CN107705418B - Paper currency facing identification method, device, equipment and readable storage medium - Google Patents
Paper currency facing identification method, device, equipment and readable storage medium Download PDFInfo
- Publication number
- CN107705418B CN107705418B CN201710936873.7A CN201710936873A CN107705418B CN 107705418 B CN107705418 B CN 107705418B CN 201710936873 A CN201710936873 A CN 201710936873A CN 107705418 B CN107705418 B CN 107705418B
- Authority
- CN
- China
- Prior art keywords
- standard
- facing
- image
- target
- vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
- G07D7/20—Testing patterns thereon
- G07D7/2016—Testing patterns thereon using feature extraction, e.g. segmentation, edge detection or Hough-transformation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- 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
-
- 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
- G07D7/20—Testing patterns thereon
- G07D7/202—Testing patterns thereon using pattern matching
- G07D7/2041—Matching statistical distributions, e.g. of particle sizes orientations
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Inspection Of Paper Currency And Valuable Securities (AREA)
Abstract
The embodiment of the invention discloses a method, a device and equipment for identifying the facing direction of paper money and a readable storage medium. The embodiment of the invention aims to solve the problems that the paper currency facing identification is difficult to uniformly process and the identification accuracy is low due to the brightness difference of the paper currency image, and realize the quick and accurate identification of the paper currency facing and type.
Description
Technical Field
The embodiment of the invention relates to the field of paper currency identification, in particular to a paper currency facing identification method, a device, equipment and a readable storage medium.
Background
In the paper money identification, the paper money facing identification is a preprocessing process of the paper money identification, and the accuracy of facing judgment directly influences the accuracy of subsequent paper money identification. However, because the paper money has various types and is often bent, stained and the like, the paper money required to be recognized has different characteristics and is difficult to be processed in a unified manner, and meanwhile, the recognition accuracy is also influenced by the brightness difference generated by the sensor for collecting the paper money image.
At present, most of paper money face recognition adopts a recognition algorithm based on a neural network, paper money is divided into a plurality of grids, then the sum of gray values of each grid is extracted to form a feature vector, and the feature vector is input into a classifier for training. However, the recognition algorithm based on the neural network needs a large number of training samples, the workload is large, and the selection of the feature vector is very sensitive to the brightness change of the image, which easily causes inaccurate recognition.
Disclosure of Invention
The embodiment of the invention provides a method, a device and equipment for identifying facing paper currency and a readable storage medium, which are used for solving the problems that the facing paper currency is difficult to uniformly process and the identification accuracy is low due to the brightness difference of a paper currency image and realizing the quick and accurate identification of the facing paper currency.
In a first aspect, an embodiment of the present invention provides a method for identifying a facing direction of a banknote, including:
acquiring a to-be-detected face gray level image of the target paper money on the to-be-identified face;
constructing an identification-oriented vector corresponding to the current identification-oriented plane according to the to-be-identified plane gray image;
and matching the facing identification vector with each standard facing vector in a standard facing library, and determining the facing of the target paper money according to a matching result.
In a second aspect, an embodiment of the present invention provides a banknote-faced recognition apparatus, including:
the to-be-detected face gray level image acquisition module is used for acquiring a to-be-detected face gray level image of the target paper money on the to-be-identified face;
an identification-oriented vector construction module, configured to construct an identification-oriented vector corresponding to the current identification-oriented plane according to the to-be-identified gray-scale image;
and the facing identification module is used for matching the facing identification vector with each standard facing vector in a standard facing library and determining the facing of the target paper money according to a matching result.
In a third aspect, an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the banknote-oriented recognition method according to any embodiment of the present invention.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, wherein the program, when executed by a processor, implements a banknote-oriented identification method according to any embodiment of the present invention.
According to the method and the device, the gray level image of the to-be-identified face of the target paper money is obtained, the face identification vector corresponding to the current to-be-identified face is constructed according to the gray level image of the to-be-identified face, then the face identification vector is matched with each standard face vector in a standard face library, the face of the target paper money is determined according to the matching result, the problems that unified processing of the face identification of the paper money is difficult and the identification accuracy is low due to the brightness difference of the paper money image are solved, and the accurate identification of the face of the paper money is realized.
Drawings
FIG. 1 is a flow chart of a banknote facing identification method according to an embodiment of the present invention;
FIG. 2 is a flow chart of an identification-oriented vector construction according to an embodiment of the present invention;
FIG. 3 is a flowchart of a gray level mean calculation according to an embodiment of the present invention;
FIG. 4 is a flowchart of an identification-vector-oriented matching according to an embodiment of the present invention;
FIG. 5 is a flow chart of another method for identifying the facing direction of a bill according to the second embodiment of the present invention;
FIG. 6 is a flowchart of a standard vector-oriented architecture according to a second embodiment of the present invention;
FIG. 7 is a structural diagram of a banknote-facing recognition apparatus according to a third embodiment of the present invention;
fig. 8 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention.
It is to be further noted that, for the convenience of description, only some but not all of the matters related to the present invention are shown in the drawings. Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the operations (or steps) as a sequential process, many of the operations can be performed in parallel, concurrently or simultaneously. In addition, the order of the operations may be re-arranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.
Example one
Fig. 1 is a flowchart of a banknote-oriented recognition method according to an embodiment of the present invention, which is applicable to a case of recognizing the banknote-oriented of a banknote, and the method can be executed by a banknote-oriented recognition device, which can be implemented by software and/or hardware, and can be generally integrated into an unmanned charging device (e.g., an ATM, a ticket vending machine, or a toll machine). The method of the embodiment specifically comprises the following steps:
and S110, acquiring a gray image of the target paper money on the to-be-identified face upwards.
Banknotes typically comprise two faces (a banknote face and a banknote back), and the orientation of the banknote is uncertain when a user inputs a target banknote into the unmanned toll device. Therefore, the banknote face matched with the user input mode needs to be used as the face to be recognized, and the identification of the banknote face of the target banknote is realized by acquiring the gray level image of the face to be recognized, which faces upward.
Typically, the grayscale image of the target banknote can be acquired by an image capture device such as a Contact Image Sensor (CIS) or a camera that stores an integrated ATM, a banknote validator, or the like.
Considering that the edges of the banknotes are worn or curled to some extent during the circulation of the banknotes, in order to ensure the accuracy of the identification process of the subsequent facing banknotes, optionally, after S110, the following steps may be further included: on each image edge (namely, four image edges of an upper image edge, a lower image edge, a left image edge and a right image edge) of the to-be-detected face gray image, pixel points with set thickness are respectively removed (namely, the target paper currency is reduced along the direction from the outer edge of the target paper currency to the center of the paper currency) so as to improve the image edge error.
Those skilled in the art can understand that, those skilled in the art can select the value of the set thickness (for example, 1mm or 2 mm) according to the actual situation, which is not limited in this embodiment, and the value of the thickness should take into account the reduction of the edge error of the image and the amount of information carried by the image itself.
And S120, constructing an identification-oriented vector corresponding to the current identification-oriented surface according to the to-be-identified surface gray level image.
The introduction of the vector can avoid the problem that the current paper money face identification based on the neural network has high sensitivity to the brightness and the darkness of the image.
In an optional implementation manner of this embodiment, the identification-oriented vector may be constructed according to the overall gray scale characteristics of the to-be-detected gray-scale image, for example, in the to-be-detected gray-scale image, the numbers of pixel points corresponding to different gray scale ranges (e.g., [0, 31], [32, 63], …, [224, 255]) are respectively counted, and the identification-oriented vector is constructed according to the number of pixel points corresponding to each gray scale range;
in another optional implementation manner of this embodiment, the to-be-measured grayscale image may be divided into equal or unequal image partitions of a set number, for example: the image blocks equally divided into 3 × 3, 5 × 5, or 9 × 9 may be obtained by calculating a mean grayscale value, a variance grayscale value, a minimum grayscale value, or a maximum grayscale value corresponding to each image block, respectively, to obtain feature values corresponding to each image block, and may be combined with the feature values of each image block to generate the facing identification vector.
And S130, matching the orientation identification vector with each standard orientation vector in the standard orientation library, and determining the orientation of the target paper currency according to the matching result.
In this embodiment, standard orientation vectors corresponding to the respective orientations of banknotes of the set denomination, set amount, and set banknote version may be constructed in advance. And determining the facing of the target paper currency by calculating the similarity between the facing identification vector and each standard facing vector in the standard facing library.
For example, for rmb, two standard facing vectors of the faces corresponding to 2015 edition of 100-element rmb, two standard facing vectors of the faces corresponding to 2005 edition of 100-element rmb, and two standard facing vectors of the faces corresponding to 1999 edition of 100-element rmb may be constructed, respectively. The standard vector-oriented construction process is matched with the identification vector-oriented construction, so that the orientation of the input RMB can be determined by matching the corresponding standard vector-oriented when a user inputs a 100-element RMB.
It will be appreciated that the standard facing library may only include vector-oriented and standard vector-oriented correspondences, for example: (note obverse, vector 1), (note reverse, vector 2), etc., so that the arrangement can directly determine the face of the currently input note, but the amount of information is small; further, the standard facing library may also include corresponding relationships between the banknote type, the banknote facing and the standard facing vector, such as: (100 yuan, note face, vector 3), (20 yuan, note back, vector 4), which increases the amount of data in the standard face library, but can obtain a rich amount of information by a single vector match.
It can be understood that if the standard orientation library stores standard orientation vectors of banknotes of multiple currencies, multiple denominations and multiple versions, after the vector matching is completed, besides the orientation information of the banknotes, the currency, the denomination and the versions of the banknotes can be simultaneously obtained.
According to the method and the device, the gray level image of the to-be-identified face of the target paper money is obtained, the face identification vector corresponding to the current to-be-identified face is constructed according to the gray level image of the to-be-identified face, then the face identification vector is matched with each standard face vector in the standard face library, the face of the target paper money is determined according to the matching result, the problems that unified processing of the face identification of the paper money is difficult and the identification accuracy rate is low due to the brightness difference of the paper money image are solved, and the face and the type of the paper money are quickly and accurately identified.
Further, fig. 2 shows a flowchart of identification vector configuration (i.e., specific operations performed by S120) according to an embodiment of the present invention, and as shown in fig. 2, the specific operations of identification vector configuration include:
s210, dividing the to-be-detected gray-scale image into at least two to-be-detected image blocks according to a set blocking strategy.
The block strategy can be set according to the actually adopted currency, face value and version of the paper money, and a uniform block strategy can also be adopted. And a uniform block strategy is adopted, so that the paper money can be uniformly processed no matter the currency, the face value or the version is different, the algorithm is simplified, and the function of the unmanned charging equipment is richer. For example, one banknote validator can only recognize the face of RMB, and at the same time, the face of other currencies such as U.S. dollars and Euros can be realized without adding an algorithm, and meanwhile, the face of banknotes with different face values of the same currency, such as 100-yuan RMB and 50-yuan RMB, can also be recognized.
Meanwhile, in the process of blocking, the block dividing strategy is set, the blocks can be equally divided or not equally divided into a set number of small blocks, and optionally, the to-be-detected face gray image is equally divided into 5 × 5 small blocks. For paper money with different currencies, denominations and versions, the image is equally divided into 25 small blocks, so that the characteristic information of the gray level image to be detected can be acquired in place, and the problem of increased calculation amount due to too many blocks can be solved.
S220, respectively calculating the gray average value corresponding to each image block to be detected according to the gray value of each pixel point in the image block to be detected.
And calculating the gray average value of each block according to the gray value of each pixel point, and combining the block average values of each block to construct an oriented identification vector, so that the oriented identification vector contains the characteristic information oriented by the paper money as much as possible. For example, the paper to be tested is equally divided into 25 small blocks, wherein the gray value of the pixel point in each small block is different from (0-255) due to the change of brightness of the image, and in the calculation of the gray average value, the gray values of all the pixel points can be selected to be averaged, and the gray average value of a part of the pixel points can also be selected to be averaged.
And S230, combining the gray level mean values of the image blocks to be detected respectively corresponding to the processed images to be detected to construct an identification-oriented vector.
It can be understood that the gray level mean values of each small block are combined to form an orientation recognition vector, and in this case, the orientation vector contains all the feature information of the target banknote orientation.
Further, fig. 3 is a flowchart of the gray scale average calculation (i.e., the specific operation performed in S220), and as shown in fig. 3, the specific operation of the gray scale average calculation includes:
s310, according to a set sampling rule, a set number of pixel points in each to-be-detected image block are obtained and used as mean value calculation pixel points.
Wherein, according to setting for the sampling rule, obtain the pixel of setting for quantity in the image block that awaits measuring and calculate the pixel as the mean value, include:
respectively acquiring horizontal sampling pixel points in the image blocks to be detected in the horizontal direction according to a first sampling interval;
in the vertical direction, according to a second sampling interval, respectively obtaining vertical sampling pixel points in the image blocks to be detected;
and S320, calculating pixel points according to the obtained mean values, and respectively calculating the gray mean values corresponding to the standard image blocks.
And taking the collection of the horizontal sampling pixel points and the vertical sampling pixel points as the mean value calculation pixel points.
In an optional embodiment, 8 pixel points in each to-be-detected image partition are selected as horizontal sampling pixel points, and 6 pixel points in each to-be-detected image partition are selected as vertical sampling pixel points. Therefore, the horizontal sampling points and the vertical sampling points are selected, and the phenomenon that the calculation amount is increased due to excessive pixel point selection is avoided. It can be understood that the selection of the pixel points is set to a set number and a sampling rule, and the operation efficiency is further improved under the condition that the characteristic information is not lost.
Further, fig. 4 shows a flowchart of identification vector matching (i.e., specific operations of S130) according to an embodiment of the present invention, and as shown in fig. 4, the specific operations of identification vector matching include:
s410, respectively calculating the similarity between the facing identification vector and each standard facing vector in the standard facing library, and acquiring a first standard facing vector which meets the set similarity condition with the facing identification vector.
Typically, the similarity calculation uses cosine similarity, assuming that the orientation-oriented identification vector is T1The normal vector is T2Then T is1And T2The cosine similarity cos θ of (c) is:
it can be understood that, when the cosine similarity is closer to 1, the more similar the feature information included in the identification-oriented vector and the standard orientation vector, the identification-oriented vector, that is, the first standard orientation vector, can be determined.
And S420, determining the face corresponding to the first standard face vector as the face of the target paper currency, and determining the paper currency type corresponding to the first standard face vector as the paper currency type of the target paper currency.
It can be understood that the first standard orientation vector contains characteristic information representing the orientation of the target banknote, and the characteristic information of the orientation of the target banknote often contains characteristic information of the type of the target banknote, for example, 2015 version of 100-element RMB is taken as an example, and the characteristic information of the front first standard orientation vector at least contains a chairman head portrait and a 100-element word, so that according to the first standard orientation vector, it can be determined that the orientation of the target banknote is the front and is 100-element RMB, but not 50-element RMB or RMB with other surface values.
Example two
Fig. 5 is a flowchart illustrating another banknote face-recognition method according to the second embodiment of the present invention, where this embodiment further includes a standard face-library construction before the first embodiment is implemented, and the method specifically includes the following steps:
and S510, acquiring at least two paper currencies corresponding to at least two paper currency types respectively as standard paper currencies.
Wherein the banknote types include: at least one banknote of at least one currency, the standard banknote, the test banknote and the slip under at least one version.
In an optional implementation manner of this embodiment, in order to further enrich the versatility of the banknote face recognition method, standard face vectors respectively corresponding to the respective faces of the test banknote and the strip paper may be further acquired, so as to further expand the banknote types that can be recognized by the method of this embodiment.
S520, constructing a standard facing vector corresponding to the target facing of the target banknote type according to the standard facing gray level image of at least two standard banknotes corresponding to the target banknote type in the target facing upwards, wherein the target facing comprises a first facing and a second facing.
In an alternative embodiment of this embodiment, the standard orientation vector may be constructed according to the above construction method for the identification vector, that is, the standard orientation vector may be constructed according to the overall gray scale characteristics of the standard orientation gray scale image, or the standard orientation vector may be constructed by first dividing the standard orientation gray scale image and using the divided gray scale characteristics.
It can be understood that due to different brightness degrees of standard oriented gray image acquisition, different standard oriented vector constructions of different paper currencies and the like, at least two standard paper currencies are necessary to be selected to ensure the integrity of the standard oriented vector database. And the target facing, i.e. the facing of the target banknote, is the front or back of the banknote. For example, if the target banknote is the 100 yuan front of 2015 edition, at least two standard banknotes need to be selected, wherein the front of one standard banknote is the first face, and the front of the other standard banknote is the second face.
And S530, storing the standard orientation vector corresponding to the target orientation of the target banknote type in a standard orientation library.
And S540, acquiring a gray image of the target paper currency on the to-be-recognized side.
And S550, constructing an identification-oriented vector corresponding to the current to-be-identified face according to the to-be-detected face gray level image.
And S560, matching the facing identification vector with each standard facing vector in the standard facing library, and determining the facing of the target paper currency according to the matching result.
The embodiment of the invention constructs the standard orientation database by storing the standard orientation vectors of the paper currency orientation of different currencies, denominations and versions, can realize the orientation identification of the paper currency of any currency, denomination and version, and realizes the unified processing of the orientation identification of the paper currency.
Further, fig. 6 shows a flowchart of a standard vector-oriented configuration (i.e., specific operations performed in S520) according to an embodiment of the present invention, and as shown in fig. 6, the specific steps of the standard vector-oriented configuration include:
s610, acquiring a standard face gray level image of a standard banknote corresponding to the type of the target banknote, wherein the standard face gray level image is upward in the target face, and the standard face gray level image serves as a current processing image.
Wherein, after S610, further comprising:
carrying out image normalization processing on at least one current processing image; and/or
After acquiring the gray-scale image of the target paper currency facing upward to be identified, the method further comprises the following steps:
and carrying out image normalization processing on the to-be-detected face gray level image.
It can be understood that the normalization processing is performed on the current processing image and the to-be-detected gray-scale-oriented image, so that under the condition that the resolution ratio of the image is changed due to different acquisition modes and different brightness changes, the algorithm does not need to be modified, and the uniform processing of the algorithm is ensured.
And S620, dividing the current processing image into at least two standard image blocks according to a set block dividing strategy.
The set blocking policy is the same as the set blocking policy in the above construction method for the identification vector. It is understood that the larger the number of blocks, the more the calculation amount increases, and therefore, it is necessary to divide the image according to the characteristics of the currently processed image. Optionally, it is equally divided into 5 x 5 small pieces.
S630, respectively calculating the gray average value corresponding to each standard image block according to the gray value of each pixel point in each standard image block.
The calculation method of the gray level mean value is the same as the calculation of the gray level mean value in the identification vector-oriented construction method, and the gray level mean value can be calculated according to the characteristics of the whole pixel points of each block or according to the characteristics of a part of the pixel points in each block.
And S640, combining the gray level mean values of the standard image blocks respectively corresponding to the current processed image to construct a middle facing vector corresponding to the target facing.
And S650, returning to execute the operation of acquiring the standard face gray-scale image of the standard face of the standard banknote corresponding to the target banknote type on the target face upwards as the current processing image until the set ending condition is met.
The termination condition is set in order to further filter whether the intermediate face vector includes the feature information of the banknote face, and may be set according to banknote faces of different denominations, and versions. For example, taking the front of 100-yuan RMB of 2015 edition as an example, the chairman avatar is important feature information for front face recognition, and when the middle face vector includes the chairman avatar, the middle face vector is determined to satisfy the setting end condition.
And S660, calculating the vector mean value of each constructed intermediate facing vector as a standard facing vector corresponding to the target facing of the target banknote type.
EXAMPLE III
Fig. 7 is a structural diagram of a banknote-facing recognition apparatus according to a third embodiment of the present invention, and as shown in fig. 7, the apparatus includes a to-be-tested facing gray image acquisition module 710, a facing recognition vector construction module 720, and a facing recognition module 730, where:
the to-be-detected face gray level image acquisition module 710 is used for acquiring a to-be-detected face gray level image of the target paper money on the to-be-identified face;
an identification-oriented vector constructing module 720, configured to construct an identification-oriented vector corresponding to the current identification-oriented plane according to the to-be-detected gray-scale image;
and the facing identification module 730 is used for matching the facing identification vector with each standard facing vector in the standard facing library and determining the facing of the target paper money according to the matching result.
According to the method and the device, the gray level image of the to-be-identified face of the target paper money is obtained, the face identification vector corresponding to the current to-be-identified face is constructed according to the gray level image of the to-be-identified face, then the face identification vector is matched with each standard face vector in the standard face library, the face of the target paper money is determined according to the matching result, the problems that unified processing of the face identification of the paper money is difficult and the identification accuracy rate is low due to the brightness difference of the paper money image are solved, and the face and the type of the paper money are accurately identified.
Further, on the basis of the foregoing embodiments, after the module for acquiring a grayscale image of the target object, the method further includes:
and the image edge improving module is used for respectively removing pixel points with set thickness on each image edge of the to-be-detected gray-scale image so as to improve the image edge error.
Further, the method also comprises the following steps:
the standard banknote obtaining module is used for obtaining at least two banknotes respectively corresponding to at least two banknote types as standard banknotes before obtaining the gray level image of the target banknote to be detected facing upwards to be identified, wherein the banknote types comprise: at least one currency, at least one version of standard banknotes, test banknotes and test strips;
the standard facing vector construction module is used for constructing a standard facing vector corresponding to the target facing of the target banknote type according to a standard facing gray image of at least two standard banknotes corresponding to the target banknote type in the target facing upwards, wherein the target facing comprises a first facing and a second facing;
and the standard oriented library construction module is used for storing a standard oriented vector corresponding to the target orientation of the target banknote type in the standard oriented library.
Specifically, the standard vector-oriented construction module includes:
the current processing image obtaining submodule is used for obtaining a standard face gray level image of a standard banknote corresponding to the type of the target banknote on the target face upwards as a current processing image;
after the current processing image is obtained, image normalization processing should be carried out on at least one current processing image; and/or performing image normalization processing on the to-be-detected face gray level image after acquiring the to-be-detected face gray level image of the target paper money on the to-be-identified face.
The first setting blocking submodule is used for dividing the current processing image into at least two standard image blocks according to a setting blocking strategy;
the first gray mean value calculating submodule is used for respectively calculating the gray mean values corresponding to the standard image blocks according to the gray values of all the pixel points in the standard image blocks;
the middle vector-oriented construction submodule is used for combining the gray average values of the standard image blocks respectively corresponding to the current processed image and constructing a middle vector-oriented corresponding to the target orientation;
a setting end condition submodule for triggering the acquiring current processing image submodule until the setting end condition is satisfied;
and the standard facing vector calculation submodule is used for calculating the vector mean value of each constructed middle facing vector as a standard facing vector corresponding to the target facing of the target banknote type.
Further, the identification-oriented vector construction module comprises:
the second setting blocking submodule is used for dividing the to-be-detected face gray level image into at least two to-be-detected image blocks according to a setting blocking strategy;
the second gray mean value calculating submodule is used for respectively calculating the gray mean value corresponding to each to-be-detected image block according to the gray value of each pixel point in the to-be-detected image block;
and the identification-oriented vector calculation submodule is used for combining the gray average values of the blocks of the images to be detected, which respectively correspond to the processed images to be detected, and constructing the identification-oriented vector.
Specifically, the second gray average calculation submodule includes;
the pixel point obtaining unit is used for obtaining a set number of pixel points in each to-be-detected image block as mean value calculation pixel points according to a set sampling rule;
wherein, obtain pixel unit, specifically be used for: respectively acquiring horizontal sampling pixel points in the image blocks to be detected in the horizontal direction according to a first sampling interval; optionally, selecting 8 pixel points in each to-be-detected image block as horizontal sampling pixel points; in the vertical direction, according to a second sampling interval, respectively obtaining vertical sampling pixel points in the image blocks to be detected; optionally, selecting 6 pixel points in each to-be-detected image block as vertical sampling pixel points;
and the calculation pixel point unit is used for calculating the gray average value corresponding to each standard image block according to the obtained average value calculation pixel points, wherein the aggregation of the horizontal sampling pixel points and the vertical sampling pixel points is used as the average value calculation pixel points.
Further, the identification-oriented module comprises:
the acquisition module is used for acquiring a first standard orientation vector submodule and respectively calculating the similarity between the orientation identification vector and each standard orientation vector in the standard orientation library, and acquiring the first standard orientation vector which meets the set similarity condition with the orientation identification vector;
and the paper currency orientation and type determination submodule is used for determining the orientation corresponding to the first standard orientation vector as the orientation of the target paper currency and determining the paper currency type corresponding to the first standard orientation vector as the paper currency type of the target paper currency.
The product can execute the method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
Example four
Fig. 8 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention, as shown in fig. 8, the electronic device includes a processor 80, a memory 81, an input device 82, and an output device 83; the number of the processors 80 in the electronic device may be one or more, and one processor 80 is taken as an example in fig. 8; the processor 80, the memory 81, the input device 82 and the output device 83 in the electronic apparatus may be connected by a bus or other means, and the bus connection is exemplified in fig. 8.
The memory 81 is used as a computer-readable storage medium for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the banknote-oriented recognition method in the embodiment of the present invention (for example, the to-be-tested gray-scale-oriented image acquisition module 710, the identification-oriented vector construction module 720, and the identification-oriented module 730 in the banknote-oriented recognition device). The processor 80 executes various functional applications and data processing of the electronic device by executing software programs, instructions, and modules stored in the memory 81, that is, realizes the above-described banknote-oriented recognition method.
The memory 81 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 81 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 81 may further include memory located remotely from the processor 80, which may be connected to the device/terminal/server via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 82 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function controls of the electronic device. The output device 83 may include a display device such as a display screen.
EXAMPLE five
The fifth embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a method for identifying a banknote.
Of course, the embodiments of the present invention provide a computer-readable storage medium, whose computer-executable instructions can perform, but are not limited to, the related operations in the banknote-oriented identification method provided in any embodiments of the present invention.
Further, it is clear to those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, by hardware, but the former is a better implementation in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the above search apparatus, each included unit and module are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (11)
1. A method of banknote face recognition, comprising:
acquiring at least two paper currencies respectively corresponding to at least two paper currency types as standard paper currencies;
constructing a standard facing vector corresponding to a target facing of a target banknote type according to a standard facing gray image of at least two standard banknotes corresponding to the target banknote type in the target facing upwards, wherein the target facing comprises a first facing and a second facing;
storing a standard facing vector corresponding to the target facing for the target banknote type in a standard facing repository;
acquiring a to-be-detected face gray level image of the target paper money on the to-be-identified face;
constructing an identification-oriented vector corresponding to the current identification-oriented plane according to the to-be-identified plane gray image;
and matching the facing identification vector with each standard facing vector in a standard facing library, and determining the facing of the target paper money according to a matching result.
2. The method according to claim 1, wherein matching the facing identification vector with each standard facing vector in a standard facing library and determining the facing of the target banknote based on the matching comprises:
respectively calculating the similarity between the identification-oriented vector and each standard orientation vector in a standard orientation library, and acquiring a first standard orientation vector meeting a set similarity condition with the identification-oriented vector;
and determining the facing corresponding to the first standard facing vector as the facing of the target paper currency, and determining the paper currency type corresponding to the first standard facing vector as the paper currency type of the target paper currency.
3. The method of claim 1, wherein the banknote types include: at least one banknote of at least one currency, the standard banknote, the test banknote and the slip under at least one version.
4. The method according to claim 1, wherein constructing a standard face vector corresponding to a target face of a target banknote type from a standard face grayscale image of at least two standard banknotes corresponding to the target banknote type with the target face up comprises:
acquiring a standard face gray level image of a standard banknote corresponding to the type of a target banknote on the target face upwards as a current processing image;
according to a set blocking strategy, dividing the current processing image into at least two standard image blocks;
respectively calculating a gray average value corresponding to each standard image block according to the gray value of each pixel point in the standard image block;
combining the gray level mean values of the standard image blocks respectively corresponding to the current processed image to construct a middle facing vector corresponding to the target facing;
returning to execute the operation of acquiring a standard face gray level image of a standard banknote corresponding to the type of the target banknote on the target face upwards as a current processing image until a set end condition is met;
calculating a vector mean of each of the constructed intermediate facing vectors as a standard facing vector corresponding to the target facing of the target banknote type.
5. The method according to claim 1, wherein constructing an identification-oriented vector corresponding to the current orientation to be identified according to the grayscale image to be identified comprises:
dividing the to-be-detected gray-scale image into at least two to-be-detected image blocks according to a set blocking strategy;
respectively calculating a gray average value corresponding to each image block to be detected according to the gray value of each pixel point in the image block to be detected;
and combining the gray level mean values of the blocks of the images to be detected respectively corresponding to the processed images to be detected to construct the identification-oriented vector.
6. The method according to claim 4, further comprising, after acquiring a standard face grayscale image of a standard banknote corresponding to a target banknote type with the target face up as a current processing image:
carrying out image normalization processing on at least one current processing image; and/or
After acquiring the gray-scale image of the target paper currency facing upward to be identified, the method further comprises the following steps:
and carrying out image normalization processing on the to-be-detected face gray level image.
7. The method of claim 5, wherein calculating a mean gray value corresponding to each of the image blocks to be tested according to the gray values of the pixels in the image block to be tested comprises:
acquiring a set number of pixel points in each to-be-detected image block as mean value calculation pixel points according to a set sampling rule;
calculating pixel points according to the obtained mean value, and respectively calculating the gray mean value corresponding to each standard image block;
wherein, according to setting for the sampling rule, obtain the pixel of setting for quantity in the image block that awaits measuring and calculate the pixel as the mean value, include:
respectively acquiring horizontal sampling pixel points in the image blocks to be detected in the horizontal direction according to a first sampling interval;
in the vertical direction, according to a second sampling interval, respectively obtaining vertical sampling pixel points in the image blocks to be detected;
and taking the collection of the horizontal sampling pixel points and the vertical sampling pixel points as the mean value calculation pixel points.
8. The method according to claim 1, further comprising, after acquiring the to-be-recognized face-up grayscale image of the target banknote with the to-be-recognized face up, the steps of:
and respectively removing pixel points with set thickness on each image edge of the to-be-detected gray-scale image so as to improve the image edge error.
9. A banknote-faced recognition apparatus, comprising:
the to-be-detected face gray level image acquisition module is used for acquiring a to-be-detected face gray level image of the target paper money on the to-be-identified face;
an identification-oriented vector construction module, configured to construct an identification-oriented vector corresponding to the current identification-oriented plane according to the to-be-identified gray-scale image;
the face recognition module is used for matching the face recognition vector with each standard face vector in a standard face library and determining the face of the target paper money according to a matching result;
the standard banknote obtaining module is used for obtaining at least two banknotes respectively corresponding to at least two banknote types as standard banknotes before obtaining the gray level image of the target banknote to be detected facing upwards to be identified, wherein the banknote types comprise: at least one currency, at least one version of standard banknotes, test banknotes and test strips;
the standard facing vector construction module is used for constructing a standard facing vector corresponding to the target facing of the target banknote type according to a standard facing gray image of at least two standard banknotes corresponding to the target banknote type in the target facing upwards, wherein the target facing comprises a first facing and a second facing;
and the standard oriented library construction module is used for storing a standard oriented vector corresponding to the target orientation of the target banknote type in the standard oriented library.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1-8 when executing the program.
11. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710936873.7A CN107705418B (en) | 2017-10-10 | 2017-10-10 | Paper currency facing identification method, device, equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710936873.7A CN107705418B (en) | 2017-10-10 | 2017-10-10 | Paper currency facing identification method, device, equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107705418A CN107705418A (en) | 2018-02-16 |
| CN107705418B true CN107705418B (en) | 2020-06-09 |
Family
ID=61184101
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710936873.7A Active CN107705418B (en) | 2017-10-10 | 2017-10-10 | Paper currency facing identification method, device, equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107705418B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108492447B (en) * | 2018-03-20 | 2020-09-15 | 深圳怡化电脑股份有限公司 | Dollar-oriented identification method, electronic device and computer-readable storage medium |
| CN112447002B (en) * | 2019-09-03 | 2024-02-02 | 深圳怡化电脑股份有限公司 | Bill discriminating method, bill discriminating apparatus, storage medium and computer apparatus |
| CN111738079A (en) * | 2020-05-19 | 2020-10-02 | 武汉卓目科技有限公司 | Banknote denomination recognition method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105654609A (en) * | 2015-12-29 | 2016-06-08 | 深圳怡化电脑股份有限公司 | Paper money processing method and paper money processing system |
| CN105894656A (en) * | 2016-03-30 | 2016-08-24 | 浙江大学 | Banknote image recognition method |
| CN106530483A (en) * | 2016-11-10 | 2017-03-22 | 深圳怡化电脑股份有限公司 | Banknote face direction identification method and device |
| CN106898085A (en) * | 2017-02-23 | 2017-06-27 | 深圳怡化电脑股份有限公司 | Identification bank note towards method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3050031B1 (en) * | 2013-09-24 | 2019-11-06 | Robert Bosch GmbH | System and method for document and article authentication |
-
2017
- 2017-10-10 CN CN201710936873.7A patent/CN107705418B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105654609A (en) * | 2015-12-29 | 2016-06-08 | 深圳怡化电脑股份有限公司 | Paper money processing method and paper money processing system |
| CN105894656A (en) * | 2016-03-30 | 2016-08-24 | 浙江大学 | Banknote image recognition method |
| CN106530483A (en) * | 2016-11-10 | 2017-03-22 | 深圳怡化电脑股份有限公司 | Banknote face direction identification method and device |
| CN106898085A (en) * | 2017-02-23 | 2017-06-27 | 深圳怡化电脑股份有限公司 | Identification bank note towards method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107705418A (en) | 2018-02-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10410292B2 (en) | Method, system, apparatus, and storage medium for realizing antifraud in insurance claim based on consistency of multiple images | |
| CN104867225B (en) | A kind of bank note towards recognition methods and device | |
| CN108875522A (en) | Face cluster methods, devices and systems and storage medium | |
| CN106952393B (en) | Paper money identification method and device, electronic equipment and storage medium | |
| CN110399890A (en) | Image-recognizing method, device, electronic equipment and readable storage medium storing program for executing | |
| CN107705418B (en) | Paper currency facing identification method, device, equipment and readable storage medium | |
| CN105894656A (en) | Banknote image recognition method | |
| CN110443350B (en) | Model quality detection method, device, terminal and medium based on data analysis | |
| CN104992496A (en) | Paper money face identification method and apparatus | |
| WO2013170663A1 (en) | Paper identifying method and related device | |
| CN107103683B (en) | Paper money identification method and device, electronic equipment and storage medium | |
| CN112052186A (en) | Target detection method, device, equipment and storage medium | |
| CN106898078B (en) | Port currency version identification method and device | |
| CN107038794B (en) | Method, device and equipment for identifying authenticity of paper money and storage medium | |
| US20080175447A1 (en) | Face view determining apparatus and method, and face detection apparatus and method employing the same | |
| CN102110323B (en) | Method and device for examining money | |
| CN109740417A (en) | Invoice type recognition methods, device, storage medium and computer equipment | |
| CN111898520A (en) | Certificate authenticity identification method and device, computer readable medium and electronic equipment | |
| CN108665603A (en) | Identify the method, apparatus and electronic equipment of bank note currency type | |
| CN107680246B (en) | Method and equipment for positioning curve boundary in paper money pattern | |
| Cai et al. | An adaptive symmetry detection algorithm based on local features | |
| CN116092230B (en) | Ticket false identifying method and system | |
| CN107958531B (en) | Paper money identification method and device and equipment | |
| CN108985350B (en) | Method and device for recognizing blurred image based on gradient amplitude sparse characteristic information, computing equipment and storage medium | |
| CN114648683B (en) | Neural network performance improving method and device based on uncertainty analysis |
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 |