CN109034067A - Commodity image reproduction detection method, system, equipment and storage medium - Google Patents
Commodity image reproduction detection method, system, equipment and storage medium Download PDFInfo
- Publication number
- CN109034067A CN109034067A CN201810844896.XA CN201810844896A CN109034067A CN 109034067 A CN109034067 A CN 109034067A CN 201810844896 A CN201810844896 A CN 201810844896A CN 109034067 A CN109034067 A CN 109034067A
- Authority
- CN
- China
- Prior art keywords
- image
- region
- commodity
- detected
- distinguishing characteristics
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of commodity image reproduction detection method, system, equipment and storage mediums, include the following steps: to acquire multiple images, to being detected on each image, to identify the corresponding merchandise classification number in multiple commodity regions and each commodity region on each image；The corresponding at least distinguishing characteristics of each image is generated according to the corresponding merchandise classification number in commodity each on each image region；A property data base is generated according to the corresponding distinguishing characteristics of each image；Image to be detected is detected, the corresponding merchandise classification number in multiple end article regions and each end article region in image to be detected is identified, the target signature of image to be detected is generated according to the corresponding merchandise classification number in end article region；It is compared according to target distinguishing characteristics with the distinguishing characteristics in property data base, judges whether image to be detected is reproduction image.The present invention can be realized the effective detection carried out to image to be detected.
Description
Technical field
The present invention relates to, and in particular, to a kind of commodity image reproduction detection method, system, equipment and storage medium.
Background technique
With the fast development of commercial economy, the raising of living standards of the people, City Brands image is established, raw in the people
Accumulation regions living can all have department stores and convenience store.
As the supplier of department stores and convenience store, such as Coca-Cola, Pepsi Co. Ltd., Procter & Gamble are all
With understand the said firm's product department stores and the display case of convenience store demand.Because commodity are in department stores and convenience
It the placement position in shop and puts quantity and directly influences whether commodity in the sales volume of department stores and convenience store.At this time supplier
Company's work that can generally dispatch officers periodically makes an inspection tour department stores and convenience store, carries out the investigation of commodity display case, and to quotient
Product shooting situation is taken pictures to record display case.
But since different department stores and convenience store use identical shelf, or even layout to duplicate there is also some,
Cause to be similar with the commodity display case of supplier same in convenience store in different department stores, such as Coca-Cola
Cola product putting in different department stores and convenience store there is a situation where it is similar.When there are inertia by some employees
In the case where, the photo for having patrolled shop shooting can be subjected to reproduction as the photo for not patrolling shop, so as to cause to department stores and just
The record of commodity ornaments situation influences the prediction to condition of sales there are great error in sharp shop.
Summary of the invention
For the defects in the prior art, the object of the present invention is to provide a kind of commodity image reproduction detection method, system,
Equipment and storage medium.
The commodity image reproduction detection method provided according to the present invention, includes the following steps:
Step S1: acquiring multiple images, to detecting in each described image, to identify in each described image
The corresponding merchandise classification number in multiple commodity regions and each commodity region；
Step S2: it is corresponding that each image is generated according to the corresponding merchandise classification number in commodity each on each image region
An at least distinguishing characteristics；
Step S3: according to one property data base of each corresponding distinguishing characteristics generation of image；
Step S4: detecting image to be detected, identifies multiple end article regions in described image to be detected
Merchandise classification number corresponding with each end article region generates institute according to the corresponding merchandise classification number in end article region
State the target distinguishing characteristics of image to be detected；
Step S5: being compared according to the target distinguishing characteristics with the distinguishing characteristics in the property data base, judgement
Whether described image to be detected is reproduction image.
Preferably, further include following steps:
When judging described image to be detected not is reproduction image, by described image to be detected and corresponding target area
Other feature is filled into the property data base；
When judging described image to be detected is reproduction image, then reproduction prompting message is issued.
Preferably, the step S2 includes the following steps:
Step S201: preset merchandise classification numeral order；
Step S202: it is corresponding that each merchandise classification number on each image is counted according to the merchandise classification numeral order
Commodity region quantity forms the corresponding one-dimensional matrix of each image, as the corresponding distinguishing characteristics of each image；
The step S4 includes the following steps:
Step S401: detecting image to be detected, identifies multiple end article areas in described image to be detected
The corresponding merchandise classification number in domain and each end article region；
Step S402: it according to the merchandise classification numeral order, counts each merchandise classification in each image to be detected and compiles
Number corresponding commodity region quantity forms the corresponding one-dimensional matrix of each image to be detected, corresponding as each image to be detected
Target distinguishing characteristics.
Preferably, the step S5 includes the following steps:
Step S501: it is asked after each element in the target distinguishing characteristics is subtracted each other with corresponding element in a distinguishing characteristics
Square generate multiple first element difference radixes；
Step S502: extraction of square root generates the target distinguishing characteristics and an area after multiple first element difference radixes are added
Fisrt feature difference radix between other feature；
Step S503: judge the fisrt feature difference radix whether less than a preset fisrt feature difference radix threshold
Value then determines the target distinguishing characteristics pair when the feature difference radix is less than the fisrt feature difference radix threshold value
The image to be detected answered is reproduction image, when the feature difference radix is more than or equal to the fisrt feature difference radix threshold value
When, then trigger step S504；
Step S504: repeating step S501 to step S503, when the target distinguishing characteristics and the characteristic
When the feature difference radix between each distinguishing characteristics in library is more than or equal to the fisrt feature difference radix threshold value, then determine
The corresponding image to be detected of the target distinguishing characteristics is not reproduction image.
Preferably, the step S2 includes the following steps:
Step S201: the surrounding in commodity region each in each image is extended to form one first with a preset area
Nine grids figure picture, the commodity region are located at the center of the first nine grids figure picture；
Step S202: judge that each preset area region whether there is a commodity region in the first nine grids figure picture；
Step S203: when a preset area region is then obtained there are when a commodity region in the first nine grids figure picture
The merchandise classification in the commodity region is numbered, and when a commodity region is not present in a preset area region, then sets the preset area
The corresponding merchandise classification number in region is a constant；
Step S204: according to preset statistics sequence, each preset area region in the first nine grids figure picture is successively counted
Corresponding merchandise classification numbers to form the corresponding one-dimensional matrix in each commodity region, special as the corresponding difference in each commodity region
Sign；
Step S205: it is repeated in and executes multiple distinguishing characteristics that step S201 to step S204 generates each image.
Preferably, the step S4 includes the following steps:
Step S401: the surrounding in end article region each in each image to be detected is extended with a preset area
One second nine grids figure picture is formed, the end article region is located at the center of the second nine grids figure picture；
Step S402: judge in the corresponding second nine grids figure picture in each end article region whether is each preset area region
There are an end article regions；
Step S403: when a preset area region is there are when an end article region in the second nine grids figure picture, then
The merchandise classification number for obtaining the end article region, when a target is not present in a preset area region in the second nine grids figure picture
When commodity region, then setting the corresponding merchandise classification number in the preset area region is a constant；
Step S404: according to the statistics sequence, each preset area area in the second nine grids figure picture is successively counted
The corresponding merchandise classification in domain numbers to form the corresponding one-dimensional matrix in each end article region, as each end article region pair
The target distinguishing characteristics answered；
Step S405: it is repeated in and executes multiple target areas that step S401 to step S404 generates each image to be detected
Other feature.
Preferably, the step S5 includes the following steps:
Step S501: the commodity image to be detected is compared with each image in the property data base, when
In the property data base in the corresponding merchandise classification number of each image inquiry less than a mesh in the commodity image to be detected
When marking the corresponding merchandise classification number in commodity region, then judge that the commodity image to be detected is not reproduction image, otherwise triggering walks
Rapid S502；
Step S502: judge every in the corresponding target distinguishing characteristics in an end article region in the commodity image to be detected
One element in an image with identical merchandise classification number the corresponding distinguishing characteristics in commodity region in each corresponding element whether
It is identical；
Step S503: when each element in the target distinguishing characteristics with an image there is identical merchandise classification to number
Commodity region correspond to each corresponding element in distinguishing characteristics it is all the same when, then determine the end article region and described image
In with identical merchandise classification number commodity region it is identical, otherwise determine that the end article region and a described image have
The commodity region of identical merchandise classification number is not identical；
Step S504: repeating step S502 to step S503, when in commodity image to be detected with a described image
End article region total amount ratio is greater than default in the identical end article region quantity in commodity region and commodity image to be detected
Proportion threshold value when, then judge the commodity image to be detected for reproduction image, otherwise determine that the commodity image to be detected is not
Reproduction image.
The commodity image reproduction detection system provided according to the present invention, for realizing the commodity image reproduction detection side
Method, comprising:
Commodity region identification module, for acquiring multiple images, to being detected in each described image, to identify often
The corresponding merchandise classification number in multiple commodity regions and each commodity region in one described image；
Distinguishing characteristics generation module, for being generated according to the corresponding merchandise classification number in commodity each on each image region
The corresponding at least distinguishing characteristics of each image；
Property data base generation module, for generating a property data base according to the corresponding distinguishing characteristics of each image；
Target distinguishing characteristics generation module identifies in described image to be detected for detecting to image to be detected
Multiple end article regions and each end article region corresponding merchandise classification number, it is corresponding according to end article region
Merchandise classification number generates the target distinguishing characteristics of described image to be detected；
Feature comparison module, for being carried out according to the distinguishing characteristics in the target distinguishing characteristics and the property data base
It compares, judges whether described image to be detected is reproduction image.
The commodity image reproduction detection device provided according to the present invention, comprising:
Processor；
Memory, wherein being stored with the executable instruction of the processor；
Wherein, the processor is configured to execute the commodity image reproduction detection via the executable instruction is executed
The step of method.
The computer readable storage medium provided according to the present invention, for storing program, described program is performed realization
The step of commodity image reproduction detection method.
Compared with prior art, the present invention have it is following the utility model has the advantages that
By acquiring multiple images in the present invention, the commodity region on each image is numbered according to preset merchandise classification
Distinguishing characteristics is generated, and summarizes the corresponding distinguishing characteristics of each image and generates a property data base, and will be according to image to be detected
It generates target distinguishing characteristics input feature vector database and carries out whether inquiry contrast judgement is out reproduction image, to realize to be checked
Effective supervision to the shooting photo for patrolling salesman is realized in the effective detection that altimetric image carries out；
When will to judge described image to be detected not in the present invention be reproduction image, by the target area of described image to be detected
Other feature is filled into the property data base, is realized the effective real-time update of the property data base, is improved to to be detected
The accuracy rate for the detection that image carries out.
Detailed description of the invention
Upon reading the detailed description of non-limiting embodiments with reference to the following drawings, other feature of the invention,
Objects and advantages will become more apparent upon:
Fig. 1 is the step flow chart of commodity image reproduction detection method in the present invention；
Fig. 2 is the step flow chart that distinguishing characteristics generates in the present invention；
Fig. 3 is the step flow chart that target distinguishing characteristics generates in the present invention；
Fig. 4 is the step flow chart of image to be detected reproduction judgement in the present invention；
Fig. 5 is the step flow chart that distinguishing characteristics generates in variation of the present invention；
Fig. 6 is the step flow chart that target distinguishing characteristics generates in variation of the present invention；
Fig. 7 is the step flow chart of image to be detected reproduction judgement in variation of the present invention；
Fig. 8 is the schematic diagram of the first nine grids figure picture in variation of the present invention；
Fig. 9 is the module diagram of commodity image reproduction detection system in the present invention；
Figure 10 is the structural schematic diagram of commodity image reproduction detection device in the present invention；And
Figure 11 is the structural schematic diagram of computer readable storage medium in the present invention.
Specific embodiment
The present invention is described in detail combined with specific embodiments below.Following embodiment will be helpful to the technology of this field
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that the ordinary skill of this field
For personnel, without departing from the inventive concept of the premise, various modifications and improvements can be made.These belong to the present invention
Protection scope.
Fig. 1 is the step flow chart of commodity image reproduction detection method in the present invention, as shown in Figure 1, provided by the invention
Commodity image reproduction detection method, includes the following steps:
Step S1: acquiring multiple images, to detecting in each described image, to identify in each described image
The corresponding merchandise classification number in multiple commodity regions and each commodity region；
Step S2: it is corresponding that each image is generated according to the corresponding merchandise classification number in commodity each on each image region
An at least distinguishing characteristics；
Step S3: a property data base is generated according to the corresponding distinguishing characteristics of image each in multiple images；
Step S4: detecting image to be detected, identifies multiple end article regions in described image to be detected
Merchandise classification number corresponding with each end article region generates institute according to the corresponding merchandise classification number in end article region
State the target distinguishing characteristics of image to be detected；
Step S5: being compared according to the target distinguishing characteristics with the distinguishing characteristics in the property data base, judgement
Whether described image to be detected is reproduction image.
In the present embodiment, it can choose summarize by the corresponding distinguishing characteristics of each image and generate a characteristic
The corresponding distinguishing characteristics of each image or multiple distinguishing characteristics can also be associated and form the property data base by library.
In the present embodiment, described image can be the shelf image shot in market, or the refrigerator figure of shooting
Picture.The commodity region is the corresponding region of commodity a certain on image, can be the region such as bottle laughable on image, food packaging
Region.
In the present embodiment, the merchandise classification number is SKU (Stock Keeping Unit, the commodity according to commodity
Class number) setting number.
In the present embodiment, by the neural network model of pre-training to each commodity region and the commodity area on image
The corresponding SKU in domain is identified, and determines that the commodity region is corresponding according to the relationship between SKU and merchandise classification number
Merchandise classification number.
By acquiring multiple images in the present invention, to the commodity region on each image according to merchandise classification number generation area
Other feature, and summarize the corresponding distinguishing characteristics of each image and generate a property data base, and mesh will be generated according to image to be detected
Mark distinguishing characteristics input feature vector database carries out whether inquiry contrast judgement is out reproduction image, to realize to image to be detected
Effective supervision to the shooting photo for patrolling salesman is realized in the effective detection carried out.
Commodity image reproduction detection method provided by the invention, further includes following steps:
When judging described image to be detected not is reproduction image, by described image to be detected and corresponding target
Distinguishing characteristics is filled into the property data base；
When judging described image to be detected is reproduction image, then reproduction prompting message is issued.
Fig. 2 is the step flow chart that distinguishing characteristics generates in the present invention, as shown in Fig. 2, in the present invention, the step S2
Include the following steps:
Step S201: preset merchandise classification numeral order；
Step S202: it is corresponding that each merchandise classification number on each image is counted according to the merchandise classification numeral order
Commodity region quantity forms the corresponding one-dimensional matrix of each image, as the corresponding distinguishing characteristics of each image；
In the present embodiment, the merchandise classification numeral order can number according to merchandise classification and carry out ascending order arrangement, institute
The quantity for stating merchandise classification number constitutes the dimension of the one-dimensional matrix, such as when the quantity of merchandise classification number is 100
When, it is dimension that the one-dimensional matrix, which is 100, when a certain merchandise classification, which numbers corresponding commodity region, to be paid no attention in commodity image,
Merchandise classification number is 0 in the corresponding element of the one-dimensional matrix.
Fig. 3 is the step flow chart that target distinguishing characteristics generates in the present invention, as shown in figure 3, in the present invention, the step
S4 includes the following steps:
Step S401: detecting image to be detected, identifies multiple end article areas in described image to be detected
The corresponding merchandise classification number in domain and each end article region；
Step S402: it according to the merchandise classification numeral order, counts each merchandise classification in each image to be detected and compiles
Number corresponding commodity region quantity forms the corresponding one-dimensional matrix of each image to be detected, corresponding as each image to be detected
Target distinguishing characteristics.
Fig. 4 is the step flow chart of image to be detected reproduction judgement in the present invention, as shown in figure 4, the step S5 includes
Following steps:
Step S501: by the distinguishing characteristics in each element and the property data base in the target distinguishing characteristics
Middle corresponding element subtracts each other the rear squared multiple first element difference radixes of generation；
Step S502: will multiple first element difference radixes be added after extraction of square root generate the target distinguishing characteristics with it is described
Fisrt feature difference radix between a distinguishing characteristics in property data base；
Step S503: judge the fisrt feature difference radix whether less than a preset fisrt feature difference radix threshold
Value then determines the target distinguishing characteristics pair when the feature difference radix is less than the fisrt feature difference radix threshold value
The image to be detected answered is reproduction image, when the feature difference radix is more than or equal to the fisrt feature difference radix threshold value
When, then trigger step S504；
Step S504: repeating step S501 to step S503, when the target distinguishing characteristics and the characteristic
When the feature difference radix between each distinguishing characteristics in library is more than or equal to the fisrt feature difference radix threshold value, then determine
The corresponding image to be detected of the target distinguishing characteristics is not reproduction image.
In the present embodiment, if the distinguishing characteristics is [M1,M2,M3,M4,M5], the target distinguishing characteristics is [N1,N2,
N3,N4,N5], then the feature difference radix F is
In the present embodiment, it is 1 or 2 that the fisrt feature difference radix threshold value, which can be set,.
Fig. 5 is the step flow chart that distinguishing characteristics generates in variation of the present invention, as shown in figure 5, in the variation, institute
Step S2 is stated to include the following steps:
Step S201: the surrounding in commodity region each in each image is extended to form one first with a preset area
Nine grids figure picture, the commodity region are located at the center of the first nine grids figure picture；
Step S202: judge that each preset area region whether there is a commodity region in the first nine grids figure picture；
Step S203: when a preset area region is then obtained there are when a commodity region in the first nine grids figure picture
The merchandise classification in the commodity region is numbered, and when a commodity region is not present in a preset area region, then sets the preset area
The corresponding merchandise classification number in region is a constant；
Step S204: according to preset statistics sequence, each preset area region in the first nine grids figure picture is successively counted
Corresponding merchandise classification numbers to form the corresponding one-dimensional matrix in each commodity region, special as the corresponding difference in each commodity region
Sign；
Step S205: it is repeated in and executes multiple distinguishing characteristics that step S201 to step S204 generates each image.
Fig. 8 is the schematic diagram of the first nine grids figure picture in variation of the present invention, as shown in figure 8, the preset area and institute
The area for stating commodity region is identical, i.e., 8 parts of surroundings for being trapped among the commodity region of the commodity region duplication is formed described
One nine grids figure picture.
In the variation, according to commodity region is identified in step S1, the central point in the commodity region identified is determined
Whether fall in a preset area region, identifies that the central point in commodity region is fallen in preset area region when existing, then sentence
The preset area region of breaking is to can extract the corresponding merchandise classification number of the pre- area there are a commodity region.
In the present embodiment, the preset statistics sequence is from left to right, successively to identify from top to bottom.For example, can be with
The target distinguishing characteristics of formation is [A1,A2,A3,A4,A5,A6,A7,A8].In the present embodiment, when preset area region is not present
When one end article region, setting the corresponding merchandise classification number in the preset area region is 0.Such as A6Region is without an image district
When domain, then sets and enable A6=0.
Fig. 6 is the step flow chart that target distinguishing characteristics generates in variation of the present invention, as shown in fig. 6, the step S4
Include the following steps:
Step S401: the surrounding in end article region each in each image to be detected is extended with a preset area
One second nine grids figure picture is formed, the end article region is located at the center of the second nine grids figure picture；
Step S402: judge in the corresponding second nine grids figure picture in each end article region whether is each preset area region
There are an end article regions；
Step S403: when a preset area region is there are when an end article region in the second nine grids figure picture, then
The merchandise classification number for obtaining the end article region, when a target is not present in a preset area region in the second nine grids figure picture
When commodity region, then setting the corresponding merchandise classification number in the preset area region is a constant；
Step S404: according to the statistics sequence, each preset area area in the second nine grids figure picture is successively counted
The corresponding merchandise classification in domain numbers to form the corresponding one-dimensional matrix in each end article region, as each end article region pair
The target distinguishing characteristics answered；
Step S405: it is repeated in and executes multiple target areas that step S401 to step S404 generates each image to be detected
Other feature.
In the variation, the target distinguishing characteristics of formation can be expressed as [B1,B2,B3,B4,B5,B6,B7,B8]。
When described image to be detected is 10,10 target distinguishing characteristics can be formed.In the present embodiment, when preset area region
It is 0 there is no the corresponding merchandise classification number in the preset area region when end article region, is set.Such as B6Region is without one
When image-region, then sets and enable B6=0.
Fig. 7 is the step flow chart of image to be detected reproduction judgement in variation of the present invention, as shown in fig. 7, the step
S5 includes the following steps:
Step S501: the commodity image to be detected is compared with each image in the property data base, when
In the property data base in the corresponding merchandise classification number of each image inquiry less than a target quotient in commodity image to be detected
When the corresponding merchandise classification number in product region, then judges that the commodity image to be detected is not reproduction image, otherwise trigger step
S502；
Step S502: judge every in the corresponding target distinguishing characteristics in an end article region in the commodity image to be detected
One element in an image with identical merchandise classification number the corresponding distinguishing characteristics in commodity region in each corresponding element whether
It is identical；
Step S503: when each element in the target distinguishing characteristics with an image there is identical merchandise classification to number
Commodity region correspond to each corresponding element in distinguishing characteristics it is all the same when, then determine the end article region and described image
In with identical merchandise classification number commodity region it is identical, otherwise determine that the end article region and a described image have
The commodity region of identical merchandise classification number is not identical；
Step S504: repeating step S502 to step S503, when in commodity image to be detected with a described image
End article region total amount ratio is greater than default in the identical end article region quantity in commodity region and commodity image to be detected
Proportion threshold value when, then judge the commodity image to be detected for reproduction image, otherwise determine that the commodity image to be detected is not
Reproduction image.
In the variation, when there are the corresponding merchandise classification numbers in a commodity region for commodity image to be detected, in feature
It is inquired in image in database not then, can directly assert that the commodity image to be detected is not reproduction image.It is only to be detected
There are the corresponding merchandise classification numbers in a commodity region when can find on an image for commodity image, just carries out step S502
To the detection of step S503.
In the variation, each element in the target distinguishing characteristics with an image there is identical merchandise classification to compile
Number commodity region correspond to each corresponding element in distinguishing characteristics it is all the same when
In the variation, preset proportion threshold value is 90%, as total in worked as end article region in commodity image to be detected
Amount is 50, and judges end article region quantity identical with the commodity region of an image in the commodity image to be detected
It is 48, then determining the image is reproduction image.
Fig. 9 is the module diagram of commodity image reproduction detection system in the present invention, for realizing the commodity image
Reproduction detection method, as shown in figure 9, commodity image reproduction detection system 100 provided by the invention, comprising:
Commodity region identification module 101, for acquiring multiple images, to being detected in each described image, with identification
The corresponding merchandise classification number in multiple commodity regions in each described image and each commodity region out；
Distinguishing characteristics generation module 102, for being numbered according to the corresponding merchandise classification in commodity each on each image region
Generate the corresponding at least distinguishing characteristics of each image；
Property data base generation module 103, for generating one according to the corresponding distinguishing characteristics of image each in multiple images
Property data base；
Target distinguishing characteristics generation module 104 identifies described image to be detected for detecting to image to be detected
On multiple end article regions and each end article region corresponding merchandise classification number, it is corresponding according to end article region
Merchandise classification number generate the target distinguishing characteristics of described image to be detected；
Feature comparison module 105, for according to the distinguishing characteristics in the target distinguishing characteristics and the property data base
It is compared, judges whether described image to be detected is reproduction image.
A kind of commodity image reproduction detection device, including processor are also provided in the embodiment of the present invention.Memory, wherein depositing
Contain the executable instruction of processor.Wherein, processor is configured to be performed commodity image and turn over via executing executable instruction
The step of clapping detection method.
As above, the commodity region on each image is compiled according to merchandise classification by acquiring multiple images in the embodiment
Number distinguishing characteristics is generated, and summarizes the corresponding distinguishing characteristics of each image and generate a property data base, and will be according to mapping to be checked
It carries out whether inquiry contrast judgement is out reproduction image as generating target distinguishing characteristics input feature vector database, is treated to realize
Effective supervision to the shooting photo for patrolling salesman is realized in the effective detection that detection image carries out.
Person of ordinary skill in the field it is understood that various aspects of the invention can be implemented as system, method or
Program product.Therefore, various aspects of the invention can be embodied in the following forms, it may be assumed that complete hardware embodiment, complete
The embodiment combined in terms of full Software Implementation (including firmware, microcode etc.) or hardware and software, can unite here
Referred to as " circuit ", " module " or " platform ".
Figure 10 is the structural schematic diagram of commodity image reproduction detection device of the invention.Basis is described referring to Figure 10
The electronic equipment 600 of the embodiment of the invention.The electronic equipment 600 that Figure 10 is shown is only an example, should not be right
The function and use scope of the embodiment of the present invention bring any restrictions.
As shown in Figure 10, electronic equipment 600 is showed in the form of universal computing device.The component of electronic equipment 600 can be with
Including but not limited to: at least one processing unit 610, at least one storage unit 620, connection different platform component (including are deposited
Storage unit 620 and processing unit 610) bus 630, display unit 640 etc..
Wherein, storage unit is stored with program code, and program code can be executed with unit 610 processed, so that processing is single
Member 610 executes various exemplary implementations according to the present invention described in this specification above-mentioned electronic prescription circulation processing method part
The step of mode.For example, processing unit 610 can execute step as shown in fig. 1.
Storage unit 620 may include the readable medium of volatile memory cell form, such as Random Access Storage Unit
(RAM) 6201 and/or cache memory unit 6202, it can further include read-only memory unit (ROM) 6203.
Storage unit 620 can also include program/utility with one group of (at least one) program module 6205
6204, such program module 6205 includes but is not limited to: operating system, one or more application program, other program moulds
It may include the realization of network environment in block and program data, each of these examples or certain combination.
Bus 630 can be to indicate one of a few class bus structures or a variety of, including storage unit bus or storage
Cell controller, peripheral bus, graphics acceleration port, processing unit use any bus structures in a variety of bus structures
Local bus.
Electronic equipment 600 can also be with one or more external equipments 700 (such as keyboard, sensing equipment, bluetooth equipment
Deng) communication, can also be enabled a user to one or more equipment interact with the electronic equipment 600 communicate, and/or with make
Any equipment (such as the router, modulation /demodulation that the electronic equipment 600 can be communicated with one or more of the other calculating equipment
Device etc.) communication.This communication can be carried out by input/output (I/O) interface 650.Also, electronic equipment 600 can be with
By network adapter 660 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network,
Such as internet) communication.Network adapter 660 can be communicated by bus 630 with other modules of electronic equipment 600.It should
Understand, although being not shown in Figure 10, other hardware and/or software module can be used in conjunction with electronic equipment 600, including unlimited
In: microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape drive and number
According to backup storage platform etc..
A kind of computer readable storage medium is also provided in the embodiment of the present invention, for storing program, program is performed
The step of commodity image reproduction detection method of realization.In some possible embodiments, various aspects of the invention may be used also
In the form of being embodied as a kind of program product comprising program code, when program product is run on the terminal device, program generation
Code is each according to the present invention described in this specification above-mentioned electronic prescription circulation processing method part for executing terminal device
The step of kind illustrative embodiments.
As it appears from the above, the program of the computer readable storage medium of the embodiment is when being executed, by acquiring multiple images,
Commodity region on each image is numbered according to merchandise classification and generates distinguishing characteristics, and it is special to summarize the corresponding difference of each image
Sign generates a property data base, and will generate target distinguishing characteristics input feature vector database according to image to be detected and carry out inquiry ratio
To judging whether be reproduction image, to realize the effective detection carried out to image to be detected, the bat to salesman is patrolled is realized
Take the photograph effective supervision of photo.
Figure 11 is the structural schematic diagram of computer readable storage medium of the invention.With reference to shown in Figure 11, basis is described
The program product 800 for realizing the above method of embodiments of the present invention can be deposited using portable compact disc is read-only
Reservoir (CD-ROM) and including program code, and can be run on terminal device, such as PC.However, of the invention
Program product is without being limited thereto, and in this document, readable storage medium storing program for executing can be any tangible medium for including or store program, should
Program can be commanded execution system, device or device use or in connection.
Program product can be using any combination of one or more readable mediums.Readable medium can be readable signal Jie
Matter or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared ray or partly lead
System, device or the device of body, or any above combination.More specific example (the non exhaustive column of readable storage medium storing program for executing
Table) it include: the electrical connection with one or more conducting wires, portable disc, hard disk, random access memory (RAM), read-only storage
Device (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
Computer readable storage medium may include in a base band or as carrier wave a part propagate data-signal,
In carry readable program code.The data-signal of this propagation can take various forms, including but not limited to electromagnetic signal,
Optical signal or above-mentioned any appropriate combination.Readable storage medium storing program for executing can also be any readable Jie other than readable storage medium storing program for executing
Matter, the readable medium can send, propagate or transmit for by instruction execution system, device or device use or and its
The program of combined use.The program code for including on readable storage medium storing program for executing can transmit with any suitable medium, including but not
It is limited to wireless, wired, optical cable, RF etc. or above-mentioned any appropriate combination.
The program for executing operation of the present invention can be write with any combination of one or more programming languages
Code, programming language include object oriented program language-Java, C++ etc., further include conventional process
Formula programming language-such as " C " language or similar programming language.Program code can be calculated fully in user
It executes in equipment, partly execute on a user device, executing, as an independent software package partially in user calculating equipment
Upper part executes on a remote computing or executes in remote computing device or server completely.It is being related to remotely counting
In the situation for calculating equipment, remote computing device can pass through the network of any kind, including local area network (LAN) or wide area network
(WAN), it is connected to user calculating equipment, or, it may be connected to external computing device (such as utilize ISP
To be connected by internet).
By acquiring multiple images in the present invention, to the commodity region on each image according to merchandise classification number generation area
Other feature, and summarize the corresponding distinguishing characteristics of each image and generate a property data base, and mesh will be generated according to image to be detected
Mark distinguishing characteristics input feature vector database carries out whether inquiry contrast judgement is out reproduction image, to realize to image to be detected
Effective supervision to the shooting photo for patrolling salesman is realized in the effective detection carried out；
When will to judge described image to be detected not in the present invention be reproduction image, by the target area of described image to be detected
Other feature is filled into the property data base, is realized the effective real-time update of the property data base, is improved to to be detected
The accuracy rate for the detection that image carries out.
Specific embodiments of the present invention are described above.It is to be appreciated that the invention is not limited to above-mentioned
Particular implementation, those skilled in the art can make various deformations or amendments within the scope of the claims, this not shadow
Ring substantive content of the invention.
Claims (10)
1. a kind of commodity image reproduction detection method, which comprises the steps of:
Step S1: acquiring multiple images, multiple in each described image to identify to detecting in each described image
The corresponding merchandise classification number in commodity region and each commodity region；
Step S2: it is corresponding at least that each image is generated according to the corresponding merchandise classification number in commodity each on each image region
One distinguishing characteristics；
Step S3: according to one property data base of each corresponding distinguishing characteristics generation of image；
Step S4: detecting image to be detected, identifies multiple end article regions in described image to be detected and every
One end article region corresponding merchandise classification number, according to the corresponding merchandise classification number in end article region generate it is described to
The target distinguishing characteristics of detection image；
Step S5: it is compared according to the target distinguishing characteristics with the distinguishing characteristics in the property data base, described in judgement
Whether image to be detected is reproduction image.
2. commodity image reproduction detection method according to claim 1, which is characterized in that further include following steps:
When judging described image to be detected not is reproduction image, described image to be detected and corresponding target are distinguished special
Sign is filled into the property data base；
When judging described image to be detected is reproduction image, then reproduction prompting message is issued.
3. commodity image reproduction detection method according to claim 1, which is characterized in that the step S2 includes following step
It is rapid:
Step S201: preset merchandise classification numeral order；
Step S202: each merchandise classification on each image is counted according to the merchandise classification numeral order and numbers corresponding commodity
Region quantity forms the corresponding one-dimensional matrix of each image, as the corresponding distinguishing characteristics of each image；
The step S4 includes the following steps:
Step S401: detecting image to be detected, identify multiple end article regions in described image to be detected and
The corresponding merchandise classification number in each end article region；
Step S402: according to the merchandise classification numeral order, each merchandise classification number pair in each image to be detected is counted
The commodity region quantity answered forms the corresponding one-dimensional matrix of each image to be detected, as the corresponding target of each image to be detected
Distinguishing characteristics.
4. commodity image reproduction detection method according to claim 3, which is characterized in that the step S5 includes following step
It is rapid:
Step S501: each element in the target distinguishing characteristics is subtracted each other with corresponding element in a distinguishing characteristics rear squared
Generate multiple first element difference radixes；
Step S502: extraction of square root generates the target distinguishing characteristics after multiple first element difference radixes are added and a difference is special
Fisrt feature difference radix between sign；
Step S503: judge the fisrt feature difference radix whether less than a preset fisrt feature difference radix threshold value, when
The feature difference radix be less than the fisrt feature difference radix threshold value when, then determine the target distinguishing characteristics it is corresponding to
Detection image is reproduction image, when the feature difference radix is more than or equal to the fisrt feature difference radix threshold value, is then touched
Send out step S504；
Step S504: repeating step S501 to step S503, when in the target distinguishing characteristics and the property data base
Each distinguishing characteristics between feature difference radix be more than or equal to the fisrt feature difference radix threshold value when, then determine described in
The corresponding image to be detected of target distinguishing characteristics is not reproduction image.
5. commodity image reproduction detection method according to claim 1, which is characterized in that the step S2 includes following step
It is rapid:
Step S201: the surrounding in commodity region each in each image is extended to form one the 1st palace with a preset area
Table images, the commodity region are located at the center of the first nine grids figure picture；
Step S202: judge that each preset area region whether there is a commodity region in the first nine grids figure picture；
Step S203: when a preset area region then obtains the quotient there are when a commodity region in the first nine grids figure picture
The merchandise classification in product region is numbered, and when a commodity region is not present in a preset area region, then sets the preset area region
Corresponding merchandise classification number is a constant；
Step S204: according to preset statistics sequence, it is corresponding successively to count each preset area region in the first nine grids figure picture
Merchandise classification number to form the corresponding one-dimensional matrix in each commodity region, as the corresponding distinguishing characteristics in each commodity region；
Step S205: it is repeated in and executes multiple distinguishing characteristics that step S201 to step S204 generates each image.
6. commodity image reproduction detection method according to claim 5, which is characterized in that the step S4 includes following step
It is rapid:
Step S401: the surrounding in end article region each in each image to be detected is extended to be formed with a preset area
One second nine grids figure picture, the end article region are located at the center of the second nine grids figure picture；
Step S402: judge that each preset area region whether there is in the corresponding second nine grids figure picture in each end article region
One end article region；
Step S403: when a preset area region is then obtained there are when an end article region in the second nine grids figure picture
The merchandise classification in the end article region is numbered, when an end article is not present in a preset area region in the second nine grids figure picture
When region, then setting the corresponding merchandise classification number in the preset area region is a constant；
Step S404: according to the statistics sequence, each preset area region pair in the second nine grids figure picture is successively counted
The merchandise classification answered numbers to form the corresponding one-dimensional matrix in each end article region, corresponding as each end article region
Target distinguishing characteristics；
Step S405: it is repeated in and executes multiple targets difference spy that step S401 to step S404 generates each image to be detected
Sign.
7. commodity image reproduction detection method according to claim 6, which is characterized in that the step S5 includes following step
It is rapid:
Step S501: the commodity image to be detected is compared with each image in the property data base, when described
In property data base in the corresponding merchandise classification number of each image inquiry less than a target quotient in the commodity image to be detected
When the corresponding merchandise classification number in product region, then judges that the commodity image to be detected is not reproduction image, otherwise trigger step
S502；
Step S502: judge in the commodity image to be detected every unitary in the corresponding target distinguishing characteristics in an end article region
Whether element and each corresponding element in the corresponding distinguishing characteristics in commodity region with identical merchandise classification number in an image are identical；
Step S503: when in the target distinguishing characteristics each element in an image with identical merchandise classification number quotient
Product region correspond to each corresponding element in distinguishing characteristics it is all the same when, then determine have in the end article region and described image
There is the commodity region of identical merchandise classification number identical, it is identical otherwise to determine that the end article region has with a described image
The commodity region of merchandise classification number is not identical；
Step S504: step S502 to step S503 is repeated, when the commodity in commodity image to be detected with a described image
End article region total amount ratio is greater than preset ratio in the identical end article region quantity in region and commodity image to be detected
When example threshold value, then the commodity image to be detected is judged for reproduction image, otherwise determines that the commodity image to be detected is not reproduction
Image.
8. a kind of commodity image reproduction detection system, for realizing commodity image reproduction described in any one of claims 1 to 7
Detection method characterized by comprising
Commodity region identification module, for acquiring multiple images, to being detected in each described image, to identify each institute
State the corresponding merchandise classification number in multiple commodity regions and each commodity region on image；
Distinguishing characteristics generation module, it is each for being generated according to the corresponding merchandise classification number in commodity each on each image region
The corresponding at least distinguishing characteristics of image；
Property data base generation module, for generating a property data base according to the corresponding distinguishing characteristics of each image；
Target distinguishing characteristics generation module identifies more in described image to be detected for detecting to image to be detected
The corresponding merchandise classification number in a end article region and each end article region, according to the corresponding commodity in end article region
Class number generates the target distinguishing characteristics of described image to be detected；
Feature comparison module, for being compared according to the target distinguishing characteristics with the distinguishing characteristics in the property data base
It is right, judge whether described image to be detected is reproduction image.
9. a kind of commodity image reproduction detection device characterized by comprising
Processor；
Memory, wherein being stored with the executable instruction of the processor；
Wherein, the processor is configured to come any one of perform claim requirement 1 to 7 institute via the execution executable instruction
The step of stating commodity image reproduction detection method.
10. a kind of computer readable storage medium, for storing program, which is characterized in that described program is performed realization power
Benefit require any one of 1 to 7 described in commodity image reproduction detection method the step of.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810844896.XA CN109034067B (en) | 2018-07-27 | 2018-07-27 | Method, system, equipment and storage medium for commodity image reproduction detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810844896.XA CN109034067B (en) | 2018-07-27 | 2018-07-27 | Method, system, equipment and storage medium for commodity image reproduction detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109034067A true CN109034067A (en) | 2018-12-18 |
| CN109034067B CN109034067B (en) | 2021-08-24 |
Family
ID=64646348
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810844896.XA Active CN109034067B (en) | 2018-07-27 | 2018-07-27 | Method, system, equipment and storage medium for commodity image reproduction detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109034067B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117576425A (en) * | 2024-01-17 | 2024-02-20 | 南京掌控网络科技有限公司 | Method and system for detecting scurrying image |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521614A (en) * | 2011-12-20 | 2012-06-27 | 中山大学 | Identification method of coping digital image |
| CN102592276A (en) * | 2011-12-20 | 2012-07-18 | 中山大学 | Image area description method and copy image detection method based on same |
| CN102930264A (en) * | 2012-09-29 | 2013-02-13 | 李炳华 | System and method for acquiring and analyzing commodity display information based on image identification technology |
| EP3182349A2 (en) * | 2015-12-18 | 2017-06-21 | Ricoh Company, Ltd. | Planogram matching |
-
2018
- 2018-07-27 CN CN201810844896.XA patent/CN109034067B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521614A (en) * | 2011-12-20 | 2012-06-27 | 中山大学 | Identification method of coping digital image |
| CN102592276A (en) * | 2011-12-20 | 2012-07-18 | 中山大学 | Image area description method and copy image detection method based on same |
| CN102930264A (en) * | 2012-09-29 | 2013-02-13 | 李炳华 | System and method for acquiring and analyzing commodity display information based on image identification technology |
| EP3182349A2 (en) * | 2015-12-18 | 2017-06-21 | Ricoh Company, Ltd. | Planogram matching |
Non-Patent Citations (2)
| Title |
|---|
| ZHILI ZHOU 等: "Effective and Efficient Global Context Verification for Image Copy Detection", 《IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY》 * |
| 谢哲 等: "基于同态补偿翻拍图像的方向预测方法", 《计算机应用》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117576425A (en) * | 2024-01-17 | 2024-02-20 | 南京掌控网络科技有限公司 | Method and system for detecting scurrying image |
| CN117576425B (en) * | 2024-01-17 | 2024-04-16 | 南京掌控网络科技有限公司 | Method and system for detecting scurrying image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109034067B (en) | 2021-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108416403B (en) | Method, system, equipment and storage medium for automatically associating commodity with label | |
| CN108647553A (en) | Rapid expansion method, system, equipment and the storage medium of model training image | |
| US11373320B1 (en) | Detecting inventory changes by comparing image data | |
| US10318917B1 (en) | Multiple sensor data fusion system | |
| CN110472515B (en) | Goods shelf commodity detection method and system | |
| CN108364005A (en) | Automatic identifying method, system, equipment and the storage medium of price tag | |
| US20150213498A1 (en) | Method and apparatus for providing product information | |
| TW202013240A (en) | Deep learning-based store realograms | |
| CN110020093A (en) | Video retrieval method, edge device, video frequency searching device and storage medium | |
| CN109784172A (en) | Ground heap commodity occupied area evaluation method, system, equipment and medium in market | |
| US11042836B1 (en) | Fusion of sensor data for detecting interactions at an inventory location | |
| US20220084102A1 (en) | Commodity recommendation method, server, shopping cart and shopping system | |
| US9811537B2 (en) | Product identification via image analysis | |
| US11887046B1 (en) | Updating machine learning systems using previous output data | |
| CN108389230A (en) | Refrigerator capacity automatic testing method, system, equipment and storage medium | |
| CN109102324B (en) | Model training method, and red packet material laying prediction method and device based on model | |
| US20210117987A1 (en) | Fraud estimation system, fraud estimation method and program | |
| Li et al. | Using Kinect for monitoring warehouse order picking operations | |
| CN109034067A (en) | Commodity image reproduction detection method, system, equipment and storage medium | |
| US9460393B2 (en) | Inference of anomalous behavior of members of cohorts and associate actors related to the anomalous behavior based on divergent movement from the cohort context centroid | |
| CN116863116A (en) | Image recognition method, device, equipment and medium based on artificial intelligence | |
| CN116580390A (en) | Price tag content acquisition method, price tag content acquisition device, storage medium and computer equipment | |
| Erke et al. | A fast calibration approach for onboard LiDAR-camera systems | |
| US20220405827A1 (en) | Platform for soliciting, processing and managing commercial activity across a plurality of disparate commercial systems | |
| CN112989957A (en) | Safety monitoring method and system suitable for equipment cabinet |
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 |