CN104318277B - A kind of merchandise sales process detection device - Google Patents
A kind of merchandise sales process detection device Download PDFInfo
- Publication number
- CN104318277B CN104318277B CN201410532855.9A CN201410532855A CN104318277B CN 104318277 B CN104318277 B CN 104318277B CN 201410532855 A CN201410532855 A CN 201410532855A CN 104318277 B CN104318277 B CN 104318277B
- Authority
- CN
- China
- Prior art keywords
- commodity
- image
- detection device
- bar code
- merchandise sales
- 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
Landscapes
- Cash Registers Or Receiving Machines (AREA)
Abstract
The invention discloses a kind of merchandise sales process detection device, including shell, conveying mechanism main motor (3), light-source head (5) and commodity disengaging conveying mechanism (8), camera (9), code reader (10), commodity detection device (11), article portal (12), commodity prize make marks laser (13) and commodity export (15).Merchandise sales process structure of the detecting device of the invention is simple, easy to use, can be convenient and effectively commodity are detected and marked.
Description
Technical field
The present invention relates to merchandise sales process detection devices.
Background technique
In the prior art, the commodity with glassine paper and bar code detect and be all made of artificial progress, inefficiency,
It cannot achieve automation to carry out.
Therefore, it is necessary to merchandise sales process detection devices to solve the above problems.
Summary of the invention
The purpose of the present invention is mention for the deficiency detected in the prior art for the commodity with glassine paper and bar code
For merchandise sales process detection device.
For achieving the above object, following technical solution can be used in merchandise sales process detection device of the present invention:
A kind of merchandise sales process detection device, including shell, conveying mechanism main motor, commodity pass in and out conveying mechanism, take the photograph
As head, code reader, commodity detection device, article portal, commodity prize make marks laser and commodity export,
The two sides of the shell respectively on be provided with article portal and commodity export, the commodity disengaging conveying mechanism setting
In the housing, one end of the commodity disengaging conveying mechanism, which connects, is provided with quotient between the article portal and commodity export
Product pass in and out at conveying mechanism, and the conveying mechanism main motor connects the commodity and passes in and out conveying mechanism, the commodity disengaging conveying
Mechanism includes conveyer belt, and the connecting pin of the commodity disengaging conveying mechanism and the article portal is provided with commodity detection device,
It is provided with camera, code reader and commodity prize above the conveyer belt to make marks laser, the camera and commodity are converted
The surface that the laser that makes marks is directed at commodity is encouraged, the code reader is directed at the bar code of the commodity.
Further, further include printer and printing paper, prize exit slot is provided on front side of the shell, it is described to beat
Print machine is arranged at the prize exit slot, and the printing paper connects the printer.
It further, further include reflection laser and baffle, the reflection laser is from the side of the commodity to quotient
Product are irradiated, and the baffle plate setting is on the route that the reflection laser irradiates the reflected light that the commodity generate.It utilizes
The light of reflection laser irradiates commodity, and light is projected on baffle by reflection.Detection baffle glazed thread reflected intensity comes area
Whether there is not glassine paper.To the commodity of same packaging, whether there is or not glassine paper, light intensity has very big difference, it is easy to distinguish.
Further, the reflection laser is linear red laser.Through testing, red laser reflecting effect compared with
It is good, it is easy to differentiate.
Further, the baffle is black plate.Black is all 0 in RGB Three Channel Color image intermediate value, reduces and makes an uproar
Acoustic jamming.
Further, the commodity prize makes marks laser as stripe laser.By stripe laser in commodity
On stamp the bar shaped marking, it is convenient that commodity are detected.
Further, the commodity detection device is photoelectric sensor.
Further, the article portal and commodity export share one outlet.
Further, the detection method of commercial glasses paper, comprising the following steps:
1) image on the baffle, is acquired using camera, the length of described image is identical as commodity length, described image
For RGB image;
2) channel R of described image, is selected;
3) vertical gradient processing, is carried out to image, obtains gradient image;
4) Threshold segmentation, is carried out to the gradient image that step 3) obtains, wherein obtain segmentation figure, the background of segmentation figure is
Black, effective coverage are white;
5), statistic procedure 3) obtained in effective coverage length, when the length of effective coverage is greater than image length
When 75%, then commercial glasses paper is complete;When the length of effective coverage is less than the 75% of image length, then commercial glasses paper
It is incomplete.Cigarette packet glassine paper detection method of the invention can facilitate the glass for judging cigarette packet by detecting reflection light length
Glass paper completely whether, be easy implement, stability is good, detection speed it is fast.
Further, it with the barcode label detection method of bar code commodity, is marked in band bar code bar code region,
The following steps are included:
1) it, takes pictures to band bar code commodity, obtains the image with bar code commodity;
2), pre-processing: gradient transformation is carried out to the image with bar code commodity, obtains gradient image;
3), barcode position in the image with bar code commodity is positioned;
4) Threshold segmentation, is carried out to bar code area image and forms binary map;
5) erosion operation, is carried out to the binary map that step 4) obtains and obtains mark shape.It is of the invention with bar code commodity
Barcode label detection method is marked in band bar code bar code region, and detection method is simple, and stability is high, is applicable in different product
The commodity of board Different Package, the relative positional accuracy for camera and with bar code commodity are of less demanding.
The utility model has the advantages that merchandise sales process structure of the detecting device of the invention is simple, and it is easy to use, it can be convenient effective
Commodity are detected and are marked.
Detailed description of the invention
Fig. 1 is the positive structural schematic diagram of merchandise sales process detection device;
Fig. 2 is merchandise sales process detection device C-C to diagrammatic cross-section;
Fig. 3 is the perspective view of merchandise sales process detection device;
Fig. 4 is the detection schematic diagram of cigarette packet glassine paper detection method;
Fig. 5 is the flow chart of cigarette packet glassine paper detection method;
Fig. 6 is the flow chart of the barcode label detection method with bar code commodity.
Specific embodiment
In the following with reference to the drawings and specific embodiments, the present invention is furture elucidated, it should be understood that these embodiments are merely to illustrate
It the present invention rather than limits the scope of the invention, after the present invention has been read, those skilled in the art are to of the invention each
The modification of kind equivalent form falls within the application range as defined in the appended claims.
It please refers to shown in Fig. 1, Fig. 2 and Fig. 3, merchandise sales process detection device of the invention, including shell, conveying mechanism
Main motor 3, commodity pass in and out conveying mechanism 8, camera 9, code reader 10, commodity detection device 11, article portal 12, commodity prize
Make marks laser 13 and commodity export 15.
Article portal 12 and commodity export 15 are provided on shell, commodity pass in and out the setting of conveying mechanism 8 in the shell, commodity
The one end for passing in and out conveying mechanism 8 connects article portal 12, and conveying mechanism main motor 3 connects commodity and passes in and out conveying mechanism 8.It is preferred that
, article portal 12 and commodity export 15 share one outlet, and only need to open on the shell one outlet can be completed task in this way,
The disengaging of commodity can be realized in the forward and reverse that conveying mechanism main motor 3 need to only be controlled.It includes defeated that commodity, which pass in and out conveying mechanism 8,
Band is sent, commodity disengaging conveying mechanism 8 and the connecting pin of article portal 12 are provided with commodity detection device 11, it is preferred that commodity inspection
Survey device 11 is photoelectric sensor, model E3Z-D81.Camera 9, code reader 10 and commodity are provided with above conveyer belt
Prize makes marks laser 13.Commodity prize makes marks laser 13 as stripe laser.Through stripe laser on commodity
The bar shaped marking is stamped, it is convenient that commodity are detected.Camera 9 and commodity the prize laser 13 that makes marks are directed at commodity 14
Surface, code reader 10 are directed at the bar code of commodity 14.
Further include printer 6 and printing paper 7, prize exit slot is provided on front side of shell, the setting of printer 6 goes out in prize
At paper mouth, printing paper 7 connects printer 6.
It further include reflection laser 1 and baffle, reflection laser 1 is irradiated commodity 14 from the side of commodity 14, gear
Plate is arranged in reflection laser 1 and irradiates on the route for the reflected light that commodity 14 generate.Quotient is irradiated using the light of reflection laser
Product, light are projected on baffle by reflection.Whether detect baffle glazed thread reflected intensity has glassine paper to distinguish.To same
The commodity of packaging, whether there is or not glassine paper, light intensity has very big difference, it is easy to distinguish.Wherein, it is preferred that reflection laser 1 is
Linear red laser.Through testing, red laser reflecting effect is preferable, is easy to differentiate.Baffle is black plate, and black is in RGB
Three Channel Color image intermediate value is all 0, reduces noise jamming.
It please refers to shown in Fig. 4 and Fig. 5, by taking cigarette packet as an example, the detection method of specific cigarette packet glassine paper is as follows:
Cigarette packet glassine paper detection method of the invention, is irradiated cigarette packet using laser in the side of cigarette packet, is swashing
Baffle is set on the route of light device reflected light, wherein baffle is black baffle.Black RGB Three Channel Color image intermediate value all
It is 0, reduces noise jamming.Laser is linear red laser.Through testing, red laser reflecting effect is preferable, is easy
It differentiates.And cigarette packet glassine paper detection method of the invention determines whether glassine paper is complete using the length of detection reflected light,
So using linear red laser.
The following steps are included:
1), using the image on camera acquisition baffle, the length of image is identical as cigarette packet length, and image is RGB image;
2) channel R of image, is selected;
3) vertical gradient processing, is carried out to image, obtains gradient image;Vertical gradient, the as gradient value of vertical direction.
Since glassine paper is not the mirror-reflection of standard after all, there are diffusing reflection situations.It makes it difficult to come out target from background separation.
As shown in Figure 1.General pattern edge is all more stable, is one horizontal line of detection based on use occasion, so ladder is only used only
The value of vertical direction is spent, formula is.
4) Threshold segmentation, is carried out to the gradient image that step 3) obtains, wherein obtain segmentation figure, the background of segmentation figure is
Black, effective coverage are white;Wherein, it is preferred that the segmentation threshold of Threshold segmentation is 3-9, this value is by test of many times institute
It obtains as a result, high stability.
5), statistic procedure 3) obtained in effective coverage length, when the length of effective coverage is greater than image length
When 75%, then cigarette packet glassine paper is complete;When the length of effective coverage is less than the 75% of image length, then cigarette packet glassine paper
It is complete.The length for counting effective coverage is more stable than statistics area, so being sentenced using the length of statistics effective coverage
Disconnected glassine paper completely whether.Length is relative to x coordinate axis, and each x coordinate is up to a point.
Preferably, it is 2 sections of images by image segmentation in step 1), step 2) -5 is respectively adopted to each section of image) to glass
Glass paper is detected, wherein image length is the length of each section of image in step 5).
2 sections of images respectively correspond the both ends of cigarette packet.The glass of general cigarette packet is all to tear since both ends, therefore 2 sections are schemed
Both ends as respectively corresponding cigarette packet can effectively determine the avulsed position of glassine paper.
Cigarette packet glassine paper detection method of the invention can facilitate the glass for judging cigarette packet by detecting reflection light length
Paper completely whether, be easy implement, stability is good, detection speed it is fast.
It please refers to shown in Fig. 6, by taking the commodity with bar code as an example, the barcode label detection of the invention with bar code commodity
Method is marked in band bar code bar code region, comprising the following steps:
1) it, takes pictures to band bar code commodity, obtains the image with bar code commodity;
2), pre-processing: gradient transformation is carried out to the image with bar code commodity, obtains gradient image;
3), barcode position in the image with bar code commodity is positioned;
4) Threshold segmentation, is carried out to bar code area image and forms binary map;
5) erosion operation, is carried out to the binary map that step 4) obtains and obtains mark shape.
Preferably, Threshold segmentation described in step 4) is ostu dynamic threshold segmentation.Ostu dynamic threshold segmentation can be certainly
It is dynamic to find threshold value, the fixed influence of flashing is eliminated, realizes image segmentation.
Wherein, gradient is transformed to the direction x in step 2) and the gradient in the direction y converts.Wherein, the direction x is band bar code commodity
Length direction, the direction y be the thickness direction with bar code commodity.
Wherein, in step 3) to the method positioned with barcode position in bar code commodity image the following steps are included:
One, gradient image in step 2) is subjected to Threshold segmentation, filters out non-edge part;
Two, image obtained in step 1 is successively subjected to closing operation processing, the stripe edge of bar code is connected to
Come;
Three, finding step 2) in largest connected region;
Four, the largest connected region bound and left and right limit that step 3) obtains are determined, the position in bar code region is obtained.From even
The approximate region of bar code is split in logical figure, but is easy to make other incoherent regional connectivities
Bar code region at segmentation is excessive, therefore is also necessary to precisely determine the bound and left and right limit in largest connected region.
Specifically, seeking variance by the gradient image in the direction x to largest connected region in step 4, being averaged for variance is asked
Value a, wherein region of the variance greater than 0.5a is bar code region, so that it is determined that the bound b in the largest connected region.It utilizes
Variance determines that the bound b stability in largest connected region is high,
Specifically, largest connected region or so limit is determined in step 4 the following steps are included:
(1), by the conduct threshold value of 0.5b;
(2), any position in bar code region is taken, and calculates bound paragraph by paragraph to the two sides of this position;
It (3), is then largest connected region when the bound that step (2) is calculated is lower than threshold value described in step (1)
Left and right limit.The left side in largest connected region is calculated on the basis of the bound b in the largest connected region of high stability
Right limit, stability are also higher.
Inventive principle: having bar code on each band bar code commodity, and texture is single in bar code, and background is also very stable.
It is suitble to=positions the band bar code commodity for detecting with label, and being suitable for Different Package.
Embodiment 1:
There is bar code on each band bar code commodity, texture is single in bar code, and background is also very stable.Be suitble to positioning with
Label detection, and it is suitable for the band bar code commodity of Different Package.It mainly include three parts, pre-processing for detection algorithm;
Bar code positioning;Label detection.
Pre-processing: mainly carrying out the transformation (direction x and y direction gradient) of gradient to target image, as in next step
Basis.
Bar code positioning: being the premise of entire algorithm success or not.Since bar code is characterized in that strip has after horizontal positioned
Very strong gradient
After (Fx-Fy) transformation, bar edges have very strong response, can filter out non-edge by fixed threshold segmentation
Part.And each stripe edge of bar code is apart from close, by can be by item after the closing operation of Gaussian Blur and suitable shape
The stripe edge connection of code is got up, and is extremely difficult to by the other parts of test tape bar code commodity so big.By searching for maximum
Connected region can be rough the region for finding bar code.It may be also comprising behind other uncorrelated some effects in this time domain
Detection, the determination limited up and down for this region.Bound uses the variance in Fx figure bar code region.Area beyond bar code
There is very strong decaying in domain, and bound can effectively be determined by setting a ratio.The determination of left and right limit is relatively difficult, in connected graph
Middle setting height ratio is it was determined that but less stable.So during (being compared with empirical value) in next step, beyond normal
Value.Parameters are then readjusted to recalculate.
Label detection: after determining bar code region, to eliminate the fixed influence of flashing, it is dynamic that ostu is carried out to area image
State Threshold segmentation forms binary map.Mark shape (shape need to meet a certain size requirement) is searched by Image erosion operation again,
The detection of label can be realized.
Barcode label detection method with bar code commodity of the invention is marked in the bar code region with bar code commodity, inspection
Survey method is simple, and stability is high, is applicable in the commodity of different brands Different Package, the relative position for camera and with bar code commodity
Required precision is not high.
Merchandise sales process structure of the detecting device of the invention is simple, easy to use, can be convenient effectively to commodity into
Row detection and label.
Claims (8)
1. a kind of merchandise sales process detection device, it is characterised in that: including shell, conveying mechanism main motor (3), commodity disengaging
Conveying mechanism (8), camera (9), code reader (10), commodity detection device (11), article portal (12), commodity prize make marks
Laser (13) and commodity export (15),
Article portal (12) and commodity export (15) are provided on the shell, commodity disengaging conveying mechanism (8) setting exists
In the shell, one end of commodity disengaging conveying mechanism (8) connects the article portal (12), the main electricity of conveying mechanism
Machine (3) connects commodity disengaging conveying mechanism (8), and commodity disengaging conveying mechanism (8) includes conveyer belt, the commodity into
The connecting pin of conveying mechanism (8) and the article portal (12) is provided with commodity detection device (11) out, the conveyer belt it is upper
Side is provided with camera (9), code reader (10) and commodity prize and makes marks laser (13), and the camera (9) and commodity are converted
The surface that the laser (13) that makes marks is directed at commodity (14) is encouraged, the code reader (10) is directed at the item of the commodity (14)
Code,
It further include reflection laser (1) and baffle, the reflection laser (1) is from the side of the commodity (14) to commodity (14)
Be irradiated, the baffle plate setting on the route that the reflection laser (1) irradiates the reflected light that the commodity (14) generate,
Following steps can be performed in the merchandise sales process detection device, the detection for commercial glasses paper:
1) image on the baffle, is acquired using camera, the length of described image is identical as commodity length, and described image is
RGB image;
2) channel R of described image, is selected;
3) vertical gradient processing, is carried out to image, obtains gradient image;
4) Threshold segmentation, is carried out to the gradient image that step 3) obtains, wherein segmentation figure is obtained, the background of segmentation figure is black,
Effective coverage is white;
5), statistic procedure 3) obtained in effective coverage length, when the length of effective coverage be greater than image length 75% when,
Then commercial glasses paper is complete;When the length of effective coverage is less than the 75% of image length, then commercial glasses paper is endless
Whole.
2. merchandise sales process detection device as described in claim 1, it is characterised in that: further include printer (6) and printing
Paper (7) is provided with prize exit slot on front side of the shell, and the printer (6) is arranged at the prize exit slot, described
Printing paper (7) connects the printer (6).
3. merchandise sales process detection device as described in claim 1, it is characterised in that: the reflection laser (1) is line
Property red laser.
4. merchandise sales process detection device as claimed in claim 3, it is characterised in that: the baffle is black plate.
5. merchandise sales process detection device as described in claim 1, it is characterised in that: the commodity prize makes marks laser
Device (13) is stripe laser.
6. merchandise sales process detection device as described in claim 1, it is characterised in that: the commodity detection device (11) is
Photoelectric sensor.
7. merchandise sales process detection device as described in claim 1, it is characterised in that: the article portal (12) and commodity
It exports (15) and shares one outlet.
8. merchandise sales process detection device as claimed in claim 3, it is characterised in that: the merchandise sales process detection dress
It sets and can also carry out following steps, the barcode label with bar code commodity is detected, be marked in band bar code bar code region:
1) it, takes pictures to band bar code commodity, obtains the image with bar code commodity;
2), pre-processing: gradient transformation is carried out to the image with bar code commodity, obtains gradient image;
3), barcode position in the image with bar code commodity is positioned;
4) Threshold segmentation, is carried out to bar code area image and forms binary map;
5) erosion operation, is carried out to the binary map that step 4) obtains and obtains mark shape.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410532855.9A CN104318277B (en) | 2014-10-10 | 2014-10-10 | A kind of merchandise sales process detection device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410532855.9A CN104318277B (en) | 2014-10-10 | 2014-10-10 | A kind of merchandise sales process detection device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104318277A CN104318277A (en) | 2015-01-28 |
CN104318277B true CN104318277B (en) | 2018-12-18 |
Family
ID=52373506
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410532855.9A Active CN104318277B (en) | 2014-10-10 | 2014-10-10 | A kind of merchandise sales process detection device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104318277B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113688650B (en) * | 2021-09-22 | 2022-06-17 | 支付宝(杭州)信息技术有限公司 | Method and device for identifying picture |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102354363A (en) * | 2011-09-15 | 2012-02-15 | 西北工业大学 | Identification method of two-dimensional barcode image on high-reflect light cylindrical metal |
CN202700831U (en) * | 2012-07-12 | 2013-01-30 | 安徽省一一通信息科技有限公司 | Automatic commodity sorting system capable of recognizing commodities with no identity |
CN204155298U (en) * | 2014-10-10 | 2015-02-11 | 南京大树智能科技股份有限公司 | A kind of merchandise sales process pick-up unit |
CN204270420U (en) * | 2014-12-25 | 2015-04-15 | 佳木斯大学 | Based on the supermarket intelligent cashier settlement device of Autonomic Computing |
-
2014
- 2014-10-10 CN CN201410532855.9A patent/CN104318277B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102354363A (en) * | 2011-09-15 | 2012-02-15 | 西北工业大学 | Identification method of two-dimensional barcode image on high-reflect light cylindrical metal |
CN202700831U (en) * | 2012-07-12 | 2013-01-30 | 安徽省一一通信息科技有限公司 | Automatic commodity sorting system capable of recognizing commodities with no identity |
CN204155298U (en) * | 2014-10-10 | 2015-02-11 | 南京大树智能科技股份有限公司 | A kind of merchandise sales process pick-up unit |
CN204270420U (en) * | 2014-12-25 | 2015-04-15 | 佳木斯大学 | Based on the supermarket intelligent cashier settlement device of Autonomic Computing |
Also Published As
Publication number | Publication date |
---|---|
CN104318277A (en) | 2015-01-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11727613B2 (en) | Systems and methods for stitching sequential images of an object | |
CN102854192B (en) | Method for detecting apple surface defect based on apple surface defect detecting system | |
CN104729426B (en) | Angle steel automatic on-line detecting system and method based on machine vision | |
US20110262036A1 (en) | Method and system for item identification | |
WO2016092646A1 (en) | Fish type determination device and fish type determination method | |
CN110892445B (en) | Drug inspection support device, drug identification device, image processing method, and program | |
CN106228546A (en) | The detection method of a kind of board and device | |
CN105866136A (en) | Camera module optical filter gluing detection apparatus and method | |
CN104458764B (en) | Curved uneven surface defect identification method based on large-field-depth stripped image projection | |
PT1770037E (en) | Process of detecting double feeding of postal items by image analysis of the upright items | |
US10534970B2 (en) | System and method for reading direct part marking (DPM) codes on objects | |
CN111239142A (en) | Paste appearance defect detection device and method | |
CN204422435U (en) | Camera module optical filter gluing pick-up unit | |
CN110308157B (en) | Method and device for detecting die cutting quality of transparent printed matter | |
CN104318277B (en) | A kind of merchandise sales process detection device | |
KR101762158B1 (en) | Appearance inspection device | |
JP5858524B2 (en) | Method for detecting the paint color of cosmetics | |
US9547784B1 (en) | Sensor arrangement | |
CN101915823A (en) | Ultraviolet detection method for shell bacteria-carrying eggs and device thereof | |
JP6878623B2 (en) | Paper leaf stain judgment device and paper leaf stain judgment method | |
JP2013246174A (en) | Inspection method of product in packaging machine | |
JP2012083965A (en) | Coin processor and coin processing method | |
KR102196114B1 (en) | Apparatus and method for inspecting steel products | |
CN205050166U (en) | Merchant quality control mark note device | |
CN201796028U (en) | Ultraviolet detection device for identifying eggs with shells carrying bacteria |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Commodity sale process detection device Effective date of registration: 20191217 Granted publication date: 20181218 Pledgee: Nanjing Bank Co., Ltd. Chengnan Branch Pledgor: Nanjing Dashu Intelligence Technology Co., Ltd. Registration number: Y2019320000367 |
|
PE01 | Entry into force of the registration of the contract for pledge of patent right |