CN1382399A - Method for recognizing impurities in in-line detection of tobacco - Google Patents
Method for recognizing impurities in in-line detection of tobacco Download PDFInfo
- Publication number
- CN1382399A CN1382399A CN 02113825 CN02113825A CN1382399A CN 1382399 A CN1382399 A CN 1382399A CN 02113825 CN02113825 CN 02113825 CN 02113825 A CN02113825 A CN 02113825A CN 1382399 A CN1382399 A CN 1382399A
- Authority
- CN
- China
- Prior art keywords
- tobacco
- foreign matter
- article
- threshold value
- image
- 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.)
- Granted
Links
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
A method for recognizing the impurities in in-line detection of tobacco includes picking up image by complementary color intensity method, that is, a filter is attached to the lens of camera to increase the color difference between tobacco and impurities, image transmission, recognizing impurities by threshold method, transmitting the recognition signal and removing the impurities.
Description
Affiliated technical field:
The invention belongs in the online detections of article such as tobacco identification and elimination method to foreign matter.
Background technology:
At present, foreign matter identification in the online detection of article and an important application occasion of elimination method are the removing foreign material from tobacco systems, the exemplary of removing foreign material from tobacco system is the tobacco separator that Honghe Cigarette Factory introduces U.S. Svision company, its course of work can be represented with following steps: 1 image capture: promptly on production line, with the tobacco leaf in the CCD line array video camera picked-up certain limit and the image of foreign matter potpourri.2 images transmission: taking the photograph image is reached in the middle of the calculator memory with the form of data with image pick-up card.3 identifications are handled: according to the data of expression tobacco leaf and foreign matter colourity in the internal memory, set the colourity threshold value and differentiate tobacco leaf and foreign matter.4 identification signals transmission: after tobacco leaf and foreign matter discerned, the signal of expression foreign matter position is transferred to ECU (Electrical Control Unit).5 reject foreign matter: ECU (Electrical Control Unit) is controlled corresponding air nozzle with foreign body eliminating according to received foreign matter position signalling.
Can identify the tobacco foreign matter exactly be the key issue that determines article foreign body eliminating system performance qualities such as tobacco.Above-mentioned removing foreign material from tobacco system is in step 3, according to R (red), the G (green) of each pixel colourity in the taking the photograph image, the difference of B (indigo plant) component, judges that this pixel belongs to tobacco or foreign matter.In the practical situation, the colourity of tobacco and foreign matter all has the scope of a change, foreign matter especially, because its diversity, its colourity and tobacco can produce to a certain degree overlapping unavoidably, and this causes difficulty for accurate differentiation of tobacco and foreign matter.And above-mentioned removing foreign material from tobacco system before tobacco and foreign matter are discerned, does not take measures to increase the colour difference of tobacco and foreign matter, and then reduces the overlapping scope of both colourities in above-mentioned steps 1.This is a weak point of this system.When tobacco and foreign matter are differentiated, be by the colourity of each pixel in the image of taking the photograph the tobacco leaf foreign matter potpourri to the computing machine is differentiated.Concrete method is: think that the chromatic value of tobacco satisfies certain limit, during differentiation, the pixel of colourity in this scope thought tobacco, do not think foreign matter at the pixel of this scope, the upper bound of this scope and lower bound then are called threshold value.Obviously, the setting of the science of threshold value is a major issue that influences discriminant accuracy.In the above-mentioned removing foreign material from tobacco system, the setting of threshold value is to carry out at the initial period that detects a collection of tobacco, for a collection of tobacco, threshold value is relatively-stationary, like this, when the colourity of this batch tobacco changes greatly, will have influence on the accuracy that tobacco and foreign matter are differentiated.This also is place that haves much room for improvement of this system.
Summary of the invention
In order to improve the correct decision rate of article such as tobacco and foreign matter, to develop the more excellent removing foreign material from tobacco system of performance.The invention provides a foreign matter recognition methods that grows tobacco etc. in the online detection of article.It has taked to increase the method for article such as tobacco in the image and foreign matter colour difference, and has proposed a kind of threshold setting method of more science when article such as tobacco and foreign matter are differentiated.
In order to realize the foregoing invention purpose, technical scheme of the present invention is: on the basis of the scheme of background technology, the 1st and the 3rd step wherein optimized respectively.In the 1st step, taked the method for " complementary colors strengthens tobacco foreign matter colour difference ".The particular content that complementary colors strengthens the method for article such as tobacco and foreign matter colourity difference is: before the camera lens to article such as tobacco and the shooting of foreign matter potpourri, add that color is the color filter of article complementary colors such as tobacco, like this, in the middle of the image of being taken the photograph, the difference of article such as tobacco and foreign matter colourity is big when just directly making a video recording than not adding color filter.With tested article is that tobacco is an example, the ultimate principle of this method is: the reflective spectral property curve of having measured tobacco with the reflection spectrum measuring instrument, the reflected light of finding tobacco mainly concentrates on wavelength visible wave band (characteristic color is yellow), and various foreign matters then do not possess this feature.So, as before adopting the promptly blue color filter of yellow complementary colors to be added on pick-up lens, the potpourri of tobacco and foreign matter is made a video recording, then the light of tobacco reflection just become very weak, and foreign matter is quite different by the filtration of color filter.Like this, take the photograph image in the middle of, it is darker relatively that the color of tobacco seems, this has just increased the colourity difference of tobacco and foreign matter.Certainly, also can adopt the method for when shooting, throwing light on to realize that complementary colors strengthens tobacco and foreign matter colour difference, be difficult to the difficulty of control flexibly, adopt the method for color filter then can overcome this difficulty but can run into the light source waveband width with blue-light source.So still the scheme with color filter is good.
In order to reach better detection effect, this method has adopted " the alert and resourceful threshold value diagnostic method of tobacco and foreign matter colourity " in the 3rd step.Alert and resourceful threshold value diagnostic method is a kind of method that adopts when the image of taking the photograph the tested article foreign matter potpourri to computing machine is handled.Its basic thought is: when article color to be detected is more shallow on a certain section time production line in the production run, the value that indicates tested article colourity center is decided height a bit, and when tobacco leaf color to be detected changes dark on another section period production line, then this value is turned down, determined in order to differentiate the colourity threshold value of tested article and foreign matter based on tested article colourity central value again.Like this, just realized concrete Changing Pattern, set the purpose of best discrimination threshold at tested article colourity on the production line.The implementation method that alert and resourceful threshold value is differentiated algorithm is: to the same width of cloth image that video camera is taken the photograph, open up the buffer memory in two zones and store its data in calculator memory.The data in zone 1 are in order to differentiate tested article and foreign matter in real time.2 in zone is in order to upgrade threshold value, concrete grammar is: the data of some width of cloth images of being taken the photograph in record a period of time in zone 2 (big or small N is doubly to zone 1), the result who goes out according to the discriminating data of these some images again, chromatic value to the pixel of the tested article of representative in the zone 2 carries out the statistical average of position and time, draw the central value of a colourity threshold value, again according to the upper and lower bound of this central value decision threshold, to form the threshold value after upgrading.Like this, system just can be at the concrete situation of change of tobacco leaf colourity on the production line, regulates automatically in order to differentiate the threshold value of article such as tobacco leaf and foreign matter.Therefore this method can reduce False Rate than the method for fixed threshold.
The invention has the beneficial effects as follows: before the potpourri of article such as tobacco and foreign matter being made a video recording and discern processing, the method that strengthens tested article foreign matter colour difference with complementary colors strengthens both colour differences earlier, makes tested article and foreign matter discern more easily.When image being discerned processing, take alert and resourceful threshold value diagnostic method to improve the accurate differentiation rate of tested article foreign matter.
Description of drawings:
Accompanying drawing 1 is for strengthening the removing foreign material from tobacco system architecture diagram of alert and resourceful threshold value diagnostic method based on complementary colors.
Accompanying drawing 2 and accompanying drawing 3 are effect contrast figures that complementary colors strengthens tobacco foreign matter colour difference method.
Accompanying drawing 4 is to comprise the system works flow process figure that alert and resourceful threshold value is differentiated algorithm.
Specific embodiment
Below in conjunction with accompanying drawing, a specific embodiment of the present invention is described: the removing foreign material from tobacco system that strengthens alert and resourceful threshold value diagnostic method based on complementary colors.
Strengthen based on complementary colors alert and resourceful threshold value diagnostic method the removing foreign material from tobacco system structured flowchart as shown in Figure 1, its course of work can be described with following steps:
1, the image capture after the complementary colors humidification: on production line, when tobacco leaf stream is transferred into the travelling belt rear end, produce high velocity air with pneumatic means, tobacco leaf stream is coupled in the air-flow and with high speed (about 5 meter per seconds) impelling, flow when unsettled at tobacco leaf, with the CCD line array video camera that is added with blue color filter before the camera lens, the tobacco in the picked-up certain limit and the image of foreign matter potpourri.The effect of color filter is to realize that complementary colors strengthens, and its effect can be found out in the contrast with accompanying drawing 2 and accompanying drawing 3: from accompanying drawing 3 as seen: the colourity of tobacco leaf and foreign matter is difference slightly, but because both colourity difference are little, is difficult to differentiation.Accompanying drawing 3 is for adopting the image after complementary colors strengthens tobacco foreign matter colour difference, and its specific implementation method is: before the camera lens of video camera, added blue color filter, and then to tobacco and the shooting of various foreign matter potpourri.In accompanying drawing 3, the colour difference of visible tobacco leaf and foreign matter has obviously increased, and it is very dark that tobacco seems, foreign matter is then brighter relatively.Contrast method that the visible complementary colors of two figure strengthens tobacco foreign matter colour difference and can make tobacco and the easier differentiation of foreign matter in the image, thereby more help next step processing.
2, image transmission: taking the photograph image is reached in the middle of the calculator memory with the form of data with image pick-up card.
3, with alert and resourceful threshold value diagnostic method image is discerned processing: to reaching the expression tobacco in the internal memory and the data of foreign matter image, with the flow process shown in the accompanying drawing 4, set the alert and resourceful threshold value that changes along with product colourity variation to be detected on the production line and differentiate tobacco and foreign matter.By above analysis as can be known this method can improve the differentiation rate than the method for fixed threshold, thereby more help the realization of systemic-function.
4, identification signal transmission: after tobacco and foreign matter discerned, the signal of expression foreign matter position is transferred to ECU (Electrical Control Unit).
5, reject foreign matter: ECU (Electrical Control Unit) is controlled corresponding air nozzle with foreign body eliminating according to received foreign matter position signalling, and continuous and so forth circulation has just realized the online detection and the rejecting of tobacco foreign matter.Method and 2 of alert and resourceful threshold value diagnostic methods that complementary colors among the present invention strengthens tobacco foreign matter colour difference all play the raising discrimination, reduce the effect of False Rate.
Claims (2)
1, the foreign matter recognition methods in the online detection of article such as tobacco may further comprise the steps:
(1) image capture: promptly on production line, with the image of article such as the tobacco in the CCD line array video camera picked-up certain limit and foreign matter potpourri; (2) image transmission: taking the photograph view data is reached in the middle of the calculator memory with image pick-up card; (3) identification is handled: according to the data of article such as expression tobacco in the internal memory and foreign matter colourity, set the colourity threshold value and differentiate article and foreign matters such as tobacco; (4) identification signal transmission: after article such as tobacco and foreign matter discerned, the signal of expression foreign matter position is transferred to ECU (Electrical Control Unit); (5) reject foreign matter: ECU (Electrical Control Unit) is controlled corresponding air nozzle with foreign body eliminating according to received foreign matter position signalling;
It is characterized in that:
The method that has adopted complementary colors to strengthen in the image capture of step (1) promptly before the camera lens to article such as tobacco and the shooting of foreign matter potpourri, adds that color is the color filter of article complementary colors such as tobacco, strengthens the colourity difference of article such as tobacco and foreign matter.
2, the foreign matter recognition methods in the online detection of article such as tobacco according to claim 1, it is characterized in that: in the identification of step (3) is handled, adopted alert and resourceful threshold value diagnostic method, promptly the same width of cloth image that video camera is taken the photograph is opened up the buffer memory in two zones and is stored its data in calculator memory.Zone 1 data are in order to differentiate article and foreign matter such as tobacco in real time.2 in zone is in order to upgrade threshold value, concrete grammar is: the data of some width of cloth images of being taken the photograph in record a period of time in zone 2 (big or small N is doubly to zone 1), the result who goes out according to the discriminating data of these some width of cloth images again, to representing the chromatic value of the pixel of article such as tobacco to carry out the statistical average of position and time in the zone 2, draw the central value of a threshold value, again according to the Upper threshold and the Lower Threshold of this central value decision threshold, to form the threshold value after upgrading.Like this, system just can regulate the threshold value of article such as differentiating tobacco and foreign matter automatically at the concrete situation of change of tested article colourities such as tobacco leaf on the production line.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021138257A CN1158021C (en) | 2002-06-04 | 2002-06-04 | Method for recognizing impurities in in-line detection of tobacco |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021138257A CN1158021C (en) | 2002-06-04 | 2002-06-04 | Method for recognizing impurities in in-line detection of tobacco |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1382399A true CN1382399A (en) | 2002-12-04 |
| CN1158021C CN1158021C (en) | 2004-07-21 |
Family
ID=4742826
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB021138257A Expired - Fee Related CN1158021C (en) | 2002-06-04 | 2002-06-04 | Method for recognizing impurities in in-line detection of tobacco |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1158021C (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1293835C (en) * | 2005-01-06 | 2007-01-10 | 重庆大学 | On line foreign matter distinguishing method for article inspection based on unit gradation uniformity |
| CN1293836C (en) * | 2005-01-06 | 2007-01-10 | 重庆大学 | Lines feature on-line identifying method for detecting foreign matter in food |
| CN1689715B (en) * | 2004-04-27 | 2010-06-02 | 豪尼制丝设备责任有限公司 | Method and apparatus for separating foreign matter from material flow |
| CN102297867A (en) * | 2011-07-20 | 2011-12-28 | 上海元一电子有限公司 | Detection system for assembly quality of wiring harness |
| CN102359962A (en) * | 2011-07-14 | 2012-02-22 | 红塔烟草(集团)有限责任公司 | Equipment for determining long tobacco stalk rate by using image analysis |
| CN106944361A (en) * | 2016-01-07 | 2017-07-14 | 西派特（北京）科技有限公司 | A kind of technology of High-Speed Automatic sorting tobacco leaf degree |
| CN111103286A (en) * | 2019-12-10 | 2020-05-05 | 红云红河烟草(集团)有限责任公司 | Cigarette stream extraction method under normal illumination condition of factory |
| CN112189877A (en) * | 2020-10-13 | 2021-01-08 | 燕山大学 | On-line detection method for tobacco shred impurities in tobacco production line |
-
2002
- 2002-06-04 CN CNB021138257A patent/CN1158021C/en not_active Expired - Fee Related
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1689715B (en) * | 2004-04-27 | 2010-06-02 | 豪尼制丝设备责任有限公司 | Method and apparatus for separating foreign matter from material flow |
| CN1293835C (en) * | 2005-01-06 | 2007-01-10 | 重庆大学 | On line foreign matter distinguishing method for article inspection based on unit gradation uniformity |
| CN1293836C (en) * | 2005-01-06 | 2007-01-10 | 重庆大学 | Lines feature on-line identifying method for detecting foreign matter in food |
| CN102359962A (en) * | 2011-07-14 | 2012-02-22 | 红塔烟草(集团)有限责任公司 | Equipment for determining long tobacco stalk rate by using image analysis |
| CN102297867A (en) * | 2011-07-20 | 2011-12-28 | 上海元一电子有限公司 | Detection system for assembly quality of wiring harness |
| CN106944361A (en) * | 2016-01-07 | 2017-07-14 | 西派特（北京）科技有限公司 | A kind of technology of High-Speed Automatic sorting tobacco leaf degree |
| CN111103286A (en) * | 2019-12-10 | 2020-05-05 | 红云红河烟草(集团)有限责任公司 | Cigarette stream extraction method under normal illumination condition of factory |
| CN111103286B (en) * | 2019-12-10 | 2022-07-01 | 红云红河烟草(集团)有限责任公司 | Cigarette stream extraction method under normal illumination condition of factory |
| CN112189877A (en) * | 2020-10-13 | 2021-01-08 | 燕山大学 | On-line detection method for tobacco shred impurities in tobacco production line |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1158021C (en) | 2004-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1158021C (en) | Method for recognizing impurities in in-line detection of tobacco | |
| CN103801514B (en) | A kind of scrap prodn. line tobacco leaf automation selecting device and method | |
| CN109269951A (en) | Floating tail-coal ash content, concentration, coarse granule detection method of content based on image | |
| CN104841646B (en) | The de-burring machine and impurity-removing method of a kind of different-time exposure and multispectral imaging | |
| CN1590988A (en) | Particle colour sorting device with display controller | |
| CN103808723A (en) | Exhaust gas blackness automatic detection device for diesel vehicles | |
| CN1293835C (en) | On line foreign matter distinguishing method for article inspection based on unit gradation uniformity | |
| CN105388162A (en) | Raw material silicon wafer surface scratch detection method based on machine vision | |
| CN105466950A (en) | Bamboo dice online detection method and system based on vision | |
| CN109146841B (en) | Visual detection method and system for filling defects of dry battery coated paper | |
| CN108563979A (en) | A method of based on the farmland image discriminating rice blast state of an illness of taking photo by plane | |
| CN110443102A (en) | Living body faces detection method and device | |
| CN115018838A (en) | Method for identifying pitting defects on surface of oxidized steel pipe material | |
| CN109345524A (en) | A kind of bearing open defect detection system of view-based access control model | |
| CN107221070A (en) | A kind of bill anti-counterfeit discrimination method recognized based on master pattern fluorescent characteristics | |
| CN110502971A (en) | Road vehicle recognition methods and system based on monocular vision | |
| CN1293836C (en) | Lines feature on-line identifying method for detecting foreign matter in food | |
| CN113820319A (en) | Textile surface defect detection device and method | |
| CN102436578A (en) | Formation method for dog face characteristic detector as well as dog face detection method and device | |
| CN108983194B (en) | Target extraction and condensation method based on ground monitoring radar system | |
| CN110991245A (en) | Real-time smoke detection method based on deep learning and optical flow method | |
| CN102004913A (en) | Method for detecting transparent plastic impurities in seed cotton | |
| Wyeth et al. | Robust adaptive vision for robot soccer | |
| CN113657250A (en) | Flame detection method and system based on monitoring video | |
| CN106128112B (en) | Night bayonet vehicle identifies grasp shoot method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C19 | Lapse of patent right due to non-payment of the annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |