CN108629886A - A kind of bank note is stained the detection method and device of grade - Google Patents
A kind of bank note is stained the detection method and device of grade Download PDFInfo
- Publication number
- CN108629886A CN108629886A CN201710159886.8A CN201710159886A CN108629886A CN 108629886 A CN108629886 A CN 108629886A CN 201710159886 A CN201710159886 A CN 201710159886A CN 108629886 A CN108629886 A CN 108629886A
- Authority
- CN
- China
- Prior art keywords
- stained
- subregion
- bank note
- detection area
- object detection
- 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
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/181—Testing mechanical properties or condition, e.g. wear or tear
- G07D7/187—Detecting defacement or contamination, e.g. dirt
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20161—Level set
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Inspection Of Paper Currency And Valuable Securities (AREA)
Abstract
The embodiment of the invention discloses the detection method and device that a kind of bank note is stained grade.The method includes：Obtain at least one object detection area of bank note tow sides image to be detected；Each object detection area is divided, obtains the detection subregion of each object detection area, and score is stained according to each object detection area of grey value characteristics determination of the detection subregion；According at least one object detection area be stained that score determines the bank note to be detected be stained grade.Bank note provided in an embodiment of the present invention is stained the detection method and device of grade, and can achieve the effect that fast and accurately determining bank note is stained grade.
Description
Technical field
The present embodiments relate to paper money recognition technical field more particularly to a kind of bank note be stained grade detection method and
Device.
Background technology
As the rapid development of sociaty and economy, daily cash transaction amount sharply increases, during paper money recognition, for
The determination that bank note is stained grade is also particularly important.
It is to judge a major criterion of fiduciary circulation to be stained grade.It includes on bank note that common bank note, which is stained situation,
Number etc. is signed or write, but cannot be circulated for such general requirement of bank note bank, it is therefore desirable to which automatic teller machine is accurate
It identifies.
In the prior art, for the detection for being stained grade of bank note, due to by sensor stability and picture noise
Deng influence, the accuracy rate for being stained grade detection of bank note is low, and algorithm is complicated, and detection efficiency is low, is that a puzzlement is ground all the time
The technical issues of studying carefully personnel.
Invention content
The embodiment of the present invention provides the detection method and device that a kind of bank note is stained grade, is fast and accurately determined with realizing
The effect for being stained grade of bank note.
In a first aspect, an embodiment of the present invention provides the detection method that a kind of bank note is stained grade, this method includes：
Obtain at least one object detection area of bank note tow sides image to be detected；
Each object detection area is divided, the detection subregion of each object detection area is obtained,
And score is stained according to each object detection area of grey value characteristics determination of the detection subregion；
According at least one object detection area be stained that score determines the bank note to be detected be stained grade.
Further, at least one object detection area for obtaining bank note tow sides image to be detected includes：
The probability for occurring being stained to the detection zone of sample bank note is for statistical analysis；
According to statistical analysis as a result, determining the mapping relations of detection level and the detection zone；
According to the mapping relations, determine and the object detection area corresponding to current detection grade.
Further, the dirt for being stained score and determining the bank note to be detected of each object detection area of the basis
Damaging grade includes：
According to each object detection area to be stained the corresponding detection level of score and the object detection area true
The fixed bank note to be detected is stained grade.
Further, described that each object detection area is divided, obtain each object detection area
Detection subregion, and determine being stained point for each object detection area according to the grey value characteristics of the detection subregion
Number includes：
The object detection area is cut into multiple detection subregions；
The gray value minimum value for counting the detection subregion, determines that the gray value minimum value is less than the institute of predetermined threshold value
It is target subregion to state detection subregion；
The pixel number for being less than the predetermined threshold value according to gray value in each target subregion determines current mesh
Mark whether subregion is to be stained subregion；
It is less than the pixel number of the predetermined threshold value and the dirt according to gray value in all target subregions
The number of subregion is damaged, determine current detection region is stained score.
Further, each gray value is less than the pixel of the predetermined threshold value to the basis in the target subregion
Number determines whether current goal subregion is to be stained subregion to include：
Gray value is less than the pixel number of the predetermined threshold value in each target subregion of statistics；
Each target subregion is corresponded to the corresponding position in the image of the bank note to be detected, it is described right to judge
Answer whether position contains gray average pattern；
If so, then when in the target subregion gray value be less than the pixel number of the predetermined threshold value to be more than first dirty
When damaging threshold value, determine that the target subregion is to be stained subregion；
If nothing, when gray value is less than the pixel number of the predetermined threshold value more than the second dirt in the target subregion
When damaging threshold value, determine that the target subregion is to be stained subregion.
Second aspect, the embodiment of the present invention additionally provide the detection device that a kind of bank note is stained grade, which includes：
Object detection area acquisition module, at least one target detection for obtaining bank note tow sides image to be detected
Region；
It is stained score determining module, for being divided to each object detection area, obtains each target
The detection subregion of detection zone, and determine each object detection area according to the grey value characteristics of the detection subregion
Be stained score；
It is stained level determination module, for being waited for according to being stained described in score determination at least one object detection area
Detection bank note is stained grade.
Further, the object detection area acquisition module includes：
Statistical analysis unit, the probability for occurring being stained for the detection zone to sample bank note are for statistical analysis；
Mapping relations establish unit, are used for according to statistical analysis as a result, determining detection level and the detection zone
Mapping relations；
Object detection area determination unit is used for according to the mapping relations, corresponding to determining and current detection grade
Object detection area.
Further, the level determination module that is stained includes：
It is stained level de-termination unit, for being stained score and the target detection according to each object detection area
What the corresponding detection level in region determined the bank note to be detected is stained grade.
Further, the score determining module that is stained includes：
Subregion cutting unit is detected, for the object detection area to be cut into multiple detection subregions；
Target subregion determination unit, the gray value minimum value for counting the detection subregion, determines the gray scale
It is target subregion to be worth minimum value and be less than the detection subregion of predetermined threshold value；
It is stained subregion judging unit, for being less than the predetermined threshold value according to gray value in each target subregion
Pixel number determine whether current goal subregion is to be stained subregion；
It is stained score determination unit, for being less than the predetermined threshold value according to gray value in all target subregions
Pixel number and the number for being stained subregion, determine current detection region is stained score.
Further, the subregion judging unit that is stained includes：
Pixel counts subelement, is less than the predetermined threshold value for counting gray value in each target subregion
Pixel number；
Pattern corresponds to subelement, for corresponding to each target subregion in the image of the bank note to be detected
Corresponding position, judges whether the corresponding position contains gray average pattern；
First judgment sub-unit, if for there is gray average pattern, when gray value is less than institute in the target subregion
When stating the pixel number of predetermined threshold value and being stained threshold value more than first, determine that the target subregion is to be stained subregion；
Second judgment sub-unit, if for without gray average pattern, when gray value is less than institute in the target subregion
When stating the pixel number of predetermined threshold value and being stained threshold value more than second, determine that the target subregion is to be stained subregion.
The embodiment of the present invention is by getting at least one object detection area of bank note tow sides image, and to all
Object detection area carries out cutting and obtains detection subregion, to obtain target detection according to the grey value characteristics of detection subregion
Region is stained score, and further obtain bank note is stained grade, solves the accuracy for being stained grade detection in the prior art
Problem low, detection speed is slow realizes the effect for being stained grade for fast and accurately determining bank note.
Description of the drawings
Fig. 1 is the flow chart for the detection method that the bank note that the embodiment of the present invention one provides is stained grade；
Fig. 2 is the flow chart for the detection method that bank note provided by Embodiment 2 of the present invention is stained grade；
Fig. 3 is the flow chart for the detection method that the bank note that the embodiment of the present invention three provides is stained grade；
Fig. 4 is the flow chart for the detection method that the bank note that the embodiment of the present invention four provides is stained grade；
Fig. 5 is the structural schematic diagram for the detection device that the bank note that the embodiment of the present invention five provides is stained grade.
Specific implementation mode
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limitation of the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
It should be mentioned that some exemplary embodiments are described as before exemplary embodiment is discussed in greater detail
The processing described as flow chart or method.Although each step is described as the processing of sequence, many of which by flow chart
Step can be implemented concurrently, concomitantly or simultaneously.In addition, the sequence of each step can be rearranged.When its operation
The processing can be terminated when completion, it is also possible to the additional step being not included in attached drawing.The processing can be with
Corresponding to method, function, regulation, subroutine, subprogram etc..
Embodiment one
Fig. 1 is the flow chart for the detection method that the bank note that the embodiment of the present invention one provides is stained grade, and the present embodiment can fit
For carrying out being stained grade detection case to bank note, the bank note that this method can be provided by the embodiment of the present invention is stained grade
Detection device executes, which can be realized by the mode of software and/or hardware, and can be integrated in cash transaction equipment.
As shown in Figure 1, the detection method that the bank note is stained grade includes：
S110, at least one object detection area for obtaining bank note tow sides image to be detected.
Wherein, the tow sides image of bank note to be detected can be the gray level image for being scanned through acquisition.Target detection area
Domain can be the region for including some specific pattern or specific position, may be used also with the region for centainly representing meaning
To be region of the gray value in certain threshold range.Entire paper coin can be reflected for the testing result of object detection area
Be stained grade, it is preferred, therefore, that can be the region being susceptible to corresponding to the position being stained.For example, it may be frontier district
Domain, white space, pattern-free or the shallower region of color.
Wherein object detection area can be the rectangle of regular shape, can also be round, ellipse and irregular component
Shape.Target area can be intercepted on the tow sides image of bank note to be detected, can also be to preset target area, it is only right
The target area of bank note is scanned the acquisition of image.Object detection area can be one, can also be multiple.
S120, each object detection area is divided, obtains detection of each object detection area
Region, and score is stained according to each object detection area of grey value characteristics determination of the detection subregion.
Wherein, can be impartial division for the division of object detection area, such as object detection area is 200*150
It can be divided into the detection subregion of multiple equalizations according to 40*30, can also be unequal division by the rectangle of pixel,
Such as the rectangle that object detection area is 203*152 pixel, it is difficult to be divided into the detection subregion of multiple equalizations, then
Certain division rule can be set to divide.
It is stained score according to each object detection area of grey value characteristics determination of the detection subregion.Its
In, grey value characteristics can detect the average value of the gray value of all pixels point in subregion, and grey value characteristics can also be
The number of the pixel of gray value in a certain range in subregion is detected, and then it is every that the average value of gray value can be utilized to determine
A object detection area is stained score.
Illustratively, when the object detection area shares 3, gray value average value is 15,17 and 225 respectively, then
It can determine that the score that is stained of 3 object detection areas is 90 points, 90 points and 10 points according to the average value of its gray value.It judges
According to that can be that there is the score for representing meaning according to what many experiments obtained, for example, wherein 90 points can illustrate that the target is examined
It surveys region to be stained, 10 points can illustrate the object detection area without being stained.
S130, according at least one object detection area be stained that score determines the bank note to be detected be stained grade.
Wherein, it can be artificially defined to be stained grade, can also be widely applied in the industry, indicate the dirt of a piece of paper coin
Damage a kind of mark of situation.
In conjunction with above-mentioned example, being stained grade can be according to the number for being determined as having the object detection area being stained, for example, can
It is stained score with definition and is determined as having the object detection area being stained in 60 points or more of target detection grade, further according to being determined as
The number for having the object detection area being stained determines the grade that is stained of bank note to be detected, and it is 5 to be such as determined as being stained, phase
It answers, it can be Pyatyi to be stained grade.Correspondingly, higher grade, illustrate bank note to be detected to be stained situation more serious.
It can also be using the size for being stained score and current goal detection zone of object detection area as being stained
The calculating basis of grade, object detection area is bigger, can distribute higher weight, and object detection area is smaller, can distribute
Lower weight.The wherein size of object detection area can be determined according to pixel number in object detection area.
The technical solution of the present embodiment, by getting at least one object detection area of bank note tow sides image,
And cutting is carried out to all object detection areas and obtains detection subregion, to obtain target inspection according to the feature of detection subregion
That surveys region is stained score, and further obtain bank note is stained grade, solves and is stained the accurate of grade detection in the prior art
Property low, problem that detection speed is slow, realize the effect for being stained grade for fast and accurately determining bank note.
Embodiment two
Fig. 2 is the flow chart for the detection method that bank note provided by Embodiment 2 of the present invention is stained grade.The present embodiment is upper
On the basis of stating embodiment, further optimization has been carried out.
As shown in Fig. 2, the detection method that the bank note is stained grade includes：
S210, the probability for occurring being stained to the detection zone of sample bank note are for statistical analysis.
Wherein, sample bank note can be a certain number of bank note, such as can be 100, and sample bank note may include having
The bank note being stained, and the bank note that is not stained, wherein it may include writing or seal writing by writing to be stained.Sample paper
Coin can be same version, the bank note of same denomination.Wherein, if version is identical, denomination is different, but the pattern of identical version
And the format of decorative pattern is all identical, can be used as same batch of sample value ratio, the bank note of same version is that same bank is same
The bank note of time distribution.
Wherein, detection zone can be by the ready-portioned region of fixed proportion, it is preferred that can be respectively to the positive and negative of bank note
Two sides bank note divides 15 detection zones, is of moderate size the advantages of this arrangement are as follows dividing region, if each detection zone
Be stained can cover it is some or all of be stained position, convenient for statistics.
The target of statistical analysis can be to determine in a certain number of bank note that the probability being stained occurs in each detection zone,
There is into the probability being stained being numbered or sorting from high to low in each detection zone again.It, can be with after statistical analysis
Object is detected as priority for occurring being stained the higher detection zone of probability, it is possible to reduce detection bank note is stained situation and is spent
Time, and the efficiency of detection can be improved.
S220, according to statistical analysis as a result, determine detection level and the detection zone mapping relations.
There is the probability being stained in each detection zone of the bank note obtained according to statistical analysis, determine detection level with it is described
The mapping relations of detection zone.Wherein, detection level can be the grade determined according to testing requirements, for example, certain bank for
The tolerance that bank note is stained is higher, then its detection level could be provided as 1 grade, and the tolerance that another bank is stained bank note
It spends relatively low, occurs any handwriting on bank note and do not collect, then its detection level could be provided as 5 grades.Different detections etc.
Grade can be set by staff in the related function module of cash transaction equipment, after setting, when there is user's access
When money, which is just detected bank note with the detection level set.Preferably, detection level may include
5 grades, detection level can be set by staff, and the benefit being arranged in this way is can to adjust different detections as needed
Grade, application are more flexible.
It determines the mapping relations of detection level and the detection zone, can be one-to-one mapping, can also be a pair
More mapping relations.Multiple detection zones can be corresponded to a detection level, can also a detection zone to multiple detections
Grade.In mapping relations, detection zone can occur being stained probability height as arrangement sequence using it according to statistical result
Number, when detection zone serial number I, illustrates the probability highest that being stained occurs in the region, successively decrease successively later.The embodiment of the present invention
In, it is preferred that establish mapping relations such as following table：
The mapping relations table of comparisons
S230, according to the mapping relations, determine and the object detection area corresponding to current detection grade.
According to the mapping relations of foundation, determine and the object detection area corresponding to current detection grade, wherein target is examined
It can be the part in all detection zones to survey region, illustratively, in conjunction with upper table, when detection level is 1 grade, target inspection
Survey region can be I, II and III, by detection level be 2 grades when, then object detection area can be I, II, III and IV, with such
It pushes away.
S240, each object detection area is divided, obtains detection of each object detection area
Region, and score is stained according to each object detection area of grey value characteristics determination of the detection subregion.
S250, corresponding detection of score and the object detection area etc. is stained according to each object detection area
The determining bank note to be detected of grade is stained grade.
Wherein, while being stained grade of bank note to be detected is calculated according to the score that is stained of each object detection area, also
It can be determined using the corresponding detection level in object detection area, it is preferred that can be with the detection of each object detection area
The weight that is stained score addition of the inverse that grade is made as object detection area, the benefit being arranged in this way are to work as target detection area
When domain includes the region of multiple detection levels, can as occur being stained general according to the affiliated detection level of object detection area
The higher object detection area weight of rate is higher, and it is relatively low to occur being stained the lower object detection area weight of probability, every to determine
The weight of a object detection area being stained when score sums it up, the numerical value being calculated in this way can more show current detection value
Ratio is stained situation, and the time is calculated and at the same time saving, and improves the efficiency that bank note is stained situation detection.
The present embodiment on the basis of the above embodiments, provides one kind during determining that bank note is stained grade, root
Object detection area is determined according to the detection level of bank note, and determines that bank note is stained according to the detection level of object detection area
The circular of grade, this method can not only determine detection level as needed, but also can effectively embody the dirt of bank note
Damage situation, convenience of calculation and efficiency is higher.
Embodiment three
Fig. 3 is the flow chart for the detection method that the bank note that the embodiment of the present invention three provides is stained grade.The present embodiment is upper
On the basis of stating embodiment, each object detection area is divided, obtains the inspection of each object detection area
Survey subregion, and according to the grey value characteristics of the detection subregion determine each object detection area be stained score into
Further optimization is gone.
As shown in figure 3, the detection method that the bank note is stained grade includes：
S310, at least one object detection area for obtaining bank note tow sides image to be detected.
S320, the object detection area is cut into multiple detection subregions.
Preferably, an object detection area can be cut into 25 detection subregions in the embodiment of the present invention, set in this way
The benefit set is can be calculated as detection unit using each detection subregion, is conducive to the data meter of each detection zone
It calculates.
The size of each detection subregion can be Wi*Hi, and wherein Wi can be the width of each detection subregion, also may be used
To be the pixel columns in each detection subregion, wherein Hi can be the height of each detection subregion, can also be every
Pixel line number in a detection subregion.Wi*Hi can indicate the size of each detection subregion, can also be each
Detect the number of pixel in subregion.
The gray value minimum value of S330, statistics the detection subregion determines that the gray value minimum value is less than default threshold
The detection subregion of value is target subregion.
It counts in each object detection area, the gray value minimum value of each pixel for detecting subregion.All
In gray value minimum value, it is target subregion to determine that it is less than the detection subregion of predetermined threshold value.Wherein, predetermined threshold value can be
One fixed value, can also be in current goal detection zone, the average value of all gray value minimum values, it is preferred that can be with
It is in current all gray value minimum values, the one third of maximum gray value, the benefit being arranged in this way is can be according to current
If the deeper pattern of color on the corresponding bank note in object detection area present position, can become larger or become smaller in overall gray value
In the case of, a reasonable value is obtained as predetermined threshold value, is avoided because predetermined threshold value is fixed value, entire bank note is stained inspection
Survey can be influenced by bank note itself pattern, lead to the phenomenon of bank note defect detection inaccuracy.
Based on the above technical solution, it is preferred that in the mistake for the gray value minimum value for counting the detection subregion
Cheng Zhong, can be sampled inspection with fixed step size, and the fixed step size can be 4.The benefit being arranged in this way is can to reduce
The data calculation amount of algorithm improves the calculating speed of algorithm, and is step-length with 4, it is ensured that if containing writing on bank note
Writing and seal writing etc. are stained all be sampled and collect, and setting step-length will not be caused excessive and missed to insult area
Sampling, improves the accuracy of bank note defect detection.
S340, the pixel number determination for being less than the predetermined threshold value according to gray value in each target subregion are worked as
Whether preceding target subregion is to be stained subregion.
After determining that some or all detection subregions are target subregion, accurately count in each target subregion
Gray value is less than the pixel number of the predetermined threshold value, and is according to how much determinations of pixel number target subregion
No is to be stained subregion.Wherein it is possible to count the pixel for determining that gray value is less than the predetermined threshold value according to many experiments
What number can be determined as being stained subregion within the scope of, for example, judgement numerical value is 100, then the gray scale of current detection subregion
When value is more than 100 less than the pixel number of predetermined threshold value, it may be determined that detection subregion is to be stained subregion.
S350, pixel number and the institute for being less than the predetermined threshold value according to gray value in all target subregions
The number for being stained subregion is stated, determine current detection region is stained score.
The pixel number that gray value in each target subregion is less than predetermined threshold value is accurately counted, calculating is summed up,
It can obtain the pixel number that gray value in all target subregions is less than predetermined threshold value.Further according to being determined as being stained subregion
Number, determine current detection region is stained score, wherein be stained score specifically can utilize following formula calculate：
Score=10*sum/ (W*H)+S/2
Wherein, score is stained score for current goal detection zone；Sum is that gray value is small in all target subregions
In the pixel number of predetermined threshold value；W*H is the pixel number of target subregion；S is the number for being stained subregion.
S360, being stained for the bank note to be detected is determined according to the score that is stained of at least one object detection area
Grade.
Preferably, the grade that is stained of bank note to be detected specifically can be using the calculating of following formula：
Grade=Σ score*weight
Wherein, grade is stained grade for bank note to be detected；Score is stained score for each object detection area；
Weight is the inverse of the detection level corresponding to each object detection area.
The present embodiment provides on the basis of the various embodiments described above and a kind of specific calculates being stained point for each target area
The algorithm of the detection level of bank note several and to be detected, the algorithm calculation amount is few compared with the prior art, and the accuracy rate of calculating is more existing
There is technology higher, and defect detection grade can be voluntarily set as needed, improves the practicability of the technical program.
On the basis of above-mentioned each technical solution, it is preferred that calculate bank note be stained grade grade be decimal when, it is excellent
Choosing, rounding processing is carried out to it, when such as grade=3.8, then the grade that is stained that current bank note can be obtained with rounding is 3 grades.It is excellent
Choosing, the grade that is stained that bank note is arranged is that 0 to 90 rank determines current bank note when calculating grade more than or equal to 10
It is 9 to be stained grade.
Example IV
Fig. 4 is the flow chart for the detection method that the bank note that the embodiment of the present invention four provides is stained grade.The present embodiment is upper
On the basis of stating embodiment, the pixel of the predetermined threshold value is less than to gray value in the basis each the target subregion
Number determines whether current goal subregion is to be stained subregion to have carried out further optimization.
As shown in figure 4, the detection method that the bank note is stained grade includes：
Gray value is less than the pixel number of the predetermined threshold value in each target subregion of S401, statistics.
S402, each target subregion is corresponded to corresponding position in the image of the bank note to be detected, judged
Whether the corresponding position containing gray average pattern, if nothing, executes S404 if so, then executing S403.
Wherein, gray average pattern can be the pattern that gray value mean value is less than a certain setting value, which can be
The deeper or shallower gray value of bank note coin face pattern color is distinguished, correspondingly, if bank note coin face pattern color is deeper, is being incited somebody to action
The gray value obtained when it is detected is relatively low accordingly, if bank note avoids pattern color shallower or without any pattern,
Its corresponding gray value is also higher, even close to 255.Each the gray value of the corresponding bank note corresponding position of target subregion is equal
Value can be obtained according to experiment statistics.
S403, when in the target subregion gray value be less than the predetermined threshold value pixel number be stained more than first
When threshold value, determine that the target subregion is to be stained subregion.
Wherein, the preferred first threshold is 100, i.e., when the pixel number that gray value is less than predetermined threshold value be more than
At 100, it is determined as being stained subregion.
S404, when in the target subregion gray value be less than the predetermined threshold value pixel number be stained more than second
When threshold value, determine that the target subregion is to be stained subregion.
Wherein, the preferred first threshold is 10, i.e., when the pixel number that gray value is less than predetermined threshold value be more than
At 10, it is determined as being stained subregion.
On the basis of above-mentioned each technical solution, the benefit that the technical program is arranged in this way is can to improve bank note to be stained
In grade detection process, the influence come to itself pattern string of bank note filters out, and obtains more accurate bank note and is stained sentencing for grade
Break as a result, improving the tolerance to bank note background decorative pattern.
Embodiment five
Fig. 5 is the structural schematic diagram for the detection device that the bank note that the embodiment of the present invention five provides is stained grade.Such as Fig. 5 institutes
Showing, the bank note is stained the detection device of grade, including：
Object detection area acquisition module 510, at least one target for obtaining bank note tow sides image to be detected
Detection zone；
It is stained score determining module 520, for being divided to each object detection area, obtains each mesh
The detection subregion of detection zone is marked, and each target detection area is determined according to the grey value characteristics of the detection subregion
Domain is stained score；
It is stained level determination module 530, for determining institute according to the score that is stained of at least one object detection area
That states bank note to be detected is stained grade.
The technical solution of the present embodiment, by getting at least one object detection area of bank note tow sides image,
And cutting is carried out to all object detection areas and obtains detection subregion, to obtain target inspection according to the feature of detection subregion
That surveys region is stained score, and further obtain bank note is stained grade, solves and is stained the accurate of grade detection in the prior art
Property low, problem that detection speed is slow, realize the effect for being stained grade for fast and accurately determining bank note.
On the basis of the various embodiments described above, the object detection area acquisition module 510 includes：
Statistical analysis unit, the probability for occurring being stained for the detection zone to sample bank note are for statistical analysis；
Mapping relations establish unit, are used for according to statistical analysis as a result, determining detection level and the detection zone
Mapping relations；
Object detection area determination unit is used for according to the mapping relations, corresponding to determining and current detection grade
Object detection area.
On the basis of the various embodiments described above, the level determination module 530 that is stained includes：
It is stained level de-termination unit, for being stained score and the target detection according to each object detection area
What the corresponding detection level in region determined the bank note to be detected is stained grade.
On the basis of the various embodiments described above, the score determining module 520 that is stained includes：
Subregion cutting unit is detected, for the object detection area to be cut into multiple detection subregions；
Target subregion determination unit, the gray value minimum value for counting the detection subregion, determines the gray scale
It is target subregion to be worth minimum value and be less than the detection subregion of predetermined threshold value；
It is stained subregion judging unit, for being less than the predetermined threshold value according to gray value in each target subregion
Pixel number determine whether current goal subregion is to be stained subregion；
It is stained score determination unit, for being less than the predetermined threshold value according to gray value in all target subregions
Pixel number and the number for being stained subregion, determine current detection region is stained score.
On the basis of the various embodiments described above, the subregion judging unit that is stained includes：
Pixel counts subelement, is less than the predetermined threshold value for counting gray value in each target subregion
Pixel number；
Pattern corresponds to subelement, for corresponding to each target subregion in the image of the bank note to be detected
Corresponding position, judges whether the corresponding position contains gray average pattern；
First judgment sub-unit, if for there is gray average pattern, when gray value is less than institute in the target subregion
When stating the pixel number of predetermined threshold value and being stained threshold value more than first, determine that the target subregion is to be stained subregion；
Second judgment sub-unit, if for without gray average pattern, when gray value is less than institute in the target subregion
When stating the pixel number of predetermined threshold value and being stained threshold value more than second, determine that the target subregion is to be stained subregion.
The said goods can perform the method that any embodiment of the present invention is provided, and have the corresponding function module of execution method
And advantageous effect.
Note that above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The present invention is not limited to specific embodiments described here, can carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out to the present invention by above example
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
May include other more equivalent embodiments, and the scope of the present invention is determined by scope of the appended claims.
Claims (10)
1. a kind of bank note is stained the detection method of grade, which is characterized in that including：
Obtain at least one object detection area of bank note tow sides image to be detected；
Each object detection area is divided, the detection subregion of each object detection area, and root are obtained
It is stained score according to each object detection area of grey value characteristics determination for detecting subregion；
According at least one object detection area be stained that score determines the bank note to be detected be stained grade.
2. according to the method described in claim 1, it is characterized in that, described obtain bank note tow sides image to be detected at least
One object detection area includes：
The probability for occurring being stained to the detection zone of sample bank note is for statistical analysis；
According to statistical analysis as a result, determining the mapping relations of detection level and the detection zone；
According to the mapping relations, determine and the object detection area corresponding to current detection grade.
3. according to the method described in claim 2, it is characterized in that, the basis each the object detection area is stained point
Number determines that the grade that is stained of the bank note to be detected includes：
Institute is determined according to the corresponding detection level of score and the object detection area that is stained of each object detection area
That states bank note to be detected is stained grade.
4. according to the method described in claim 1, it is characterized in that, described divide each object detection area,
The detection subregion of each object detection area is obtained, and is determined each according to the grey value characteristics of the detection subregion
The score that is stained of the object detection area includes：
The object detection area is cut into multiple detection subregions；
The gray value minimum value for counting the detection subregion, determines that the gray value minimum value is less than the inspection of predetermined threshold value
Survey subregion is target subregion；
The pixel number for being less than the predetermined threshold value according to gray value in each target subregion determines current goal
Whether region is to be stained subregion；
It is less than the pixel number of the predetermined threshold value according to gray value in all target subregions and described is stained son
The number in region, determine current detection region is stained score.
5. according to the method described in claim 4, it is characterized in that, the basis each gray value is small in the target subregion
Determine whether current goal subregion is to be stained subregion to include in the pixel number of the predetermined threshold value：
Gray value is less than the pixel number of the predetermined threshold value in each target subregion of statistics；
Each target subregion is corresponded to the corresponding position in the image of the bank note to be detected, judges the corresponding position
It sets and whether contains gray average pattern；
If so, then when in the target subregion gray value be less than the predetermined threshold value pixel number be stained threshold more than first
When value, determine that the target subregion is to be stained subregion；
If nothing, when gray value is stained threshold less than the pixel number of the predetermined threshold value more than second in the target subregion
When value, determine that the target subregion is to be stained subregion.
6. a kind of bank note is stained the detection device of grade, which is characterized in that including：
Object detection area acquisition module, at least one target detection area for obtaining bank note tow sides image to be detected
Domain；
It is stained score determining module, for being divided to each object detection area, obtains each target detection
The detection subregion in region, and the dirt of each object detection area of grey value characteristics determination according to the detection subregion
Damage score；
Be stained level determination module, for according at least one object detection area be stained score determine it is described to be detected
Bank note is stained grade.
7. device according to claim 6, which is characterized in that the object detection area acquisition module includes：
Statistical analysis unit, the probability for occurring being stained for the detection zone to sample bank note are for statistical analysis；
Mapping relations establish unit, are used for according to statistical analysis as a result, determining the mapping of detection level and the detection zone
Relationship；
Object detection area determination unit, for according to the mapping relations, determining and the target corresponding to current detection grade
Detection zone.
8. device according to claim 7, which is characterized in that the level determination module that is stained includes：
It is stained level de-termination unit, for being stained score and the object detection area according to each object detection area
What corresponding detection level determined the bank note to be detected is stained grade.
9. device according to claim 6, which is characterized in that the score determining module that is stained includes：
Subregion cutting unit is detected, for the object detection area to be cut into multiple detection subregions；
Target subregion determination unit, the gray value minimum value for counting the detection subregion, determines the gray value most
Small value is target subregion less than the detection subregion of predetermined threshold value；
It is stained subregion judging unit, the picture for being less than the predetermined threshold value according to gray value in each target subregion
Vegetarian refreshments number determines whether current goal subregion is to be stained subregion；
It is stained score determination unit, the pixel for being less than the predetermined threshold value according to gray value in all target subregions
Point number and the number for being stained subregion, determine current detection region is stained score.
10. device according to claim 9, which is characterized in that the subregion judging unit that is stained includes：
Pixel counts subelement, and the pixel of the predetermined threshold value is less than for counting gray value in each target subregion
Point number；
Pattern corresponds to subelement, the correspondence in image for each target subregion to be corresponded to the bank note to be detected
Position, judges whether the corresponding position contains gray average pattern；
First judgment sub-unit, if for there is gray average pattern, when in the target subregion gray value be less than it is described pre-
If the pixel number of threshold value is stained threshold value more than first, determine that the target subregion is to be stained subregion；
Second judgment sub-unit, if for without gray average pattern, when in the target subregion gray value be less than it is described pre-
If the pixel number of threshold value is stained threshold value more than second, determine that the target subregion is to be stained subregion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710159886.8A CN108629886B (en) | 2017-03-17 | 2017-03-17 | Method and device for detecting stain grade of paper money |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710159886.8A CN108629886B (en) | 2017-03-17 | 2017-03-17 | Method and device for detecting stain grade of paper money |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108629886A true CN108629886A (en) | 2018-10-09 |
| CN108629886B CN108629886B (en) | 2020-06-09 |
Family
ID=63686846
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710159886.8A Active CN108629886B (en) | 2017-03-17 | 2017-03-17 | Method and device for detecting stain grade of paper money |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108629886B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102021212020A1 (en) * | 2021-10-26 | 2023-04-27 | Robert Bosch Gesellschaft mit beschränkter Haftung | Method for detecting soiling on a lens unit of a camera of an agricultural working machine |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06318246A (en) * | 1993-05-06 | 1994-11-15 | Oki Electric Ind Co Ltd | Paper sheets recognizing device |
| JP2007034448A (en) * | 2005-07-25 | 2007-02-08 | Fujitsu Ltd | Paper sheet processing equipment |
| US20080130943A1 (en) * | 2006-12-05 | 2008-06-05 | Canon Kabushiki Kaisha | Image processing apparatus, control method therefor, program, and storage medium |
| CN101506851A (en) * | 2006-08-31 | 2009-08-12 | 光荣株式会社 | Paper sheet identification device and paper sheet identification method |
| CN101901511A (en) * | 2009-05-27 | 2010-12-01 | 株式会社东芝 | Document handling apparatus |
| CN103324946A (en) * | 2013-07-11 | 2013-09-25 | 广州广电运通金融电子股份有限公司 | Method and system for identifying and classifying paper money |
| CN103489255A (en) * | 2013-08-23 | 2014-01-01 | 广州智萃电子科技有限公司 | Device and method for detecting defect of bank note |
| CN105225335A (en) * | 2015-10-28 | 2016-01-06 | 深圳怡化电脑股份有限公司 | A kind of stained recognition methods of paper money number and system |
| CN106355739A (en) * | 2016-08-18 | 2017-01-25 | 深圳怡化电脑股份有限公司 | Method and device for detecting new or old paper money |
-
2017
- 2017-03-17 CN CN201710159886.8A patent/CN108629886B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06318246A (en) * | 1993-05-06 | 1994-11-15 | Oki Electric Ind Co Ltd | Paper sheets recognizing device |
| JP2007034448A (en) * | 2005-07-25 | 2007-02-08 | Fujitsu Ltd | Paper sheet processing equipment |
| CN101506851A (en) * | 2006-08-31 | 2009-08-12 | 光荣株式会社 | Paper sheet identification device and paper sheet identification method |
| US20080130943A1 (en) * | 2006-12-05 | 2008-06-05 | Canon Kabushiki Kaisha | Image processing apparatus, control method therefor, program, and storage medium |
| CN101901511A (en) * | 2009-05-27 | 2010-12-01 | 株式会社东芝 | Document handling apparatus |
| CN103324946A (en) * | 2013-07-11 | 2013-09-25 | 广州广电运通金融电子股份有限公司 | Method and system for identifying and classifying paper money |
| CN103489255A (en) * | 2013-08-23 | 2014-01-01 | 广州智萃电子科技有限公司 | Device and method for detecting defect of bank note |
| CN105225335A (en) * | 2015-10-28 | 2016-01-06 | 深圳怡化电脑股份有限公司 | A kind of stained recognition methods of paper money number and system |
| CN106355739A (en) * | 2016-08-18 | 2017-01-25 | 深圳怡化电脑股份有限公司 | Method and device for detecting new or old paper money |
Non-Patent Citations (1)
| Title |
|---|
| 许万征: ""纸币多光谱图像分析方法"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108629886B (en) | 2020-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106412573A (en) | Method and device for detecting lens stain | |
| US9031294B2 (en) | Region segmented image data creating system and feature extracting system for histopathological images | |
| CN104867225B (en) | A kind of bank note towards recognition methods and device | |
| CN115249246B (en) | Optical glass surface defect detection method | |
| CN106529559A (en) | Pointer-type circular multi-dashboard real-time reading identification method | |
| CN110163853A (en) | A kind of detection method of edge defect | |
| CN106355739B (en) | A kind of method and device that detection bank note is new and old | |
| CN103268481A (en) | Method for extracting text in complex background image | |
| CN104992496A (en) | Paper money face identification method and apparatus | |
| CN115049671B (en) | Method and system for detecting cloth surface defects based on computer vision | |
| CN106355744B (en) | A kind of recognition methods of Indonesian Rupiah value of money and device | |
| CN115272305B (en) | Button hole defect detection method | |
| CN116067671B (en) | A method, system and medium for testing vehicle paint quality | |
| CN106875540A (en) | A kind of banknote thickness abnormity detection method and device | |
| CN106780965B (en) | A kind of Paper Currency Identification and device | |
| CN101299242B (en) | Method and device for determining threshold value in human body skin tone detection | |
| CN108629886A (en) | A kind of bank note is stained the detection method and device of grade | |
| CN106934922A (en) | A kind of paper currency detecting method and device | |
| Talbot et al. | Elliptical distance transforms and the object splitting problem | |
| Einstein et al. | Reproducibility and accuracy of interactive segmentation procedures for image analysis in cytology | |
| WO2019007164A1 (en) | Banknote graffiti detection method and device | |
| CN107301718B (en) | A kind of image matching method and device | |
| CN109448012A (en) | A kind of method for detecting image edge and device | |
| Schier et al. | Automated counting of yeast colonies using the fast radial transform algorithm | |
| CN111507234B (en) | Cell flow detection method |
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 |