting system and method |
| CN113877836B (en) * | 2021-11-05 | 2023-05-09 | 江苏昱博自动化设备有限公司 | Intelligent identification sorting system based on visual detection system |
| CN114852428A (en) * | 2022-06-07 | 2022-08-05 | 安徽格拉泰科机械有限公司 | Conveying system capable of rapidly switching packaging forms |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104507814A (en) * | 2012-08-13 | 2015-04-08 | 莱特拉姆有限责任公司 | Parcel handling methods |
| CN104511436A (en) * | 2013-09-28 | 2015-04-15 | 沈阳新松机器人自动化股份有限公司 | Express sorting method and system based on robot visual servo technology |
| CN105292892A (en) * | 2015-11-11 | 2016-02-03 | 江苏汇博机器人技术有限公司 | Automatic storage system of industrial robot |
| CN109794435A (en) * | 2019-01-18 | 2019-05-24 | 金陵科技学院 | Fruit quality detection system and method based on deep learning |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102144125B1 (en) * | 2014-01-14 | 2020-08-12 | 한국전자통신연구원 | Apparatus and Method Measuring Mail Size and Acquiring Mail Information |
| FR3031046B1 (en) * | 2014-12-24 | 2018-06-29 | Solystic | INSTALLATION FOR THE SEPARATION AND INDIVIDUALIZATION OF HETEROGENEOUS POSTAL OBJECTS |
-
2019
- 2019-07-17 CN CN201910645494.1A patent/CN110328157B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104507814A (en) * | 2012-08-13 | 2015-04-08 | 莱特拉姆有限责任公司 | Parcel handling methods |
| CN104511436A (en) * | 2013-09-28 | 2015-04-15 | 沈阳新松机器人自动化股份有限公司 | Express sorting method and system based on robot visual servo technology |
| CN105292892A (en) * | 2015-11-11 | 2016-02-03 | 江苏汇博机器人技术有限公司 | Automatic storage system of industrial robot |
| CN109794435A (en) * | 2019-01-18 | 2019-05-24 | 金陵科技学院 | Fruit quality detection system and method based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110328157A (en) | 2019-10-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110328157B (en) | Intelligent sorting device based on machine vision and sorting method thereof | |
| CN109279373B (en) | Flexible unstacking and stacking robot system and method based on machine vision | |
| CN106124520A (en) | A kind of full-automatic loading and unloading and the device of vision-based detection | |
| CN106829430B (en) | Automatic detection and conveying production line for multiple connecting lines of cross shaft parts | |
| CN107264900A (en) | Industrial robot automatic processing method and device for logistics system | |
| CN106742380B (en) | Automatic detection packaging equipment for automobile connecting rod | |
| CN109748219B (en) | Spin sealing device for classifying and conveying bottle bodies according to bottle cap colors | |
| CN110404785B (en) | Automatic sorting line wrapping and loading system and method based on three-dimensional vision | |
| CN202038662U (en) | Annular sorting and conveying device for multiple products | |
| CN114260210B (en) | But parallel movement's electronic commerce goods picking device | |
| CN107737915B (en) | A kind of casting and forming production system | |
| CN212190961U (en) | Multi-station automatic conveying system of stamping equipment | |
| CN109650307B (en) | Tightening transmission system for conveying bottle bodies according to bottle cap color classification | |
| CN107497702A (en) | It is automatic to weigh interflow sorter and article joining ajusts method | |
| CN112090778A (en) | Categorised vanning system of ammeter | |
| CN208666416U (en) | Logistics automatic distribution system | |
| CN111054641A (en) | Automatic separation conveyor of discernment | |
| CN215707625U (en) | Automatic detection, kicking and overturning positioning equipment for packaging box | |
| CN111375571A (en) | Automatic classification production line | |
| CN105855179B (en) | Materials-sorting system, Material Sorting machine used and Material Sorting method | |
| CN204564592U (en) | The automatic Rough Inspection production line of silicon ingot | |
| CN210001131U (en) | robot system for automatic material pouring of optical module | |
| CN112591415A (en) | Sign indicating number device is beaten in sample bottle intelligence evacuation | |
| CN206937310U (en) | A kind of discharging manipulator of end of tape positioner | |
| CN207104308U (en) | A kind of flexible assembly line equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221226 Address after: Room 1050, 10th Floor, Building A, No. 27, Xinfa Road, Suzhou Industrial Park, Suzhou Area, China (Jiangsu) Pilot Free Trade Zone, Wuzhong District, Suzhou City, Jiangsu Province, 215000 Patentee after: Suzhou Ruiwo Intelligent Technology Co.,Ltd. Address before: No. 1 Park Avenue Jiangning Hongjing District of Nanjing city of Jiangsu province in 211167 Patentee before: NANJING INSTITUTE OF TECHNOLOGY |