CN109858491A - Commodity recognition method, storage medium, vending machine and pattern recognition device - Google Patents
Commodity recognition method, storage medium, vending machine and pattern recognition device Download PDFInfo
- Publication number
- CN109858491A CN109858491A CN201811579609.3A CN201811579609A CN109858491A CN 109858491 A CN109858491 A CN 109858491A CN 201811579609 A CN201811579609 A CN 201811579609A CN 109858491 A CN109858491 A CN 109858491A
- Authority
- CN
- China
- Prior art keywords
- commodity
- taken
- vending machine
- identifier information
- recognition method
- 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
- Control Of Vending Devices And Auxiliary Devices For Vending Devices (AREA)
Abstract
The invention discloses a kind of commodity recognition method, storage medium, vending machine and pattern recognition devices.The commodity recognition method is taken the first identifier information that label is identified on commodity the following steps are included: obtaining；It obtains and is taken the second identifier information for identifying label on commodity；The merchandise news for being taken commodity is determined according to the first identifier information and the second identifier information.The commodity recognition method of vending machine according to the present invention, the merchandise news for being taken commodity is determined according to first identifier information and second identifier information, and the accuracy and reliability of commodity identification can be improved in dual identification guarantee.One reduces influence of the ambient to image recognition, to improve recognition accuracy；Two can recognize the range of commodity to increase image recognition, can also be put into vending machine and be sold without the commodity of great amount of images training；Meanwhile it being compared to the high cost of RFID label tag, cost is relatively low for the identification label that the present invention uses.
Description
Technical field
The present invention relates to unmanned retail technology fields, commodity recognition method, storage in particular to a kind of vending machine
Medium, vending machine and pattern recognition device.
Background technique
Due to the higher cost of RFID (Radio Frequency Identification, radio frequency identification) label, so,
Many businessmans wish to identify the commodity in selling cabinet by way of image recognition, the program need will packaging or appearance have it is bright
The commodity of aobvious feature carry out a large amount of image training, so that the image recognition rate of commodity is improved, but this image recognition scheme
Identifiable commodity coverage is relatively narrow, and discrimination is lower, and also has the disadvantages that
1, it only can be directed to appearance rule and have the commodity of more apparent difference each other；At present using it is more be for knowing
Other bottled drink, the bottle cap or body of beverage for identification, and must have obvious characteristic differentiation between every kind of bottle cap, otherwise can not
Identification, such as be all two kinds of mineral waters of white bottle cap, then it can not be put into a selling cabinet and be sold simultaneously；
2, the irregular commodity of appearance, such as various packed foods can not be identified, since the angle put is different, appearance shape
Shape is also different, does not fit through image at present and is identified；
3, the commodity infrequently sold are not suitable for carrying out image recognition, due to image recognition need a large amount of image data into
Row training, could improve discrimination, and workload is huge, and the commodity infrequently sold are also not suitable for investment great effort and go to be instructed
Practice, is generally trained just for common commodity.
Summary of the invention
The present invention is directed to solve one of above-mentioned technical problem in the prior art at least to a certain extent.For this purpose, this hair
A kind of bright commodity recognition method for proposing vending machine, cost is relatively low, and the accuracy rate of identification is higher.
The invention also provides a kind of computer readable storage mediums.
The invention also provides a kind of identified using the commodity recognition method to be taken the vending machines of commodity.
The invention also provides a kind of identified using the commodity recognition method to be taken the pattern recognition devices of commodity.
The commodity recognition method of vending machine according to an embodiment of the present invention identifies on commodity the following steps are included: obtaining to be taken
The first identifier information of label；It obtains and is taken the second identifier information for identifying label on commodity；According to the first identifier information
The merchandise news for being taken commodity is determined with the second identifier information.
The commodity recognition method of vending machine according to an embodiment of the present invention, according to first identifier information and second identifier information
Determine the merchandise news for being taken commodity, the accuracy and reliability of commodity identification can be improved in dual identification guarantee.One reduces
Influence of the ambient to image recognition, to improve recognition accuracy；Two can recognize commodity to increase image recognition
Range can also be put into vending machine without the commodity of great amount of images training and be sold；Meanwhile being compared to RFID label tag
High cost, cost is relatively low for the identification label that the present invention uses.
According to some embodiments of the present invention, the first identifier information is colouring information, and the second identifier information is
Graph text information.
Further, described that the commodity that taken are determined according to the first identifier information and the second identifier information
Merchandise news, comprising:
The colouring information of acquisition is compared with color libraries to generate corresponding first discrimination of the colouring information, and
The first discrimination is filtered out higher than commodity corresponding to pre-set color threshold value, the commodity obtained will be compared according to identification based on color
Rate is ranked up from high to low, obtains corresponding commodity sequence T1, T2 ... Tn；
The graph text information of acquisition is compared with graphics and text library to generate corresponding second discrimination of the graph text information, and
It filters out the second discrimination and is higher than commodity corresponding to default picture and text threshold value, the commodity obtained will be compared according to identification based on picture and text
Rate is ranked up from high to low, obtains corresponding commodity sequence t1, t2 ... tn；
According to T1, T2 ..., Tn and t1, t2 ... tn determines the merchandise news for being taken commodity.
Further, described according to T1, T2 ... the Tn and t1, t2 ... tn determines the merchandise news for being taken commodity,
It include: to judge whether T1 and t1 is same commodity；If T1 and t1 is same commodity, it is determined that described to be taken commodity for T1.
Further, described according to T1, T2 ... the Tn and t1, t2 ... tn determines the merchandise news for being taken commodity,
If further include: T1 and t1 be different commodity, by any of any of T1, T2 ... Tn and t1, t2 ... tn according to
Secondary comparison, when occurring then determining that the corresponding commodity of the first identical comparison result are described to be taken quotient when first comparison result is identical
Product.
Further, described according to T1, T2 ... the Tn and t1, t2 ... tn determines the merchandise news for being taken commodity,
If further include: without identical comparison result, confirm that T1 is taken commodity to be described.
Optionally, the graph text information is the combined information of character and digit.
The computer readable storage medium of embodiment according to a second aspect of the present invention is stored thereon with computer program, institute
When stating program and being executed by processor, the commodity recognition method is realized.
The vending machine of embodiment according to a third aspect of the present invention is identified using the commodity recognition method and is taken commodity.
The pattern recognition device of embodiment according to a fourth aspect of the present invention comprising:
Receiving module, the picture of the identification label for receiving vending machine transmission, open the door request, door signal；
Analysis module is converted to identifiable format for the picture for identifying label to be analyzed and processed；
Identification module identifies the commodity that identification label is directed toward using the commodity recognition method；
Judgment module judges the merchandise news of user's purchase for comparing the merchandise news for front and back of closing the door；
Billing module, the merchandise news for being bought according to user carry out checkout and withhold.
Detailed description of the invention
Fig. 1 is the inside side schematic diagram in right-side-out side and the back side for identifying label；
Fig. 2 is the schematic diagram of the commodity for posting identification label；
Fig. 3 is the schematic diagram that another posts the commodity of identification label；
Fig. 4 is the schematic diagram of vending machine；
Fig. 5 is the logic chart of the vending machine commodity recognition method of one embodiment of the invention；
Fig. 6 is the schematic diagram of the pattern recognition device of one embodiment of the invention.
Appended drawing reference:
Identify label 1, first identifier information 11, second identifier information 12, front 13, the back side 14, commodity 2, vending machine 3,
Cabinet body 31, cabinet door 32, camera 33, shelf 34, door lock 35, headlamp 36, pattern recognition device 4, receiving module 41, analysis mould
Block 42, identification module 43, judgment module 44, billing module 45, pushing module 46.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to is used to explain the present invention, and is not considered as limiting the invention.
In the present invention unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation " etc.
Term shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or integral；It can be mechanical connect
It connects, is also possible to be electrically connected or can communicate with each other；It can be directly connected, can also indirectly connected through an intermediary, it can
To be the interaction relationship of connection or two elements inside two elements.For the ordinary skill in the art,
The specific meanings of the above terms in the present invention can be understood according to specific conditions.
The commodity recognition method of vending machine according to an embodiment of the present invention is described in detail below with reference to Fig. 1-Fig. 6.
Shown in referring to Fig.1, the commodity recognition method of vending machine according to an embodiment of the present invention the following steps are included:
(1) it obtains and is taken the first identifier information 11 for identifying label 1 on commodity；
(2) it obtains and is taken the second identifier information 12 for identifying label 1 on commodity；
(3) merchandise news for being taken commodity is determined according to first identifier information 11 and second identifier information 12.
Specifically, being provided with identification label 1 on each commodity 2 referring to shown in Fig. 2-Fig. 4, Fig. 1 is identification label 1
The front 13 (i.e. outside side) of one embodiment, identification label 1 has first identifier information 11 and second identifier information 12, knows
(i.e. inside side) the whole face gluing of the back side 14 or interval gluing of distinguishing label 1, so that identification label 1 arbitrarily can be pasted freely and be taken office
On the commodity 2 for shape of anticipating.Identification label 1 can be the anti-material torn to pieces of any one waterproof, such as PE, BOPP class film adhesive sticker
Deng, identification the label back side 14 glue be removable plastic, in this way, will not be on commodity 2 after identification label 1 is torn from commodity 2
Offset printing is left, it is easy to use.
Identification label 1 is common material label used in the present invention, compared with RFID label tag, is known used in the present invention
Cost is relatively low for distinguishing label 1, is conducive to promote and apply.
Identification label 1 shown in FIG. 1 is bar shaped label, and bar shaped label is set in outside commodity 2, certainly, identifies label 1
It can be made other patterns, but at least need around 2 four sides of commodity, so that the camera 33 in vending machine 3 can take easily
Identify the first identifier information 11 and second identifier information 12 on label 1.
It, could be right after being taken the merchandise news of commodity according to first identifier information 11 and second identifier information 12 are determining
The subsequent operations such as the commodity 2 carry out that checkout is withholdd.The commodity recognition method of vending machine of the invention and traditional image recognition side
Formula is compared, dual identification, it is ensured that discrimination is high, and the accuracy rate identified is high.
General image recognition, is influenced by light, it is possible that identification mistake, in the present invention, in addition to identification first
Except identification information 11, also identification second identifier information 12 reduces ambient and knows to image to further confirm that commodity 2
Other influence, to improve recognition accuracy.
As shown in Figure 1-Figure 3, first identifier information 11 and second identifier information 12 are integrated on identification label 1, and will be known
Distinguishing label 1 is attached on commodity 2, in this way, without carrying out great amount of images training to commodity 2, only by distinguishing the on identification label 1
One identification information 11 and second identifier information 12, can determine merchandise news, greatly reduce workload, instruct without image
The experienced commodity 2 seldom sold, which can also be put into vending machine 3, to be sold.That is, it is only necessary to which identification label 1 is known
Not, it is identified without the external appearance characteristic to goods themselves, thus the shape of commodity is unrestricted, expands recognizable quotient
The type of product.
In the description of the present invention, it is to be understood that, term " first ", " second " are used for description purposes only, and cannot
It is interpreted as indication or suggestion relative importance or implicitly indicates the quantity of indicated technical characteristic.Define as a result, " the
One ", the feature of " second " can explicitly or implicitly include one or more of the features.In the description of the present invention,
The meaning of " plurality " is at least two, such as two, three etc., unless otherwise specifically defined.
The commodity recognition method of vending machine according to an embodiment of the present invention is believed according to first identifier information 11 and second identifier
Breath 12 determines the merchandise news for being taken commodity, and the accuracy and reliability of commodity identification can be improved in dual identification guarantee.One drops
Low influence of the ambient to image recognition, to improve recognition accuracy；Two can recognize quotient to increase image recognition
The range of product can also be put into vending machine without the commodity of great amount of images training and be sold；Meanwhile being compared to RFID
The high cost of label, 1 cost of identification label that the present invention uses are generally lower, it might even be possible to very lower than RFID label tag cost
One of.
In some embodiments of the invention, first identifier information 11 is colouring information, and second identifier information 12 is picture and text
Information.Optionally, graph text information is the combined information of character and digit.That is, the present invention other than identifying color, is gone back
It identifies that alphanumeric to further confirm that commodity 2, reduces influence of the ambient to image recognition, it is accurate that identification can be improved
Rate.Since accuracy rate improves, can increase image recognition can recognize the range of commodity 2, pack irregular or packet
The commodity 2 of no significant difference can recognize between dress, conveniently sell extensive stock 2.
As Figure 1-Figure 4, identify label 1 front 13 (i.e. outside side) have color as background color, and by up to
Lower print has expired clearly letter and number, and camera 33 can take the color identified on label 1, letter and number easily.
Further, above-mentioned (3) step is taken commodity according to first identifier information 11 and the determination of second identifier information 12
Merchandise news, comprising:
1., the colouring information of acquisition and color libraries compared, the first discrimination is higher than commodity corresponding to p% according to identification
Rate is ranked up from high to low, obtains corresponding commodity sequence T1, T2 ... Tn, wherein n >=1；
2., the graph text information of acquisition and graphics and text library compared, the second discrimination is higher than commodity corresponding to q% according to identification
Rate is ranked up from high to low, obtains corresponding commodity sequence t1, t2 ... tn, wherein n >=1；
3., according to T1, T2 ..., Tn and t1, t2 ... tn determine the merchandise news for being taken commodity.
The commodity number in tn it should be pointed out that the commodity amount in T1, T2 ... Tn needs not be equal to t1, t2 ...
Amount.For example, the first discrimination be higher than 80% corresponding to commodity be ranked up from high to low according to discrimination, obtain corresponding quotient
Product sort T1, T2, T3, T4, T5, the second discrimination be higher than 80% corresponding to commodity arranged from high to low according to discrimination
Sequence obtains corresponding commodity sequence t1, t2, t3.That is, " n " here is it is merely meant that the first discrimination is right higher than p% institute
The commodity answered are at least one, and the second discrimination is higher than commodity corresponding to q% and is at least one.For convenience, may be used
Being defined as T1, T2 so that the first discrimination is higher than the sequence of commodity corresponding to p% ..., Tm, it is right that the second discrimination is higher than q% institute
The commodity sequence answered is defined as t1, t2 ... tn, wherein m can be equal to n, and m can also be not equal to n.
It should be noted that graphics and text library is alphanumeric at this time when graph text information is the combined information of character and digit
Library.Optionally, 80≤p < 100,80≤q < 100, wherein p can be equal to q, and p can also be not equal to q；P, the value of q is according to reality
The image recognition rate for testing data is set.
Further, according to T1, T2 ..., Tn and t1, t2 ... tn determines the merchandise news for being taken commodity, comprising: 4.,
Judge whether T1 and t1 is same commodity；If T1 and t1 is same commodity, it is determined that being taken commodity is T1.
Further, according to T1, T2 ..., Tn and t1, t2 ... tn determines the merchandise news for being taken commodity, further includes:
If 5., T1 and t1 be different commodity, any of any of T1, T2 ... Tn and t1, t2 ... tn are successively compared
It is right, when there is first comparison result identical (i.e. Tx and ty are directed toward same commodity), it is determined that the first identical comparison result
Corresponding commodity are to be taken commodity.
When successively comparing any of any of T1, T2 ... Tn and t1, t2 ... tn, there are many compare
Sequentially, such as in the first comparison sequence, compare first whether T1=t1? if it is not, compare again whether T1=t2? if it is not, again
Do you compare t1=T2? if it is not, comparing T2=t3 again? if it is not, comparing t2=T3 again? until having identical result Tx=ty (i.e. Tx and ty
It is directed toward same commodity), then terminate to compare, output result is the commodity.
Comparing T1 and t1, t2 first ..., is any of tn identical in second of comparison sequence? if it is not, again
Comparing T2 and t1, t2 ..., is any of tn identical? if it is not, compare again T3 whether with any in t1, t2 ... tn
It is a identical? ... is any of tn identical if it is not, comparing Tn and t1, t2 again ...? until there is identical result Tx=ty
(i.e. Tx and ty are directed toward same commodity), then terminate to compare, and output result is the commodity.
Comparing t1 and T1, T2 first ..., is any of Tn identical in the third comparison sequence? if it is not, again
Comparing t2 and T1, T2 ..., is any of Tn identical? if it is not, compare again t3 whether with any in T1, T2 ... Tn
It is a identical? ... is any of Tn identical if it is not, comparing tn and T1, T2 again ...? until there is identical result ty=Tx
(i.e. ty and Tx are directed toward same commodity), then terminate to compare, and output result is the commodity.
The purpose of comparison be in order to find out whether there is Tx and ty to be directed toward same commodity (wherein, x and y can be equal, can also
With unequal), which kind of comparison sequence is limitation of the present invention is not construed as using.Certainly, also there are many unshowned comparison is suitable
Sequence will not enumerate here.
Further, according to T1, T2 ..., Tn and t1, t2 ... tn determines the merchandise news for being taken commodity, further includes:
If 6., without identical comparison result, confirm that T1 is to be taken commodity.
In a particular embodiment, the priority of color libraries is higher than the priority of graphics and text library or the priority height of graphics and text library
In the priority of color libraries, the priority of color libraries and the priority of graphics and text library are different, to guarantee when probability is identical, moreover it is possible to
Enough identification results.
The computer readable storage medium of embodiment according to a second aspect of the present invention, is stored thereon with computer program, should
When program is executed by processor, the commodity recognition method of above-mentioned vending machine may be implemented.
The vending machine 3 of embodiment according to a third aspect of the present invention, is identified using above-mentioned commodity recognition method and is taken quotient
Product.Referring to shown in Fig. 4, vending machine 3 includes cabinet body 31 and cabinet door 32, and when cabinet door 32 is opened, the commodity 2 in cabinet body 31 are exposed to
Outside, customer can buy goods 2, when cabinet door 32 is closed, and can prevent commodity 2 from losing.There is the shelf for placing commodity 2 in cabinet body 31
34, shelf 34 can be multilayer, to increase by 2 quantity of commodity that can be placed in cabinet body 31.
Camera 33 is provided in cabinet body 31, vending machine 3 is obtained 2 picture of commodity in cabinet body 31 using camera 33, taken the photograph
The position that visual field face shelf 34 in cabinet body 31 is set as first 33, to increase the vision of commodity 2 in accessed cabinet body 31
Area.As shown in figure 4, the top of every layer of shelf 34 is provided with a camera 33, the commodity 2 on every layer of shelf 34 are thereby guaranteed that
It can be taken by camera 33.
User enough buys the process of commodity 2 using vending machine 3 are as follows: the authentication to user by (such as user use
Mobile phone barcode scanning opens cabinet door 32) when, cabinet door 32 is opened after once being identified to commodity 2 in vending machine 3, and chosen in user
The commodity 2 in vending machine 3 are identified again when closing cabinet door 32 after commodity 2, it is laggard with the commodity 2 for determining that user takes away
It settles accounts on row backstage.
It is provided with door lock 35 in cabinet body 31 and cabinet door 32, after the authentication to user passes through, door lock 35 could be beaten
It opens, to promote the safety of selling cabinet 3.Headlamp 36 can be set in cabinet body 31, to increase the brightness in cabinet body 31, facilitate Gu
Visitor buys goods.
One embodiment of vending machine commodity recognition method of the invention, the commodity recognition method are described referring to Fig. 5
The following steps are included:
S1: having detected whether trigger signal, if so, entering step S2；If it is not, repeating step S1；
S2: it takes pictures to commodity 2, and picture is uploaded to cloud；
S3: cloud obtains the colouring information and alphanumeric information that label 1 is identified on commodity 2；
S4: colouring information is compared with color libraries, and the first discrimination is higher than commodity corresponding to p% according to discrimination
Sequence is ranked up from high to low, obtains corresponding commodity sequence T1, T2 ... Tn；Simultaneously by alphanumeric information and alphabetical number
Character library is compared, and the second discrimination is higher than commodity corresponding to q% and is ranked up from high to low according to discrimination, is corresponded to
Commodity sort t1, t2 ... tn, wherein n >=1；
S5: comparing T1 and whether t1 is directed toward same commodity, if so, the S9 that gos to step；If it is not, entering step S6；
S6: with T1 and t1 highest priority, successively comparing step by step, i.e., by any of T1, T2 ... Tn and t1,
Any of t2 ... tn is successively compared, if there is first identical comparison result, go to step S8；If it is not, entering step
Rapid S7；
S7: T1 is confirmed as the commodity, go to step S10；
S8: by first identical commodity, the commodity are confirmed as, go to step S10；
S9: confirm that the commodity 2 are the commodity that T1 and t1 is directed toward；
S10: calling the information of the commodity, carries out checkout and withholds；
S11: process terminates.
Trigger signal in above-mentioned steps S1 can be door signal.
In step s 2, the commodity 2 on 33 pairs of the camera of every layer of shelf 34 corresponding shelves 34 are taken pictures.
In step s 4, color libraries and alphanumeric library are compared simultaneously, and are arranged from high to low according to discrimination
Sequence.
In step s 6, successively compare color and alphanumeric from high to low according to discrimination as a result, carrying out downwards in turn
Compare, i.e., compare first whether T1=t1? if it is not, compare again whether T1=t2? if it is not, comparing t1=T2 again? if it is not, comparing again
T2=t3? if it is not, comparing t2=T3 again? until there is identical result Tx=ty, then terminate to compare, output result is the commodity.
In step s 6, the priority of color libraries is higher than alphanumeric library, and two priority must be different, prevents from
It can not identification result when now probability is identical；In this way, in step s 6, in the result of high discrimination, color libraries and alphanumeric
Without being equal to each other as a result, the priority due to color libraries is higher than alphanumeric library in library, it is determined that T1 is the commodity 2.Certainly,
According to the actual situation, also the adjustable priority for alphanumeric library is higher than color libraries, at this time in step s 6, when height identifies
In the result of rate, without being equal to each other as a result, the priority due to alphanumeric library is higher than face in color libraries and alphanumeric library
Color library, it is determined that t1 is the commodity 2.
One embodiment of color libraries is as shown in table 1, and each color represents a kind of merchandise news, further, Mei Gese
System can represent a kind of commodity major class, for example, colour system can be divided into red colour system, yellow class, green system, blue series etc.；Wherein, red
Colour system can segment again are as follows: pink, pinkish red, pink, crabapple red, garnet, bright cherry-red, pale rose colour, bright red, dark reddish purple, kermes etc.；
Each color in color libraries is associated with a kind of merchandise news, may include but is not limited to the following information of commodity: commodity name, product
Board, taste, weight/net content；For example, pink represents commodity: monkey mushroom shortbread type biscuit 96g in river, merchandise news in color libraries
Example is as shown in table 1.
1 color libraries of table
The one embodiment in alphanumeric library is as shown in table 2.Each letter represents a commodity major class, and each number represents
A kind of merchandise news, for example, A1 represents commodity: monkey mushroom shortbread type biscuit 96g in river；Wherein, position also can be interchanged in letter and number
It sets, at this point, each number can represent a commodity major class, each letter can represent a kind of merchandise news, certainly, letter and number
Other forms text or figure can also be changed to；It is possible to further the selected section text in letter and number combines or
More characters or pattern is added, with information such as Expiration Date/shelf-life/dates of manufacture for indicating commodity.
The commodity of color libraries and the commodity in alphanumeric library correspond, i.e. one in each merchandise news corresponding color library
Kind color, and a character and digit in corresponding alphanumeric library.When colouring information and alphanumeric information direction are same
When commodity, the accuracy rate highest of commodity is taken.
2 alphanumeric library of table
Image recognition server is established in cloud, the alphanumeric in color and alphanumeric library to above-mentioned color libraries, is led to
A large amount of sample trainings and study are crossed, discrimination is improved, to carry out the image recognition of the commodity 2 with identification label 1.
To further illustrate commodity recognition method of the invention, it is exemplified below, for example, it is assumed that being put into a Fig. 2
Chef Kang's sauerkraut beef noodle barreled (emerald green C3 label), set p=q=90, then recognition result in color libraries is greater than
90% result exports from high to low according to discrimination, it may be assumed that T1=emerald green (96%), T2=emerald green (93%), T3=
Light green (91%)；Result by recognition result in alphanumeric library greater than 90% exports from high to low according to discrimination, it may be assumed that t1=
C3 (98%), t2=O3 (92%), t3=C8 (91.5%), t4=O8 (90.5%)；By comparing T1 and word in color libraries
T1 in female digital library, corresponding commodity 2 are Chef Kang's sauerkraut beef noodle barreled, then confirm that commodity 2 are Chef Kang's sauerkraut
Beef noodle barreled, calls its merchandise news, quantity and unit price etc., and cloud carries out checkout and withholds.
It should be noted that colouring information is only a kind of form of first identifier information 11, graph text information is only the second mark
Know a kind of form of information 12, in some unshowned embodiments, first identifier information 11 and second identifier information 12 may be used also
To be commodity pattern, special symbol, text, various forms of numbers (such as Arabic font etc.) etc., these identification informations are equal
It can be substituted into commodity recognition method of the invention, no longer repeat one by one here.
Referring to shown in Fig. 6, the pattern recognition device 4 of embodiment according to a fourth aspect of the present invention comprising:
Receiving module 41, the picture of the identification label 1 for receiving the transmission of vending machine 3, open the door request, door signal；
Analysis module 42 is converted to identifiable format for the picture for identifying label 1 to be analyzed and processed；
Identification module 43 identifies the commodity 2 that identification label 1 is directed toward using above-mentioned commodity recognition method；
Judgment module 44 judges the merchandise news of user's purchase for comparing the merchandise news for front and back of closing the door；
Billing module 45, the merchandise news for being bought according to user carry out checkout and withhold.
In some alternative embodiments, pattern recognition device 4 further include: pushing module 46, pushing module 46 is by commodity 2
Recommendation information be pushed to vending machine 3 and/or the mobile terminal of user.
When user opens cabinet door 32 using mobile phone barcode scanning, receiving module 41 receives enabling request, and door lock 35 is unlocked,
Cabinet door 32 is opened, before opening cabinet door 32, camera 33 shoots commodity picture, and picture is uploaded to receiving module 41, with
Afterwards, analysis module 42 will include to identify that the commodity picture of label 1 is analyzed and processed, and is converted to identifiable format, identifies
Module 43 once identifies commodity picture in vending machine 3, to determine the type of merchandize and quantity in cabinet body 31.Open cabinet door
After 32, when closing cabinet door 32 after user has chosen commodity 2, camera 33 shoots commodity picture again, and again by picture
It is uploaded to receiving module 41, analysis module 42 will include to identify that the commodity picture of label 1 is analyzed and processed again, be converted to
Identifiable format, identification module 43 identify the commodity picture in vending machine 3 again, remaining in cabinet body 31 to determine
Type of merchandize and quantity, so that it is determined that 2 type and quantity of commodity that user takes away, billing module 45 so as to taken commodity into
It settles accounts on row backstage.
In some alternative embodiments, can also by camera 33 capture user take commodity 2 when dynamic menu,
To which Dynamic Recognition is taken the merchandise news of commodity, and commodity are taken to carry out backstage clearing this part.
Pattern recognition device 4 further include: processor, for storage processor executable instruction memory and be stored in
On memory and the image recognition program that can run on a processor, wherein, can be real when image recognition program is performed processor
Existing above-mentioned commodity recognition method.
A kind of server, including above-mentioned pattern recognition device 4.
A kind of image identification system, comprising: vending machine 3, the mobile terminal of user, above-mentioned server, wherein server is held
The above-mentioned commodity recognition method of row.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means particular features, structures, materials, or characteristics described in conjunction with this embodiment or example
It is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms need not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be any
One or more embodiment or examples in can be combined in any suitable manner.In addition, those skilled in the art can illustrate this
Different embodiments or examples described in book are engaged and are combined.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (10)
1. a kind of commodity recognition method of vending machine, which comprises the following steps:
It obtains and is taken the first identifier information for identifying label on commodity；
It obtains and is taken the second identifier information for identifying label on commodity；
The merchandise news for being taken commodity is determined according to the first identifier information and the second identifier information.
2. the commodity recognition method of vending machine according to claim 1, which is characterized in that the first identifier information is face
Color information, the second identifier information are graph text information.
3. the commodity recognition method of vending machine according to claim 2, which is characterized in that described according to the first identifier
Information and the second identifier information determine the merchandise news for being taken commodity, comprising:
The colouring information of acquisition is compared with color libraries to generate corresponding first discrimination of the colouring information, and screened
Out the first discrimination be higher than pre-set color threshold value corresponding to commodity, by based on color compares acquisition commodity according to discrimination by
It is high to low to be ranked up, obtain corresponding commodity sequence T1, T2 ... Tn；
The graph text information of acquisition is compared with graphics and text library to generate corresponding second discrimination of the graph text information, and screened
The second discrimination is higher than commodity corresponding to default picture and text threshold value out, will be compared based on picture and text the commodity of acquisition according to discrimination by
It is high to low to be ranked up, obtain corresponding commodity sequence t1, t2 ... tn；
According to T1, T2 ..., Tn and t1, t2 ... tn determines the merchandise news for being taken commodity.
4. the commodity recognition method of vending machine according to claim 3, which is characterized in that described according to T1, T2 ... the Tn
The merchandise news for being taken commodity is determined with t1, t2 ... tn, comprising:
Judge whether T1 and t1 is same commodity；
If T1 and t1 is same commodity, it is determined that described to be taken commodity for T1.
5. the commodity recognition method of vending machine according to claim 4, which is characterized in that described according to T1, T2 ... the Tn
The merchandise news for being taken commodity is determined with t1, t2 ... tn, further includes:
If T1 and t1 is different commodity, any of any of T1, T2 ... Tn and t1, t2 ... tn are successively compared
It is right, when occurring then determining that the corresponding commodity of the first identical comparison result are described to be taken commodity when first comparison result is identical.
6. the commodity recognition method of vending machine according to claim 5, which is characterized in that described according to T1, T2 ... the Tn
The merchandise news for being taken commodity is determined with t1, t2 ... tn, further includes:
If confirming that T1 is taken commodity to be described without identical comparison result.
7. the commodity recognition method of vending machine according to claim 2, which is characterized in that the graph text information is that letter adds
The combined information of number.
8. a kind of computer readable storage medium, which is characterized in that be stored thereon with computer program, described program is by processor
When execution, commodity recognition method described in any one of claim 1-7 is realized.
9. a kind of vending machine, which is characterized in that identified and taken using commodity recognition method described in any one of claim 1-7
Commodity.
10. a kind of pattern recognition device characterized by comprising
Receiving module, the picture of the identification label for receiving vending machine transmission, open the door request, door signal；
Analysis module is converted to identifiable format for the picture for identifying label to be analyzed and processed；
Identification module identifies the quotient that identification label is directed toward using commodity recognition method described in any one of claim 1-7
Product；
Judgment module judges the merchandise news of user's purchase for comparing the merchandise news for front and back of closing the door；
Billing module, the merchandise news for being bought according to user carry out checkout and withhold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811579609.3A CN109858491B (en) | 2018-12-24 | 2018-12-24 | Commodity identification method, storage medium, vending machine and image identification device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811579609.3A CN109858491B (en) | 2018-12-24 | 2018-12-24 | Commodity identification method, storage medium, vending machine and image identification device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109858491A true CN109858491A (en) | 2019-06-07 |
| CN109858491B CN109858491B (en) | 2023-04-07 |
Family
ID=66891957
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811579609.3A Active CN109858491B (en) | 2018-12-24 | 2018-12-24 | Commodity identification method, storage medium, vending machine and image identification device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109858491B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110837824A (en) * | 2019-08-09 | 2020-02-25 | 达闼科技（北京）有限公司 | Commodity identification method for vending device, vending device and storage medium |
| CN111768564A (en) * | 2020-06-02 | 2020-10-13 | 合肥美的智能科技有限公司 | Intelligent device and control method thereof |
| CN111784910A (en) * | 2020-06-02 | 2020-10-16 | 合肥美的智能科技有限公司 | Control system and method of intelligent device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107748973A (en) * | 2017-09-29 | 2018-03-02 | 时瑞科技（深圳）有限公司 | Product identification system and automatic identification are sold or storage method |
| CN108053204A (en) * | 2017-12-20 | 2018-05-18 | 杭州几禾科技有限公司 | Automatic settlement method and sell equipment |
| CN108171172A (en) * | 2017-12-27 | 2018-06-15 | 惠州Tcl家电集团有限公司 | Self-help shopping method, self-service sale device and computer readable storage medium |
| WO2018137136A1 (en) * | 2017-01-24 | 2018-08-02 | 广州市蚁道互联网有限公司 | Vending machine and operation method thereof |
-
2018
- 2018-12-24 CN CN201811579609.3A patent/CN109858491B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018137136A1 (en) * | 2017-01-24 | 2018-08-02 | 广州市蚁道互联网有限公司 | Vending machine and operation method thereof |
| CN107748973A (en) * | 2017-09-29 | 2018-03-02 | 时瑞科技（深圳）有限公司 | Product identification system and automatic identification are sold or storage method |
| CN108053204A (en) * | 2017-12-20 | 2018-05-18 | 杭州几禾科技有限公司 | Automatic settlement method and sell equipment |
| CN108171172A (en) * | 2017-12-27 | 2018-06-15 | 惠州Tcl家电集团有限公司 | Self-help shopping method, self-service sale device and computer readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 丁伟利等: "一种基于新型标签识别的购物导航系统", 《光电工程》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110837824A (en) * | 2019-08-09 | 2020-02-25 | 达闼科技（北京）有限公司 | Commodity identification method for vending device, vending device and storage medium |
| CN111768564A (en) * | 2020-06-02 | 2020-10-13 | 合肥美的智能科技有限公司 | Intelligent device and control method thereof |
| CN111784910A (en) * | 2020-06-02 | 2020-10-16 | 合肥美的智能科技有限公司 | Control system and method of intelligent device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109858491B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11151427B2 (en) | Method and apparatus for checkout based on image identification technique of convolutional neural network | |
| CN109858491A (en) | Commodity recognition method, storage medium, vending machine and pattern recognition device | |
| CN109308434B (en) | Electronic price tag, server and system | |
| CN102016931B (en) | Method and system for controlling production of items | |
| US9996780B2 (en) | Specific bar code and forming method of the same | |
| CN102930264B (en) | Based on commodity display information acquisition and analysis system and the method for image recognition technology | |
| CN108074330A (en) | Inspection system, check device and recreation currency | |
| CN108364047B (en) | Electronic price tag, electronic price tag system and data processing method | |
| CN109635705B (en) | Commodity identification method and device based on two-dimensional code and deep learning | |
| CN102831528A (en) | Commodity anti-counterfeiting method and commodity anti-counterfeiting system | |
| CN103646334A (en) | Automatic identification anti-fake method | |
| CN109241883B (en) | Goods returning control method and device based on vending machine | |
| JP6136257B2 (en) | Vending machine management apparatus, data processing method thereof, and program | |
| US8783554B2 (en) | Information reading apparatus, commodity sales information processing apparatus, and pasted object | |
| CN106384142A (en) | Random color two-dimensional code based anti-counterfeiting method and device | |
| CN202870897U (en) | Commodity anti-fake system | |
| CN109312929A (en) | Heating device | |
| CN111222389A (en) | Method and system for analyzing commodities on commercial and super goods shelf | |
| US11748787B2 (en) | Analysis method and system for the item on the supermarket shelf | |
| CN110837824A (en) | Commodity identification method for vending device, vending device and storage medium | |
| CN108268912A (en) | The mark system and method for article storage state in a kind of school bag | |
| CN211827326U (en) | System of unique identification | |
| CN109359706A (en) | Merchandise news intelligent identifying system and method | |
| CN108280497A (en) | A kind of recognition methods of commodity packaging information | |
| CN105488532B (en) | A kind of license plate screening technique and 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 |