WO2021239000A1

WO2021239000A1 - Method and apparatus for identifying motion blur image, and electronic device and payment device

Info

Publication number: WO2021239000A1
Application number: PCT/CN2021/096147
Authority: WO
Inventors: 何炜雄; 李志荣; 窦川川; 梁明杰
Original assignee: 支付宝(杭州)信息技术有限公司
Priority date: 2020-05-27
Filing date: 2021-05-26
Publication date: 2021-12-02
Also published as: CN111368813B; CN113743173A; CN111368813A

Abstract

Disclosed are a method and apparatus for identifying a motion blur image, and an electronic device and a payment device. The method comprises: acquiring an image sequence; starting from an nth target image P_n in the image sequence, performing conformance determination on the next target image P_n+1 thereof, wherein n is a positive integer greater than or equal to 1; if the target image P_n+1 does not meet conformance, determining that there is motion blur in the target image P_n+1; and if the target image P_n+1 meets conformance, updating the state of the target image P_n+1 according to the state of the target image P_n, and determining whether there is motion blur in the target image P_n+1 according to the state of the target image P_n+1, wherein the target image is an image in the image sequence that includes a target area, and the nth target image P_n is in an initial state.

Description

Motion blur image recognition method, device, electronic equipment and payment equipment

Technical field

This application relates to the field of computer technology, and in particular to a method, device, electronic equipment, and payment equipment for recognizing motion blurred images.

Background technique

With the gradual maturity of pattern recognition technology, the technology of identifying biological individuals based on biological signs (ie, biometrics) has begun to be applied and promoted in areas that require identity verification, such as fingerprint/face recognition-based mobile phone unlocking, fingerprint door locks , Pay by face, etc.

At present, the market has launched a payment method that can complete quick payment by swiping the face, officially entering the era of "relying on the face to eat". However, it is difficult to identify similar faces and facial features are easily affected by external factors and easily attacked by confirming the identity through face recognition. In contrast, the iris is superior to the human face in terms of individual characteristics, stability, and resistance to aggression. Therefore, multi-modal identity recognition that integrates human face and iris has become a development trend. Because the iris has very high requirements for image quality, it is necessary to judge and eliminate the blurred image to obtain a high-quality iris image and then recognize it.

Summary of the invention

In view of this, the embodiments of this specification provide a method, device, electronic device and payment device for recognizing a motion blurred image, so as to more quickly and accurately recognize whether there is a motion blurred image in an image sequence.

The embodiment of this specification adopts the following technical solution: the embodiment of this specification provides a method for recognizing a motion blurred image, the method includes: acquiring an image sequence; starting from the nth target image P _{n in} the image sequence, The next target image P _n+1 is subject to compliance determination; n is a positive integer greater than or equal to 1; if the target image P _n+1 does not meet compliance, it is determined that the target image P _n+1 has motion blur; if object image P _{n + 1} satisfy the compliance, the state of the object image P _{n + 1} is updated according to the state of the target image P _n, P and determines the target image according to a state of the object image P _{n + 1} _n+1 whether there is motion blur; wherein, the target image is an image that includes a target area in the image sequence, and the nth target image P _n is an initial state.

The embodiment of this specification also provides a method for recognizing a motion blurred image, which includes: acquiring an image sequence; starting from the nth target image P _n in the image sequence, the next target image P _n+1 Perform compliance determination; n is a positive integer greater than or equal to 1; if the target image P _n+1 meets compliance, it is determined that the target image P _n+1 has motion blur; otherwise, it is determined that the target image P _{n+1 is} not There is motion blur; wherein, the target image is an image containing a target area in the image sequence.

The embodiment of this specification provides a device for recognizing a motion blur image, the device includes: an acquisition module, the acquisition module is used to acquire an image sequence; a judgment module, the judgment module is used to obtain an image from the acquisition module Start with the n-th target image P _n in the sequence, and perform compliance judgment on the next target image P _n+1 ; n is a positive integer greater than or equal to 1; and, if the target image P _n+1 does not meet the compliance, it is determined that the target image P _{n + 1} motion blur exists; if the target image P _{n + 1} satisfy the compliance, the state of the image P _{n + 1} is updated in accordance with the target state of the target image P _n, and in accordance with the said object image P _{n + 1} determines the state of the object image P _{n + 1} if there is a motion blur; wherein the target image is an image in the sequence of images including the target region, and the n-th object image P _n is the initial state.

The embodiment of this specification also provides a device for recognizing a motion blurred image, the device includes: an acquisition unit, the acquisition unit is used to acquire an image sequence; a determination unit, the determination unit is used to acquire the Start with the nth target image P _n in the image sequence, and perform compliance judgment on the next target image P _n+1 ; n is a positive integer greater than or equal to 1; and, if the target image P _n+1 does not meet the compliance , It is determined that the target image P _n+1 has motion blur; otherwise, it is determined that the target image P _n+1 does not have motion blur; wherein, the target image is an image that includes a target area in the image sequence.

The embodiment of the present specification also provides an electronic device, including: at least one processor and a memory, the memory stores a program, and is configured to execute the above-mentioned motion blur image recognition method by the at least one processor.

The embodiments of the present specification also provide a computer-readable storage medium that stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, the aforementioned motion-blurred image recognition method is implemented.

The embodiment of this specification also provides a payment device, which includes: a collection unit for collecting facial video data; an image screening unit for screening out images to be recognized without motion blur according to the above-mentioned method; and an image recognition unit, It is used for recognizing the screened out image to be recognized; the payment unit is used for determining whether to perform a payment operation according to the recognition result of the image recognition unit.

The above-mentioned at least one technical solution adopted in the embodiments of this specification can achieve the following beneficial effects: The motion blur image recognition method provided by the embodiments of this specification is used to remove images of poor quality caused by motion blur in an image sequence. Compared with traditional methods, Perform feature extraction or coding on the region of interest on a single image to determine whether the image has motion blur. The method in this specification is to use the position information of the region of interest in the image sequence (such as the position of the human eye) in the image and Time information is used to determine whether the image has motion blur, which is more robust to factors such as illumination and image noise. In addition, the method also has the advantages of simple process, no dependence on complex calculations, and short calculation time.

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of this specification or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments described in this specification. For those of ordinary skill in the art, without creative labor, other drawings can be obtained from these drawings:

Fig. 1 is a main flow chart of a method for recognizing a motion blurred image provided by an embodiment of this specification;

FIG. 2 is a specific flowchart of a method for recognizing a motion blurred image provided by an embodiment of this specification;

FIG. 3 is an example of the judgment result of some images in the image sequence obtained by the embodiment of this specification;

FIG. 4 is a schematic structural diagram of a motion blur image recognition device provided by an embodiment of this specification;

Figure 5 is a schematic structural diagram of a payment device provided by an embodiment of the specification;

FIG. 6 is a main flow chart of a method for recognizing a motion blurred image according to another embodiment of this specification;

FIG. 7 is a schematic structural diagram of a motion blur image device provided by another embodiment of this specification.

Detailed ways

As described in the background art, since the iris has very high requirements for image quality, it is necessary to determine and exclude the blurred image to obtain a high-quality iris image. In the fast payment scenario, since the payment process requires very high timeliness, most of the entire process is the transition time from customer movement to stationary. Therefore, how to exclude poor quality images due to motion blur from the image sequence is very important.

The traditional motion blur determination method mainly determines whether the image has motion blur by performing gradient calculation or encoding on the region of interest on a single image. However, under the premise of calculating the region of interest in the original image, this type of method has higher additional computational complexity, which leads to a long time-consuming judgment. At the same time, such methods are less robust to factors such as illumination and image noise.

In order to enable those skilled in the art to better understand the technical solutions in this specification, the following will clearly and completely describe the technical solutions in the embodiments of this specification in conjunction with the drawings in the embodiments of this specification. Obviously, the described The embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of this specification, all other embodiments obtained by a person of ordinary skill in the art without creative work shall fall within the protection scope of this application.

In this article, the term motion blur (motion blur): a static scene or a series of pictures like movies or fast moving objects in an animation cause obvious fuzzy drag traces; the term human eye positioning: locate the human eye in the image The position, the result is generally given by the horizontal and vertical coordinates, length and width of the circumscribed rectangular frame of the human eye in the upper left corner of the image; the term Intersection-over-Union (IoU): refers to the difference between the candidate frame and the original marked frame Overlap rate, that is, the ratio of their intersection and union. The term region of interest (region of interest): that is, the region that needs to be processed is outlined in the form of boxes, circles, ellipses, and irregular polygons from the processed image.

FIG. 1 is a main flow chart of a method for recognizing a motion blurred image provided by an embodiment of this specification. The method includes: S102: Obtain an image sequence.

In this step, an image sequence in the video data can be obtained based on the collected video data. The image sequence generally contains multiple frames of images, for example, each video data contains about 20 or 30 frames of images. As an example, in a quick payment scenario, the user's facial video data can be collected through a corresponding collection device, where each facial video data can include about 20 frames of images.

S104: Starting from the n-th target image P _n in the image sequence, perform compliance judgment on the next target image P _n+1.

As a specific example, before this step and after step S102 (acquiring an image sequence), the method of this specification may further include: determining the nth target image P _{n in} the image sequence and its next The target image P _n+1 . As an example, the target area contained in the target image is the position of the human eye, and those skilled in the art can detect the position of the human eye through a corresponding eye positioning algorithm. Regarding the eye location algorithm, but not limited to the use of the eye detection algorithm based on the deep convolutional network, the embodiment of this specification does not limit this, and those skilled in the art can select a suitable algorithm to detect the image sequence containing people according to actual needs. Image of eye position. In the embodiment of this specification, when the position of the human eye is detected, the rectangular frame in the image can be used to indicate the position of the human eye. Among them, the first image in which the target area is detected can be used as the target image P _n , and the next image in which the target area is detected can be used as the target image P _n+1 .

In this step, n is a positive integer greater than or equal to 1. As an example, an example to n = 1, i.e., from the first object images P _n (including the image of the target area), beginning with the next line with its determination target image P _2. It should be noted here that in the embodiments of this specification, the compliance determination mainly refers to the compliance of two adjacent target images, and not necessarily the comparison between two adjacent frames in the image sequence. . In some special cases, the current frame image is the target image, and the next frame image may not be the target image. The next frame image is the target image. At this time, the current target image is P _n , and the next frame of the current target image is P n. One frame of image is its next target image P ₂ .

Regarding compliance, in the embodiments of this specification, the compliance may be compliance in the space-time dimension. Specifically, the conformity may refer to the temporal and spatial conformity of two adjacent target images. Among them, the temporal compliance can be the time interval between two adjacent target images; the spatial compliance can be the consistency between two adjacent target images, generally the corresponding target in the target image Regional consistency.

The following is a further detailed description of how to perform compliance determination for this step.

In the example of this specification, _{the time interval t between the n-th target image P n} and the next target image P _n+1 can be compared first, if t>the first set threshold, then the target image P _{n+1 can be determined} Does not meet compliance. If t ≤ the first set threshold, compare _{the consistency r between the nth target image P n} and the next target image P _n+1 , if r <the second set threshold, then determine the target image P _{n +1} does not meet the compliance; if r≥the second set threshold, it is determined that the target image P _n+1 meets the compliance.

In the foregoing, the first set threshold may be a preset time threshold. For example, the first set threshold may be 300 milliseconds. Those skilled in the art can also flexibly set the first set threshold according to actual conditions. For example, an appropriate first threshold can be set based on past experience, or the first threshold can be set reasonably according to the hardware device's conditions, such as acquisition accuracy. The first set threshold is not specifically limited. As an example, the time interval t between the n-th target image P _n and the next target image P _n+1 can also be determined by the number of frames or the frame sequence number, etc., to reflect the current target image frame and its previous target The length of time between images.

In the above, when the target area is the human eye position, the human eye position is represented by the rectangular frame in the image, and _{the rectangular frame size R 1} corresponding to the human eye position in the _{first target image P n and the next target image can be calculated separately} The _{size of the rectangular frame R 2} corresponding to the position of the human eye in _{P n+1} . The consistency r between the target image P _n and the next target image P _n+1 can be measured by the intersection ratio, that is, the overlap ratio between the candidate frame and the original marked frame. It can be calculated according to the following formula:

r=R _n+1 /R _n ;

Among them, R _n+1 is the rectangular frame data corresponding to the target area in the target image P _n+1 _{; R n} is the rectangular frame data corresponding to the target area in the target image P _n.

It should be noted that although the embodiment of this specification selects the intersection ratio as the measure of consistency, for those skilled in the art, other measures can also be selected to indicate the consistency of the target regions corresponding to two adjacent target images. For example, the ratio of the size of the rectangular frame or the pixel distance of the center of the rectangular frame can also be used to reflect the consistency of the human eye position corresponding to the current target image and the human eye position corresponding to the previous target image.

In the above, the second set threshold may be 0.8, 0.9, etc., that is, when _{the consistency r between the target image P n} and the next target image P _n+1 reaches 80% or more, the compliance is satisfied. Those skilled in the art can also set other appropriate second set thresholds according to actual needs, such as setting appropriate second set thresholds based on past experience, or reasonably based on hardware equipment conditions, such as acquisition accuracy, etc. The second set threshold is set, and the embodiment of this specification does not specifically limit how to set the second set threshold.

This step may be executed cyclically, for example: start with the first object images P ₁ starts, to its next target image P ₁ P ₂ a target image for compliance determination, if the conformity is performed to meet the step S106, or step S108 is performed ; Then use the target image P ₂ to determine the compliance of its next target image P ₃ , if the compliance is met, perform step S106, otherwise perform step S108; then use the target image P ₃ for the next target image P ₄ Perform compliance determination, if compliance is met, step S106 is executed, otherwise, step S108...sequentially cyclically determine.

S106: If the target image P _n+1 does not satisfy the compliance, it is determined that the target image P _n+1 has motion blur.

In this step, if it is determined in step S104 that the target image P _n+1 does not meet the compliance, that is, whether _{the time interval between the target images P n} and P _n+1 does not meet the first set threshold, or the target image P _n If the consistency with P _n+1 does not meet the second set threshold, it is determined that the target image P _n+1 does not meet the compliance, that is, there is motion blur.

S108: If the target image P _{n + 1} satisfy the compliance, the state of the object image P _{n + 1} is updated according to the state of the target image P _n, P of the target image and determining the state of the object image P _{n + 1} _{Whether n+1} is motion blur.

In this step, the state of the object image P _n to an initial state, the target state image P _{n + 1} represents the updated target image P _{n + 1} is m times in a row to meet compliance target image. Specifically, taking n=1 as an example, take the target image P ₁ to make a compliance judgment for its next target image P _2. If the target image P ₂ satisfies the compliance judgment, at this time, since the target image P _n is the initial state, For example, the initial state may be the target image _{P. 1} number m = 1, then the state of the object image P ₂ is updated according to the initial state of the object image P _n when the target image P-number update ₂ may be as m = 2; and then in its next target image P ₂ P ₃ a target image for compliance determination, if the target image P ₃ satisfy the compliance determination, the status update empathy object image P _3, P-number target image may be updated to ₃ m = 3; when following the target image and then its P ₃ P ₄ a target image for compliance determination, if the target image P ₄ satisfies compliance determination, the object image P empathy update state _4, the target image P may be ₄ The number of is updated to m=4, and it is determined and updated in turn.

It should be noted that, in the above, if the target image P ₂ does not meet the compliance determination, _{the state of the target image P 1} can be updated to m=0 ( _{the initial state of the target image P 1} is m=1), then according to the target When the initial state of the image P _n updates the state of the target image P ₂ , the state of the target image P ₂ can be updated to m = 1; at this time, the target image P _{2 is} used to determine the conformity of the next target image P _{3, if} If the target image P ₃ satisfies the compliance determination, _{the number of the update P 3} is m = 1. If the target image P ₃ does not meet the compliance determination, _{the state of the target image P 2} can be updated to m = 0 in the same way. P-state image ₃ is updated to m = 1.

As an example, in this step, according to the state of the object image P _{n + 1} determines the target image P _{n + 1} if motion blur exists, specifically comprising: if m> a third set threshold, it is determined that the object image P _n+1 does not have motion blur; otherwise, it is determined that the target image P _n+1 has motion blur. The third set threshold value can be 1, 2 or 3, and can also be flexibly set by those skilled in the art according to actual conditions, which is not limited in the embodiment of this specification.

In order to describe in more detail a method for recognizing a motion blur image in this specification, a specific practical example is used for description below. Referring to FIG. 2, FIG. 2 is a specific flowchart of a method for recognizing a motion blurred image provided by an embodiment of this specification. In this specific implementation, the following steps are performed:

S210: Initialize parameters.

In this step, the time threshold T (the first set threshold), the overlap rate threshold D (ie, the second set threshold), and the frame threshold M are set, and the frame counter n=0 is initialized.

S220: Acquire a single frame image in the image sequence.

In this step, an image sequence in the video data can be obtained according to the collected video data. The image sequence contains multiple frames of images, and each frame of image is obtained in turn.

S230: Human eye positioning.

In this step, for the acquired frame image, starting from the first frame image, it is detected whether the image contains the position of the human eye. If there is no human eye in the image, set n=0, repeat S220-S230, that is, detect the first image. Whether the two frames of images contain the position of the human eye until the image containing the position of the human eye is detected, the rectangular frame in the image is used to indicate the position of the human eye.

S240: Frame time determination.

In this step, the frame image including the human eye position detected in step S230, if the frame image is the second frame image, then the image is denoted as n=1, and the time of the frame image is denoted as t ₁ , And execute step S270. After the status is updated, execute S220-230 again, that is, to detect whether the third frame of image contains the position of the human eye, if it contains the position of the human eye, then the third frame of The time of is recorded as t ₂ , _{and the difference between t 2} and t ₁ is compared. If t ₂ -t ₁ >T, it is determined that the current frame image has motion blur. At this time, let n=0, record the current image (the third frame of image) as n=1, and record the time of the current image as t ₁ , and execute step S270. After the status is updated, execute S220-230 again, that is, detect the first Whether the four frames of images contain the position of the human eye, if it contains the position of the human eye, the fourth frame of image is taken as the current image, the time of the current image is recorded as t ₂ , _{the difference between t 2} and t ₁ is compared, and so on Determine until t ₂ -t ₁ ≤T.

S250: Consistency determination.

In this step, when the fourth frame of image is used as the current image, if t ₂ -t ₁ ≤ T, the rectangular frame corresponding to the position of the human eye in the third frame of image is marked as R ₁ , and the current frame (fourth The rectangular frame corresponding to the position of the human eye contained in the frame image is denoted as R ₂ . Then calculate the overlap ratio r=R ₂ /R ₁ . If r<D, it is determined that the current image has motion blur. At this time, let n=0, record the current image (the fourth frame of image) as n=1, and record the time of the current image as t ₁ , and execute step S270. After the status is updated, execute S220-S250 again (that is, for the fifth The frame contains the image of the human eye position, and the frame time determination and consistency determination are performed again, until r≥D.

When the fifth frame image is used as the current frame image, t ₂ -t ₁ ≤ T and r ≥ D, the current frame count is increased by 1, that is, the current frame is marked as n = 2, and step S270 is executed to proceed to the state After updating, perform S220-S250 again, if the sixth frame of image is used as the current frame of image, if t ₂ -t ₁ ≤ T and r ≥ D, then the current frame count is increased by 1, that is, the current frame is denoted as n = 3, otherwise n=0.

S260: Motion blur determination.

If n>M, it is determined that there is no motion blur in the current frame of image. The frame threshold M here is the third set threshold. If M=1, then in step S250, the frame image corresponding to n=2 and n=3 does not have motion blur; if M=2, then In step S250, the frame image corresponding to n=3 does not have motion blur. Referring to FIG. 3, FIG. 3 is an example of the judgment result of partial frame images in the image sequence obtained in the embodiment of this specification. Figure 3 shows seven frame images containing human eyes. Taking M = 1 as an example, the result of determining whether there is motion blur is shown in Figure 3, that is, n = 0, n = 1 (that is, n ≤ M) represents an image with motion blur, n=2, n=3 (ie n>M) represents an image without motion blur. Those skilled in the art can set the frame threshold M according to the actual situation.

As mentioned above, the motion-blurred image recognition method provided by the embodiments of this specification is used to remove images of poor quality caused by motion blur in an image sequence. Compared with the traditional method, the feature extraction or coding of the region of interest on a single image is performed. To determine whether the image has motion blur, the method of this specification is to determine the position information and time information (mainly through frame time determination and consistency determination) of the region of interest in the image sequence (such as the position of the human eye) in the image. Judge whether the image has motion blur, it has better robustness to factors such as illumination and image noise. In addition, the method also has the advantages of simple process, no dependence on complex calculations, and short calculation time.

Based on the same idea, the embodiment of this specification also provides a motion blur image recognition device. FIG. 4 is a schematic diagram of the structure of a motion blur image recognition device provided by an embodiment of this specification. 4, the apparatus comprising: an acquisition module 401, acquisition module 401 for acquiring a sequence of images; decision block 402, decision block 402 for acquiring module 401 acquires from the n-th object image P _n starts, its Perform compliance determination for the next target image P _n+1 ; n is a positive integer greater than or equal to 1; and, if the target image P _n+1 does not meet the compliance, it is determined that the target image P _n+1 has motion blur; if object image P _{n + 1} satisfy the compliance, the state of the object image P _{n + 1} is updated according to the state of the target image P _n, and the determination of the target image P _{n + 1} in accordance with the state of the object image P _{n + 1} Whether there is motion blur; wherein, the target image is an image containing a target area in the image sequence, and the nth target image P _n is an initial state.

Further, the decision block 402 for starting the n-th object image P _n, for compliance to its next determination target image P _{n + 1,} comprises: comparing the target n-th image P _n to its next target image P _{n The time interval t between +1} ; if t>the first set threshold, it is determined that the target image P _n+1 does not meet the compliance.

Further, the judging module 402 is configured to start from the nth target image P _n and make a compliance judgment for the next target image P _n+1 , and specifically includes: if t ≤ the first set threshold, compare the nth target image P n The consistency r between the target image P _n and its next target image P _n+1 ; if r<the second set threshold, it is determined that the target image P _n+1 does not meet the compliance; if r≥the second set Threshold, it is determined that the target image P _n+1 satisfies compliance.

Further, the determination module calculates r according to the following formula: r=R _n+1 /R _n ; where R _n+1 is the rectangular frame data corresponding to the target area in the target image P _n+1 _{; R n} is the target image P _n The rectangular frame data corresponding to the target area in.

Further, as shown in FIG. 4, the device further includes a detection module 400, which is used to determine the nth target image P _n and the next target image P _{n+1 in} the image sequence; wherein, the The target area contained in the target image is the position of the human eye.

Further, if the target image P _{n + 1} satisfy the compliance, the state of the image P _{n + 1} after updating the target object image represented by P _{n + 1} is m times continuously satisfied compliance target image.

Further, decision block 402 determines that the object image P _{n + 1} according to whether there is motion blur state of the object image P _{n + 1,} specifically comprising: if m> a third set threshold, it is determined that the object image P _{n + 1} is not There is motion blur; otherwise, it is determined that the target image P _n+1 has motion blur.

For the specific implementation of the device, please refer to the above description of the method, which will not be repeated here.

Based on the same idea, an embodiment of this specification also provides an electronic device, including: at least one processor and a memory, the memory stores a program and is configured to execute the motion blur of this specification by the at least one processor Image recognition method.

Based on the same idea, the embodiments of this specification also provide a computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, the movement of this specification is realized. Recognition method of blurred images.

The embodiment of this specification also provides a payment device. Referring to FIG. 5, FIG. 5 is a schematic structural diagram of the payment device provided by the embodiment of the specification. As shown in FIG. Face video data; image screening unit 502, used to screen out images to be recognized without motion blur according to the aforementioned motion blur image recognition method; image recognition unit 503, used to recognize the screened images to be recognized; payment unit 504 , Used to determine whether to perform a payment operation according to the recognition result of the image recognition unit 503.

In the embodiment of this specification, the payment device may be a face-swiping payment device, specifically a multi-modal identification method based on the face and the iris of the human eye to realize the payment operation. The image screening unit 502 of the payment device uses the aforementioned motion-blur image recognition method to obtain images without motion blur from the collected multi-frame face images to perform face and eye iris recognition, and then determine whether to proceed according to the recognition results Payment operation. As an example, if multiple images to be identified are filtered out, one of the images to be identified can be selected for identification, which can be specifically determined by the configuration of the payment device, which is not limited in this application.

For the specific implementation of how to filter the motion-blurred images in the payment device, please refer to the above description of the motion-blurred image recognition method, which will not be repeated here.

This specification also provides a method for recognizing a motion blurred image according to another embodiment. Referring to FIG. 6, FIG. 6 is a main flow chart of a method for recognizing a motion blurred image according to another embodiment of this specification. The method for recognizing a motion blur image provided by this embodiment includes: S602: acquiring an image sequence; S604: starting from the nth target image P _n in the image sequence, conforming to the next target image P _n+1 Judgment; n is a positive integer greater than or equal to 1; S606: If the target image P _n+1 meets compliance, it is determined that the target image P _n+1 has motion blur; S608: If the target image P _n+1 does not meet the compliance, Then it is determined that there is no motion blur in the _{target image P n+1.}

Wherein, the target image is an image containing a target area in the image sequence.

The foregoing steps S602-S606 are the same as steps S102-S106. For specific embodiments of steps S602-S606, please refer to the above description of steps S102-S106, which will not be repeated here. The difference between this embodiment and the foregoing embodiment is that in step S608 of this embodiment, if the target image P _n+1 does not meet the compliance, it is determined that the target image P _n+1 does not have motion blur. In other words, in this embodiment, only be determined from the result of the target image P _{n + 1} for compliance determination target image P _{n + 1} if there is a motion blur, further simplifying the decision process more quickly identify the image sequence The presence of motion-blurred images.

This specification also provides another embodiment of a motion blur image recognition device. Referring to FIG. 7, FIG. 7 is a schematic structural diagram of a motion blur image device provided by another embodiment of this specification. As shown in FIG. 7, the apparatus of this embodiment includes: an obtaining unit 701, which is used to obtain an image sequence; The image P _n starts, and the next target image P _n+1 is determined for compliance; n is a positive integer greater than or equal to 1; and, if the target image P _n+1 does not meet the compliance, then the target image P is determined _n+1 has motion blur; otherwise, it is determined that the target image P _n+1 does not have motion blur; wherein, the target image is an image that includes a target area in the image sequence.

For the specific embodiment of the device, please refer to the description of the method above, which will not be repeated here.

The specific embodiments of this specification have been described above, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than in the embodiments and still achieve desired results. In addition, the processes depicted in the drawings do not necessarily have to be in the specific order or sequential order shown in order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer-readable storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and for related parts, please refer to the part of the description of the method embodiments.

The apparatus, equipment, non-volatile computer readable storage medium, and method provided in the embodiments of this specification correspond to each other. Therefore, the apparatus, equipment, and non-volatile computer storage medium also have beneficial technical effects similar to the corresponding method. The beneficial technical effects of the method have been described in detail above, therefore, the beneficial technical effects of the corresponding device, equipment, and non-volatile computer storage medium will not be repeated here.

In the 1990s, the improvement of a technology can be clearly distinguished between hardware improvements (for example, improvements in circuit structures such as diodes, transistors, switches, etc.) and software improvements (improvements in method flow). However, with the development of technology, the improvement of many methods and processes of today can be regarded as a direct improvement of the hardware circuit structure. Designers almost always get the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that the improvement of a method flow cannot be realized by the hardware entity module. For example, a programmable logic device (Programmable Logic Device, PLD) (such as a Field Programmable Gate Array (Field Programmable Gate Array, FPGA)) is such an integrated circuit whose logic function is determined by the user's programming of the device. It is programmed by the designer to "integrate" a digital system on a piece of PLD, without having to ask the chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly realized by using "logic compiler" software, which is similar to the software compiler used in program development and writing, but before compilation The original code must also be written in a specific programming language, which is called Hardware Description Language (HDL), and there is not only one type of HDL, but many types, such as ABEL (Advanced Boolean Expression Language) , AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description), etc., currently most commonly used It is VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. It should also be clear to those skilled in the art that just a little bit of logic programming of the method flow in the above-mentioned hardware description languages and programming into an integrated circuit can easily obtain the hardware circuit that implements the logic method flow.

The controller can be implemented in any suitable manner. For example, the controller can take the form of, for example, a microprocessor or a processor and a computer-readable medium storing computer-readable program codes (such as software or firmware) executable by the (micro)processor. , Logic gates, switches, application specific integrated circuits (ASICs), programmable logic controllers and embedded microcontrollers. Examples of controllers include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicon Labs C8051F320, the memory controller can also be implemented as part of the memory control logic. Those skilled in the art also know that, in addition to implementing the controller in a purely computer-readable program code manner, it is entirely possible to program the method steps to make the controller use logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded logic. The same function can be realized in the form of a microcontroller or the like. Therefore, such a controller can be regarded as a hardware component, and the devices included in it for realizing various functions can also be regarded as a structure within the hardware component. Or even, the device for realizing various functions can be regarded as both a software module for realizing the method and a structure within a hardware component.

The systems, devices, modules, or units explained in the foregoing embodiments may be specifically implemented by computer chips or entities, or implemented by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cell phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or Any combination of these devices.

For the convenience of description, when describing the above device, the functions are divided into various units and described separately. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and/or hardware.

Those skilled in the art should understand that the embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, the embodiments of this specification may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of this specification may adopt the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

This specification is described with reference to flowcharts and/or block diagrams of methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and/or block in the flowchart and/or block diagram, and the combination of processes and/or blocks in the flowchart and/or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing equipment to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing equipment are used to generate It is a device that realizes the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing equipment to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction device. The device implements the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

These computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operation steps are executed on the computer or other programmable equipment to produce computer-implemented processing, so as to execute on the computer or other programmable equipment. The instructions provide steps for implementing functions specified in a flow or multiple flows in the flowchart and/or a block or multiple blocks in the block diagram.

In a typical configuration, the computing device includes one or more processors (CPU), input/output interfaces, network interfaces, and memory.

The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and/or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). Memory is an example of computer readable media.

Computer-readable media include permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical storage, Magnetic cartridges, magnetic tape storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices. According to the definition in this article, computer-readable media does not include transitory media, such as modulated data signals and carrier waves.

It should also be noted that the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or equipment including a series of elements includes not only those elements, but also Other elements that are not explicitly listed, or include elements inherent to such processes, methods, commodities, or equipment. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, commodity or equipment that includes the element.

This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification can also be practiced in distributed computing environments. In these distributed computing environments, tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, as for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for related parts, please refer to the part of the description of the method embodiment.

The above descriptions are only examples of this specification, and are not intended to limit this application. For those skilled in the art, this application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included in the scope of the claims of this application.

Claims

A method for recognizing motion blurred images, the method includes:

Get image sequence;

Starting from the n-th target image P n in the image sequence, perform compliance judgment on the next target image P n+1 ; n is a positive integer greater than or equal to 1;

If the target image P n+1 does not meet the compliance, it is determined that the target image P n+1 has motion blur;

If the target image P n + 1 satisfy the compliance, the state of the object image P n + 1 is updated according to the state of the target image P n, and the determination target image according to a state of the object image P n + 1 P n+1 whether there is motion blur;

Wherein, the target image is an image that includes a target area in the image sequence, and the nth target image P n is an initial state.
According to the method of claim 1, starting from the nth target image P n , the compliance determination for the next target image P n+1 specifically includes:

Compare the time interval t between the nth target image P n and the next target image P n+1;

If t>the first set threshold, it is determined that the target image P n+1 does not satisfy the compliance.
According to the method of claim 2,

If t ≤ the first set threshold, compare the consistency r between the nth target image P n and the next target image P n+1;

If r<the second set threshold, it is determined that the target image P n+1 does not meet the compliance; if r ≥ the second set threshold, it is determined that the target image P n+1 meets the compliance.
According to the method of claim 3 , the consistency r between the n-th target image P n and the next target image P n+1 is calculated according to the following formula:

r=R n+1 /R n ;

Among them, R n+1 is the rectangular frame data corresponding to the target area in the target image P n+1 ; R n is the rectangular frame data corresponding to the target area in the target image P n.
The method according to claim 1, after acquiring the image sequence, and starting from the n-th target image P n in the image sequence, the compliance determination is performed on the next target image P n+1, before, The method also includes:

Determine the n-th target image P n and its next target image P n+1 in the image sequence;

Wherein, the target area included in the target image is the position of the human eye.
1 5 A method according to any one of the claims, if the target image P n + 1 satisfy the compliance, the state of the image P n + 1 after updating the target object image P n + represents an m-th consecutive Meet the target image for compliance.
The method according to claim 6, according to the state of the object image P n + 1 determines the target image P n + 1 if motion blur exists, comprises:

If m>the third set threshold, it is determined that the target image P n+1 does not have motion blur; otherwise, it is determined that the target image P n+1 has motion blur.
A method for recognizing motion blurred images, the method includes:

Get image sequence;

Starting from the n-th target image P n in the image sequence, perform compliance judgment on the next target image P n+1 ; n is a positive integer greater than or equal to 1;

If the target image P n+1 meets the compliance, it is determined that the target image P n+1 has motion blur; otherwise, it is determined that the target image P n+1 does not have motion blur;

Wherein, the target image is an image containing a target area in the image sequence.
A motion blur image recognition device, the device includes:

An acquisition module, the acquisition module is used to acquire an image sequence;

A determination module, the determination module is used to start from the nth target image P n in the image sequence acquired by the acquisition module, and make a compliance determination for the next target image P n+1 ; n is greater than or equal to 1 Positive integer; and,

If the target image P n+1 does not meet the compliance, it is determined that the target image P n+1 has motion blur;

If the target image P n + 1 satisfy the compliance, the state of the object image P n + 1 is updated according to the state of the target image P n, and determines the target image according to a state of the object image P n + 1 P n+1 whether there is motion blur;

Wherein, the target image is an image that includes a target area in the image sequence, and the nth target image P n is an initial state.
The device according to claim 9, wherein the determination module is configured to start from the nth target image P n and make a compliance determination for the next target image P n+1 , which specifically includes:

Compare the time interval t between the nth target image P n and the next target image P n+1;

If t>the first set threshold, it is determined that the target image P n+1 does not satisfy the compliance.
The device according to claim 10, wherein the determination module is configured to start from the nth target image P n and perform a compliance determination for the next target image P n+1 , and specifically further includes:

If t ≤ the first set threshold, compare the consistency r between the nth target image P n and the next target image P n+1;

If r<the second set threshold, it is determined that the target image P n+1 does not meet the compliance; if r ≥ the second set threshold, it is determined that the target image P n+1 meets the compliance.
The device according to claim 11, wherein the determination module calculates r according to the following formula:

r=R n+1 /R n ;

Among them, R n+1 is the rectangular frame data corresponding to the target area in the target image P n+1 ; R n is the rectangular frame data corresponding to the target area in the target image P n.
The device according to claim 9, further comprising:

A detection module, which is used to determine the nth target image P n and its next target image P n+1 in the image sequence;

Wherein, the target area included in the target image is the position of the human eye.
9-1 apparatus according to any of claims 13, if the target image P n + 1 satisfy the compliance, the object image P n + 1 denotes the state after updating the target image P n + m-1 consecutive times Meet the target image for compliance.
The apparatus according to claim 14, the determining module determines that the object image P n + 1 according to whether there is motion blur state of the object image P n + 1, specifically comprising:

If m>the third set threshold, it is determined that the target image P n+1 does not have motion blur; otherwise, it is determined that the target image P n+1 has motion blur.
A motion blur image recognition device, the device includes:

An acquiring unit, the acquiring unit is used to acquire an image sequence;

A judging unit, the judging unit is configured to start from the n-th target image P n in the image sequence acquired by the acquiring unit, and make a compliance judgment for the next target image P n+1 ; n is greater than or equal to 1 Positive integer; and,

If the target image P n+1 does not meet the compliance, it is determined that the target image P n+1 has motion blur; otherwise, it is determined that the target image P n+1 does not have motion blur;

Wherein, the target image is an image containing a target area in the image sequence.
An electronic device, comprising: at least one processor and a memory, the memory stores a program, and is configured to execute the motion blur image recognition according to any one of claims 1 to 7 by the at least one processor Method, or implement the motion-blurred image recognition method of claim 8.
A computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, the motion blur image according to any one of claims 1 to 7 is realized Recognition method, or realize the recognition method of motion blur image according to claim 8.
A payment device including:

The collection unit is used to collect face video data;

An image screening unit, configured to screen out images to be recognized without motion blur according to the method described in any one of claims 1 to 7 or the method shown in claim 8;

The image recognition unit is used to recognize the screened out images to be recognized;

The payment unit is used to determine whether to perform a payment operation according to the recognition result of the image recognition unit.