Image processing method, image processing system, video monitoring system

Abstract

The application provides an image processing method, an image processing system and a video monitoring system. The image processing method comprises: acquiring multiple images of the same target; displaying multiple image processing options, wherein each image processing option comprises multiple weight coefficients, each weight coefficient corresponds to a score index, and each weight coefficient is used to represent the proportion of the score index corresponding thereto; based on a selection operation of a user on one of the multiple image processing options, displaying an image matched with the selected one of the image processing options. The image processing method of the application can meet different display requirements of different users for images in different scenarios, so as to improve the use experience of the user.

Description

Technical Field

[0001] This application relates to the field of image technology, and in particular to an image processing method, an image processing system, and a video surveillance system. Background Technology

[0002] Image processing of video streams containing captured targets is a key technology in video surveillance systems. By identifying, detecting, tracking, and evaluating targets within the video stream, and then outputting the results, images can be displayed to the user. Image processing techniques typically evaluate the acquired image based on one or more dimensions, such as target angle, image sharpness, or the degree of target occlusion. The images output through image processing techniques form the basis for subsequent downstream tasks such as target attribute analysis, target recognition, or target counting, enabling precise operations.

[0003] Image processing technology is widely used in scenarios such as, but not limited to, subway entrances and exits, security checkpoints, turnstiles, lobby entrances and exits, indoor corridors, and traffic roads. In these different application scenarios, users have varying preferences and needs regarding the images displayed by the image processing technology. However, existing image processing methods produce images that are too generic to meet diverse user needs, leading to a diminished user experience. Summary of the Invention

[0004] This application provides an image processing method, an image processing system, and a video surveillance system. The image processing method of this application can meet the different display needs of various users, thereby improving the user experience. Specifically, this application includes the following technical solutions:

[0005] In a first aspect, this application provides an image processing method, which includes: acquiring multiple images of the same target; displaying multiple image processing options, wherein each image processing option includes multiple weight coefficients, each weight coefficient corresponds to a scoring index, and each weight coefficient is used to represent the weight of its corresponding scoring index; and displaying an image that matches one of the selected image processing options based on a user's selection operation of one of the multiple image processing options.

[0006] In the image processing method of this application, different image processing options are used to achieve image display that tends to different evaluation indicators. Users can select the corresponding image processing options based on different image display needs and preferences, so that the displayed image can meet the user's image display needs and preferences, thereby achieving the user's customized image display effect, satisfying the user's different preferences and different display needs for images in different scenarios, and improving the user's user experience.

[0007] One implementation method further includes: evaluating the image quality of each image based on multiple scoring metrics to obtain a single image quality score corresponding to each scoring metric.

[0008] In this implementation, each acquired image is evaluated individually based on multiple scoring indicators and multiple weight coefficients corresponding to these indicators. This allows for evaluation of each image from multiple dimensions, ensuring the accuracy of the output and displayed image and thus improving the reliability of the image processing method.

[0009] One implementation method further includes: weighting the image quality score of each image based on multiple weight coefficients within one of the image processing options selected by the user, to obtain the total score for each image.

[0010] In this implementation, each image is weighted and calculated based on its individual image quality score for each rating indicator and multiple weight coefficients within the image processing options. This yields a comprehensive image quality score for each image, where each weight coefficient represents the proportion of each rating indicator among all rating indicators. When a user selects different image processing options based on their image display needs and preferences, the weight of a particular rating indicator relative to all rating indicators can be adjusted. This allows for adjustments to the perceived bias of an output image relative to multiple rating indicators, resulting in different dimensions of image output and enhancing the user experience.

[0011] One implementation method further includes: displaying the image with the highest total image quality score among multiple images.

[0012] In this implementation, after receiving one of the image processing options selected by the user, the image processing method of this application performs a weighted calculation on multiple image quality individual scores corresponding to each image based on multiple weight coefficients corresponding to the image processing option, thereby obtaining a comprehensive image quality score for each image. Then, by selecting the image with the highest total image quality score from among the multiple images—that is, selecting the image with the highest matching degree to one of the image processing options selected by the user—the method outputs and displays the image that best meets the user's needs and preferences, further enhancing the user experience and satisfying user requirements.

[0013] One implementation involves one or more weighting coefficients within each image processing option and one or more of any other image processing option.

[0014] In this implementation, at least one weight coefficient in any image processing option differs from at least one weight coefficient in any other image processing option. This results in different weight coefficient values ​​for the same scoring metric across different image processing options. In other words, different image processing options have different weighted calculation effects for each scoring metric, thus causing different image processing options to favor different scoring metrics and output images with varying effects. This satisfies users' different image preferences or display needs in different scenarios. Furthermore, multiple image processing options are displayed to the user and can be selected by the user. When the user selects any one of the multiple image processing options, images with different effects and preferences are output without requiring the user to sequentially set the values ​​of all weight coefficients in the same image processing option. This not only improves the display effect of user-customized images to meet different user needs but also simplifies the user customization process and operation, further enhancing the user experience.

[0015] One implementation involves giving one weight coefficient greater than the others for the same image processing option.

[0016] In this implementation, for the same image processing option, by setting the value of any one weight coefficient to be greater than the values ​​of the other weight coefficients, the weight of the scoring indicator with the largest weight coefficient value can be made greater than the weight of the other scoring indicators. This allows the output image to highlight the scoring indicator, enhances the image's tendency to correspond to different preferences and display needs, and further improves the user experience.

[0017] One implementation involves having equal weighting coefficients for one of a plurality of image processing options.

[0018] In this implementation, all weight coefficients in one of the image processing options have the same value, that is, the weight of the scoring index corresponding to each weight coefficient is the same, so that the image output by the image processing method can be compatible with each scoring index and present the effect of comprehensive display of the output image.

[0019] One implementation method further includes: displaying multiple weight coefficients of the selected image processing option based on a user's selection operation of one of multiple image processing options; and displaying an image that matches the selected image processing option based on a user's numerical adjustment operation of one or more weight coefficients of the selected image processing option.

[0020] In this implementation, upon receiving a user's selection of one of the image processing options, the system can simultaneously display each weight coefficient of that selected option to the user. This allows the user to clearly understand the weight of each weight coefficient corresponding to a different scoring metric within the current image processing option. Furthermore, the values ​​of one or more weight coefficients for one of the image processing options can be further customized by the user to enhance the interaction between the image processing method and the user, making the output and displayed images more aligned with the user's display needs and preferences.

[0021] One implementation involves a sum of multiple weight coefficients equal to 1, with each weight coefficient ranging from 0 to 1.

[0022] One implementation is that the sum of multiple weight coefficients within each image processing option is 1, and each weight coefficient is between 0 and 1.

[0023] In this implementation, when a user adjusts the value of at least one weight coefficient in each image processing option, the values ​​of other weight coefficients within the same image processing option are adjusted synchronously. This satisfies the requirement that the sum of the multiple weight coefficients within each image processing option equals 1, and the value of each weight coefficient is greater than or equal to 0 and less than or equal to 1. In other words, by defining the relationship between all weight coefficients within each image processing option, when a user adjusts or changes the value of at least one weight coefficient, the values ​​of the remaining weight coefficients are adjusted synchronously. This changes the weight of each scoring indicator corresponding to each weight coefficient, ensuring that the captured image output by the image processing method of this application meets the user's preferences and needs, further enhancing the user experience of the image processing method.

[0024] One implementation method further includes: acquiring multiple images of the same target, wherein the target is a person or a vehicle.

[0025] In this implementation, the image processing method can be used to acquire and process human face images, as well as to acquire and process vehicle body images, thereby outputting captured images that meet user preferences. This not only enhances the user experience but also expands the application scenarios and scope of the image processing method of this application.

[0026] One implementation method involves a human being as the target, multiple images being multiple face images, and multiple scoring metrics including one or more of the following: face sharpness, face 3D rotation angle, face landmarks, and face size.

[0027] In this implementation, each acquired face image is scored based on one or more of the following criteria: face clarity, face 3D rotation angle, face key points, and face size. This allows for the evaluation of each acquired face image from one or more dimensions, thereby meeting different user preferences for captured images and display needs, and further enhancing the user experience.

[0028] In one implementation, multiple image processing options include one or more of the following: face sharpness option, face 3D rotation angle option, face key point option, and face size option. Specifically, in the face sharpness option, a weight coefficient corresponding to face sharpness is greater than any other weight coefficient in the face sharpness option; in the face 3D rotation angle option, a weight coefficient corresponding to face 3D rotation angle is greater than any other weight coefficient in the face 3D rotation angle option; in the face key point option, a weight coefficient corresponding to face key points is greater than any other weight coefficient in the face sharpness option; and in the face size option, a weight coefficient corresponding to face size is greater than any other weight coefficient in the face size option.

[0029] In this implementation, the weight coefficients corresponding to face clarity, 3D face rotation angle, facial landmarks, and face size are respectively maximized among the face clarity, 3D face rotation angle, facial landmarks, and face size options. When the user selects any one of these options, the image processing method prioritizes face image selection based on these factors, thus ensuring the output face image leans towards the scoring criteria corresponding to the weight coefficients, satisfying user preferences and display needs in different scenarios. Furthermore, setting separate weight coefficients for face clarity, 3D face rotation angle, facial landmarks, and face size allows for adaptive adjustment of all weight coefficients within the selected image processing option based on user selection, making the image processing method more intuitive and user-friendly, and enhancing the user experience.

[0030] One implementation method further includes: acquiring one or more images based on a video stream; performing target matching on one or more images in the video stream based on a first condition to acquire one or more images corresponding to the same target, wherein the first condition includes a similarity greater than or equal to a first threshold.

[0031] In this implementation, by performing target matching on the same target, all video frames containing that target in the video stream can be obtained, ensuring that the obtained images correspond to the same target, thereby improving the comprehensiveness and accuracy of image acquisition. When the similarity of targets in multiple obtained images is greater than or equal to a first threshold, that is, the target information in the images meets a first condition, it can be determined that the images in each video frame that meet the first condition correspond to the same target. Target matching based on target similarity can improve the accuracy and efficiency of target matching in the image processing method of this application, further enhancing the user experience of using the image processing method.

[0032] Secondly, this application provides an image processing system, including an acquisition module and an interaction module. The acquisition module is used to acquire one or more images of the same target. The interaction module is used to display multiple image processing options and receive a user's selection operation on one of the multiple image processing options. The interaction module is also used to display an image that matches the image processing option selected by the user after receiving the user's selection operation on one of the image processing options.

[0033] The image processing system of this application offers different image processing options to achieve image display that is biased towards different evaluation indicators. Users can actively select one of the multiple image processing options through the interactive module, so that the image evaluation system can output images according to the user's needs and preferences, thereby achieving user-customized effects, meeting the user's different display preferences for images in different scenarios, and improving the user experience.

[0034] In one implementation, the image processing system further includes an evaluation module, which is used to evaluate the image quality of one or more images based on multiple scoring indicators and one of the image processing options selected by the user, so as to obtain a total image quality score corresponding to each image. Each image processing option includes multiple weight coefficients, each weight coefficient corresponds to a scoring indicator, and each weight coefficient is used to represent the proportion of its corresponding scoring indicator.

[0035] In this implementation, when the evaluation module assesses the image quality of each image based on each scoring metric, each weight coefficient represents the proportion of each scoring metric among all scoring metrics. At this point, by selecting different image processing options, the user can apply different weighted calculations to each image, adjusting the proportion of a scoring metric corresponding to that weight coefficient relative to all scoring metrics. This allows for adjusting the bias of the output image relative to multiple scoring metrics, thereby achieving different output and display effects for the image.

[0036] Thirdly, this application provides a video surveillance system, including a camera and an image processing system provided by any of the above implementations. The camera is used to acquire images and transmits the acquired video stream to the image processing system through the interaction module of the image processing system.

[0037] The video surveillance system of this application is equipped with the image processing system provided by any of the above implementations. Therefore, the video surveillance system of this application has all the possible beneficial effects of the image processing system provided by any of the above implementations. Attached Figure Description

[0038] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0039] Figure 1 This is a schematic diagram of the planar structure of the video surveillance system provided in the embodiments of this application;

[0040] Figure 2 This is a planar structural block diagram of the image processing system provided in the embodiments of this application;

[0041] Figure 3 This is a schematic diagram of the planar structure of the image processing system provided in the embodiments of this application;

[0042] Figure 4 A flowchart of the image processing method provided in the embodiments of this application;

[0043] Figure 5 This is a flowchart illustrating the workflow of the image processing method provided in the embodiments of this application.

[0044] Figure 6 This is a schematic diagram illustrating the workflow of the image processing method provided in the embodiments of this application;

[0045] Figure 7 A flowchart of the image processing method provided in the embodiments of this application;

[0046] Figure 8 This is a flowchart illustrating the process of evaluating multiple images using the image processing method provided in this embodiment of the application.

[0047] Figure 9 This is a schematic diagram illustrating the image quality evaluation of each image based on multiple scoring indicators in the image processing method provided in the embodiments of this application;

[0048] Figure 10A planar structural schematic diagram showing multiple image processing options for the image processing method provided in the embodiments of this application;

[0049] Figure 11 This is a schematic diagram of the weight coefficient configuration when the image processing option is a comprehensive display in the image processing method provided in the embodiments of this application;

[0050] Figure 12 This is a schematic diagram of multiple weighting coefficients in different image processing options in the image processing method provided in the embodiments of this application;

[0051] Figure 13 This is a flowchart illustrating the image processing method provided in an embodiment of this application. Detailed Implementation

[0052] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. Obviously, the described embodiments are merely some, and not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection claimed in this application.

[0053] Please see Figure 1 , Figure 1 This is a schematic diagram of the planar structure of the video surveillance system 1000 provided in an embodiment of this application. This application provides a video surveillance system 1000, which can be applied to fields such as industry, security, transportation, and daily life. Specifically, for example, but not limited to, monitoring systems are installed at subway entrances and exits, security checkpoints and exits, parking lot entrances and exits, lobby entrances and exits, industrial parks, indoor corridors, and highways. Figure 1 In the illustrated embodiment, the video surveillance system 1000 includes an image processing system 100 and one or more cameras 1001. The cameras 1001 are used to capture images or video streams and transmit the captured images or video streams to the image processing system 100. The images or video streams captured by the cameras 1001 can be transmitted to the image processing system 100 directly via a data cable connection or other means, or they can be transmitted to the image processing system 100 via wireless communication, network transmission, or other means.

[0054] The image processing system 100 analyzes and processes images or video streams captured by the camera 1001 to output images containing targets. For example, the image processing system 100 can process images or video streams captured by the camera 1001 to output captured images containing targets, which can then be used for subsequent target attribute analysis, recognition, counting, and other operations. Here, a target can be understood as an object to be monitored in the video surveillance system 1000; that is, a target can be at least one of a person or a vehicle. For example, when the video surveillance system 1000 is applied in the transportation field, the target can be vehicles in scenarios such as highways or parking lot entrances and exits. When the video surveillance system 1000 is applied in a security system, the target can be people in scenarios such as subway entrances and exits or security checkpoints.

[0055] Please see Figure 2 , Figure 2 This is a planar structural block diagram of the image processing system 100 provided in an embodiment of this application. Figure 2 In the illustrated embodiment, the image processing system 100 includes an acquisition module 10 and an interaction module 20. The acquisition module 10 is used to acquire one or more images of the same target, and the interaction module 20 is used to display multiple image processing options and receive a user's selection operation for one of the multiple image processing options. The interaction module 20 is also used to display an image that matches one of the image processing options selected by the user after receiving the user's selection operation for one of the image processing options.

[0056] For example, the image processing system 100 further includes an evaluation module 30, which is used to evaluate the image quality of one or more images based on multiple scoring indicators and one image processing option selected by the user, to obtain a total image quality score corresponding to each image. Each image processing option includes multiple weight coefficients, each weight coefficient corresponding to a scoring indicator, and each weight coefficient representing the proportion of its corresponding scoring indicator. Specifically, each weight coefficient representing the proportion of its corresponding scoring indicator can be understood as the proportion of a scoring indicator corresponding to a weight coefficient to all scoring indicators. This proportion can be understood as the ratio between a weight coefficient and the sum of all weight coefficients.

[0057] For example, the functions that the image processing system 100 can perform can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. That is, the image processing system 100 of this application can be implemented as a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware.

[0058] When implemented using software, it can be implemented wholly or partially as a computer program product. The computer program product includes one or more computer instructions. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. The computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0059] For example, the computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital versatile disc (DVD)), or a semiconductor medium (e.g., solid-state disk (SSD)). In this application embodiment, the computer-readable storage medium can be a non-volatile storage medium, that is, a non-transient storage medium.

[0060] Please see Figure 3 , Figure 3 This is a schematic planar structure diagram of the image processing system 100 provided in an embodiment of this application. Figure 3 In the illustrated embodiment, the image processing system 100 can be implemented through a combination of software and hardware. Specifically, the image processing system 100 includes a memory 101 and a processor 102, which can work together to implement some or all of the functions of one or more of the acquisition module 10, the interaction module 20, and the evaluation module 30 in the image processing system 100.

[0061] The processor 102 and memory 101 are connected via, but not limited to, a communication bus 105. The communication bus 105 can be divided into an address bus, a data bus, a control bus, etc. Figure 3 In the illustrated embodiment, the communication bus 105 is shown by a thick solid line, but the actual structural shape of the communication bus 105 is not limited to this.

[0062] For example, memory 101 is used to store data and program instructions. Specifically, memory 101 stores program instructions and data used by the image processing system 100 of this application to implement monitoring and snapshot functions. Memory 101 can be a read-only memory (ROM), a random access memory (RAM), an electrically erasable programmable read-only memory (EEPROM), an optical disc (including a compact disc read-only memory (CD-ROM), a compressed optical disc, a laser disc, a digital versatile optical disc, a Blu-ray disc, etc.), a magnetic disk storage medium, or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto.

[0063] For example, the processor 102 can execute program code stored in the memory 101. This program code may include one or more software modules, and the image processing system 100 can utilize the processor 102 and the program code in the memory 101. Specifically, the processor 102 is used to call data and program instructions from the memory 101, enabling the various functional modules and devices within the image processing system 100 to work collaboratively, thereby enabling the image processing system 100 to perform monitoring and image capture functions.

[0064] For example, processor 102 can be a general-purpose central processing unit (CPU), a network processor (NP), a microprocessor, or one or more integrated circuits, such as application-specific integrated circuits (ASICs), programmable logic devices (PLDs), or combinations thereof. Here, PLD can be understood as a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), generic array logic (GAL), or any combination thereof.

[0065] For example, there can be multiple processors 102, each of which can be a single-core processor or a multi-core processor. A processor 102 can be understood as one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions).

[0066] For example, each processor 102 may also be one or more CPUs.

[0067] For example, memory 101 may exist independently, or memory 101 may be integrated with processor 102.

[0068] For example, the image processing system 100 also includes an input device 103 and an output device 104, which can be used to implement the functions of the interaction module 20 in the image processing system 100.

[0069] The input device 103 can be used to receive user input commands, or it can also be used to receive one or more images acquired by the acquisition module 10 in the image processing system 100. The output device 104 can be used to display a video stream or display one or more images, or it can also be used to display instruction information that can be selected by the user.

[0070] For example, input device 103 may be, but is not limited to, at least one or more of a mouse, keyboard, touchscreen device, or sensing device, for inputting user commands. Output device 203 may be a display device for displaying video streams or one or more images, or for displaying selectable command information to the user, etc.

[0071] For example, input device 103 and output device 104 may also use any transceiver-like device for communication interfaces to communicate with other devices or communication networks. For example, when image processing system 100 is a cloud processor, it may, but is not limited to, receive one or more images captured by camera 1001 through acquisition module 10 via communication interface, or it may, but is not limited to, output one or more images and display them on a display screen via communication interface.

[0072] The communication interface includes wired communication interfaces and may also include wireless communication interfaces (not shown in the figure). Wired communication interfaces may be, for example, Ethernet interfaces, which can be optical interfaces, electrical interfaces, or combinations thereof. Wireless communication interfaces may be wireless local area network (WLAN) interfaces, cellular network communication interfaces, or combinations thereof.

[0073] It should be noted that, in Figure 3The embodiments shown are merely illustrative examples of one possible implementation of the types, communication methods, and arrangement of functional modules and devices within the image processing system 100, and do not limit the types, communication methods, and arrangement of functional modules and devices within the image processing system 100 in the embodiments shown in this application to this extent. In other embodiments of this application, the types, communication methods, and arrangement of functional modules and devices within the image processing system 100 can be adjusted according to actual application scenarios or actual design requirements, and this application does not specifically limit them.

[0074] Conventional image processing methods produce single-format images that cannot meet users' diverse display needs and preferences, thus failing to satisfy users' different capture requirements and reducing their overall experience.

[0075] This application provides an image processing method that can be applied to the image processing system 100 provided in the above embodiments, or it can be implemented through the functional devices or apparatuses in the image processing system 100 provided in the above embodiments. The image processing method of this application can meet the different display needs of different users for images, thereby improving the user experience of using the image processing method of this application.

[0076] The image processing method of this application can be applied to a video surveillance system 1000. For example, it can acquire and display the optimal captured image of a target from the images corresponding to the real-time monitored video stream to improve monitoring capabilities. This image processing method can also be applied to a video conferencing system. For example, it can acquire and display captured images of participants, such as attendees, from the images corresponding to the real-time conference video stream, and provide facial image quality evaluation results for user reference, analysis, and use. This image processing method can also be applied to mobile phone photography, such as selecting or capturing the most suitable image during focusing, or selecting or capturing the most suitable image during continuous shooting, or in a facial recognition system, such as selecting or capturing the most suitable facial image when recording a face unlock image, etc.

[0077] Next, in this application specification, the image processing system will be described exemplarily using the example of the image processing system applied to a video surveillance system with a human target.

[0078] Please see Figure 4 , Figure 4 This is a flowchart illustrating an image processing method provided in an embodiment of this application. This application provides an image processing method that can be implemented using the image processing system 100 provided in any of the above embodiments, or the image processing method provided in this application can be executed using the image processing system 100 provided in any of the above embodiments.

[0079] Specifically, the image processing method of this application includes the following steps:

[0080] S100. Acquire multiple images of the same target P;

[0081] Please combine Figure 4 See also Figure 5 and Figure 6 , Figure 5 This is a flowchart illustrating the workflow of the image processing method provided in the embodiments of this application. Figure 6 This is a schematic diagram illustrating the workflow of the image processing method provided in the embodiments of this application. Figures 4-6 In the illustrated embodiment, step S100, "acquiring multiple images of the same target P," can be achieved by directly or indirectly acquiring one or more images of the same target P through the acquisition module 10. The acquisition module 10 can be a camera 1001 provided by the video surveillance system 1000.

[0082] Specifically, when target P appears within the shooting range of camera 1001, camera 1001 will capture images of target P in real time to obtain a video stream of target P. The video stream of target P acquired by camera 1001 can be understood as several frames of images acquired during the period when camera 1001 captures images of target P. The period during which target P is captured can be understood as the period from the first frame of target P's appearance to the last frame of target P's appearance.

[0083] For example, one or more images of the same target P acquired by the acquisition module 10 can be understood as cropped images of several frames in the video stream containing several small images of the target P's face, or as partial images containing other parts of the target P. In this specification, an image is referred to as a small image containing the target P's face. For clarity and illustration, in this specification, several initial large images of the video stream captured by the acquisition module 10 when shooting the target P are represented as initial image U0, and small images of the target P's face after cropping several initial images are represented as image U. Image U can also be understood as the face image U of the target P described in this specification.

[0084] For example, the acquisition module 10 is used to acquire at least two images U of the same target P. It can also be understood that the video stream acquired by the camera 1001 shooting the target P is transmitted to the acquisition module 10. The acquisition module 10 detects the face of the same target P in several frames of initial images U0 in the video stream, thereby acquiring at least two face images U of the same target P, that is, acquiring at least two images U of the same target P.

[0085] In other words, in Figures 4-6In the illustrated embodiment, the acquisition module 10 can be used to directly acquire at least two images U of the target P. In this case, the acquisition module 10 can be a hardware device such as a camera 1001 to directly capture images U. The acquisition module 10 can also be used to detect the same target P in several initial images U0 of a video stream to acquire at least two images U of the same target P. In this case, the acquisition module 10 can be an internal software module that receives and processes the video stream captured by an external hardware device on the target P. The specific implementation of the acquisition module 10 is not limited in this application embodiment.

[0086] Understandable, Figure 6 In the embodiment shown in (b), the acquisition module 10 detects and captures images of the same target P to obtain a video stream of the target P. It can acquire several initial images U0 of the target P from the first frame image that appears in the acquisition module 10 to the last frame image that disappears in the acquisition module 10. By processing such as cropping the several initial images U0, several face images U of the target P can be acquired, that is, several images U of the target P can be acquired.

[0087] Please see Figure 7 , Figure 7 This is a flowchart illustrating the image processing method provided in an embodiment of this application. Figure 7 In the illustrated embodiment, step S100, "acquiring multiple images of the same target P," further includes:

[0088] S101. Acquire one or more images based on the video stream;

[0089] S102. Based on a first condition, perform target matching on one or more images in the video stream to obtain one or more images U corresponding to the same target P, wherein the first condition includes a similarity greater than or equal to a first threshold.

[0090] When the similarity of target P in multiple acquired images U is greater than or equal to a first threshold, meaning the target P information in image U satisfies a first condition, it can be determined that the images U in the video stream that meet this first condition correspond to the same target P. The similarity of target P can be understood as, but is not limited to, determining whether target P in different images of the video stream is the same target P based on, but not limited to, the external features, movement trajectories, etc., of target P in different images of the video stream, and the degree of overlap in their external features, movement trajectories, etc. For example, the external features of target P include, but are not limited to, clothing, hairstyle, facial features, etc.

[0091] The similarity of the target P is greater than or equal to the first threshold, which can be understood as the degree of overlap of the external features or movement trajectory of the target P in different images in the video stream being greater than or equal to the first threshold.

[0092] Understandably, by performing target P matching on the same target P, all video frames containing that target P in the video stream can be obtained, ensuring that the obtained images correspond to the same target P, thereby improving the comprehensiveness and accuracy of image acquisition. When the similarity of target P in multiple obtained images is greater than or equal to a first threshold, that is, the target P information in the images meets the first condition, it can be determined that the images in each video frame that meet the first condition correspond to the same target P. Target P matching based on target P similarity can improve the accuracy and efficiency of target P matching in the image processing method of this application, further enhancing the user experience of using the image processing method.

[0093] S200. Display multiple image processing options, wherein each image processing option includes multiple weight coefficients, each weight coefficient corresponds to a scoring index, and each weight coefficient is used to represent the weight of its corresponding scoring index.

[0094] S300: Based on the user's selection of one of a number of image processing options, display an image that matches one of the selected image processing options.

[0095] For example, please continue reading Figure 4 - Figure 6 In step S200, “display multiple image processing options”, and in step S300, “display an image that matches one of the selected image processing options based on the user’s selection operation of one of the multiple image processing options”, an image can be implemented or executed by the interaction module 20 in the image processing system 100, or it can run in a hardware device or apparatus such as a display in the image processing device 200.

[0096] Specifically, such as Figure 6 As shown in (d), the interaction module 20 is also used to display at least one image processing option, which the user can select based on the displayed multiple image processing options to suit their needs or preferences. For example, the multiple image processing options can also be displayed to the user via a display or the like in the image processing system 100. Furthermore, the user can also select any one of the at least one image processing option from the interaction module 20.

[0097] In the image processing method of this application, different image processing options are used to achieve image display that tends to different evaluation indicators. Users can select the corresponding image processing options based on different image display needs and preferences, so that the displayed image can meet the user's image display needs and preferences, thereby achieving the user's customized image display effect, satisfying the user's different preferences and different display needs for images in different scenarios, and improving the user's user experience.

[0098] It should be noted that, in Figures 4-6 In the illustrated embodiments, only a human target P and a face image U to be processed are used as examples for illustrative purposes. However, this does not limit the image processing method shown in the embodiments of this application to only be used for processing face images U. In other embodiments of this application, the image processing method can also be used to process at least two images U of a vehicle or other target P and output a captured image that meets the user's needs.

[0099] In other words, the multiple scoring indicators in the image processing method of this application can be adjusted according to different application scenarios or actual design requirements, and the embodiments of this application do not specifically limit them.

[0100] Understandably, the image processing system 100 can be used to acquire and process images of human faces, as well as images of vehicle bodies, and then output captured images that meet user preferences. This not only enhances the user experience but also expands the application scenarios and scope of the image processing system 100 of this application.

[0101] Furthermore, the image processing system 100 of this application can be used to implement the image processing method provided in any of the above embodiments. Therefore, the image processing system 100 of this application has all the possible beneficial effects of the image processing method provided in any of the above implementations.

[0102] Furthermore, since the video surveillance system 1000 of this application is equipped with the image processing system 100 provided by any of the above implementations, the video surveillance system 1000 of this application possesses all the possible beneficial effects of the image processing system 100 provided by any of the above implementations.

[0103] For example, please refer to Figures 4-7 See also Figure 8 , Figure 8 This is a flowchart illustrating the workflow of the image processing method provided in this application embodiment, which evaluates multiple images. Figure 8In the embodiment shown, the image processing system 100 further includes an evaluation module 30. The evaluation module 30 is used to evaluate the acquired multiple images before step S300, which is "based on the user's selection operation of one of the multiple image processing options, display an image that matches one of the selected image processing options", so as to output an image that matches one of the image processing options selected by the user, and can be displayed to the user through the interaction module 20.

[0104] For example, such as Figure 8 As shown, before step S300 "displaying an image that matches one of the selected image processing options based on the user's selection of one of the multiple image processing options", the following step is also included:

[0105] Image quality is evaluated for each image based on multiple scoring metrics to obtain a single image quality score corresponding to each scoring metric.

[0106] Specifically, the evaluation module 30 is used to perform a single-item image quality evaluation on each image U according to each scoring index, so as to obtain a single-item image quality score corresponding to each scoring index. By performing a single-item image quality evaluation on each acquired image based on multiple scoring indices and multiple weight coefficients corresponding to the multiple scoring indices, each image can be evaluated from multiple dimensions to ensure the accuracy of the output and displayed image, thereby improving the image processing reliability of the image processing method.

[0107] Please see Figure 9 , Figure 9 This is a schematic diagram illustrating the image quality evaluation of each image based on multiple scoring indicators in the image processing method provided in this application embodiment. Figure 9 In the illustrated embodiment, when the target P is a person, at least two images U are at least two face images, and multiple scoring metrics include at least one of face sharpness, face 3D rotation angle, face landmarks, and face size. Among these, in... Figure 9 In the embodiments shown in (a)-(d), the acquisition module 10 acquires three images U of the target P, which are respectively illustrated as the first image U1, the second image U2 and the third image U3.

[0108] exist Figure 9 In the embodiment shown in (a), facial sharpness is used to evaluate the clarity of faces in each acquired image U. Facial sharpness can be understood as the clarity of image U as directly observed by the user's eyes. The sharpness of the first image U1, the second image U2, and the third image U3 decreases sequentially. The evaluation module 30 performs a single-item image quality evaluation on each image U based on facial sharpness to obtain a single-item image quality score for facial sharpness corresponding to each image U. Figure 9 In the embodiment shown in (a), the image quality score of each image U based on the face clarity by the evaluation module 30 is illustrated as the first score a.

[0109] exist Figure 9 In the embodiment shown in (b), the three-dimensional face rotation angle is used to evaluate whether the pose of the target P's face in each image U is upright. Specifically, the more upright the face pose, the higher the accuracy of face recognition for the target P. The three-dimensional face rotation angle includes yaw, pitch, and roll, and the values ​​of each of the yaw, pitch, and roll angles can be between -90° and +90°.

[0110] Among them, the closer the yaw angle, pitch angle, and roll angle are to 0 degrees, the more upright the face in image U is. For example... Figure 9 As shown in (b), the absolute value of the yaw angle in the first image U1 is greater than 0 degrees, and the roll angle and pitch angle are both 0 degrees. The absolute value of the roll angle in the second image U2 is greater than 0 degrees, and the yaw angle and pitch angle are both 0 degrees. The absolute value of the pitch angle in the third image U3 is greater than 0 degrees, and the yaw angle and roll angle are both 0 degrees.

[0111] Evaluation module 30 performs a single-item image quality evaluation on each image U based on the three-dimensional rotation angle of the face, to obtain a single-item image quality score for each image U corresponding to the three-dimensional rotation angle of the face. Figure 9 In the embodiment shown in (b), the image quality score of each image U based on the three-dimensional rotation angle of the face by the evaluation module 30 is illustrated as the second score b.

[0112] exist Figure 9 In the embodiment shown in (c), facial landmarks are used to evaluate the visibility of the face of target P in each image U. Specifically, the more facial landmarks detected in image U, the higher the visibility of the face of target P in image U, i.e., the smaller the occluded area of ​​the face. Facial landmarks can be understood as facial features that are helpful for face recognition or face detection. Facial landmarks may include at least one of the following: eyes, nose, mouth, ears, chin, and eyebrows.

[0113] like Figure 9As shown in (c), all facial key points are visible in the first image U1, which can be understood as a high degree of visibility of the target P's face in the first image U1, with a small obscured area. In the second image U2, at least part of the target P's nose and mouth are obscured, thus the visibility of the target P's face in the second image U2 is lower than that in the first image U1. In the third image U3, the target P's mouth is completely obscured, and most of its nose is obscured, thus the visibility of the target P's face in the third image U3 is lower than that in both the first image U1 and the second image U2. That is, the visibility of the target P's face decreases sequentially in the first image U1, the second image U2, and the third image U3.

[0114] Evaluation module 30 performs a single-item image quality evaluation on each image U based on facial key points to obtain a single-item image quality score for each image U corresponding to the facial key points. Figure 9 In the embodiment shown in (c), the image quality score of each image U based on the facial key points by the evaluation module 30 is illustrated as the third score c.

[0115] Understandably, by detecting and recognizing at least one of the eyes, nose, mouth, ears, chin, and eyebrows in the face image of target P, it is possible to effectively and accurately identify the appearance features of target P, and thus distinguish the key facial features of different targets P.

[0116] It should be noted that, in Figure 9 In the embodiment shown in (c), only possible facial key points are described as examples, and the types and number of facial key points provided in this application embodiment are not limited to this. In other embodiments of this application, facial key points can be adjusted and set according to the user's actual needs. For example, when the user is more concerned about the details of a certain key feature or region of the face of the target P, the facial features can be decomposed into more key points. This application embodiment does not specifically limit this.

[0117] exist Figure 9 In the embodiment shown in (d), face size is used to evaluate whether the face area of ​​target P in each image U is appropriate. Face size can be understood as face area. Specifically, by setting a certain face area threshold, it is used to determine whether the face size of target P in each acquired image U is too large, or whether the face size of target P is too large. Figure 9 As shown in (d), the face sizes in the first image U1, the second image U2, and the third image U3 decrease sequentially. By evaluating the face sizes, low-quality face images that are too far away, have poor angles, or are too large can be filtered out, thereby improving the accuracy of the evaluation.

[0118] Evaluation module 30 performs a single-item image quality evaluation on each image U based on face size, to obtain a single-item image quality score for each image U corresponding to the face size. Figure 9 In the embodiment shown in (d), the image quality score of each image U based on the face size by the evaluation module 30 is illustrated as the fourth score d.

[0119] Understandable, Figure 9 In the embodiments shown in (a)-(d), each acquired face image is scored based on one or more of the following as scoring indicators: face clarity, face 3D rotation angle, face key points, and face size. This allows for the evaluation of each acquired face image from one or more dimensions, thereby meeting different user preferences for captured images and display needs, and further enhancing the user experience.

[0120] For example, the evaluation module 30 performs a single image quality score for each image U based on each scoring index. The evaluation can be performed by manual scoring or by a trained neural network, etc. This application embodiment does not specifically limit this.

[0121] Please continue reading. Figure 8 Before step S300, "based on the user's selection of one of a plurality of image processing options, display an image that matches one of the selected image processing options," the following step is also included:

[0122] Based on multiple weighting coefficients within one of the image processing options selected by the user, the image quality score for each image is weighted and calculated to obtain the total score for each image.

[0123] exist Figure 8 In the illustrated embodiment, in each of the at least one image processing options, each weight coefficient corresponds to a scoring indicator. This can be understood as the number of weight coefficients being the same as the number of scoring indicators, and each weight coefficient being used to indicate the proportion of one of the scoring indicators to all the scoring indicators.

[0124] In this embodiment, for each image processing option, the weight coefficient corresponding to facial sharpness is represented as sharpness weight, and the coefficient of sharpness weight is represented as first coefficient A. The weight coefficient corresponding to the three-dimensional rotation angle of the face is represented as angle weight, and the coefficient of angle weight is represented as second coefficient B. The weight coefficient corresponding to facial key points is represented as key point weight, and the coefficient of key point weight is represented as third coefficient C. The weight coefficient corresponding to facial size is represented as size weight, and the coefficient of size weight is represented as fourth coefficient D.

[0125] For example, in Figure 8In the illustrated embodiment, the evaluation module 30 performs score fusion processing on the individual image quality score and the corresponding weight coefficient for each image quality score of each image U to obtain the total image quality score F for each image U. Specifically, the total image quality score F for each image U is F = f(a, A, b, B, c, C, d, D), where f() can be understood as a fusion operation function.

[0126] For example, the total image quality score F for each image U satisfies the condition: F = a × A + b × B + c × C + d × D.

[0127] Understandably, when the evaluation module 30 evaluates the image quality of each image U using F = a × A + b × B + c × C + d × D, the user can adjust the value of at least one weight coefficient through the interaction module 20. This allows the user to adjust the weight of a scoring indicator corresponding to that weight coefficient among all scoring indicators, thereby ensuring that the captured image output by the interaction module 20 meets the user's needs, capture preferences, and requirements, and ultimately improves the user experience.

[0128] For example, the sum of multiple weight coefficients within each image processing option is 1, and each weight coefficient is between 0 and 1. Specifically, in Figure 8 In the illustrated embodiment, for each image processing option, the sharpness weight, angle weight, key point weight, and size weight satisfy the condition: A+B+C+D=1, and the values ​​of the first coefficient A, the second coefficient B, the third coefficient C, and the fourth coefficient D are all between 0 and 1.

[0129] Understandably, when a user adjusts the value of at least one weight coefficient in each image processing option, the values ​​of other weight coefficients within the same image processing option should be adjusted synchronously to meet the requirement that the sum of multiple weight coefficients within each image processing option is equal to 1, and the value of each weight coefficient is between 0 and 1.

[0130] In other words, by defining the relationship between all weight coefficients within each image processing option, when a user adjusts or changes the value of at least one weight coefficient, the values ​​of the remaining weight coefficients can be adjusted simultaneously. This allows the weight of the scoring indicators corresponding to each weight coefficient to be changed, so that the captured images output by the image processing method of this application meet the user's preferences and needs, further improving the user experience of the image processing method.

[0131] Understandably, the evaluation module 30 is used to perform a weighted calculation of each image quality score based on the weight coefficients corresponding to each scoring indicator in an image processing option, in order to obtain the total image quality score F corresponding to each image U. By performing a weighted calculation on each image based on the image quality score corresponding to each scoring indicator and multiple weight coefficients within the image processing option, a comprehensive image quality score for each image can be obtained.

[0132] Each weight coefficient represents the proportion of each rating indicator among all rating indicators. When a user selects different image processing options based on their image display needs and preferences, the weight of a rating indicator corresponding to that weight coefficient can be adjusted relative to all rating indicators. This allows for adjustments to the bias of the output image compared to multiple rating indicators, resulting in different dimensions of image output and improving the user experience.

[0133] For example, before step S300 "displaying an image that matches one of the selected image processing options based on the user's selection of one of a plurality of image processing options", the following step is also included:

[0134] Select the image with the highest total image quality score from among multiple images.

[0135] Understandably, upon receiving one of the image processing options selected by the user, the image processing method of this application performs a weighted calculation of multiple individual image quality scores for each image based on multiple weight coefficients corresponding to that image processing option, thereby obtaining a comprehensive image quality score for each image. Then, by selecting the image with the highest total image quality score from among the multiple images—that is, selecting the image with the highest matching degree to one of the image processing options selected by the user—and outputting and displaying it, the method can output and display the image that best meets the user's needs and preferences, further enhancing the user experience and satisfying user requirements.

[0136] exist Figure 8 In the illustrated embodiment, the evaluation module 30 is used to evaluate the image quality of each image U to obtain a total image quality score F for each image U (e.g., ...). Figure 5 As shown in the figure, the evaluation module 30 is used to evaluate the image quality of several face images U collected from the target P, so as to obtain the total image quality score F for each image U corresponding to the same target P.

[0137] The evaluation module 30 is used to evaluate the image quality of each image U based on multiple scoring indicators and at least one image processing option, so as to obtain the total image quality score F for each image U. Each image processing option includes multiple weighting coefficients, each weighting coefficient corresponding to a scoring indicator, and each weighting coefficient representing the proportion of its corresponding scoring indicator among all scoring indicators.

[0138] For example, after the evaluation module 30 evaluates the image quality of each image U of the same target P and obtains the total image quality score F of each image U, the evaluation module 30 can select the image U with the largest total image quality score F among at least two images U, and transmit the selected image U with the largest total image quality score F to the interaction module 20.

[0139] The evaluation module 30 and the related programs or algorithms that perform their respective functions can be stored in the memory of the image U processing device. When the image U is evaluated for image quality and the image U with the highest total image quality score F is selected, the processor 102 of the image U processing device can call the related programs or algorithms in the memory 101 to realize the functions of the evaluation module 30 in evaluating the image U for image quality and selecting the image U with the highest total image quality score F.

[0140] For example, the interaction module 20 is used to display the image U selected by the selection module 40 and show it to the user. The image U with the highest total image quality score F selected by the selection module 40 can be displayed to the user through the output device 104 of the interaction module 20.

[0141] Understandably, by evaluating the image quality of each image U based on multiple scoring metrics and corresponding weight coefficients, each image U can be evaluated from multiple dimensions to ensure the accuracy of the output captured image, thereby improving the performance and reliability of the image processing method. Furthermore, each weight coefficient in an image processing option corresponds to a scoring metric. When the evaluation module 30 evaluates the image quality of each image U based on each scoring metric, each weight coefficient represents the proportion of each scoring metric among all scoring metrics.

[0142] Understandably, by adjusting the value of at least one weighting coefficient, the proportion of a scoring indicator corresponding to that weighting coefficient among all scoring indicators can be adjusted, thereby adjusting the tendency of the output image U relative to multiple scoring indicators, and thus achieving the output of different dimensions of the captured image.

[0143] Furthermore, users can actively input at least one weighting coefficient through the interaction module 20 to adjust its value. This allows the image evaluation system to output captured images according to the user's needs and preferences, achieving user-customized effects and meeting different user preferences and display requirements in various scenarios, thus enhancing the user experience. Simultaneously, multiple weighting coefficients for at least one image processing option are displayed to the user through the interaction module 20, allowing the user to clearly understand the value of each weighting coefficient for each scoring indicator. This enables the user to know which scoring indicator the currently output captured image is more aligned with and the characteristics of the output captured image.

[0144] Please refer to the following: Figure 10 and Figure 11 , Figure 10 A planar structural schematic diagram showing multiple image processing options for the image processing method provided in the embodiments of this application. Figure 11 This diagram illustrates the weighting coefficient configuration when the image processing option is a comprehensive display in the image processing method provided in this embodiment. Figure 10 and Figure 11 In the illustrated embodiment, multiple image processing options are displayed to the user and made available for selection by the user through the interaction module 20 in the image processing method. When the user selects one of the image processing options and combines it with the image quality score output by the evaluation module 30, the user can adjust the value of one or more weighting coefficients among multiple weighting coefficients.

[0145] For example, in a plurality of image processing options, multiple weight coefficients for one of the image processing options are equal. Specifically, such as... Figure 10 and Figure 11 As shown, when multiple weight coefficients in one of the image processing options are equal, that is, the values ​​of sharpness weight, angle weight, key point weight and size weight satisfy the condition: A=B=C=D.

[0146] Among them, Figure 10 and Figure 11 In the illustrated embodiment, when all weight coefficients in one of the image processing options are equal so that each rating indicator of each image U has the same weight, then a “comprehensive display” can be shown to the user through the interaction module 20, but is not limited to, so that the user can select the image processing option.

[0147] It is understandable that all weight coefficients in at least one image processing option have the same value, that is, the weight of the scoring index corresponding to each weight coefficient is the same, so that the captured image output by the image processing method can be compatible with each scoring index and present the best overall effect of the captured image.

[0148] For example, one or more weight coefficients within each image processing option differ from one or more in any other image processing option. Specifically, the value of each weight coefficient corresponding to different scoring indicators in each image processing option can be adjusted based on different user needs, thereby achieving further differentiation settings between multiple image processing options and resulting in different output effects for captured images. That is, multiple weight coefficients corresponding to the same scoring indicator can be partially set to the same value or all set to different values ​​in different image processing options.

[0149] Understandably, at least one weight coefficient in any image processing option is different from at least one weight coefficient in any other image processing option, resulting in different values ​​for the weight coefficients corresponding to the same scoring indicator in different image processing options. In other words, different image processing options have different weighted calculation effects for each scoring indicator, thus causing different image processing options to favor different scoring indicators to output captured images with different effects, meeting users' different preferences or display needs for captured images in different scenarios.

[0150] Multiple image processing options are displayed to the user through the interaction module 20 and can be selected by the user. When the user selects any one of the multiple image processing options, the captured image with different effects and preferences can be output. This eliminates the need for the user to set the values ​​of all the weight coefficients in the same image processing option in turn. While improving the effect of the captured image displayed by the user to meet different user needs, it also simplifies the user customization process and operation, further enhancing the user experience.

[0151] Please see Figure 12 , Figure 12 This diagram illustrates multiple weighting coefficients among different image processing options in the image processing method provided in this embodiment. The multiple image processing options include one or more of the following: face sharpness option, face 3D rotation angle option, face key point option, and face size option.

[0152] exist Figure 12 In the illustrated embodiment, the image processing options also include a face clarity option, a face 3D rotation angle option, a face key point option, and a face size option, which are displayed as clarity, angle, key points, and size respectively on the display page of the interaction module 20, so that the user can select any one of the image processing options.

[0153] For example, one of the weight coefficients of at least one image processing option is greater than the other weight coefficients of the same image processing option.

[0154] Understandably, for the same image processing option, by setting the value of any one weight coefficient to be greater than the values ​​of the other weight coefficients, the weight of the scoring indicator with the largest weight coefficient value can be made greater than the weight of the other scoring indicators. This allows the output captured image to highlight the scoring indicator, enhances the tendency of the captured image to meet different preferences and display needs, and further improves the user experience.

[0155] For example, such as Figure 12 As shown in (a), in the face clarity option, one weighting coefficient corresponding to face clarity is greater than any other weighting coefficient in the face clarity option. That is, when the user selects the displayed clarity option, the value of the first coefficient A, which can be adjusted in the face clarity option, is greater than the value of the second coefficient B, the third coefficient C, and the fourth coefficient D, respectively, meaning that the value of the first coefficient A is the largest.

[0156] For example, such as Figure 12 As shown in (b), one of the weighting coefficients corresponding to the 3D face rotation angle in the face rotation angle option is greater than any other weighting coefficient in the face rotation angle option. That is, when the user selects the displayed angle option, the value of the second coefficient B in the face rotation angle option that can be adjusted is greater than the first coefficient A, the third coefficient C, and the fourth coefficient D, respectively, meaning that the value of the second coefficient B is the largest.

[0157] For example, such as Figure 12 As shown in (c), in the facial landmark options, one weight coefficient corresponding to a facial landmark is greater than any other weight coefficient in the facial clarity options. That is, when the user selects the displayed landmark options, the value of the third coefficient C, which can be adjusted in the facial landmark options, is greater than the first coefficient A, the second coefficient B, and the fourth coefficient D, respectively, meaning that the value of the third coefficient C is the largest.

[0158] For example, such as Figure 12 As shown in (d), in the face size option, one weighting coefficient corresponding to the face size is greater than any other weighting coefficient in the face size option. That is, when the user selects the displayed size option, the value of the fourth coefficient D, which can adjust the face size option, is greater than the first coefficient A, the second coefficient B, and the third coefficient C, respectively, meaning that the value of the fourth coefficient D is the largest.

[0159] For example, when a user selects one of the image processing options to maximize the value of the weight coefficient corresponding to one of the rating indicators, the values ​​of the other weight coefficients for the same image processing option selected by the user can be different.

[0160] Specifically, such as Figure 12 As shown in (d), when the user selects the display size option, the value of the fourth coefficient D can be set to 8, while the values ​​of the first coefficient A, the second coefficient B, and the third coefficient C can be different. That is, the values ​​of the other multiple weighting coefficients can be adjusted according to the application scenario and actual design requirements, or the display requirements of the captured image, etc., and this application embodiment does not specifically limit them.

[0161] Understandable, Figure 12 In the embodiments shown in (a)-(d), the weight coefficients corresponding to face clarity, face 3D rotation angle, face key points, and face size are respectively the largest among the face clarity option, face 3D rotation angle, face key points, and face size options. When the user selects any one of the face clarity option, face 3D rotation angle option, face key points option, and face size option, the image processing method can select face images with face clarity, face 3D rotation angle, face key points, and face size as evaluation priorities, respectively, so that the output face images tend to the scoring indicators corresponding to the weight coefficients, thereby meeting the user's preferences and display needs in different scenarios.

[0162] Meanwhile, options for face clarity, face 3D rotation angle, face key points, and face size are set separately. These options can adaptively adjust the values ​​of all weight coefficients within the selected image processing options based on the user's choices, making the image processing method of this application more intuitive and user-friendly, and improving the user experience.

[0163] For example, please refer to Figure 12 See also Figure 13 , Figure 13 This is a flowchart illustrating the image processing method provided in an embodiment of this application. Figure 12 and Figure 13 In the illustrated embodiment, step S300, "based on the user's selection of one of a plurality of image processing options, displaying an image that matches one of the selected image processing options," further includes the following steps:

[0164] S301. Based on the user's selection of one of a plurality of image processing options, display one or more weight coefficients of the selected image processing option;

[0165] S302. Based on the user's numerical adjustment operation of one or more weight coefficients in one of the image processing options, display an image that matches one of the selected image processing options.

[0166] Specifically, such as Figure 12 (d) and Figure 13As shown, each weight coefficient in one of the image processing options can be manually entered by the user to adjust the effect of each weight coefficient in an image processing option. When the value of each weight coefficient can be manually adjusted for the corresponding image processing option, the interface displayed to the user in the interaction module 20 can be shown as a custom setting.

[0167] Understandably, upon receiving a user's selection of an image processing option, the system can simultaneously display each weight coefficient of that selected option to the user. This allows the user to clearly understand the weight of each coefficient corresponding to a different scoring metric within the current image processing option. Furthermore, the values ​​of one or more weight coefficients for one image processing option can be further customized by the user to enhance the interaction between the image processing method and the user, making the output and displayed images more aligned with the user's display needs and preferences.

[0168] It should be noted that, in Figure 12 In the provided embodiments, only one possible embodiment of the display of one image processing option and the adjustment method of multiple weight coefficients in each image processing option in the interaction module 20 is described as an example, but it is not limited to the adjustment method of at least one weight coefficient in at least one image processing option provided in the embodiments of this application.

[0169] In other embodiments of this application, the display interface of the interaction module 20, the display method, arrangement order and naming of multiple image processing options, the adjustment method of at least one weight coefficient of each image processing option and the adjustment method of the remaining multiple weight coefficients can all be adjusted according to the application scenario, the actual needs of the user, and the actual design requirements of the image processing method. This application embodiment does not make specific limitations in this regard.

[0170] Of course, the above-described embodiments can be applied individually or in combination. The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications are also considered to be within the scope of protection of this application.

Claims

1. An image processing method, characterized in that, The image processing method includes: Acquire multiple images of the same target; Multiple image processing options are displayed, wherein each image processing option includes multiple weight coefficients, each weight coefficient corresponds to a scoring index, and each weight coefficient is used to represent the weight of the scoring index corresponding to it; Based on the user's selection of one of the plurality of image processing options, an image that matches one of the selected image processing options is displayed.

2. The image processing method according to claim 1, characterized in that, The method further includes: Image quality is evaluated for each image based on the multiple scoring indicators to obtain a single image quality score corresponding to each scoring indicator.

3. The image processing method according to claim 2, characterized in that, The image processing method further includes: Based on the multiple weight coefficients within one of the image processing options selected by the user, the image quality score for each image is weighted and calculated to obtain the total score for each image.

4. The image processing method according to claim 3, characterized in that, The method further includes: Display the image with the highest total image quality score among the plurality of images.

5. The image processing method according to any one of claims 1-4, characterized in that, One or more weighting coefficients within each of the image processing options are different from one or more weighting coefficients in any other image processing option.

6. The image processing method according to any one of claims 1-5, characterized in that, One of the weighting coefficients for the same image processing option is greater than the other weighting coefficients.

7. The image processing method according to any one of claims 1-6, characterized in that, One of the plurality of image processing options has the same weight coefficient.

8. The image processing method according to any one of claims 1-7, characterized in that, The method further includes: Based on the user's selection of one of the plurality of image processing options, display the plurality of weight coefficients of the selected image processing option; Based on the user's numerical adjustment of one or more of the weight coefficients in one of the image processing options, an image that matches the selected image processing option is displayed.

9. The image processing method according to claim 8, characterized in that, The sum of the plurality of weight coefficients is 1, and each of the weight coefficients is between 0 and 1.

10. The image processing method according to any one of claims 1-9, characterized in that, The method further includes: Acquire multiple images of the same target, wherein the target is a person or a vehicle.

11. The image processing method according to claim 10, characterized in that, The target is a person, and the multiple images are multiple facial images; The multiple scoring indicators include several of the following: facial clarity, facial 3D rotation angle, facial landmarks, and facial size.

12. The image processing method according to claim 11, characterized in that, The plurality of image processing options includes one or more of the following: face sharpness option, face 3D rotation angle option, face landmark option, and face size option, wherein: In the face clarity option, one of the weighting coefficients corresponding to the face clarity is greater than any other weighting coefficient in the face clarity option; In the face 3D rotation angle option, one of the weighting coefficients corresponding to the face 3D rotation angle is greater than any other weighting coefficient in the face 3D rotation angle option; In the facial key point option, one of the weighting coefficients corresponding to the facial key point is greater than any other weighting coefficient in the facial clarity option; In the face size option, one of the weighting coefficients corresponding to the face size is greater than any other weighting coefficient in the face size option.

13. The image processing method according to any one of claims 1-12, characterized in that, The method further includes: One or more of the images are acquired based on the video stream; Based on a first condition, target matching is performed on one or more images in the video stream to obtain one or more images corresponding to the same target, wherein the first condition includes a similarity greater than or equal to a first threshold.

14. An image processing system, characterized in that, The image processing system includes: The acquisition module is used to acquire one or more images of the same target; An interactive module is used to display multiple image processing options and receive a user's selection operation for one of the multiple image processing options; The interaction module is also configured to, after receiving a user's selection operation for one of the image processing options, display an image that matches one of the image processing options selected by the user.

15. The image processing system according to claim 14, characterized in that, The image processing system further includes an evaluation module, which is used to evaluate the image quality of one or more images based on the plurality of scoring indicators and one of the image processing options selected by the user, to obtain a total image quality score corresponding to each image, wherein: Each of the image processing options includes multiple weight coefficients, each weight coefficient corresponds to a scoring indicator, and each weight coefficient is used to represent the weight of its corresponding scoring indicator.

16. The image processing system according to claim 15, characterized in that, The evaluation module is used to perform a single image quality evaluation on each image according to each of the scoring indicators, so as to obtain a single image quality score corresponding to each of the scoring indicators; The evaluation module is further configured to perform a weighted calculation on each image quality individual score according to the weight coefficient corresponding to each of the scoring indicators in one of the image processing options, so as to obtain the total image quality score corresponding to each of the images.

17. A video surveillance system, characterized in that, The system includes a camera and an image processing system as described in any one of claims 14-16, wherein the camera is used to acquire images and transmit the acquired video stream to the image processing system via an interactive module of the image processing system.

Images