CN104535580A - Cigarette filter stick built-in object detection method based on machine vision technology - Google Patents
Cigarette filter stick built-in object detection method based on machine vision technology Download PDFInfo
- Publication number
- CN104535580A CN104535580A CN201410851370.6A CN201410851370A CN104535580A CN 104535580 A CN104535580 A CN 104535580A CN 201410851370 A CN201410851370 A CN 201410851370A CN 104535580 A CN104535580 A CN 104535580A
- Authority
- CN
- China
- Prior art keywords
- filter stick
- baffle plate
- cigarette filter
- industrial camera
- light source
- 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.)
- Pending
Links
Abstract
The invention discloses a cigarette filter stick built-in object detection method based on a machine vision technology. After the characteristic of a cigarette filter stick containing built-in objects is deeply learned, the detection mode for the cigarette filter stick containing the built-in objects is provided in an innovative mode, and through a special image collecting mode and a special projection transformation method, deficiency and deviation of the objects inside the filter stick are detected. According to the method, the difficulty of image processing can be greatly lowered, the detection precision is high, the quality of the filter stick and cigarettes can be guaranteed, and the method has broad application spaces in the similar detection process of the built-in objects.
Description
Technical field
The present invention relates to cigarette-testing method field, specifically a kind of cigarette filter stick inner matter detection method based on machine vision technique.
Background technology
Filter stick is the primary raw material producing cigaratte filter, and it can be made into the filter tip of cigarette in cigarette process through operations such as cuttings.In order to the objectionable constituent of security especially in selective removal flue gas improving smoking, current domestic a lot of cigar mill all adds some special inner matters in cigaratte filter, ratio is if any the activated charcoal of special adsorption effect and have secreting of opsonic action to moisten ball etc. to some organ of human body, reaches the object of " lowering harm and decreasing coking " with this.
Add in the filter stick production run of inner matter this, inner matter disappearance, the inner matter defect such as larger out of position in filter stick may being occurred, for ensureing the quality of filter stick and cigarette, needing to carry out quality testing to filter stick; At present a kind of perfect detection method also be there is no to such special cigarette filter stick, be therefore badly in need of a kind of detection method, accurately detect qualified cigarette filter stick.
summary of the inventionthe object of this invention is to provide a kind of cigarette filter stick inner matter detection method based on machine vision technique, to solve the problem that prior art cigarette filter stick detects existing defects.
In order to achieve the above object, the technical solution adopted in the present invention is:
A kind of cigarette filter stick inner matter detection method based on machine vision technique, it is characterized in that: adopt bias light radiation modality, LED strip shape light source, filter stick to be detected and industrial camera are installed on same level line, wherein filter stick is placed in a groove with the light tight baffle plate of groove, and LED strip shape light source and industrial camera are placed in the forward and backward side of baffle plate respectively; When LED strip shape light source is from ensureing during baffle plate frontside illuminated that light only passes in filter stick, industrial camera gathers filter stick image at the rear side of baffle plate and is sent to computing machine; Computing machine carries out gray processing, area-of-interest location, projective transformation process to filter stick image, finally judges filter stick to be detected whether existing defects according to drop shadow curve's waveform situation.
The invention has the beneficial effects as follows:
The feature utilizing object transmittance different of the present invention's innovation, adopts the method such as back lighting, projective transformation to detect containing secreting the cigarette filter stick moistening ball.Because different objects transmittance is different, ambient light illumination mode can make filter stick when secreting the mistake of profit segment, skew and normal filter stick produces notable difference, the process such as vertical projection conversion are utilized this species diversity to be changed into the data representation of image expression and quantification intuitively, be convenient to the detection of cigarette filter stick, thus ensure that filter stick and quality cigarette.The inventive method is a kind of novel cigarette filter stick inner matter detection mode, also can be generalized in similar inner matter detection simultaneously, has a wide range of applications space.
Accompanying drawing explanation
Fig. 1 be gather under common illumination include 4 secrete profit ball cigarette filter stick image.
Fig. 2 is light source, filter stick and industrial camera scheme of installation.
Fig. 3 is the intact filter stick image that industrial camera collects.
Fig. 4 is the horizontal projection curve map of filter stick image.
Fig. 5 is the area-of-interest of filter stick image.
Fig. 6 carries out vertical projection conversion etc. to area-of-interest to process the drop shadow curve obtained.
Fig. 7 is that secreting in filter stick moistens the locating effect figure of ball in drop shadow curve.
Fig. 8 is drop shadow curve in filter stick under certain ball deletion condition and locating effect figure.
Fig. 9 is drop shadow curve in filter stick in the larger situation of certain ball skew and locating effect figure.
Embodiment
A kind of cigarette filter stick inner matter detection method based on machine vision technique, adopt bias light radiation modality, LED strip shape light source, filter stick to be detected and industrial camera are installed on same level line, wherein filter stick is placed in a groove with the light tight baffle plate of groove, and LED strip shape light source and industrial camera are placed in the forward and backward side of baffle plate respectively; When LED strip shape light source is from ensureing during baffle plate frontside illuminated that light only passes in filter stick, industrial camera gathers filter stick image at the rear side of baffle plate and is sent to computing machine; Computing machine carries out gray processing, area-of-interest location, projective transformation process to filter stick image, finally judges filter stick to be detected whether existing defects according to drop shadow curve's waveform situation.
Under common white light front illuminated, the cigarette filter stick image that industrial camera collects as shown in Figure 1, can see in figure that filter stick outside surface is neatly intact, but whether is beyond recognition out filter stick inside containing secreting profit ball, is also beyond recognition out number and the position thereof of ball.
Consider that to secrete profit ball poor than the light transmission of filter stick, the present invention proposes employing back lighting mode, and light source, filter stick, industrial camera mounting structure are as shown in Figure 2.Be placed in by filter stick on a light tight baffle plate, baffle plate is with groove sizable with filter stick, and LED light source is aimed at filter stick and irradiated from baffle plate side, baffle plate makes light only pass from filter stick, and industrial camera gathers image at filter stick opposite side.Containing the position of secreting profit ball, light is through less, and in the filter stick image gathered, this place's grey scale pixel value is also relatively low, in the dash area assembled, as Fig. 3 is the filter stick image that industrial camera gathers.
Greyscale transformation is carried out to collection image, and the location of area-of-interest (ROI) is carried out to gray level image.Location adopts the method for horizontal projection, with the row pixel average gray value of gray level image for feature carries out the projection of horizontal direction, drop shadow curve as shown in Figure 4, because filter stick part gray-scale value is higher, in drop shadow curve, the correspondence position projection value of filter stick is comparatively large, with the mean value of curve for reference value, can well be partitioned into filter stick area-of-interest, part higher than reference value is area-of-interest, as shown in Figure 5.
In the area-of-interest of filter stick image, adopt vertical projection method to detect secreting profit ball in filter stick, vertical projection curve as shown in Figure 6.Because secrete profit ball in the dash area assembled in filter stick image, the projection value corresponding in drop shadow curve's top shadow part is also little, and drop shadow curve's image shows as the trough that four amplitudes are larger, and crest and trough numerical value differ greatly.By stages finds its correspondence of the minimum value meeting certain threshold value in each interval to be exactly the position of ball in filter stick, locating effect as shown in Figure 7, in drop shadow curve vertical white line be find secrete profit ball position.
For there is the filter stick that segment loses or ball offsets, its vertical projection curve and ball locating effect are respectively as shown in FIG. 8 and 9.In Fig. 8, in filter stick, second segment loses, so there is not the trough satisfied condition in the detection zone of second ball corresponding in drop shadow curve.In like manner, in Fig. 9, the 3rd ball offset by precalculated position, also there is not the trough satisfied condition, i.e. filter stick existing defects in the detection zone of corresponding drop shadow curve's the 3rd ball.
Claims (1)
1. the cigarette filter stick inner matter detection method based on machine vision technique, it is characterized in that: adopt bias light radiation modality, LED strip shape light source, filter stick to be detected and industrial camera are installed on same level line, wherein filter stick is placed in a groove with the light tight baffle plate of groove, and LED strip shape light source and industrial camera are placed in the forward and backward side of baffle plate respectively; When LED strip shape light source is from ensureing during baffle plate frontside illuminated that light only passes in filter stick, industrial camera gathers filter stick image at the rear side of baffle plate and is sent to computing machine; Computing machine carries out gray processing, area-of-interest location, projective transformation process to filter stick image, finally judges filter stick to be detected whether existing defects according to drop shadow curve's waveform situation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410851370.6A CN104535580A (en) | 2014-12-31 | 2014-12-31 | Cigarette filter stick built-in object detection method based on machine vision technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410851370.6A CN104535580A (en) | 2014-12-31 | 2014-12-31 | Cigarette filter stick built-in object detection method based on machine vision technology |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104535580A true CN104535580A (en) | 2015-04-22 |
Family
ID=52851152
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410851370.6A Pending CN104535580A (en) | 2014-12-31 | 2014-12-31 | Cigarette filter stick built-in object detection method based on machine vision technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104535580A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104807829A (en) * | 2015-04-17 | 2015-07-29 | 武汉易视维科技有限公司 | Visual detection system and method for filter |
CN106056617A (en) * | 2016-05-31 | 2016-10-26 | 中国电子科技集团公司第四十研究所 | Cigarette carton sealing machine hot melt adhesive detection method |
CN106091936A (en) * | 2016-06-01 | 2016-11-09 | 中国电子科技集团公司第四十研究所 | A kind of cellophane offset detecting device based on machine vision technique and method |
CN106226316A (en) * | 2016-08-31 | 2016-12-14 | 江苏大学 | A kind of single camera wide visual field vision thread detecting device and detection method thereof |
CN106896112A (en) * | 2015-12-21 | 2017-06-27 | 深圳市青铜科技有限公司 | Flake class terminal is in vision-based detection by means of the optical imaging method of shadow shield |
CN109682305A (en) * | 2019-02-27 | 2019-04-26 | 厦门烟草工业有限责任公司 | Filter stick implosion pearl axial position detection device and method |
CN110947647A (en) * | 2019-11-28 | 2020-04-03 | 中国电子科技集团公司第四十一研究所 | Bead blasting filter stick on-line measuring device |
CN111084412A (en) * | 2019-11-28 | 2020-05-01 | 中国电子科技集团公司第四十一研究所 | Online detection device for fragrant line filter stick |
CN111229637A (en) * | 2020-01-07 | 2020-06-05 | 江苏科技大学 | Automatic detection device and detection method for inner hole burrs of electronic connector |
CN116843692A (en) * | 2023-09-04 | 2023-10-03 | 青岛冠宝林活性炭有限公司 | Regenerated active carbon state detection method based on artificial intelligence |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070189597A1 (en) * | 2005-08-23 | 2007-08-16 | Limer Daniel J | Machine vision counting system apparatus and method |
CN101893580A (en) * | 2010-06-10 | 2010-11-24 | 北京交通大学 | Digital image based detection method of surface flaw of steel rail |
CN102759531A (en) * | 2012-07-25 | 2012-10-31 | 深圳市华星光电技术有限公司 | Automatic optical detection device |
CN103884631A (en) * | 2014-04-02 | 2014-06-25 | 川渝中烟工业有限责任公司 | Quick testing method for particle sizes of aromatic beads for tobacco |
-
2014
- 2014-12-31 CN CN201410851370.6A patent/CN104535580A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070189597A1 (en) * | 2005-08-23 | 2007-08-16 | Limer Daniel J | Machine vision counting system apparatus and method |
CN101893580A (en) * | 2010-06-10 | 2010-11-24 | 北京交通大学 | Digital image based detection method of surface flaw of steel rail |
CN102759531A (en) * | 2012-07-25 | 2012-10-31 | 深圳市华星光电技术有限公司 | Automatic optical detection device |
CN103884631A (en) * | 2014-04-02 | 2014-06-25 | 川渝中烟工业有限责任公司 | Quick testing method for particle sizes of aromatic beads for tobacco |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104807829A (en) * | 2015-04-17 | 2015-07-29 | 武汉易视维科技有限公司 | Visual detection system and method for filter |
CN106896112A (en) * | 2015-12-21 | 2017-06-27 | 深圳市青铜科技有限公司 | Flake class terminal is in vision-based detection by means of the optical imaging method of shadow shield |
CN106056617A (en) * | 2016-05-31 | 2016-10-26 | 中国电子科技集团公司第四十研究所 | Cigarette carton sealing machine hot melt adhesive detection method |
CN106091936A (en) * | 2016-06-01 | 2016-11-09 | 中国电子科技集团公司第四十研究所 | A kind of cellophane offset detecting device based on machine vision technique and method |
CN106226316A (en) * | 2016-08-31 | 2016-12-14 | 江苏大学 | A kind of single camera wide visual field vision thread detecting device and detection method thereof |
CN109682305B (en) * | 2019-02-27 | 2024-04-05 | 厦门烟草工业有限责任公司 | Device and method for detecting axial position of filter stick implosion beads |
CN109682305A (en) * | 2019-02-27 | 2019-04-26 | 厦门烟草工业有限责任公司 | Filter stick implosion pearl axial position detection device and method |
CN110947647A (en) * | 2019-11-28 | 2020-04-03 | 中国电子科技集团公司第四十一研究所 | Bead blasting filter stick on-line measuring device |
CN111084412B (en) * | 2019-11-28 | 2022-02-25 | 中国电子科技集团公司第四十一研究所 | Online detection device for fragrant line filter stick |
CN111084412A (en) * | 2019-11-28 | 2020-05-01 | 中国电子科技集团公司第四十一研究所 | Online detection device for fragrant line filter stick |
CN111229637A (en) * | 2020-01-07 | 2020-06-05 | 江苏科技大学 | Automatic detection device and detection method for inner hole burrs of electronic connector |
CN116843692A (en) * | 2023-09-04 | 2023-10-03 | 青岛冠宝林活性炭有限公司 | Regenerated active carbon state detection method based on artificial intelligence |
CN116843692B (en) * | 2023-09-04 | 2023-12-01 | 青岛冠宝林活性炭有限公司 | Regenerated active carbon state detection method based on artificial intelligence |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104535580A (en) | Cigarette filter stick built-in object detection method based on machine vision technology | |
CN109028233B (en) | Kitchen oil smoke concentration dividing method, oil smoke image recognition system and range hood | |
CN104135660B (en) | A kind of dirty detection method of camera module and detection system | |
CN102261892B (en) | Rapid width measurement system for tobacco shreds | |
CN104598915A (en) | Gesture recognition method and gesture recognition device | |
CN106871570A (en) | A kind of device based on various dietary regimens in single multispectral imaging unit detection refrigerating chamber | |
CN204064961U (en) | Scarce of cigarette short and detection device for foreign matter | |
CN102269710A (en) | Rapid lossless prediction device of fresh port validity based on multispectral imaging | |
CN103679790B (en) | A kind of Bunker coal tonnage detection method based on image processing techniques | |
CN103033476A (en) | Equipment and method for online detecting tobacco leaves with stems based on infrared imaging | |
CN104568989A (en) | Glass substrate defect detecting method | |
CN101726265A (en) | Online detection method of parallelism of spring by machine vision system | |
CN101937505B (en) | Target detection method and equipment and used image acquisition device thereof | |
CN111665251A (en) | Visual detection method for surface defects | |
CN103903284A (en) | Multi-circle detecting method based on least square method | |
CN102706879A (en) | Detection system of lip surface quality of lip-shaped sealing member | |
CN202074946U (en) | Cigarette case appearance detecting device | |
CN105973903B (en) | A kind of Oral liquid bottle lid detection method | |
CN202939619U (en) | Coal rock identification system based on image discrete multi-wavelet transformation | |
CN103942792A (en) | Impurity detecting method in medicine detection robot based on time domain features of sequence images | |
CN109915867B (en) | Data processing method of oil smoke sensor | |
CN107967448B (en) | Early fire smoke real-time detection method and system | |
CN106372644A (en) | Image identification method used for wheel hub sorting system | |
CN110210364A (en) | A kind of control method, terminal and computer storage medium | |
CN203078780U (en) | Cigarette filter tip hollow pattern integrity detection device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150422 |