A method for evaluating image clarity, related device, equipment and storage medium

Abstract

The present application discloses a method for evaluating image clarity, and the application scenarios include at least various types of terminals, such as mobile phones, computers, and vehicle-mounted terminals. It includes: obtaining a first image group from an image sequence, and each frame of the image sequence includes an image corresponding to a target object; extracting feature points from the first image group to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image, and the first feature point and the second feature point have a one-to-one correspondence; determining a first motion speed of the target object according to the first feature point set and the second feature point set; determining a clarity evaluation result of the second image according to the size relationship between the first motion speed and the blur speed threshold. The present application also provides an apparatus, a device, and a storage medium. The present application can effectively define image clarity, thereby helping to screen out images with higher clarity for subsequent processing.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an image clarity evaluation method, related devices, equipment and storage medium. Background Art

[0002] With the continuous development of image processing technology, its application is becoming more and more common in people's daily life, and the requirements for image clarity are gradually increasing. There are many factors that may cause image blur during the acquisition, compression, transmission and storage of images and videos. Image blur is an important factor in image degradation, so it is necessary to identify the clarity of the image.

[0003] At present, there is a grayscale variance algorithm for detecting image clarity. This algorithm uses the grayscale average value of all pixels in the image as a reference, calculates the difference of the grayscale value of each pixel and then calculates the square sum. Then it is normalized using the total number of pixels. It represents the average degree of grayscale change in the image. The greater the average degree of grayscale change, the clearer the image.

[0004] The inventors have found that the existing solutions have at least the following problems: although the existing detection methods can determine the clarity of images to a certain extent, they still have limitations. For example, in the palm print image recognition process, the existing solutions are difficult to extract images with higher texture clarity from multiple frames of continuous images for recognition. For this, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiment of the present application provides an evaluation method, related device, equipment and storage medium for image clarity. The present application can effectively define image clarity, thereby helping to screen out images with higher clarity for subsequent processing.

[0006] In view of this, the present application provides, on one hand, a method for evaluating image clarity, comprising:

[0007] Acquire a first image group from an image sequence, wherein the image sequence is acquired by a target camera, each frame of the image sequence includes an image corresponding to the target object, and the first image group includes a first image and a second image, wherein the second image is a subsequent frame of the image adjacent to the first image;

[0008] Extracting feature points from the first image group to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image, wherein the first feature points included in the first feature point set and the second feature points included in the second feature point set have a one-to-one correspondence;

[0009] Determining a first motion speed of the target object according to the first feature point set and the second feature point set;

[0010] The clarity evaluation result of the second image is determined according to the magnitude relationship between the first motion speed and the blur speed threshold.

[0011] Another aspect of the present application provides an image clarity evaluation device, comprising:

[0012] An acquisition module is used to acquire a first image group from an image sequence, wherein the image sequence is acquired by a target camera, each frame of the image sequence includes an image corresponding to the target object, and the first image group includes a first image and a second image, wherein the second image is a subsequent frame of the image adjacent to the first image;

[0013] A processing module, configured to extract feature points from the first image group to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image, wherein the first feature points included in the first feature point set and the second feature points included in the second feature point set have a one-to-one correspondence;

[0014] A determination module, configured to determine a first motion speed of the target object according to the first feature point set and the second feature point set;

[0015] The determination module is further used to determine the clarity evaluation result of the second image according to the magnitude relationship between the first motion speed and the blur speed threshold.

[0016] In one possible design, in another implementation of another aspect of the embodiment of the present application,

[0017] A processing module, specifically used to extract feature points from the first image to obtain a first feature point set for the target object, wherein the first feature point set includes K first feature points, where K is an integer greater than 1;

[0018] Feature points are extracted from the second image to obtain a second feature point set for the target object, wherein the second feature point set includes K second feature points.

[0019] In one possible design, in another implementation of another aspect of the embodiment of the present application,

[0020] A determination module, specifically used to obtain a distance value corresponding to each first feature point in the first feature point set, and to obtain a distance value corresponding to each second feature point in the second feature point set, wherein the distance value represents a distance between the feature point and a camera plane corresponding to the target camera;

[0021] Obtaining the two-dimensional coordinates corresponding to each first feature point in the first feature point set, and obtaining the two-dimensional coordinates corresponding to each second feature point in the second feature point set, wherein the two-dimensional coordinates include the horizontal coordinate and the vertical coordinate of the feature point in the pixel coordinate system;

[0022] Determine an average distance of feature points according to the distance value corresponding to each first feature point, the distance value corresponding to each second feature point, the two-dimensional coordinates corresponding to each first feature point, and the two-dimensional coordinates corresponding to each second feature point;

[0023] A first motion speed of the target object is determined according to an average distance between feature points and an inter-frame duration.

[0024] In one possible design, in another implementation of another aspect of the embodiment of the present application,

[0025] A determination module, specifically configured to calculate, for each first feature point in the first feature point set, the abscissa of the first feature point in the spatial coordinate system according to the distance value corresponding to the first feature point, the abscissa of the first feature point in the pixel coordinate system, the abscissa of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0026] For each first feature point in the first feature point set, the vertical coordinate of the first feature point in the spatial coordinate system is calculated according to the distance value corresponding to the first feature point, the vertical coordinate of the first feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0027] For each second feature point in the second feature point set, the abscissa of the second feature point in the spatial coordinate system is calculated according to the distance value corresponding to the second feature point, the abscissa of the second feature point in the pixel coordinate system, the abscissa of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0028] For each second feature point in the second feature point set, the vertical coordinate of the second feature point in the spatial coordinate system is calculated according to the distance value corresponding to the second feature point, the vertical coordinate of the second feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0029] The average distance of the feature points is determined according to the horizontal coordinate of each first feature point in the spatial coordinate system, the vertical coordinate of each first feature point in the spatial coordinate system, the horizontal coordinate of each second feature point in the spatial coordinate system, the vertical coordinate of each second feature point in the spatial coordinate system, the distance value corresponding to each first feature point, and the distance value corresponding to each second feature point.

[0030] In one possible design, in another implementation of another aspect of the embodiment of the present application, the first feature point set includes K first feature points, and the second feature point set includes K second feature points, where K is an integer greater than 1;

[0031] A determination module, specifically configured to construct K groups of feature point pairs according to the first feature point set and the second feature point set, wherein each group of feature point pairs includes a first feature point and a second feature point having a corresponding relationship;

[0032] For each pair of feature point pairs in the K groups of feature point pairs, the feature point distance corresponding to the feature point pair is calculated according to the abscissa of the first feature point in the spatial coordinate system, the abscissa of the second feature point in the spatial coordinate system, the ordinate of the first feature point in the spatial coordinate system, the ordinate of the second feature point in the spatial coordinate system, the distance value corresponding to the first feature point, and the distance value corresponding to the second feature point in the feature point set;

[0033] The feature point distances corresponding to each feature point pair in the K groups of feature point pairs are averaged to obtain the average feature point distance.

[0034] In one possible design, in another implementation of another aspect of the embodiment of the present application, the first feature point set includes K first feature points, and the second feature point set includes K second feature points, where K is an integer greater than 1;

[0035] A determination module, specifically configured to construct K groups of feature point pairs according to the first feature point set and the second feature point set, wherein each group of feature point pairs includes a first feature point and a second feature point having a corresponding relationship;

[0036] For each pair of feature point pairs in the K groups of feature point pairs, the feature point distance corresponding to the feature point pair is calculated according to the abscissa of the first feature point in the spatial coordinate system, the abscissa of the second feature point in the spatial coordinate system, the ordinate of the first feature point in the spatial coordinate system, the ordinate of the second feature point in the spatial coordinate system, the distance value corresponding to the first feature point, and the distance value corresponding to the second feature point in the feature point set;

[0037] The feature point distances corresponding to each feature point pair in the K groups of feature point pairs are sorted in descending order to obtain a feature point distance sequence, wherein the feature point distance sequence includes K feature point distances;

[0038] Select P feature point distances from the feature point distance sequence, where P is an integer greater than or equal to 1 and less than K;

[0039] The distances of P feature points are averaged to obtain the average distance of the feature points.

[0040] In one possible design, in another implementation of another aspect of the embodiment of the present application,

[0041] A determination module, specifically configured to, if K is an odd number, take the [(K+1) / 2]th feature point distance, the Q feature point distances before the [(K+1) / 2]th feature point distance, and the Q feature point distances after the [(K+1) / 2]th feature point distance in the feature point distance sequence as P feature point distances, where Q is an integer greater than or equal to 1;

[0042] If K is an even number, the (K / 2)th feature point distance in the feature point distance sequence, the (K / 2+1)th feature point distance in the feature point distance sequence, the Q feature point distances before the (K / 2)th feature point distance, and the Q feature point distances after the (K / 2+1)th feature point distance are collectively taken as P feature point distances.

[0043] In one possible design, in another implementation of another aspect of the embodiment of the present application, the image clarity evaluation device further includes a collection module, a recognition module, and an adjustment module;

[0044] A collection module, configured to collect a first image sequence through a target camera when the target fixture moves at a first speed, wherein the first image sequence includes at least one image;

[0045] A recognition module, used for recognizing images in a first image sequence to obtain a first accuracy rate;

[0046] an adjustment module, configured to adjust the first speed movement to a second speed movement if the first accuracy rate is greater than or equal to an accuracy rate threshold, wherein the second speed movement is greater than the first movement speed;

[0047] The acquisition module is further used to acquire a second image sequence through the target camera when the target fixture moves at a second speed, wherein the second image sequence includes at least one image;

[0048] The recognition module is further used to recognize images in the second image sequence to obtain a second accuracy rate;

[0049] The determination module is further configured to determine the first speed motion as a blur speed threshold if the second accuracy rate is less than the accuracy rate threshold.

[0050] In one possible design, in another implementation of another aspect of the embodiment of the present application,

[0051] The target camera uses global exposure mode;

[0052] The maximum exposure time of the target camera is less than or equal to 5 milliseconds;

[0053] The lens transmittance of the target camera is greater than or equal to 80%;

[0054] The unified glare value of the fill light provided by the target camera is less than or equal to level 22.

[0055] In one possible design, in another implementation of another aspect of the embodiment of the present application,

[0056] a determination module, specifically configured to determine that the clarity evaluation result of the second image is an unsharp image if the first motion speed is greater than or equal to a blur speed threshold;

[0057] If the first motion speed is less than the blur speed threshold, it is determined that the clarity evaluation result of the second image is a clear image.

[0058] In one possible design, in another implementation of another aspect of the embodiment of the present application,

[0059] The acquisition module is further used to acquire a second image group from the image sequence after determining that the clarity evaluation result of the second image is an unsharp image, wherein the second image group includes the third image and the fourth image, and the fourth image is a subsequent frame image adjacent to the third image;

[0060] The processing module is further used to extract feature points from the second image group to obtain a third feature point set corresponding to the third image and a fourth feature point set corresponding to the fourth image, wherein the third feature points included in the third feature point set and the fourth feature points included in the fourth feature point set have a one-to-one correspondence;

[0061] The determination module is further used to determine a second motion speed of the target object according to the third feature point set and the fourth feature point set;

[0062] The determination module is further used to determine the clarity evaluation result of the fourth image according to the magnitude relationship between the second motion speed and the blur speed threshold.

[0063] In one possible design, in another implementation of another aspect of the embodiment of the present application, the image clarity evaluation device further includes a starting module and a detection module;

[0064] A start module, used for starting a target camera in response to a trigger instruction for an image acquisition control before acquiring a first image group from an image sequence;

[0065] The acquisition module is further used to acquire an original image sequence through a target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1;

[0066] A detection module is used to perform target area detection on the original image sequence to obtain M frames of continuous images including the target object, and use the M frames of continuous images as an image sequence, wherein the target object is a palm, and M is an integer greater than 1 and less than or equal to N;

[0067] The recognition module is further configured to, after determining that the clarity evaluation result of the second image is a clear image, perform a similarity comparison between the second image and T recorded images in the database to obtain T similarity results, wherein each recorded image is a palm image pre-recorded into the database, and T is an integer greater than or equal to 1;

[0068] The determination module is also used to determine the palmprint recognition result according to the T similarity result.

[0069] In one possible design, in another implementation of another aspect of the embodiment of the present application, the image clarity evaluation device further includes a storage module;

[0070] The start module is further used to start the target camera in response to a trigger instruction for the image entry control before acquiring the first image group from the image sequence;

[0071] The acquisition module is further used to acquire an original image sequence through a target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1;

[0072] The detection module is further used to perform target area detection on the original image sequence to obtain M frames of continuous images including the target object, and use the M frames of continuous images as the image sequence, wherein the target object is a palm, and M is an integer greater than 1 and less than or equal to N;

[0073] The storage module is used to store the second image in a database after determining that the clarity evaluation result of the second image is a clear image.

[0074] In one possible design, in another implementation of another aspect of the embodiment of the present application,

[0075] The acquisition module is further used to acquire an original image sequence through a target camera before acquiring the first image group from the image sequence, wherein the original image sequence includes N frames of images, where N is an integer greater than 1;

[0076] The detection module is further used to perform target area detection on the original image sequence, obtain M frames of continuous images including the target object, and use the M frames of continuous images as an image sequence, wherein the target object is a license plate, and M is an integer greater than 1 and less than or equal to N;

[0077] The recognition module is further used to perform optical character recognition (OCR) on the second image to obtain the target text;

[0078] The determination module is also used to match the target text with T input texts in the database to obtain a license plate recognition result, wherein each input text is license plate information pre-entered into the database, and T is an integer greater than or equal to 1.

[0079] On the other hand, the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned methods when executing the computer program.

[0080] Another aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the above-mentioned methods are implemented.

[0081] Another aspect of the present application provides a computer program product, including a computer program, which implements the above-mentioned methods when executed by a processor.

[0082] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0083] In an embodiment of the present application, a method for evaluating image clarity is provided. First, an image sequence containing a target object is captured by a target camera, thereby obtaining a first image group from the image sequence. The first image group includes a first image and a second image, and the second image is a subsequent frame image adjacent to the first image. Based on this, feature points are extracted from the images in the first image group respectively to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image. Therefore, the first motion speed of the target object can be determined based on the first feature point set and the second feature point set, and finally, the clarity evaluation result of the second image is determined based on the size relationship between the first motion speed and the blur speed threshold. In the above manner, the motion speed of the target object in the image can be detected using the previous and next two frames of images. Based on the size relationship between the motion speed of the target object and the blur speed threshold, the image clarity can be effectively defined, thereby facilitating the selection of images with higher clarity from the image sequence for subsequent processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 A schematic diagram of the architecture of an image processing system in an embodiment of the present application;

[0085] Figure 2 A schematic diagram of a pinhole imaging model in an embodiment of the present application;

[0086] Figure 3 A schematic diagram of a flow chart of an image clarity evaluation method in an embodiment of the present application;

[0087] Figure 4 A schematic diagram of a flow chart for determining the clarity of a target object in an embodiment of the present application;

[0088] Figure 5A schematic diagram of a flow chart for calculating the moving speed of a target object in an embodiment of the present application;

[0089] Figure 6 A schematic diagram of extracting a plurality of sets of feature point pairs based on two adjacent frames of images in an embodiment of the present application;

[0090] Figure 7 A schematic diagram of selecting distances of P feature points in an embodiment of the present application;

[0091] Figure 8 A schematic diagram of a flow chart for determining a fuzzy speed threshold in an embodiment of the present application;

[0092] Fig. 9 This is a schematic diagram showing the effect of comparing the exposure mode and the exposure time in the embodiment of the present application;

[0093] Fig.10 A schematic diagram of implementing palm recognition in an embodiment of the present application;

[0094] Fig.11 A schematic diagram of implementing palm input in an embodiment of the present application;

[0095] Fig.12 A schematic diagram of realizing license plate recognition in an embodiment of the present application;

[0096] Fig.13 A schematic diagram of an image clarity evaluation device in an embodiment of the present application;

[0097] Fig.14 A schematic diagram of the structure of a terminal device in an embodiment of the present application;

[0098] Fig.15 A structural diagram of a server in an embodiment of the present application. DETAILED DESCRIPTION

[0099] The embodiment of the present application provides an evaluation method, related device, equipment and storage medium for image clarity. The present application can effectively define image clarity, thereby helping to screen out images with higher clarity for subsequent processing.

[0100] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "corresponding to" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0101] Images and videos usually contain a large amount of visual information, and visual information itself has intuitive and efficient description capabilities. With the rapid development of information technology, the application of images and videos has gradually spread to all fields of human society. In the fields of computer image processing, computer vision and machine learning, there are usually certain requirements for image clarity. The clarity here mainly refers to the clarity of the detailed patterns and their boundaries on the image.

[0102] In order to screen out images that meet the definition requirements, this application provides an image definition evaluation method, which is applied to Figure 1 The image processing system shown in the figure, as shown in the figure, the image processing system system includes a terminal device 110, or, the image processing system system includes a terminal device 110 and a server 120. Among them, the client can be run on the terminal device 110 in the form of a browser, or it can be run on the terminal device 110 in the form of an independent application (application, APP), etc., and the specific presentation form of the client is not limited here. The server 120 involved in this application can be an independent physical server, or a server 120 cluster or distributed system composed of multiple physical servers 120, or it can be a cloud server 120 that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms. The terminal device 110 can be a smart phone, a tablet computer, a laptop, a PDA, a personal computer, a smart TV, a smart watch, a car device, a wearable device, etc., but is not limited to this. The terminal device 110 and the server 120 may be connected directly or indirectly via wired or wireless communication, which is not limited in this application. The number of servers 120 and terminal devices 110 is also not limited.

[0103] Exemplarily, the solution provided by the present application can be completed independently by a terminal device. First, an image sequence containing a target object (e.g., a palm) is collected by the terminal device, and then two consecutive frames of images are extracted from the image sequence. Thus, the terminal device can extract feature points of the two frames of images respectively, and then calculate the motion speed of the target object based on the extracted feature points. The terminal device can determine the clarity of the next frame of image based on the relationship between the motion speed and the blur speed threshold. If the clarity meets the standard, the terminal device can use the clear image for subsequent processing.

[0104] Exemplarily, the solution provided by the present application can be completed by the cooperation of a terminal device and a server. First, an image sequence containing a target object (for example, a palm) is collected by a terminal device, and then the image sequence is sent to a server, which extracts two consecutive frames of images from the image sequence. Thus, the server can extract feature points from the two frames of images respectively, and then calculate the motion speed of the target object based on the extracted feature points. The server can determine the clarity of the next frame of image based on the relationship between the motion speed and the blur speed threshold. If the clarity meets the standard, the server can use the clear image for subsequent processing.

[0105] Since this application involves some terms related to professional fields, they will be explained below for ease of understanding.

[0106] (1) Palm area detection: This refers to using object detection technology to locate the finger gaps in the image, thereby extracting the image of the palm area from the image.

[0107] (2) Palmprint recognition: Identify the identity information of different objects based on the image of the palmprint area.

[0108] (3) Cross-device registration and identification: Register and identify on two devices with significant differences. For example, register on a mobile phone and identify through an IoT device.

[0109] (4) Color image: A color image captured by a color sensor using natural light. It is usually used for face or palm optimization and comparative recognition in face or palm payment.

[0110] (5) Infrared image: An infrared image captured by an infrared sensor using infrared light imaging. It is usually used for liveness detection in face scanning or palm payment.

[0111] (6) refers to selecting a set of color images, depth images, and infrared images that meet the preconditions of the liveness detection and contrast recognition algorithms. Among them, palm scanning requires the selection of color images and infrared images. Face scanning requires the selection of color images, infrared images, and depth images.

[0112] For example, after the selection is passed, the color image can be selected based on the palm angle, size, centering and clarity of the color image, and the infrared image can be selected based on the brightness of the infrared image.

[0113] Exemplarily, after the optimization is passed, the color image can be optimized based on the palm angle, size, centering and clarity of the color image, the infrared image can be optimized based on the brightness of the infrared image, and the depth image can be optimized based on the completeness of the depth image.

[0114] (7) Preferred frame: a set of color images and infrared images that are preferred for palm scanning. Or, a set of color images, depth images, and infrared images that are preferred for face scanning. Preferred payment frames can be used for liveness detection and comparative identification.

[0115] (8) Proximity Sensor (PSensor): A sensor that detects the presence of a nearby object without any physical contact. A proximity sensor typically emits an electromagnetic field or a beam of electromagnetic radiation (e.g., infrared) and looks for changes in the field or return signal. The object being sensed is often referred to as the target of the proximity sensor.

[0116] (9) Camera intrinsic parameters: Parameters used to describe the conversion relationship between the three-dimensional coordinates of a real-world object when it is imaged on a camera sensor and the pixel coordinates after imaging. This conversion relationship can be expressed as follows: Figure 2 The pinhole imaging model shown in the figure is represented. As shown in the figure, in the imaging plane, O' is the origin and the coordinate axis x' and the coordinate axis y' form the pixel coordinate system O'-x'-y'. Point P represents a point in three-dimensional space, and point P' represents the point projected onto the imaging plane.

[0117] Based on the principle of similar triangles, the following formula can be obtained:

[0118] Z / f=L' / L;

[0119] Where Z represents the actual distance between the target and the camera, f represents the focal length (i.e., the distance between the focus and the imaging plane), L' represents the pixel width, and L represents the actual object width.

[0120] (10) Principal point: represents the projection point or intersection point of the camera’s optical axis in the imaging plane, which can be obtained through camera calibration.

[0121] (11) Camera coordinate system: combined Figure 2 The camera coordinate system refers to the pixel coordinate system with the camera optical center O as the coordinate origin, the optical axis as the z-axis, and the x-axis and y-axis parallel to the camera imaging. The three-dimensional coordinates obtained by transforming the depth map and the camera intrinsic parameters are the coordinates in the camera coordinate system Oxyz.

[0122] In combination with the above introduction, the evaluation method of image clarity in this application will be introduced below. Figure 3 The image definition evaluation method in the embodiment of the present application may be executed by a computer device, which may be a terminal device or a server, including:

[0123] 210. Acquire a first image group from an image sequence, wherein the image sequence is acquired by a target camera, each frame of the image sequence includes an image corresponding to the target object, and the first image group includes a first image and a second image, wherein the second image is a subsequent frame of the image adjacent to the first image;

[0124] In one or more embodiments, a target camera collects multiple consecutive frames of images, and then performs target region detection on each frame of image, and detects the multiple consecutive frames of images containing the target object as an image sequence. Based on this, two consecutive frames of images can be extracted from the image sequence as a first image group, wherein the previous frame of image in the first image group is taken as a "first image", and the next frame of image in the first image group is taken as a "second image".

[0125] It should be noted that the target objects involved in this application are not limited to palms, faces, license plates, etc.

[0126] 220. Extract feature points from the first image group to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image, wherein the first feature points included in the first feature point set and the second feature points included in the second feature point set have a one-to-one correspondence;

[0127] In one or more embodiments, feature points are extracted from the first image and the second image, respectively, thereby obtaining a first feature point set corresponding to the first image and a second feature point set corresponding to the second image. The first feature point set includes at least K first feature points, and the second feature point set includes at least K second feature points, wherein there is a corresponding relationship between the first feature points and the second feature points, and therefore, there are K pairs of matching feature points.

[0128] 230. Determine a first motion speed of the target object according to the first feature point set and the second feature point set;

[0129] In one or more embodiments, a first motion speed of the target object is calculated based on the first feature point set and the second feature point set, wherein the first motion speed represents the motion speed for the second image, and both the first feature point set and the second feature point set are extracted from the target object.

[0130] Specifically, the motion speed is calculated based on the comprehensive pixel displacement of the feature points in the two frames of images and the distance between the feature points in the two frames of images and the camera plane, wherein the distance between the camera plane and the feature points can be measured by PSensor.

[0131] 240. Determine a clarity evaluation result of the second image according to a magnitude relationship between the first motion speed and the blur speed threshold.

[0132] In one or more embodiments, by comparing the motion speed and the blur speed threshold, a motion blur-related judgment can be made, thereby achieving preferred recognition of the second image. If the first motion speed is greater than or equal to the blur speed threshold, the clarity evaluation result of the second image is determined to be an unsharp image, that is, the motion speed of the second image is too fast, which easily leads to a low success rate of image recognition. If the first motion speed is less than the blur speed threshold, the clarity evaluation result of the second image is determined to be a clear image, that is, the clear image can be used as a preferred frame for subsequent processing.

[0133] For ease of understanding, take the target object as a palm as an example. Figure 4 , Figure 4 This is a flow chart of determining the clarity of a target object in an embodiment of the present application, as shown in the figure, specifically:

[0134] In step A1, the blur speed threshold is a value calibrated in a laboratory environment. During the research and development process, the corresponding relationship between motion speed and image clarity (or image blur) can be tested to find the motion speed corresponding to the minimum required clarity for image recognition, and it can be used as the blur speed threshold in subsequent use.

[0135] In step A2, under the usage environment, the movement speed of the palm can be calculated by using the feature points extracted from two adjacent frames of images.

[0136] In step A3, under the usage environment, the movement speed of the palm is compared with the blur speed threshold. If the movement speed of the palm is greater than or equal to the blur speed threshold, it is determined that the palm motion is blurred.

[0137] In an embodiment of the present application, a method for evaluating image clarity is provided. In the above manner, the motion speed of the target object in the image can be detected using the two frames of images, and the image clarity can be effectively defined based on the size relationship between the motion speed of the target object and the blur speed threshold, thereby helping to screen out images with higher clarity from the image sequence for subsequent processing.

[0138] Optionally, in the above Figure 3On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, feature point extraction is performed on the first image group to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image, which may specifically include:

[0139] Extracting feature points from the first image to obtain a first feature point set for the target object, wherein the first feature point set includes K first feature points, where K is an integer greater than 1;

[0140] Feature points are extracted from the second image to obtain a second feature point set for the target object, wherein the second feature point set includes K second feature points.

[0141] In one or more embodiments, a method for obtaining a feature point set based on a feature extraction method is introduced. As can be seen from the above embodiments, feature points need to be extracted from the first image and the second image respectively. If the number of feature points of the two frames of images is equal, the motion speed of the target object can be calculated based on the two frames of images. If the number of feature points of the two frames of images is not equal, two consecutive frames of images are obtained again from the image sequence.

[0142] The following are three ways to extract feature points:

[0143] 1. Scale-invariant feature transform (SIFT);

[0144] Extracting feature points based on SIFT mainly includes four steps. The first step is to construct a scale space, detect extreme points, and obtain scale invariance. The second step is to filter feature points and achieve precise positioning. The third step is to assign direction values ​​to feature points. The fourth step is to generate feature descriptors to describe feature points.

[0145] After the feature descriptors of the two frames are generated, the Euclidean distance of the key point feature vectors can be used as a similarity determination metric for the feature points in the two frames. For example, take a feature point in the first image and traverse to find the two closest key points in the second image. Among these two key points, if the closest distance divided by the next closest distance is less than the threshold, they are determined to be a pair of matching points.

[0146] 2. Speeded up robust features (SURF);

[0147] Extracting feature points based on SURF mainly includes four steps. The first step is to detect feature points using the Hessian matrix. The second step is to filter feature points and achieve precise positioning. The third step is to assign direction values ​​to feature points. The fourth step is to generate feature descriptors to describe feature points.

[0148] SURF simplifies some operations in SIFT. SURF simplifies the Gaussian second-order differential in SIFT so that the convolution smoothing operation only needs to be converted into addition and subtraction operations, making the SURF algorithm robust and low in time complexity.

[0149] 3. Features based on FAST and BRIEF (Oriented FAST and Rotated BRIEF, ORB);

[0150] Based on the features from accelerated segment test (FAST) and binary robust independent elementary features (BRIEF), ORB feature points are extended and can be used for real-time feature detection. The feature point extraction based on ORB mainly includes two steps. The first step is directional FAST feature point detection. The second step is BRIEF feature description.

[0151] It should be noted that this application uses ORB features as an example. However, in practical applications, other feature point extraction methods may also be used, and ORB, SIFT, and SURF are all illustrative and should not be construed as limiting this application.

[0152] Secondly, in the embodiment of the present application, a method for obtaining a feature point set based on a feature extraction method is provided. Through the above method, feature points for the target object can be extracted using methods such as ORB, SIFT or SURF, thereby improving the flexibility and feasibility of the solution. In addition, the use of ORB can also improve the efficiency of feature point extraction.

[0153] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, determining the first motion speed of the target object according to the first feature point set and the second feature point set may specifically include:

[0154] Obtaining a distance value corresponding to each first feature point in the first feature point set, and obtaining a distance value corresponding to each second feature point in the second feature point set, wherein the distance value represents a distance between the feature point and a camera plane corresponding to the target camera;

[0155] Obtaining the two-dimensional coordinates corresponding to each first feature point in the first feature point set, and obtaining the two-dimensional coordinates corresponding to each second feature point in the second feature point set, wherein the two-dimensional coordinates include the horizontal coordinate and the vertical coordinate of the feature point in the pixel coordinate system;

[0156] Determine an average distance of feature points according to the distance value corresponding to each first feature point, the distance value corresponding to each second feature point, the two-dimensional coordinates corresponding to each first feature point, and the two-dimensional coordinates corresponding to each second feature point;

[0157] A first motion speed of the target object is determined according to an average distance between feature points and an inter-frame duration.

[0158] In one or more embodiments, a method for calculating the moving speed of a target object is introduced. As can be seen from the above embodiments, the image sequence is collected by a target camera. Therefore, based on the location of the target camera, the following can be constructed: Figure 2 The camera plane shown in the figure is based on this. Using PSensor, the distance from the target object to the camera plane can be measured, that is, the distance from each first feature point to the camera plane and the distance from each second feature point to the camera plane can be obtained respectively. Taking the target object as a palm as an example, the surface of the palm is usually parallel to the surface of the camera lens. Therefore, it can be considered that the distance on the palm surface is basically the same as the vertical distance to the lens. That is, it can be considered that the distance values ​​of each first feature point are equal, and the distance values ​​of each second feature point are equal.

[0159] In addition, based on the pixel coordinate system, the two-dimensional coordinates of each first feature point and the two-dimensional coordinates of each second feature point can also be determined, and then the average distance of the feature points can be calculated based on the two-dimensional coordinates and the distance value. Therefore, the moving speed of the target object can be calculated in the following way:

[0160]

[0161] Where v represents the speed of movement. mean Represents the average distance of feature points. t represents the time interval between the previous and next two frames.

[0162] For example, the average distance of the feature points calculated based on the first feature point set and the second feature point set is 5 centimeters (cm). The time interval between the first image and the second image (i.e., the frame length) is 0.1 seconds (s). Based on formula (1), the first motion speed can be calculated to be 50 cm / s.

[0163] For ease of understanding, take the target object as a palm as an example. Figure 5 , Figure 5This is a flow chart of calculating the moving speed of a target object in an embodiment of the present application, as shown in the figure, specifically:

[0164] In step B1, two consecutive frames of images are buffered.

[0165] In step B2, feature point detection is performed on the palm in the image.

[0166] In step B3, feature point matching is performed based on feature point sets of two adjacent image frames.

[0167] In step B4, the three-dimensional positions of the spatial points corresponding to the matching feature points of two adjacent image frames are calculated.

[0168] In step B5, the distances between all corresponding matching feature points are calculated, thereby obtaining an average distance of the feature points.

[0169] In step B6, the movement speed of the palm is calculated based on the average distance between feature points and the time between frames.

[0170] Secondly, in the embodiment of the present application, a method for calculating the moving speed of the target object is provided. In the above method, the distance between the feature point and the camera plane and the position in the pixel coordinate system can be used to calculate the distance between each pair of feature points, and based on this, the moving speed of the target object can be determined. Thus, the feasibility and operability of the solution are improved.

[0171] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, according to the distance value corresponding to each first feature point, the distance value corresponding to each second feature point, the two-dimensional coordinates corresponding to each first feature point, and the two-dimensional coordinates corresponding to each second feature point, determining the average distance of the feature points may specifically include:

[0172] For each first feature point in the first feature point set, the abscissa of the first feature point in the spatial coordinate system is calculated according to the distance value corresponding to the first feature point, the abscissa of the first feature point in the pixel coordinate system, the abscissa of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0173] For each first feature point in the first feature point set, the vertical coordinate of the first feature point in the spatial coordinate system is calculated according to the distance value corresponding to the first feature point, the vertical coordinate of the first feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0174] For each second feature point in the second feature point set, the abscissa of the second feature point in the spatial coordinate system is calculated according to the distance value corresponding to the second feature point, the abscissa of the second feature point in the pixel coordinate system, the abscissa of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0175] For each second feature point in the second feature point set, the vertical coordinate of the second feature point in the spatial coordinate system is calculated according to the distance value corresponding to the second feature point, the vertical coordinate of the second feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0176] The average distance of the feature points is determined according to the horizontal coordinate of each first feature point in the spatial coordinate system, the vertical coordinate of each first feature point in the spatial coordinate system, the horizontal coordinate of each second feature point in the spatial coordinate system, the vertical coordinate of each second feature point in the spatial coordinate system, the distance value corresponding to each first feature point, and the distance value corresponding to each second feature point.

[0177] In one or more embodiments, a method for converting feature points from a pixel coordinate system to a spatial coordinate system is introduced. It can be seen from the above embodiments that before calculating the average distance of the feature points, each first feature point and each second feature point need to be converted to a spatial coordinate system, that is, the horizontal coordinate and the vertical coordinate of each feature point in the spatial coordinate system are obtained.

[0178] Specifically, the horizontal and vertical coordinates of the feature points in the spatial coordinate system can be calculated as follows:

[0179] X 1i =Z 1i (x 1i -x0) / f; Formula (2)

[0180] X 2i =Z 2i (x 2i -x0) / f; Formula (3)

[0181] Y 1i =Z 1i (y 1i -y0) / f; Formula (4)

[0182] Y 2i =Z 2i (y 2i -y0) / f; Formula (5)

[0183] Among them, x0 represents the horizontal coordinate of the principal point in the pixel coordinate system, y0 represents the vertical coordinate of the principal point in the pixel coordinate system, and f represents the focal length of the target camera in the pixel coordinate system.

[0184] X1i Represents the horizontal coordinate of the i-th first feature point in the spatial coordinate system. 1i Indicates the distance value corresponding to the i-th first feature point. 1i Represents the horizontal coordinate of the i-th first feature point in the pixel coordinate system.

[0185] Y 1i Represents the ordinate of the i-th first feature point in the spatial coordinate system. 1i Indicates the distance value corresponding to the i-th first feature point. 1i Represents the ordinate of the i-th first feature point in the pixel coordinate system.

[0186] X 2i Z represents the horizontal coordinate of the i-th second feature point in the spatial coordinate system. 2i Indicates the distance value corresponding to the i-th second feature point. 2i Represents the horizontal coordinate of the i-th second feature point in the pixel coordinate system.

[0187] Y 2i Represents the ordinate of the i-th second feature point in the spatial coordinate system. 2i Indicates the distance value corresponding to the i-th second feature point. 2i Represents the ordinate of the i-th second feature point in the pixel coordinate system.

[0188] Thus, we get the first feature point (C 1i ) in the space coordinate system is (X 1i ,Y 1i ), the i-th second feature point (C 2i ) in the space coordinate system is (X 2i ,Y 2i ). Based on the coordinates of each first feature point in the spatial coordinate system, the coordinates of each second feature point in the spatial coordinate system, the distance values ​​of each first feature point and the distance values ​​of each second feature point, the average distance of the feature points can be calculated.

[0189] Again, in the embodiment of the present application, a method for converting feature points from a pixel coordinate system to a spatial coordinate system is provided. Through the above method, considering that the real scene is a 3D scene, after converting the feature points corresponding to the target object from 2D positions to 3D positions, a more accurate distance can be calculated, thereby obtaining a motion speed that is more consistent with the actual motion of the target object, which is conducive to improving the reliability of image optimization.

[0190] Optionally, in the above Figure 3On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, the first feature point set includes K first feature points, and the second feature point set includes K second feature points, where K is an integer greater than 1;

[0191] Determining the average distance of feature points according to the abscissa of each first feature point in the spatial coordinate system, the ordinate of each first feature point in the spatial coordinate system, the abscissa of each second feature point in the spatial coordinate system, the ordinate of each second feature point in the spatial coordinate system, the distance value corresponding to each first feature point, and the distance value corresponding to each second feature point may specifically include:

[0192] Constructing K groups of feature point pairs according to the first feature point set and the second feature point set, wherein each group of feature point pairs includes a first feature point and a second feature point having a corresponding relationship;

[0193] For each pair of feature point pairs in the K groups of feature point pairs, the feature point distance corresponding to the feature point pair is calculated according to the abscissa of the first feature point in the spatial coordinate system, the abscissa of the second feature point in the spatial coordinate system, the ordinate of the first feature point in the spatial coordinate system, the ordinate of the second feature point in the spatial coordinate system, the distance value corresponding to the first feature point, and the distance value corresponding to the second feature point in the feature point set;

[0194] The feature point distances corresponding to each feature point pair in the K groups of feature point pairs are averaged to obtain the average feature point distance.

[0195] In one or more embodiments, a method for calculating the average distance of feature points is introduced. From the above embodiments, it can be seen that K first feature points are extracted from the first image, and K second feature points are extracted from the second image, and then a global brute force matching method is used to achieve feature point matching between the previous and next two frames (for example, the first image and the second image).

[0196] Specifically, for ease of understanding, see Figure 6 , Figure 6 This is a schematic diagram of extracting several groups of feature point pairs based on two adjacent frames of images in an embodiment of the present application. As shown in the figure, the left figure is the previous frame of image (for example, the first image), and the right figure is the next frame of image (for example, the second image). It can be seen that there are corresponding feature points between the previous and next two frames of image. The figure shows 10 groups of feature point pairs, and for the sake of convenience, feature points with the same number are regarded as the same group of feature point pairs. Based on this, the i-th group of feature point pairs includes the i-th first feature point (C 1i ) and the i-th second feature point (C 2i), then, the coordinates of each first feature point and each second feature point in the spatial coordinate system can be calculated by using formula (2) to formula (5), and thus, the feature point distance corresponding to the feature point pair can be calculated by the following method:

[0197]

[0198] Among them, L i Indicates the feature point distance corresponding to the i-th set of feature point pairs. 2i represents the horizontal coordinate of the i-th second feature point in the spatial coordinate system. 1i Y represents the horizontal coordinate of the i-th first feature point in the spatial coordinate system. 2i Y represents the ordinate of the i-th second feature point in the spatial coordinate system. 1i Represents the ordinate of the i-th first feature point in the spatial coordinate system. 2i Represents the distance value corresponding to the i-th second feature point. 1i Indicates the distance value corresponding to the i-th first feature point.

[0199] Therefore, the average distance of feature points can be calculated as follows:

[0200]

[0201] Among them, L mean Indicates the average distance of feature points. L i represents the feature point distance corresponding to the i-th feature point pair. K represents the total number of feature point pairs. i represents the i-th feature point pair in the K-th feature point pair.

[0202] Furthermore, in an embodiment of the present application, a method for calculating the average distance of feature points is provided. In the above method, feature points of two frames of images are matched to obtain K groups of feature point pairs. Thus, the corresponding feature point distance is calculated based on each group of feature point pairs, thereby improving the feasibility and operability of the solution.

[0203] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, the first feature point set includes K first feature points, and the second feature point set includes K second feature points, where K is an integer greater than 1;

[0204] Determining the average distance of feature points according to the abscissa of each first feature point in the spatial coordinate system, the ordinate of each first feature point in the spatial coordinate system, the abscissa of each second feature point in the spatial coordinate system, the ordinate of each second feature point in the spatial coordinate system, the distance value corresponding to each first feature point, and the distance value corresponding to each second feature point may specifically include:

[0205] Constructing K groups of feature point pairs according to the first feature point set and the second feature point set, wherein each group of feature point pairs includes a first feature point and a second feature point having a corresponding relationship;

[0206] For each pair of feature point pairs in the K groups of feature point pairs, the feature point distance corresponding to the feature point pair is calculated according to the abscissa of the first feature point in the spatial coordinate system, the abscissa of the second feature point in the spatial coordinate system, the ordinate of the first feature point in the spatial coordinate system, the ordinate of the second feature point in the spatial coordinate system, the distance value corresponding to the first feature point, and the distance value corresponding to the second feature point in the feature point set;

[0207] The feature point distances corresponding to each feature point pair in the K groups of feature point pairs are sorted in descending order to obtain a feature point distance sequence, wherein the feature point distance sequence includes K feature point distances;

[0208] Select P feature point distances from the feature point distance sequence, where P is an integer greater than or equal to 1 and less than K;

[0209] The distances of P feature points are averaged to obtain the average distance of the feature points.

[0210] In one or more embodiments, another method for calculating the average distance of feature points is introduced. From the above embodiments, it can be seen that K first feature points are extracted from the first image, and K second feature points are extracted from the second image, and then a global brute force matching method is used to achieve feature point matching between the previous and next two frames (for example, the first image and the second image).

[0211] Specifically, in combination with the above Figure 6 And the content described in formula (6), the feature point distance corresponding to each feature point pair in the K groups of feature point pairs can be calculated. Based on this, the K feature point distances can be sorted in order from large to small, so as to obtain a feature point distance sequence. Select some feature point distances in the feature point distance sequence for averaging calculation. This application takes the selection of P feature point distances from the feature point distance sequence as an example, and the average feature point distance can be calculated in the following way:

[0212]

[0213] Among them, L mean Indicates the average distance of feature points. L r represents the distance between the feature points corresponding to the rth set of feature point pairs, and L r Calculation and L i The calculation method of is the same. P represents the total number of selected feature point pairs. r represents the rth feature point pair in P groups of feature point pairs.

[0214] It should be noted that the P value is usually proportional to the K value, for example, P = 0.2 K. 0.2 is an empirical value. Usually, when the speed stability requirement is high, the empirical value can be increased. After averaging, the front and rear speeds tend to be smooth and will not change dramatically.

[0215] Furthermore, in an embodiment of the present application, another method for calculating the average distance of feature points is provided. By the above method, selecting some feature point distances for calculation can reduce the amount of calculation, thereby improving the calculation efficiency.

[0216] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, selecting P feature point distances from the feature point distance sequence may specifically include:

[0217] If K is an odd number, the distance of the [(K+1) / 2]th feature point, the distance of the Q feature points before the [(K+1) / 2]th feature point, and the distance of the Q feature points after the [(K+1) / 2]th feature point in the feature point distance sequence are taken together as P feature point distances, where Q is an integer greater than or equal to 1;

[0218] If K is an even number, the (K / 2)th feature point distance in the feature point distance sequence, the (K / 2+1)th feature point distance in the feature point distance sequence, the Q feature point distances before the (K / 2)th feature point distance, and the Q feature point distances after the (K / 2+1)th feature point distance are collectively taken as P feature point distances.

[0219] In one or more embodiments, a method for selecting P feature point distances is introduced. It can be seen from the aforementioned embodiments that P feature point distances can be selected from the feature point distance sequence according to a certain ratio, wherein the P feature point distances here belong to the middle part of the feature point distance sequence. Exemplarily, assuming that the feature point distance sequence includes 100 feature point distances, taking the ratio of 20% as an example, the 41st to 60th feature point distances in the feature point distance sequence package can be used as P feature point distances. The following will be combined with the diagrams to explain the processing methods when the K value is an odd number or an even number.

[0220] 1. K is an odd number;

[0221] Specifically, for ease of understanding, see Figure 7 , Figure 7 A schematic diagram of selecting the distances of P feature points in an embodiment of the present application is shown in FIG. Figure 7As shown in the figure (A), assume that K is 11, that is, the feature point distance sequence includes the feature point distances from No. 1 to No. 11. Assume that P is 5, then Q is 2. Based on this, the [(K+1) / 2]th feature point distance (that is, the feature point distance No. 6) is taken as the center, and Q feature point distances are taken forward and backward respectively, and then taken together with the feature point distance No. 6 as the P feature point distances.

[0222] It can be seen that the P feature point distances include the distances from feature point No. 4 to feature point No. 8.

[0223] 2. K is an even number;

[0224] Specifically, for ease of understanding, see Figure 7 , Figure 7 A schematic diagram of selecting the distances of P feature points in an embodiment of the present application is shown in FIG. Figure 7 As shown in the middle (B), assume that K is 10, that is, the feature point distance sequence includes feature point distances 1 to 10. Assuming P is 4, Q is 1. Based on this, the (K / 2)th feature point distance (that is, feature point distance 5) is used as the left starting point, and Q feature point distances are taken forward. The (K / 2+1)th feature point distance (that is, feature point distance 6) is used as the right starting point, and Q feature point distances are taken backward. Then, together with the feature point distances 5 and 6, they are taken as P feature point distances.

[0225] It can be seen that the P feature point distances include the distance from feature point No. 4 to the distance from feature point No. 7.

[0226] Furthermore, in an embodiment of the present application, a method for selecting P feature point distances is provided. By the above method, selecting some feature point distances for calculation can reduce the amount of calculation, thereby improving the calculation efficiency. At the same time, selecting the feature point distances of the middle section for calculation can calculate a more accurate average distance of the feature points.

[0227] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, another optional embodiment provided by the embodiment of the present application may further include:

[0228] When the target fixture moves at a first speed, a first image sequence is captured by the target camera, wherein the first image sequence includes at least one image;

[0229] recognizing images in a first image sequence to obtain a first accuracy rate;

[0230] If the first accuracy rate is greater than or equal to the accuracy rate threshold, adjusting the first speed movement to a second speed movement, wherein the second speed movement is greater than the first speed movement;

[0231] When the target fixture moves at a second speed, a second image sequence is captured by the target camera, wherein the second image sequence includes at least one image;

[0232] recognizing images in a second image sequence to obtain a second accuracy rate;

[0233] If the second accuracy is less than the accuracy threshold, the first speed motion is determined as the blur speed threshold.

[0234] In one or more embodiments, a method for determining a blur speed threshold is introduced. It can be seen from the aforementioned embodiments that since there is a corresponding relationship between the motion speed and the image clarity, the blur speed threshold can be calibrated in a laboratory environment. Taking the target object as a palm as an example, based on this, it is necessary to prepare a target camera for palm scanning and a target jig (i.e., a prosthetic hand motion jig) in a laboratory environment, wherein the prosthetic hand motion jig has some palm prints on the palm for identification. However, since the target jig is not a human arm, the liveness detection function needs to be turned off during the test.

[0235] Specifically, the target fixture is set to pass through the camera module at a speed of 10 cm / s to 100 cm / s, and the accuracy corresponding to each speed is recorded respectively. Assuming that the first speed is 10 cm / s, when the target fixture moves at the first speed, the first image sequence is collected by the target camera. Based on this, the palm prints in the first image sequence are recognized to obtain the first accuracy. If the first accuracy is greater than or equal to the accuracy threshold (for example, 99.99%), it means that a higher recognition accuracy can be achieved at the first speed. Therefore, the movement speed of the target fixture can be accelerated, that is, it moves at a second speed, and the second image sequence is collected by the target camera. Based on this, the palm prints in the second image sequence are recognized to obtain the second accuracy. If the second accuracy is less than the accuracy threshold (for example, 99.99%), it means that a good recognition effect cannot be achieved at the second speed, so the first speed movement is determined as the fuzzy speed threshold.

[0236] For ease of understanding, please refer to Table 1, which is a diagram showing the corresponding relationship between the target fixture movement speed and the recognition accuracy.

[0237] Table 1

[0238] Movement speed Accuracy 10cm / s 99.99％ 20cm / s 99.99％ 30cm / s 99.99％ 40cm / s 99.99％ 50cm / s 99.99％ 60cm / s 99.97％ 70cm / s 99.96％ 80cm / s 99.94％ 90cm / s 99.91％ 100cm / s 99.86％

[0239] Taking Table 1 as an example, assuming that the accuracy threshold is 99.99%, it can be seen that when the target fixture's movement speed reaches 60cm / s, the qualified accuracy cannot be achieved. Therefore, "50cm / s" can be used as the fuzzy speed threshold. Furthermore, more experimental values ​​can be set based on "50cm / s" and "60cm / s", for example, "51cm / s", "52cm / s", "53cm / s", etc., to find a more accurate fuzzy speed threshold.

[0240] For ease of understanding, take the target object as a palm as an example. Figure 8 , Figure 8 This is a flow chart of determining the fuzzy speed threshold in an embodiment of the present application, as shown in the figure, specifically:

[0241] In step C1, a palm camera and a prosthetic hand motion jig are prepared in a laboratory environment.

[0242] In step C2, the maximum exposure time of the camera may be set to 5 milliseconds.

[0243] In step C3, the liveness detection function is turned off to ensure that the prosthetic hand motion fixture can be recognized and processed.

[0244] In step C4, the prosthetic hand motion fixture moves at a certain interval speed (eg, 10 cm / s) and passes by the camera module.

[0245] In step C5, the camera module performs palm swiping recognition in various motion states and records the recognition results.

[0246] In step C6, the highest speed at which the recognition pass rate reaches a pass rate threshold (eg, 99.99%) is used as the fuzzy speed threshold.

[0247] Secondly, in the embodiment of the present application, a method for determining a blur speed threshold is provided. Through the above method, the motion speed that meets the minimum clarity requirement can be tested using the experimental environment, and the motion speed is used as the blur speed threshold. Compared with directly setting the empirical value as the blur speed threshold, the blur speed threshold calibrated based on the test results has better adaptability and better accuracy.

[0248] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, the target camera adopts a global exposure mode;

[0249] The maximum exposure time of the target camera is less than or equal to 5 milliseconds;

[0250] The lens transmittance of the target camera is greater than or equal to 80%;

[0251] The unified glare value of the fill light provided by the target camera is less than or equal to level 22.

[0252] In one or more embodiments, parameters related to the target camera are introduced. As can be seen from the above embodiments, the camera parameters will also affect the quality of the collected image. The ability of the camera system to combat motion blur is related to the exposure method, exposure time, lens hardware parameters and fill light quality, which will be explained below.

[0253] 1. Exposure method;

[0254] Specifically, the exposure mode of the target camera is a hardware fixed parameter. The target camera provided in this application can adopt a global exposure mode. For ease of introduction, please refer to Fig. 9 , Fig. 9 This is a schematic diagram of the effect of comparing the exposure mode and the exposure time in the embodiment of the present application, wherein: Fig. 9 Figures (A) and (C) in the figure are a set of comparisons between the effects of global exposure and row exposure. Fig. 9 Figures (B) and (D) in the figure are another set of comparisons between global exposure and row exposure. It can be seen that under global exposure, the exposure time of each row of the image can be relatively consistent. Therefore, compared with row exposure, the image collected by global exposure will not be offset or skewed.

[0255] 2. Camera exposure time;

[0256] Specifically, the maximum exposure time of the target camera is a software adjustable parameter. The maximum exposure time of the target camera provided in this application can be 5 milliseconds. For ease of introduction, please refer to Fig. 9 , Fig. 9 Figures (A) and (B) in the figure are a comparison of the effects of low exposure time and high exposure time. Fig. 9 Figures (C) and (D) in the figure are another set of comparisons between low exposure time and high exposure time. It can be seen that the longer the exposure time, the brighter the image and the more noise. When shooting moving objects, long exposure time will cause moving objects to have a trailing image, and the blur effect caused by the image will be more obvious. Therefore, try to use a shorter exposure time.

[0257] 3. Lens hardware parameters;

[0258] Specifically, the lens hardware parameters are hardware fixed parameters. Since the better the transmittance of the lens, the clearer the captured image. Therefore, the target camera provided in this application can select a lens with a transmittance of more than 80%.

[0259] 4. Fill light quality;

[0260] Specifically, the quality of fill light can be determined by the fill light device provided by the target camera. Therefore, the unified glare rating (UGR) can be used as an indicator to measure the quality of fill light. UGR is divided into six levels, namely level 13, level 16, level 19, level 22, level 25 and level 28. Among them, most occasions need to be less than or equal to level 22. Therefore, the fill light device provided by the target camera in this application corresponds to UGR less than or equal to level 22.

[0261] Different models of cameras usually have different ISP setting parameters, etc. Therefore, in order to obtain a blur speed threshold that is more in line with the actual situation, the corresponding blur speed threshold can be calibrated for different models of cameras. For ease of understanding, please refer to Table 2, which is a schematic diagram of the relationship between camera models and blur speed thresholds.

[0262] Table 2

[0263] Camera Model Exposure method Exposure duration Lens transmittance UGR Blur speed threshold X123_01 Global Exposure 5 ms 80％ Level 16 40cm / s X000_05 Global Exposure 5 ms 85％ Level 19 50cm / s X999_08 Global Exposure 3 ms 90％ Level 19 60cm / s

[0264] The values ​​of the blur speed thresholds shown in Table 2 are only for reference and should not be construed as limiting the present application.

[0265] Secondly, in the embodiment of the present application, parameters related to the target camera are provided. By adopting the above parameters, a better quality image sequence can be captured, which is conducive to image optimization.

[0266] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, determining the clarity evaluation result of the second image according to the magnitude relationship between the first motion speed and the blur speed threshold may specifically include:

[0267] If the first motion speed is greater than or equal to the blur speed threshold, determining that the clarity evaluation result of the second image is an unsharp image;

[0268] If the first motion speed is less than the blur speed threshold, it is determined that the clarity evaluation result of the second image is a clear image.

[0269] In one or more embodiments, a method for determining image clarity based on motion speed is introduced. As can be seen from the above embodiments, by comparing the motion speed and the blur speed threshold, a motion blur-related judgment can be made, thereby achieving optimal recognition of the second image.

[0270] Specifically, if the first motion speed is greater than or equal to the blur speed threshold, the clarity evaluation result of the second image is determined to be a non-clear image. That is, it is determined that the target object is motion blurred and the image quality is unusable. Therefore, the next set of two consecutive frames of images can be selected for processing such as steps 210 to 240. If the first motion speed is less than the blur speed threshold, the clarity evaluation result of the second image is determined to be a clear image. That is, it is determined that the target object is not motion blurred and the image quality is usable. Therefore, the second image can be used as a preferred frame for subsequent processing (e.g., recognition, comparison, matching, etc.).

[0271] Secondly, in the embodiment of the present application, a method for determining image clarity based on motion speed is provided. By comparing the motion speed and the blur speed threshold, it is possible to determine whether the next frame of image is clear enough, thereby improving the feasibility and operability of the solution.

[0272] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, after determining that the clarity evaluation result of the second image is a non-clear image, the following may also be included:

[0273] Acquire a second image group from the image sequence, wherein the second image group includes a third image and a fourth image, and the fourth image is a subsequent frame image adjacent to the third image;

[0274] Extracting feature points from the second image group to obtain a third feature point set corresponding to the third image and a fourth feature point set corresponding to the fourth image, wherein the third feature points included in the third feature point set and the fourth feature points included in the fourth feature point set have a one-to-one correspondence;

[0275] Determining a second motion speed of the target object according to the third feature point set and the fourth feature point set;

[0276] The clarity evaluation result of the fourth image is determined according to the magnitude relationship between the second motion speed and the blur speed threshold.

[0277] In one or more embodiments, a method for processing an image group that does not meet the clarity condition is introduced. From the above embodiments, if the first motion speed is less than the blur speed threshold, then the second image group can be obtained from the image sequence and processed in a manner similar to steps 210 to 240.

[0278] Specifically, the previous frame image in the second image group is taken as the "third image", and the next frame image in the second image group is taken as the "fourth image". Feature points are extracted from the third image and the fourth image respectively, thereby obtaining a third feature point set corresponding to the third image and a fourth feature point set corresponding to the fourth image. Among them, the third feature point set includes at least R third feature points, and the fourth feature point set includes at least R fourth feature points, wherein there is a corresponding relationship between the third feature point and the fourth feature point, and therefore, there are R pairs of matching feature points. Based on this, the second motion speed of the target object is calculated according to the third feature point set and the fourth feature point set. If the second motion speed is greater than or equal to the blur speed threshold, the clarity evaluation result of the fourth image is determined to be a non-clear image. If the second motion speed is less than the blur speed threshold, the clarity evaluation result of the fourth image is determined to be a clear image.

[0279] It is understandable that the R value and the K value may be equal or different, and this is not limited here.

[0280] Again, in the embodiment of the present application, a method for processing an image group that does not meet the definition condition is provided. In the above method, if an image group taken out from the image sequence does not meet the image optimization condition, the next image group can be further extracted from the image sequence for detection, thereby improving the feasibility of image optimization.

[0281] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, before acquiring the first image group from the image sequence, the following may also be included:

[0282] In response to a trigger instruction for an image acquisition control, a target camera is started;

[0283] Collecting an original image sequence through a target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1;

[0284] Performing target region detection on the original image sequence to obtain M frames of continuous images including the target object, and taking the M frames of continuous images as an image sequence, wherein the target object is a palm, and M is an integer greater than 1 and less than or equal to N;

[0285] After determining that the clarity evaluation result of the second image is a clear image, the method may further include:

[0286] Performing a similarity comparison between the second image and T input images in the database to obtain T similarity results, wherein each input image is a palm image pre-input into the database, and T is an integer greater than or equal to 1;

[0287] According to the T similarity result, the palmprint recognition result is determined.

[0288] In one or more embodiments, a method for realizing palmprint recognition based on clear images is introduced. It can be seen from the above embodiments that if the target object is a palm, it is also necessary to obtain a corresponding infrared image to realize liveness detection. When the liveness detection is satisfied, the target area detection is performed on the original image sequence captured by the target camera to obtain an image sequence including M frames of continuous images. Among them, the original image sequence includes N frames of images, so M≤N.

[0289] It should be noted that the You Only Look Once (YOLO) network, or the region convolutional neural network (RCNN), or the fast region convolutional neural network (Fast RCNN), or the single shot multiBox detector (SSD) can be used as the target detection network to perform target area detection on the original image sequence. The target detection technology is used to locate the finger seam points, thereby extracting the palm area image. Then, the technical solution provided in this application is used to perform feature point extraction, feature point matching, motion speed calculation, clarity evaluation and other processing on the palm area image, which will not be repeated here.

[0290] Specifically, for ease of understanding, see Fig.10 , Fig.10 A schematic diagram of palm recognition in an embodiment of the present application is shown in the figure. As shown in the figure, the target camera captures the palm of object A, thereby collecting an original image sequence. An image sequence including the palm is extracted from the original image sequence by the target detection model. Based on this, the image clarity evaluation provided by the present application is used to evaluate the images in the image sequence, so as to screen out a clear palm image (for example, the second image). Therefore, the palm print features are extracted for the clear palm image, and then the extracted palm print features are matched with the T palm print features stored in the database, that is, a similarity comparison between the clear image and the T input images is achieved. Thus, T similarity results are obtained, and the palm print recognition result is determined based on the T similarity results. That is, in the case of a successful match, the palm print recognition result is the identified object identity, and in the case of a failed match, the palm print recognition result is used to prompt the palm to be swiped again.

[0291] It is understandable that the database stores the mapping relationship between the object identifier and the palm print features of the input image. Therefore, the corresponding object identifier can be queried based on the matched palm print features to identify the identity of the palm scan object.

[0292] Again, in the embodiment of the present application, a method for realizing palmprint recognition based on clear images is provided. Through the above method, the clarity of the palm in the image can be effectively defined, thereby eliminating blurred images caused by excessive movement speed, thereby improving the pass rate of palm scanning.

[0293] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, before acquiring the first image group from the image sequence, the following may also be included:

[0294] In response to a trigger instruction for an image recording control, a target camera is started;

[0295] Collecting an original image sequence through a target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1;

[0296] Performing target region detection on the original image sequence to obtain M frames of continuous images including the target object, and taking the M frames of continuous images as an image sequence, wherein the target object is a palm, and M is an integer greater than 1 and less than or equal to N;

[0297] After determining that the clarity evaluation result of the second image is a clear image, the method may further include:

[0298] The second image is stored in the database.

[0299] In one or more embodiments, a method for implementing palm print recording based on clear images is introduced. It can be seen from the above embodiments that if the target object is a palm, it is also necessary to obtain a corresponding infrared image to achieve liveness detection. When the liveness detection is satisfied, the target area detection is performed on the original image sequence captured by the target camera to obtain an image sequence including M frames of continuous images. Among them, the original image sequence includes N frames of images, so M≤N.

[0300] It should be noted that the YOLO network, or RCNN, or Fast RCNN, or SSD, etc. can be used as the target detection network to detect the target area of ​​the original image sequence. The target detection technology is used to locate the finger seam points, thereby extracting the palm area image. Then, the technical solution provided by this application is used to extract feature points, match feature points, calculate motion speed, evaluate clarity, etc. on the palm area image, which will not be described in detail here.

[0301] Specifically, for ease of understanding, see Fig.12 , Fig.11This is a schematic diagram of implementing palm entry in an embodiment of the present application. As shown in the figure, the target camera captures the palm of subject A, thereby acquiring an original image sequence. An image sequence including the palm is extracted from the original image sequence by the target detection model. Based on this, the image clarity evaluation provided by the present application is used to evaluate the images in the image sequence, thereby screening out a clear palm image (e.g., the second image). Therefore, the second image can be stored in a database, or the palm print features extracted from the second image can be stored in a database for subsequent recognition.

[0302] Again, in the embodiment of the present application, a method for implementing palm print entry based on clear images is provided. Through the above method, the clarity of the palm in the image can be effectively defined, thereby eliminating blurred images caused by excessive movement speed, thereby improving the accuracy of palm print features in the database.

[0303] Optionally, in the above Figure 3 On the basis of the corresponding embodiments, in another optional embodiment provided by the embodiment of the present application, before acquiring the first image group from the image sequence, the following may also be included:

[0304] Collecting an original image sequence through a target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1;

[0305] Perform target area detection on the original image sequence to obtain M frames of continuous images including the target object, and use the M frames of continuous images as an image sequence, wherein the target object is a license plate, and M is an integer greater than 1 and less than or equal to N;

[0306] After determining that the clarity evaluation result of the second image is a clear image, the method may further include:

[0307] Performing optical character recognition (OCR) on the second image to obtain target text;

[0308] The target text is matched with T input texts in the database to obtain the license plate recognition result, where each input text is the license plate information pre-entered into the database, and T is an integer greater than or equal to 1.

[0309] In one or more embodiments, a method for realizing license plate recognition based on clear images is introduced. As can be seen from the above embodiments, the target object can be a license plate, and the target area detection is performed on the original image sequence captured by the target camera to obtain an image sequence including M frames of continuous images. Among them, the original image sequence includes N frames of images, so M≤N.

[0310] It should be noted that the YOLO network, or RCNN, or Fast RCNN, or SSD, etc. can be used as the target detection network to detect the target area of ​​the original image sequence. The target detection technology is used to locate the license plate, thereby extracting the license plate area image. Then, the technical solution provided by this application is used to extract feature points, match feature points, calculate motion speed, evaluate clarity, etc. on the license plate area image, which will not be described in detail here.

[0311] Specifically, for ease of understanding, see Fig.13 , Fig.12 This is a schematic diagram of implementing license plate recognition in an embodiment of the present application. As shown in the figure, the target camera shoots the vehicle, thereby collecting the original image sequence. The image sequence including the license plate is extracted from the original image sequence by the target detection model. Based on this, the image clarity evaluation provided by the present application is used to evaluate the images in the image sequence, so as to screen out a clear license plate image (for example, the second image). Then, optical character recognition (OCR) is performed on the clear license plate image to obtain the target text (for example, Guangdong B11111). The extracted target text is then matched with the T input texts stored in the database, thereby obtaining the license plate recognition result. The license plate recognition result is used to indicate whether the match is successful. If the match is successful, the vehicle can be released. On the contrary, if the match is recognized, the vehicle is prohibited from passing.

[0312] Again, in an embodiment of the present application, a method for realizing license plate recognition based on a clear image is provided. Through the above method, the clarity of the license plate in the image can be effectively defined, thereby eliminating blurred images caused by excessive movement speed, thereby improving the accuracy of license plate recognition.

[0313] The image clarity evaluation device in this application is described in detail below. Fig.13 , Fig.13 This is a schematic diagram of an embodiment of an image clarity evaluation device in an embodiment of the present application. The image clarity evaluation device 30 includes:

[0314] An acquisition module 310 is used to acquire a first image group from an image sequence, wherein the image sequence is acquired by a target camera, each frame of the image sequence includes an image corresponding to the target object, and the first image group includes a first image and a second image, wherein the second image is a subsequent frame of the first image;

[0315] The processing module 320 is used to extract feature points from the first image group to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image, wherein the first feature points included in the first feature point set and the second feature points included in the second feature point set have a one-to-one correspondence relationship;

[0316] A determination module 330, configured to determine a first motion speed of the target object according to the first feature point set and the second feature point set;

[0317] The determination module 330 is further configured to determine a clarity evaluation result of the second image according to a magnitude relationship between the first motion speed and the blur speed threshold.

[0318] In an embodiment of the present application, an image clarity evaluation device is provided. With the above device, the motion speed of the target object in the image can be detected using the two frames of images, and the image clarity can be effectively defined based on the size relationship between the motion speed of the target object and the blur speed threshold, thereby helping to screen out images with higher clarity from the image sequence for subsequent processing.

[0319] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application,

[0320] The processing module 320 is specifically configured to extract feature points from the first image to obtain a first feature point set for the target object, wherein the first feature point set includes K first feature points, where K is an integer greater than 1;

[0321] Feature points are extracted from the second image to obtain a second feature point set for the target object, wherein the second feature point set includes K second feature points.

[0322] In the embodiment of the present application, an image clarity evaluation device is provided. With the above device, feature points for the target object can be extracted using methods such as ORB, SIFT or SURF, thereby improving the flexibility and feasibility of the solution. In addition, using ORB can also improve the efficiency of feature point extraction.

[0323] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application,

[0324] The determination module 330 is specifically used to obtain a distance value corresponding to each first feature point in the first feature point set, and to obtain a distance value corresponding to each second feature point in the second feature point set, wherein the distance value represents a distance between the feature point and a camera plane corresponding to the target camera;

[0325] Obtaining the two-dimensional coordinates corresponding to each first feature point in the first feature point set, and obtaining the two-dimensional coordinates corresponding to each second feature point in the second feature point set, wherein the two-dimensional coordinates include the horizontal coordinate and the vertical coordinate of the feature point in the pixel coordinate system;

[0326] Determine an average distance of feature points according to the distance value corresponding to each first feature point, the distance value corresponding to each second feature point, the two-dimensional coordinates corresponding to each first feature point, and the two-dimensional coordinates corresponding to each second feature point;

[0327] A first motion speed of the target object is determined according to an average distance between feature points and an inter-frame duration.

[0328] In an embodiment of the present application, an image clarity evaluation device is provided. With the above device, the distance between each pair of feature points can be calculated using the distance between the feature point and the camera plane and the position in the pixel coordinate system, and based on this, the movement speed of the target object can be determined. Thus, the feasibility and operability of the solution are improved.

[0329] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application,

[0330] The determination module 330 is specifically configured to calculate, for each first feature point in the first feature point set, the abscissa of the first feature point in the spatial coordinate system according to the distance value corresponding to the first feature point, the abscissa of the first feature point in the pixel coordinate system, the abscissa of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system.

[0331] For each first feature point in the first feature point set, the vertical coordinate of the first feature point in the spatial coordinate system is calculated according to the distance value corresponding to the first feature point, the vertical coordinate of the first feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0332] For each second feature point in the second feature point set, the abscissa of the second feature point in the spatial coordinate system is calculated according to the distance value corresponding to the second feature point, the abscissa of the second feature point in the pixel coordinate system, the abscissa of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0333] For each second feature point in the second feature point set, the vertical coordinate of the second feature point in the spatial coordinate system is calculated according to the distance value corresponding to the second feature point, the vertical coordinate of the second feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system;

[0334] The average distance of the feature points is determined according to the horizontal coordinate of each first feature point in the spatial coordinate system, the vertical coordinate of each first feature point in the spatial coordinate system, the horizontal coordinate of each second feature point in the spatial coordinate system, the vertical coordinate of each second feature point in the spatial coordinate system, the distance value corresponding to each first feature point, and the distance value corresponding to each second feature point.

[0335] In the embodiment of the present application, an image clarity evaluation device is provided. With the above device, considering that the real scene is a 3D scene, after converting the feature points corresponding to the target object from 2D positions to 3D positions, a more accurate distance can be calculated, thereby obtaining a motion speed that is more consistent with the real motion of the target object, which is conducive to improving the reliability of image optimization.

[0336] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application, the first feature point set includes K first feature points, and the second feature point set includes K second feature points, where K is an integer greater than 1;

[0337] The determination module 330 is specifically configured to construct K groups of feature point pairs according to the first feature point set and the second feature point set, wherein each group of feature point pairs includes a first feature point and a second feature point having a corresponding relationship;

[0338] For each pair of feature point pairs in the K groups of feature point pairs, the feature point distance corresponding to the feature point pair is calculated according to the abscissa of the first feature point in the spatial coordinate system, the abscissa of the second feature point in the spatial coordinate system, the ordinate of the first feature point in the spatial coordinate system, the ordinate of the second feature point in the spatial coordinate system, the distance value corresponding to the first feature point, and the distance value corresponding to the second feature point in the feature point set;

[0339] The feature point distances corresponding to each feature point pair in the K groups of feature point pairs are averaged to obtain the average feature point distance.

[0340] In an embodiment of the present application, an image clarity evaluation device is provided. The above device is used to match feature points of two frames of images to obtain K groups of feature point pairs. Thus, the corresponding feature point distance is calculated based on each group of feature point pairs, thereby improving the feasibility and operability of the solution.

[0341] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application, the first feature point set includes K first feature points, and the second feature point set includes K second feature points, where K is an integer greater than 1;

[0342] The determination module 330 is specifically configured to construct K groups of feature point pairs according to the first feature point set and the second feature point set, wherein each group of feature point pairs includes a first feature point and a second feature point having a corresponding relationship;

[0343] For each pair of feature point pairs in the K groups of feature point pairs, the feature point distance corresponding to the feature point pair is calculated according to the abscissa of the first feature point in the spatial coordinate system, the abscissa of the second feature point in the spatial coordinate system, the ordinate of the first feature point in the spatial coordinate system, the ordinate of the second feature point in the spatial coordinate system, the distance value corresponding to the first feature point, and the distance value corresponding to the second feature point in the feature point set;

[0344] The feature point distances corresponding to each feature point pair in the K groups of feature point pairs are sorted in descending order to obtain a feature point distance sequence, wherein the feature point distance sequence includes K feature point distances;

[0345] Select P feature point distances from the feature point distance sequence, where P is an integer greater than or equal to 1 and less than K;

[0346] The distances of P feature points are averaged to obtain the average distance of the feature points.

[0347] In an embodiment of the present application, a device for evaluating image clarity is provided. By using the device, selecting some feature point distances for calculation can reduce the amount of calculation, thereby improving the calculation efficiency.

[0348] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application,

[0349] The determination module 330 is specifically configured to, if K is an odd number, take the [(K+1) / 2]th feature point distance, the Q feature point distances before the [(K+1) / 2]th feature point distance, and the Q feature point distances after the [(K+1) / 2]th feature point distance in the feature point distance sequence as P feature point distances, where Q is an integer greater than or equal to 1;

[0350] If K is an even number, the (K / 2)th feature point distance in the feature point distance sequence, the (K / 2+1)th feature point distance in the feature point distance sequence, the Q feature point distances before the (K / 2)th feature point distance, and the Q feature point distances after the (K / 2+1)th feature point distance are collectively taken as P feature point distances.

[0351] In an embodiment of the present application, an image clarity evaluation device is provided. By using the above device, selecting some feature point distances for calculation can reduce the amount of calculation, thereby improving the calculation efficiency. At the same time, selecting the feature point distances of the middle section for calculation can calculate a more accurate average distance of the feature points.

[0352] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application, the image clarity evaluation device 30 further includes a collection module 340, a recognition module 350 and an adjustment module 360;

[0353] A collection module 340 is used to collect a first image sequence through a target camera when the target fixture moves at a first speed, wherein the first image sequence includes at least one image;

[0354] A recognition module 350, configured to recognize images in a first image sequence and obtain a first accuracy rate;

[0355] An adjustment module 360, configured to adjust the first speed movement to a second speed movement if the first accuracy rate is greater than or equal to an accuracy rate threshold, wherein the second speed movement is greater than the first movement speed;

[0356] The acquisition module 340 is further configured to acquire a second image sequence through the target camera when the target fixture moves at a second speed, wherein the second image sequence includes at least one image;

[0357] The recognition module 350 is further used to recognize the images in the second image sequence to obtain a second accuracy rate;

[0358] The determination module 330 is further configured to determine the first speed motion as a blur speed threshold if the second accuracy rate is less than the accuracy rate threshold.

[0359] In the embodiment of the present application, an image clarity evaluation device is provided. By using the above device, the motion speed that meets the minimum clarity requirement can be tested in an experimental environment, and the motion speed is used as the blur speed threshold. Compared with directly setting the empirical value as the blur speed threshold, the blur speed threshold calibrated based on the test result has better adaptability and better accuracy.

[0360] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application,

[0361] The target camera uses global exposure mode;

[0362] The maximum exposure time of the target camera is less than or equal to 5 milliseconds;

[0363] The lens transmittance of the target camera is greater than or equal to 80%;

[0364] The unified glare value of the fill light provided by the target camera is less than or equal to level 22.

[0365] In the embodiment of the present application, parameters related to the target camera are provided. By adopting the above parameters, a better quality image sequence can be captured, which is conducive to image optimization.

[0366] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application,

[0367] The determination module 330 is specifically configured to determine that the clarity evaluation result of the second image is an unsharp image if the first motion speed is greater than or equal to a blur speed threshold;

[0368] If the first motion speed is less than the blur speed threshold, it is determined that the clarity evaluation result of the second image is a clear image.

[0369] In an embodiment of the present application, an image clarity evaluation device is provided. By using the above device, after comparing the motion speed and the blur speed threshold, it can be determined whether the next frame of image is clear enough, thereby improving the feasibility and operability of the solution.

[0370] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application,

[0371] The acquisition module 310 is further configured to acquire a second image group from the image sequence after determining that the clarity evaluation result of the second image is an unsharp image, wherein the second image group includes a third image and a fourth image, and the fourth image is a subsequent frame image adjacent to the third image;

[0372] The processing module 320 is further configured to extract feature points from the second image group to obtain a third feature point set corresponding to the third image and a fourth feature point set corresponding to the fourth image, wherein the third feature points included in the third feature point set and the fourth feature points included in the fourth feature point set have a one-to-one correspondence;

[0373] The determination module 330 is further configured to determine a second motion speed of the target object according to the third feature point set and the fourth feature point set;

[0374] The determination module 330 is further configured to determine a clarity evaluation result of the fourth image according to a magnitude relationship between the second motion speed and the blur speed threshold.

[0375] In an embodiment of the present application, an image clarity evaluation device is provided. With the above device, if an image group taken out from an image sequence does not meet the image optimization condition, the next image group can be further extracted from the image sequence for detection, thereby improving the feasibility of image optimization.

[0376] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application, the image clarity evaluation device 30 further includes a starting module 370 and a detection module 380;

[0377] A start module 370 is used to start a target camera in response to a trigger instruction for an image acquisition control before acquiring a first image group from an image sequence;

[0378] The acquisition module 340 is further used to acquire an original image sequence through a target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1;

[0379] A detection module 380 is used to perform target area detection on the original image sequence to obtain M frames of continuous images including the target object, and use the M frames of continuous images as an image sequence, wherein the target object is a palm, and M is an integer greater than 1 and less than or equal to N;

[0380] The recognition module 350 is further configured to, after determining that the clarity evaluation result of the second image is a clear image, perform a similarity comparison between the second image and T recorded images in the database to obtain T similarity results, wherein each recorded image is a palm image pre-recorded into the database, and T is an integer greater than or equal to 1;

[0381] The determination module 330 is further used to determine the palmprint recognition result according to the T similarity result.

[0382] In the embodiment of the present application, an image clarity evaluation device is provided. The above device can effectively define the clarity of the palm in the image, thereby eliminating the blurred image caused by too fast movement speed, thereby improving the pass rate of palm scanning.

[0383] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application, the image clarity evaluation device 30 further includes a storage module 390;

[0384] The start module 370 is further configured to start the target camera in response to a trigger instruction for the image entry control before acquiring the first image group from the image sequence;

[0385] The acquisition module 340 is further used to acquire an original image sequence through a target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1;

[0386] The detection module 380 is further used to perform target area detection on the original image sequence to obtain M frames of continuous images including the target object, and use the M frames of continuous images as the image sequence, wherein the target object is a palm, and M is an integer greater than 1 and less than or equal to N;

[0387] The storage module 390 is used to store the second image in a database after determining that the clarity evaluation result of the second image is a clear image.

[0388] In the embodiment of the present application, an image clarity evaluation device is provided. The above device can effectively define the clarity of the palm in the image, thereby eliminating the blurred image caused by too fast movement speed, thereby improving the accuracy of the palm print features in the database.

[0389] Optionally, in the above Fig.13 On the basis of the corresponding embodiment, in another embodiment of the image clarity evaluation device 30 provided in the embodiment of the present application,

[0390] The acquisition module 340 is further configured to acquire an original image sequence through a target camera before acquiring the first image group from the image sequence, wherein the original image sequence includes N frames of images, where N is an integer greater than 1;

[0391] The detection module 380 is further used to perform target area detection on the original image sequence, obtain M frames of continuous images including the target object, and use the M frames of continuous images as the image sequence, wherein the target object is a license plate, and M is an integer greater than 1 and less than or equal to N;

[0392] The recognition module 350 is further used to perform optical character recognition (OCR) on the second image to obtain a target text;

[0393] The determination module 330 is also used to match the target text with T input texts in the database to obtain a license plate recognition result, wherein each input text is license plate information pre-entered into the database, and T is an integer greater than or equal to 1.

[0394] In an embodiment of the present application, an image clarity evaluation device is provided. The above device can effectively define the clarity of the license plate in the image, thereby eliminating the blurred image caused by excessive movement speed, thereby improving the accuracy of license plate recognition.

[0395] The present application also provides an image clarity evaluation device, which can be deployed in a terminal device, such as Fig.14For the sake of convenience, only the parts related to the embodiments of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiments of the present application. In the embodiments of the present application, the terminal device is a mobile phone as an example for explanation:

[0396] Fig.14 FIG. 1 is a block diagram showing a partial structure of a mobile phone related to a terminal device provided in an embodiment of the present application. Fig.14 The mobile phone includes: a radio frequency (RF) circuit 410, a memory 420, an input unit 430, a display unit 440, a sensor 450, an audio circuit 460, a wireless fidelity (WiFi) module 470, a processor 480, and a power supply 490. Those skilled in the art will understand that Fig.14 The mobile phone structure shown in the figure does not constitute a limitation on the mobile phone, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0397] Combine the following Fig.14 A detailed introduction to the various components of the mobile phone:

[0398] The RF circuit 410 can be used for receiving and sending signals during the process of sending and receiving information or making calls. In particular, after receiving the downlink information of the base station, it is sent to the processor 480 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit 410 includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (LNA), a duplexer, etc. In addition, the RF circuit 410 can also communicate with the network and other devices through wireless communication. The above-mentioned wireless communication can use any communication standard or protocol, including but not limited to the global system of mobile communication (GSM), general packet radio service (GPRS), code division multiple access (CDMA), wideband code division multiple access (WCDMA), long term evolution (LTE), email, short messaging service (SMS), etc.

[0399] The memory 420 can be used to store software programs and modules. The processor 480 executes various functional applications and data processing of the mobile phone by running the software programs and modules stored in the memory 420. The memory 420 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory 420 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0400] The input unit 430 can be used to receive input digital or character information, and to generate key signal input related to the user settings and function control of the mobile phone. Specifically, the input unit 430 may include a touch panel 431 and other input devices 432. The touch panel 431, also known as a touch screen, can collect the user's touch operation on or near it (such as the user's operation on the touch panel 431 or near the touch panel 431 using any suitable object or accessory such as a finger, stylus, etc.), and drive the corresponding connection device according to a pre-set program. Optionally, the touch panel 431 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it to the processor 480, and can receive and execute commands sent by the processor 480. In addition, the touch panel 431 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic waves. In addition to the touch panel 431, the input unit 430 may further include other input devices 432. Specifically, the other input devices 432 may include but are not limited to one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a mouse, a joystick, etc.

[0401] The display unit 440 may be used to display information input by the user or information provided to the user and various menus of the mobile phone. The display unit 440 may include a display panel 441. Optionally, the display panel 441 may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc. Further, the touch panel 431 may cover the display panel 441. When the touch panel 431 detects a touch operation on or near it, it is transmitted to the processor 480 to determine the type of touch event. Subsequently, the processor 480 provides a corresponding visual output on the display panel 441 according to the type of touch event. Although in Fig.14 In the embodiment, the touch panel 431 and the display panel 441 are used as two independent components to realize the input and output functions of the mobile phone, but in some embodiments, the touch panel 431 and the display panel 441 can be integrated to realize the input and output functions of the mobile phone.

[0402] The mobile phone may also include at least one sensor 450, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel 441 according to the brightness of the ambient light, and the proximity sensor may turn off the display panel 441 and / or the backlight when the mobile phone is moved to the ear. As a type of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in all directions (generally three axes), and can detect the magnitude and direction of gravity when stationary. It can be used for applications that identify the posture of the mobile phone (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; as for other sensors that can be configured in the mobile phone, such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc., they will not be repeated here.

[0403] The audio circuit 460, the speaker 461, and the microphone 462 can provide an audio interface between the user and the mobile phone. The audio circuit 460 can transmit the received audio data to the speaker 461 after converting the received audio data into an electrical signal, which is converted into a sound signal for output; on the other hand, the microphone 462 converts the collected sound signal into an electrical signal, which is received by the audio circuit 460 and converted into audio data, and then the audio data is output to the processor 480 for processing, and then sent to another mobile phone through the RF circuit 410, or the audio data is output to the memory 420 for further processing.

[0404] WiFi is a short-range wireless transmission technology. The mobile phone can help users send and receive emails, browse web pages and access streaming media through the WiFi module 470. It provides users with wireless broadband Internet access. Fig.14A WiFi module 470 is shown, but it is understandable that it is not an essential component of the mobile phone and can be omitted as needed without changing the essence of the invention.

[0405] The processor 480 is the control center of the mobile phone. It uses various interfaces and lines to connect various parts of the entire mobile phone. It executes various functions of the mobile phone and processes data by running or executing software programs and / or modules stored in the memory 420, and calling data stored in the memory 420. Optionally, the processor 480 may include one or more processing units; optionally, the processor 480 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 480.

[0406] The mobile phone also includes a power source 490 (such as a battery) for supplying power to various components. Optionally, the power source can be logically connected to the processor 480 through a power management system, thereby implementing functions such as charging, discharging, and power consumption management through the power management system.

[0407] Although not shown, the mobile phone may also include a camera, a Bluetooth module, etc., which will not be described in detail here.

[0408] The steps performed by the terminal device in the above embodiment can be based on the Fig.14 The terminal device structure shown.

[0409] The present application also provides an image clarity evaluation device, which can be deployed on a server. Fig.15 : is a schematic diagram of a server structure provided in an embodiment of the present application. The server 500 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 522 (for example, one or more processors) and a memory 532, and one or more storage media 530 (for example, one or more mass storage devices) storing application programs 542 or data 544. Among them, the memory 532 and the storage medium 530 can be short-term storage or permanent storage. The program stored in the storage medium 530 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the server. Furthermore, the central processing unit 522 can be configured to communicate with the storage medium 530 to execute a series of instruction operations in the storage medium 530 on the server 500.

[0410] The server 500 may also include one or more power supplies 526, one or more wired or wireless network interfaces 550, one or more input and output interfaces 558, and / or one or more operating systems 541, such as Windows Server 2003, ... TM , Mac OS X TM , Unix TM ,Linux TM , FreeBSD TM etc.

[0411] The steps performed by the server in the above embodiment can be based on the Fig.15 The server structure shown.

[0412] A computer device is also provided in an embodiment of the present application, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the methods described in the above embodiments are implemented.

[0413] A computer-readable storage medium is also provided in an embodiment of the present application, on which a computer program is stored. When the computer program is executed by a processor, the steps of the methods described in the above embodiments are implemented.

[0414] A computer program product is also provided in an embodiment of the present application, including a computer program, which, when executed by a processor, implements the steps of the methods described in the above embodiments.

[0415] It can be understood that in the specific implementation of this application, user information, palm print information, license plate information and other related data are involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0416] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0417] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0418] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0419] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0420] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store computer programs.

[0421] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for evaluating image clarity, characterized in that: include: Acquire a first image group from an image sequence, wherein the image sequence is acquired by a target camera, each frame of the image sequence includes an image corresponding to the target object, the first image group includes a first image and a second image, and the second image is a subsequent frame of the image adjacent to the first image; Extracting feature points from the first image group to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image, wherein the first feature points included in the first feature point set and the second feature points included in the second feature point set have a one-to-one correspondence; Obtaining a distance value corresponding to each first feature point in the first feature point set, and obtaining a distance value corresponding to each second feature point in the second feature point set, wherein the distance value represents a distance between a feature point and a camera plane corresponding to the target camera; Obtaining the two-dimensional coordinates corresponding to each first feature point in the first feature point set, and obtaining the two-dimensional coordinates corresponding to each second feature point in the second feature point set, wherein the two-dimensional coordinates include the horizontal coordinate and the vertical coordinate of the feature point in the pixel coordinate system; For each first feature point in the first feature point set, calculate the abscissa of the first feature point in the spatial coordinate system according to the distance value corresponding to the first feature point, the abscissa of the first feature point in the pixel coordinate system, the abscissa of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system; For each first feature point in the first feature point set, calculate the vertical coordinate of the first feature point in the spatial coordinate system according to the distance value corresponding to the first feature point, the vertical coordinate of the first feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system; For each second feature point in the second feature point set, calculate the abscissa of the second feature point in the spatial coordinate system according to the distance value corresponding to the second feature point, the abscissa of the second feature point in the pixel coordinate system, the abscissa of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system; For each second feature point in the second feature point set, calculate the vertical coordinate of the second feature point in the spatial coordinate system according to the distance value corresponding to the second feature point, the vertical coordinate of the second feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system; Determine the average distance of the feature points according to the abscissa of each first feature point in the spatial coordinate system, the ordinate of each first feature point in the spatial coordinate system, the abscissa of each second feature point in the spatial coordinate system, the ordinate of each second feature point in the spatial coordinate system, the distance value corresponding to each first feature point, and the distance value corresponding to each second feature point; Determining a first motion speed of the target object according to the average distance of the feature points and the inter-frame duration; The clarity evaluation result of the second image is determined according to the magnitude relationship between the first motion speed and the blur speed threshold.

2. The evaluation method according to claim 1, characterized in that: The extracting feature points from the first image group to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image includes: Extracting feature points from the first image to obtain the first feature point set for the target object, wherein the first feature point set includes K first feature points, where K is an integer greater than 1; Feature points are extracted from the second image to obtain the second feature point set for the target object, wherein the second feature point set includes K second feature points.

3. The evaluation method according to claim 1, characterized in that: The first feature point set includes K first feature points, and the second feature point set includes K second feature points, wherein K is an integer greater than 1; The determining the average distance of the feature points according to the abscissa of each first feature point in the spatial coordinate system, the ordinate of each first feature point in the spatial coordinate system, the abscissa of each second feature point in the spatial coordinate system, the ordinate of each second feature point in the spatial coordinate system, the distance value corresponding to each first feature point, and the distance value corresponding to each second feature point includes: Constructing K groups of feature point pairs according to the first feature point set and the second feature point set, wherein each group of feature point pairs includes a first feature point and a second feature point having a corresponding relationship; For each set of feature point pairs in the K sets of feature point pairs, the feature point distance corresponding to the feature point pair is calculated according to the abscissa of the first feature point in the spatial coordinate system, the abscissa of the second feature point in the spatial coordinate system, the ordinate of the first feature point in the spatial coordinate system, the ordinate of the second feature point in the spatial coordinate system, the distance value corresponding to the first feature point, and the distance value corresponding to the second feature point in the feature point set; The feature point distances corresponding to each feature point pair in the K groups of feature point pairs are averaged to obtain the feature point average distance.

4. The evaluation method according to claim 1, characterized in that: The first feature point set includes K first feature points, and the second feature point set includes K second feature points, wherein K is an integer greater than 1; The determining the average distance of the feature points according to the abscissa of each first feature point in the spatial coordinate system, the ordinate of each first feature point in the spatial coordinate system, the abscissa of each second feature point in the spatial coordinate system, the ordinate of each second feature point in the spatial coordinate system, the distance value corresponding to each first feature point, and the distance value corresponding to each second feature point includes: Constructing K groups of feature point pairs according to the first feature point set and the second feature point set, wherein each group of feature point pairs includes a first feature point and a second feature point having a corresponding relationship; For each set of feature point pairs in the K sets of feature point pairs, the feature point distance corresponding to the feature point pair is calculated according to the abscissa of the first feature point in the spatial coordinate system, the abscissa of the second feature point in the spatial coordinate system, the ordinate of the first feature point in the spatial coordinate system, the ordinate of the second feature point in the spatial coordinate system, the distance value corresponding to the first feature point, and the distance value corresponding to the second feature point in the feature point set; The feature point distances corresponding to each feature point pair in the K groups of feature point pairs are sorted in descending order to obtain a feature point distance sequence, wherein the feature point distance sequence includes K feature point distances; Selecting P feature point distances from the feature point distance sequence, wherein P is an integer greater than or equal to 1 and less than K; The distances of the P feature points are averaged to obtain the average distance of the feature points.

5. The evaluation method according to claim 4, characterized in that: The selecting P feature point distances from the feature point distance sequence comprises: If K is an odd number, the [(K+1) / 2]th feature point distance in the feature point distance sequence, the Q feature point distances before the [(K+1) / 2]th feature point distance, and the Q feature point distances after the [(K+1) / 2]th feature point distance are taken together as the P feature point distances, where Q is an integer greater than or equal to 1; If K is an even number, the (K / 2)th feature point distance in the feature point distance sequence, the (K / 2+1)th feature point distance in the feature point distance sequence, the Q feature point distances before the (K / 2)th feature point distance, and the Q feature point distances after the (K / 2+1)th feature point distance are collectively taken as the P feature point distances.

6. The evaluation method according to claim 1, characterized in that: The method further comprises: When the target fixture moves at a first speed, a first image sequence is captured by the target camera, wherein the first image sequence includes at least one image; recognizing images in the first image sequence to obtain a first accuracy rate; If the first accuracy rate is greater than or equal to an accuracy rate threshold, adjusting the first speed movement to a second speed movement, wherein the second speed movement is greater than the first movement speed; When the target fixture moves at a second speed, a second image sequence is captured by the target camera, wherein the second image sequence includes at least one image; recognizing images in the second image sequence to obtain a second accuracy rate; If the second accuracy is less than the accuracy threshold, the first speed motion is determined as the blur speed threshold.

7. The evaluation method according to claim 1, characterized in that: The target camera adopts a global exposure mode; The maximum exposure time of the target camera is less than or equal to 5 milliseconds; The lens transmittance of the target camera is greater than or equal to 80%; The unified glare value of the fill light device provided by the target camera is less than or equal to level 22.

8. The evaluation method according to any one of claims 1 to 7, characterized in that: The step of determining the clarity evaluation result of the second image according to the magnitude relationship between the first motion speed and the blur speed threshold comprises: If the first motion speed is greater than or equal to the blur speed threshold, determining that the clarity evaluation result of the second image is an unsharp image; If the first motion speed is less than the blur speed threshold, it is determined that the clarity evaluation result of the second image is a clear image.

9. The evaluation method according to claim 8, characterized in that: After determining that the clarity evaluation result of the second image is an unclear image, the method further includes: Acquire a second image group from the image sequence, wherein the second image group includes a third image and a fourth image, and the fourth image is a subsequent frame image adjacent to the third image; Extracting feature points from the second image group to obtain a third feature point set corresponding to the third image and a fourth feature point set corresponding to the fourth image, wherein the third feature points included in the third feature point set and the fourth feature points included in the fourth feature point set have a one-to-one correspondence; Determining a second motion speed of the target object according to the third feature point set and the fourth feature point set; The clarity evaluation result of the fourth image is determined according to the magnitude relationship between the second motion speed and the blur speed threshold.

10. The evaluation method according to claim 8, characterized in that: Before acquiring the first image group from the image sequence, the method further includes: In response to a trigger instruction for an image acquisition control, the target camera is started; Acquire an original image sequence through the target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1; Performing target area detection on the original image sequence to obtain M frames of continuous images including the target object, and using the M frames of continuous images as the image sequence, wherein the target object is a palm, and M is an integer greater than 1 and less than or equal to N; After determining that the clarity evaluation result of the second image is a clear image, the method further includes: Performing a similarity comparison between the second image and T recorded images in the database to obtain T similarity results, wherein each recorded image is a palm image pre-recorded into the database, and T is an integer greater than or equal to 1; The palmprint recognition result is determined according to the T similarity result.

11. The evaluation method according to claim 8, characterized in that: Before acquiring the first image group from the image sequence, the method further includes: In response to a trigger instruction for an image entry control, the target camera is started; Acquire an original image sequence through the target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1; Performing target area detection on the original image sequence to obtain M frames of continuous images including the target object, and using the M frames of continuous images as the image sequence, wherein the target object is a palm, and M is an integer greater than 1 and less than or equal to N; After determining that the clarity evaluation result of the second image is a clear image, the method further includes: The second image is stored in a database.

12. The evaluation method according to claim 8, characterized in that: Before acquiring the first image group from the image sequence, the method further includes: Acquire an original image sequence through the target camera, wherein the original image sequence includes N frames of images, where N is an integer greater than 1; Performing target area detection on the original image sequence to obtain M frames of continuous images including the target object, and using the M frames of continuous images as the image sequence, wherein the target object is a license plate, and M is an integer greater than 1 and less than or equal to N; After determining that the clarity evaluation result of the second image is a clear image, the method further includes: Performing optical character recognition (OCR) on the second image to obtain a target text; The target text is matched with T input texts in the database to obtain a license plate recognition result, wherein each input text is license plate information pre-entered into the database, and T is an integer greater than or equal to 1.

13. An image clarity evaluation device, characterized in that: include: an acquisition module, configured to acquire a first image group from an image sequence, wherein the image sequence is acquired by a target camera, each frame of the image sequence includes an image corresponding to a target object, and the first image group includes a first image and a second image, wherein the second image is a subsequent frame of the image adjacent to the first image; a processing module, configured to extract feature points from the first image group to obtain a first feature point set corresponding to the first image and a second feature point set corresponding to the second image, wherein the first feature points included in the first feature point set and the second feature points included in the second feature point set have a one-to-one correspondence; a determination module, configured to obtain a distance value corresponding to each first feature point in the first feature point set, and to obtain a distance value corresponding to each second feature point in the second feature point set, wherein the distance value represents a distance between a feature point and a camera plane corresponding to the target camera; The determination module is further configured to obtain the two-dimensional coordinates corresponding to each first feature point in the first feature point set, and obtain the two-dimensional coordinates corresponding to each second feature point in the second feature point set, wherein the two-dimensional coordinates include the horizontal coordinate and the vertical coordinate of the feature point in the pixel coordinate system; The determination module is further configured to calculate, for each first feature point in the first feature point set, the horizontal coordinate of the first feature point in the spatial coordinate system according to the distance value corresponding to the first feature point, the horizontal coordinate of the first feature point in the pixel coordinate system, the horizontal coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system; The determination module is further configured to calculate, for each first feature point in the first feature point set, the vertical coordinate of the first feature point in the spatial coordinate system according to the distance value corresponding to the first feature point, the vertical coordinate of the first feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system; The determination module is further configured to calculate, for each second feature point in the second feature point set, the horizontal coordinate of the second feature point in the spatial coordinate system according to the distance value corresponding to the second feature point, the horizontal coordinate of the second feature point in the pixel coordinate system, the horizontal coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system, to obtain the horizontal coordinate of the second feature point in the spatial coordinate system; The determination module is further configured to calculate, for each second feature point in the second feature point set, the vertical coordinate of the second feature point in the spatial coordinate system according to the distance value corresponding to the second feature point, the vertical coordinate of the second feature point in the pixel coordinate system, the vertical coordinate of the principal point in the pixel coordinate system, and the focal length of the target camera in the pixel coordinate system, to obtain the vertical coordinate of the second feature point in the spatial coordinate system; The determination module is further used to determine the average distance of the feature points according to the abscissa of each first feature point in the spatial coordinate system, the ordinate of each first feature point in the spatial coordinate system, the abscissa of each second feature point in the spatial coordinate system, the ordinate of each second feature point in the spatial coordinate system, the distance value corresponding to each first feature point, and the distance value corresponding to each second feature point; The determination module is further used to determine a first motion speed of the target object according to the average distance of the feature points and the inter-frame duration; The determination module is further used to determine the clarity evaluation result of the second image according to the magnitude relationship between the first motion speed and the blur speed threshold.

14. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for evaluating image clarity according to any one of claims 1 to 12 are implemented.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for evaluating image clarity according to any one of claims 1 to 12 are implemented.

16. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for evaluating image clarity described in any one of claims 1 to 12 are implemented.

Images