CN109946229B - Intelligent digital double-pull-wire detection system and detection method for cigarette strip package - Google Patents

Intelligent digital double-pull-wire detection system and detection method for cigarette strip package Download PDF

Info

Publication number
CN109946229B
CN109946229B CN201910137821.2A CN201910137821A CN109946229B CN 109946229 B CN109946229 B CN 109946229B CN 201910137821 A CN201910137821 A CN 201910137821A CN 109946229 B CN109946229 B CN 109946229B
Authority
CN
China
Prior art keywords
speed camera
detection
control cpu
main control
data
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
Application number
CN201910137821.2A
Other languages
Chinese (zh)
Other versions
CN109946229A (en
Inventor
李志勇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Wit Electronic Science & Technology Co ltd
Original Assignee
Nanjing Wit Electronic Science & Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing Wit Electronic Science & Technology Co ltd filed Critical Nanjing Wit Electronic Science & Technology Co ltd
Priority to CN201910137821.2A priority Critical patent/CN109946229B/en
Publication of CN109946229A publication Critical patent/CN109946229A/en
Application granted granted Critical
Publication of CN109946229B publication Critical patent/CN109946229B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Image Analysis (AREA)
  • Studio Devices (AREA)

Abstract

The invention relates to an intelligent digital double-stay wire detection system and a detection method for cigarette strip packages, wherein the system comprises a microcontroller, a high-speed camera and an LED light source, wherein the high-speed camera is connected with the microcontroller, the LED light source is positioned at the front end of the high-speed camera, the optical center of the LED light source is intersected with the optical center of the high-speed camera, the center of the high-speed camera is aligned with a reference stay wire, and the microcontroller comprises a main control CPU, a memory and an LED touch screen, wherein the memory and the LED touch screen are connected with the main control CPU. The DSP processing technology is adopted, so that the operation speed is high, the processing capacity is high, the complex algorithm can be adapted, and the existence or the offset of each type of double-pull wire of the strip package can be accurately detected; the high-speed camera adopts a full-view image visual detection mode, supports multi-window application, and freely sets a detection range; and the touch screen display function is used for displaying the running state detected by the current product and various running data in real time.

Description

Intelligent digital double-pull-wire detection system and detection method for cigarette strip package
Technical Field
The invention relates to the technical field of tobacco packaging quality detection, in particular to an intelligent digital double-pull-wire detection system and method for cigarette strip packages.
Background
In the production process of cigarettes, a layer of transparent packaging paper is wrapped outside the packaging box, so that the packaging box has the effects of sealing and moisture proofing, and the packaging box is convenient for users to unpack through the stay wires, and the quality of cigarettes in the strip box is seriously affected due to the defects, deviation and the like of the stay wires.
At present, various cigarette enterprises attach importance to the quality defect of CV strip wrapping stay wires, and various strip wrapping stay wire detection devices are correspondingly installed, and a micro-motion sensor mode with a simple mechanical principle is adopted to detect whether stay wires exist or not, so that the deviation of the stay wires cannot be detected; after package forming, installing a strip package visual detection device, wherein the detection is installed after the cigarette package forming, the price is high, the detection is missed, the false detection phenomenon exists, and the transparent pull wire detection is obviously insufficient; by adopting the image detection method, the light source attenuation is faster, the detection precision is low, the detection is not sensitive enough, and the daily maintenance and adjustment are more troublesome.
Disclosure of Invention
The invention aims to provide an intelligent digital double-pull-wire detection system and method for cigarette strip packages, which adopt an image vision technology to shoot an image at a full visual angle, and the detection precision freely determines the detection range by a window-drawing form so as to meet the quality standard of the strip package double-pull-wire process.
In order to achieve the above purpose, the technical scheme of the invention is as follows:
The utility model provides a cigarette strip package intelligence digital double-stay wire detecting system, includes microcontroller and with high-speed camera and the LED light source that microcontroller links to each other, the LED light source is located the front end of high-speed camera and the light core of LED light source intersects with the light core of high-speed camera, and the reference stay wire is aimed at to the center of high-speed camera, microcontroller include master control CPU and with memory and the LED touch-sensitive screen that master control CPU is connected.
In the above scheme, the microcontroller further comprises a serial deserializer connected between the high-speed camera and the main control CPU, and used for converting serial signals transmitted by the high-speed camera into parallel signals recognized by the main control CPU.
In the scheme, the LED touch screen is provided with the detection windows with the same number as the stay wires, and the detection windows respectively detect the corresponding stay wires and display detection data.
In the above scheme, the microcontroller further comprises a network chip, wherein the network chip is electrically connected with the main control CPU, is externally connected with the PC, and is used for transmitting the image data of the main control CPU to the PC.
An intelligent digital double-pull-wire detection method for cigarette strip packages comprises the following steps:
S1, initializing, namely, retrieving data of a data memory, initializing a window, and synchronizing flash control of the LED light source and the high-speed camera image acquisition;
s2, acquiring images, wherein a high-speed camera is matched with an LED light source to shoot and convert acquired image signals into LVDS serial signals to be transmitted;
S3, transmitting image signals, converting LVDS serial signals into parallel signals through a serial deserializer and transmitting the parallel signals to a main control CPU through a bus transceiver;
S4, performing image calculation and analysis, performing geometric normalization processing on the size of the image stay wire by the main control CPU, and sending the image data to the LED touch screen for display.
In step S4, the geometric normalization processing data includes first detection window data and second detection window data, average filtering is performed on the first detection window data and the second detection window data in sequence, binary entropy is obtained, whether the binary entropy is within a set range is determined, and if the binary entropy value is lower than a preset value, a rejection signal is sent.
In the above scheme, the calculation formula of the binary entropy is H (X) =Σ -P (i, j) ×lgp (i, j), where i is the gray value of the current pixel, j is the mean value of the 8 neighborhood of the current pixel, the ranges of i and j are 0-255, P (i, j) is the probability that the gray value i of the current pixel and the mean value j of the 8 neighborhood of the current pixel occur, and meanwhile, the gray value of the current pixel and the mean value of the 8 neighborhood of the current pixel are accumulated in a segmented manner to form a new binary array with the size of 25×25.
The intelligent digital double-pull-wire detection system and the detection method for the cigarette strip packets adopt a DSP processing technology, have high operation speed and high processing capacity, can adapt to complex algorithms, and can accurately detect whether double pull wires of various types of strip packets exist or not; the high-speed camera adopts a full-view image visual detection mode, supports multi-window application, and freely sets a detection range; and the touch screen display function is used for displaying the running state detected by the current product and various running data in real time.
Drawings
FIG. 1 is a schematic diagram of an intelligent digital dual-wire detection system according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for intelligent digital dual-pull wire detection in an embodiment of the invention;
reference numerals illustrate: 1. the device comprises a high-speed camera, 2 LED light sources, 3, a microcontroller, 4 and a reference stay wire.
Detailed Description
The technical scheme of the invention is further described in detail below with reference to the accompanying drawings and examples.
As shown in figure 1, the intelligent digital double-stay wire detection system for cigarette strip packages comprises a microcontroller 3, a high-speed camera 1 and an LED light source 2, wherein the high-speed camera 1 and the LED light source 2 are connected with the microcontroller 3. The LED light source 2 is located at the front end of the high-speed camera 1, the optical center of the LED light source 2 is intersected with the optical center of the high-speed camera 1, and the center of the high-speed camera 1 is aligned with the reference stay wire 4. The microcontroller 3 comprises a main control CPU, a serial de-serializer, a memory and an LED touch screen, wherein the memory and the LED touch screen are respectively and electrically connected with the main control CPU, and the serial de-serializer is connected between the high-speed camera 1 and the main control CPU and is used for converting serial signals transmitted by the high-speed camera 1 into parallel signals recognized by the main control CPU. The memory comprises a synchronous dynamic random access memory, a program memory and a data memory, wherein the data memory is used for storing setting parameters of a window, the program memory is used for storing image processing software program codes, and the synchronous dynamic random access memory is used for storing collected image data.
When a strip bag stay wire to be detected passes through a view finding area of the high-speed camera 1, the LED light source 2 illuminates the stay wire, the high-speed camera 1 starts to shoot images, and when the strip bag cellophane to be detected passes through an image acquisition area, the high-speed camera 1 is matched with the LED light source 2 to finish image acquisition, and a continuous shooting mode is adopted, wherein the acquisition speed of each image is 5-6ms. The strip package cellophane image signals shot by the high-speed camera 1 are transmitted to the synchronous dynamic random access memory through a special data line and an image acquisition card, the main control CPU performs geometric normalization processing on the size of the image pull wire through image processing software, so that the pull wire size is consistent, and calculation data is displayed on the LED touch screen.
The LED touch screen is provided with a first detection window and a second detection window, two stay wires, such as a 0-wire stay and a 1-wire stay, a detection value window and a setting value window are correspondingly arranged on the 0-wire stay, and a detection value window and a setting value window are correspondingly arranged on the 1-wire stay; the detection value is used for displaying a real-time value of the current stay wire on-line detection, the set value is a threshold value, the set value is a reference value for detecting the stay wire, and when the detection value is lower than the reference value, the set value is removed; the first detection window is used for setting the detection range of the No. 0 stay wire, the second detection window is used for setting the detection range of the No. 1 stay wire, the test interval between the No. 0 stay wire and the No. 1 stay wire can be set according to actual conditions, and the maximum adjustable range is 10mm.
In order to enable a user to conveniently and simply operate and set the system, the microcontroller 3 further comprises a network chip, wherein the network chip is electrically connected with the main control CPU in a pair and is externally connected with the PC, and the network chip is used for transmitting image data of the main control CPU to the PC to complete operation control.
In an embodiment, the image sensor in the high-speed camera 1 adopts a PYTHON1300 chip, and adopts LVDS output, the main control CPU adopts a 600MHz DSP BFU33 chip of ADI company, and the serial deserializer adopts a DS92LV1212A chip.
The main control CPU is electrified, data in the data memory is called, the PYTHON1300 is initialized and set through a 12C bus, and the flash speed of the LED light source 2 is controlled through an internal timer; the PYTHON1300 collects image signals and converts parallel signals into LVDS serial signals, the DS92LV1212A receives the LVDS serial signals transmitted by the PYTHON1300 and converts the LVDS serial signals into parallel signals, the bus transceiver transmits the parallel signals to the PPI interface of the main control CPU, and the DSP sends processed image data to the LED touch screen through the serial ports and can be set through the LED touch screen. The DSP may also upload the image data to the PC through the network chip DM9000, and set the DSP through the PC.
An intelligent digital double-pull-wire detection method for cigarette strip packages comprises the following steps:
Firstly, initializing, namely, retrieving data of a data memory, initializing a window, and synchronizing flash control of image acquisition of an LED light source 2 and a high-speed camera 1;
step two, image acquisition, wherein the high-speed camera 1 is matched with the LED light source 2 to shoot and convert the acquired image signals into LVDS serial signals for transmission;
step three, transmitting image signals, converting LVDS serial signals into parallel signals through a serial deserializer and transmitting the parallel signals to a main control CPU through a bus transceiver;
And step four, image calculation and analysis, wherein the main control CPU performs geometric normalization processing on the size of the image stay wire and sends the image data to the LED touch screen for display.
In the fourth step, the data subjected to geometric normalization processing comprises first detection window data and second detection window data, average filtering is carried out on the first detection window data and the second detection window data respectively in sequence, binary entropy is obtained, whether the data are in a set range or not is judged, and if the data are lower than a preset value, a rejection signal is sent. The binary entropy is calculated by a formula of H (X) =sigma-P (i, j) ×lgP (i, j), wherein i is a current pixel gray value, j is an average value of a current pixel 8 neighborhood, the range of i and j is 0-255, P (i, j) is the probability that the current pixel gray value i and the average value j of the current pixel 8 neighborhood appear, and meanwhile, the current pixel gray value i and the average value j of the current pixel 8 neighborhood are calculated in a segmented accumulated mode to form a new binary array with the size of 25×25.
The DSP is powered on for resetting, data in a data memory are read, PYTHON1300 is initialized and set through an I2C bus, an internal timer is set to control an LED light source 2 to flash, when a high-speed camera 1 is normal, PPI of the DSP is interrupted and set, PPI data are waited for, whether the PPI data are valid data or not is judged, when the data are valid, a first detection window is opened, average filtering is carried out on an acquired image of a No. 0 stay wire, the binary entropy of the acquired image is obtained, and when the binary entropy of the No. 0 stay wire is not in a set range, a rejection signal is sent out; and secondly, opening a second detection window, carrying out mean value filtering on the acquired image of the No. 1 stay wire, solving the binary entropy of the acquired image, and sending out a rejection signal when the binary entropy of the No. 1 stay wire is not in the set range. When the first detection window does not collect images, the image data processing can be directly carried out by entering the second detection window.
In one embodiment, the binary entropy is calculated as: h (X) =Σ -P (i, j) ×lgp (i, j), where i is the current pixel gray value, the range is 0 to 255, j is the mean value of the current pixel 8 neighborhood, the range value is 0 to 255, the binary array P (i, j) is the current pixel gray value i, the probability that the mean value j of the current pixel 8 neighborhood appears, and the array size is 256×256. Because the DSP is too slow for the floating point operations of the array, each window is more than about 100 milliseconds, and two windows are more than 200 milliseconds, real-time detection cannot be achieved.
The invention optimizes the binary array, and specifically optimizes: the probabilities of the current pixel gray values 0-15 are added to the new array P_IMG (0), the probabilities of the current pixel gray values 16-25 are added to the new array P_IMG (1), the probabilities of the current pixel gray values 26-35 are added to the new array P_IMG (2), and so on, and finally the probabilities of the current pixel gray values 246-255 are added to the new array P_IMG (24). Similarly, the probability of 0-15 of the average value of the current pixel 8 neighborhood is added to the new array P_AVG (0), the probability of 16-25 of the average value of the current pixel 8 neighborhood is added to the new array P_AVG (1), the probability of 26-35 of the average value of the current pixel 8 neighborhood is added to the new array P_AVG (2), and the like, finally, the probability of 246-255 of the average value of the current pixel 8 neighborhood is added to the new array P_AVG (24), the size of the obtained new binary array is 25 x 25, and the size is reduced by 100 times compared with the size of the original array, namely the speed is increased by 100 times, the real-time performance of the system is improved, and the detection speed is improved.
The intelligent digital double-pull-wire detection system and the detection method for the cigarette strip packets adopt a DSP processing technology, have high operation speed and high processing capacity, can adapt to complex algorithms, and can accurately detect whether double pull wires of various types of strip packets exist or not; the high-speed camera adopts a full-view image visual detection mode, supports multi-window application, and freely sets a detection range; and the touch screen display function displays the running state and various running data of the current product detection in real time, and the quality detection standard of the strip-package double-pull-wire process is achieved.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.

Claims (2)

1. An intelligent digital double-pull-wire detection system for cigarette strip bags is characterized in that: the LED light source (2) is positioned at the front end of the high-speed camera (1), the optical center of the LED light source (2) is intersected with the optical center of the high-speed camera (1), the center of the high-speed camera (1) is aligned with the reference stay wire (4), and the microcontroller (3) comprises a main control CPU, a memory and an LED touch screen, wherein the memory is connected with the main control CPU;
The microcontroller (3) further comprises a serial de-serializer, wherein the serial de-serializer is connected between the high-speed camera (1) and the main control CPU and is used for converting serial signals transmitted by the high-speed camera (1) into parallel signals recognized by the main control CPU;
The LED touch screen is provided with detection windows with the same number as the stay wires, and the detection windows respectively detect the corresponding stay wires and display detection data;
The microcontroller (3) further comprises a network chip, wherein the network chip is electrically connected with the main control CPU, is externally connected with the PC and is used for transmitting the image data of the main control CPU to the PC;
the system performs the following steps when detecting:
S1, initializing, calling data of a data memory, initializing a window, and synchronizing flash control of image acquisition of an LED light source (2) and a high-speed camera (1);
S2, acquiring images, shooting by the aid of a high-speed camera (1) matched with an LED light source (2), and converting acquired image signals into LVDS serial signals for transmission;
S3, transmitting image signals, converting LVDS serial signals into parallel signals through a serial deserializer and transmitting the parallel signals to a main control CPU through a bus transceiver;
S4, performing image calculation and analysis, performing geometric normalization processing on the size of the image stay wire by a main control CPU, and sending image data to an LED touch screen for display;
the geometric normalization processing data comprise first detection window data and second detection window data, average filtering is respectively carried out on the first detection window data and the second detection window data in sequence, binary entropy is obtained, whether the data are in a set range or not is judged, and if the binary entropy value is lower than a preset value, a rejection signal is sent;
The binary entropy calculation formula is H (X) =Σ -P (i, j) ×lgp (i, j), wherein i is a current pixel gray value, j is a mean value of a current pixel 8 neighborhood, the ranges of i and j are 0-255, P (i, j) is the probability that the current pixel gray value i and the mean value j of the current pixel 8 neighborhood appear, and meanwhile, the current pixel gray value and the mean value of the current pixel neighborhood are accumulated in a segmented mode to form a new binary array with the size of 25×25.
2. The intelligent digital double-pull-wire detection system for cigarette packets according to claim 1, wherein: the high-speed camera (1) is an image sensor with a model PYTHON1300 chip.
CN201910137821.2A 2019-02-25 2019-02-25 Intelligent digital double-pull-wire detection system and detection method for cigarette strip package Active CN109946229B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910137821.2A CN109946229B (en) 2019-02-25 2019-02-25 Intelligent digital double-pull-wire detection system and detection method for cigarette strip package

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910137821.2A CN109946229B (en) 2019-02-25 2019-02-25 Intelligent digital double-pull-wire detection system and detection method for cigarette strip package

Publications (2)

Publication Number Publication Date
CN109946229A CN109946229A (en) 2019-06-28
CN109946229B true CN109946229B (en) 2024-07-16

Family

ID=67006882

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910137821.2A Active CN109946229B (en) 2019-02-25 2019-02-25 Intelligent digital double-pull-wire detection system and detection method for cigarette strip package

Country Status (1)

Country Link
CN (1) CN109946229B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202089280U (en) * 2011-01-30 2011-12-28 沈阳中科朗运科技有限公司 Visual detection apparatus of small cigarette package and strip package transparent paper drawing line
CN209858406U (en) * 2019-02-25 2019-12-27 南京文易特电子科技有限公司 Intelligent digital double-pull-wire detection system for cigarette carton and pack

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7406184B2 (en) * 2002-07-03 2008-07-29 Equinox Corporation Method and apparatus for using thermal infrared for face recognition
CN100361138C (en) * 2005-12-31 2008-01-09 北京中星微电子有限公司 Method and system of real time detecting and continuous tracing human face in video frequency sequence
CN107801041B (en) * 2011-06-16 2019-08-13 Ge视频压缩有限责任公司 Decoder, decoding and encoded video method
CN203178196U (en) * 2013-02-05 2013-09-04 厦门烟草工业有限责任公司 Tear-tape detector based on image processing
CN203921363U (en) * 2014-07-08 2014-11-05 林锦超 A kind of cigarette-packaging machine barrel appearance quality detecting device
JP6257042B2 (en) * 2014-08-27 2018-01-10 日本電信電話株式会社 Quantum key distribution system and quantum key distribution method
CN104354935B (en) * 2014-10-22 2017-03-22 中国电子科技集团公司第四十一研究所 Defect detection device for drawn wire on transparent packaging material
CN204726757U (en) * 2015-05-14 2015-10-28 湖南中烟工业有限责任公司 To the instrument that false proof coding and the bracing wire on transparent paper of cigarette packet surface detect

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202089280U (en) * 2011-01-30 2011-12-28 沈阳中科朗运科技有限公司 Visual detection apparatus of small cigarette package and strip package transparent paper drawing line
CN209858406U (en) * 2019-02-25 2019-12-27 南京文易特电子科技有限公司 Intelligent digital double-pull-wire detection system for cigarette carton and pack

Also Published As

Publication number Publication date
CN109946229A (en) 2019-06-28

Similar Documents

Publication Publication Date Title
CN105136804B (en) The acquisition of high-speed band steel surface image and image processing apparatus and method
CN101482518A (en) On-line quality detection system for movable band-shaped material
CN104717442B (en) Automatic switching method of more video formats to VESA agreement 1600X1200 resolution ratio 60Hz frame-rate videos
CN102034098A (en) Automatic bolt sorting system and automatic bolt sorting method
CN202935608U (en) Cigarette detector based on machine vision technology
US20220321840A1 (en) Method for transmitting high bandwidth camera data through serdes links
CN104354935B (en) Defect detection device for drawn wire on transparent packaging material
CN105118262A (en) Community-oriented high-efficiency remote meter reading method and system
CN209858406U (en) Intelligent digital double-pull-wire detection system for cigarette carton and pack
CN109946229B (en) Intelligent digital double-pull-wire detection system and detection method for cigarette strip package
WO2024055698A1 (en) Package inspection method and system and electronic device
CN110723342A (en) Cigar quality detection system and detection method
CN112432953A (en) Cigarette packet missing and reverse detection method based on machine vision technology
CN105430297B (en) The automatic control system that more video formats are changed to IIDC protocol videos form
CN107786866A (en) A kind of binocular vision image synthesis system and method
CN212354503U (en) Cigar quality detection system
CN212953668U (en) Intelligent device for detecting scene personnel of elevator car
CN209296607U (en) Intelligent cloth examination device
CN216581216U (en) Tobacco bale breaking detecting system based on photoelectricity dynamic scanning
CN202940909U (en) Double-transmission mode monitoring device based on DSP and FPGA
CN211766772U (en) Binary channels packagine machine detection device that acts as go-between
CN111800556B (en) Image communication device and control method thereof
CN114936174B (en) Image processing and fusion computing method based on ground unmanned platform
CN207379461U (en) A kind of more stranding distance classification image collecting devices of twisted-pair feeder
CN111741256B (en) Moving target detection method for power transmission line external damage prevention visualization device

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