Camera video quality diagnosis method, device and storage medium

Abstract

The present invention discloses a camera video quality diagnosis method, device and storage medium, which belongs to the field of video surveillance / video management. The method comprises: obtaining a first video file recorded by a camera to be tested and a second video file recorded by a drone camera, wherein the first video file and the second video file are respectively recorded by the camera to be tested and the drone camera under the same shooting conditions; obtaining a first image and a second image within a preset time range from the first video file and the second video file respectively; and determining the video quality diagnosis result of the camera to be tested by performing feature analysis on the first image and the second image. The technical solution of the present invention can improve the accuracy of video quality diagnosis without optimizing the accuracy of the current video quality diagnosis algorithm and without performing additional modifications on ordinary cameras.

Description

Technical Field

[0001] The present invention relates to the field of video monitoring / video management, and in particular to a camera video quality diagnosis method, equipment and storage medium. Background Art

[0002] A large number of cameras are connected to the video management platform. How to determine whether the camera image is abnormal is a problem encountered during the operation of the video management platform. The existing technology directly performs algorithm feature analysis on the camera image. The video management platform submits the video stream or image of the camera to be analyzed to the quality diagnosis module. The quality diagnosis module performs feature analysis on the image through the algorithm. Different algorithms can be used to analyze the problems corresponding to the algorithm (such as snow, scrolling, blur, color cast, etc.).

[0003] However, the accuracy of the diagnostic results currently determined by various algorithms is low, and false positives often occur. Summary of the invention

[0004] The main purpose of the embodiments of the present invention is to provide a camera video quality diagnosis method, device and storage medium, aiming to improve the accuracy of video quality diagnosis.

[0005] To achieve the above-mentioned purpose, an embodiment of the present invention provides a camera video quality diagnosis method, which includes the following steps: obtaining a first video file recorded by a camera to be tested and a second video file recorded by a drone camera, wherein the first video file and the second video file are respectively recorded by the camera to be tested and the drone camera under the same shooting conditions; obtaining a first image and a second image within a preset time range from the first video file and the second video file respectively; and determining a video quality diagnosis result of the camera to be tested by performing feature analysis on the first image and the second image.

[0006] To achieve the above-mentioned purpose, an embodiment of the present invention also proposes a camera video quality diagnosis device, a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory. When the program is executed by the processor, the steps of the aforementioned camera video quality diagnosis method are realized.

[0007] To achieve the above objectives, the present invention provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the aforementioned camera video quality diagnosis method.

[0008] The camera video quality diagnosis method, device and storage medium proposed in the embodiments of the present invention use the second image taken by a drone camera in the same shooting environment as the camera to be tested as a standard, analyze and compare the first image taken by the camera to be tested with the second image taken by the drone camera, and diagnose the video quality of the camera to be tested, thereby improving the accuracy of video quality diagnosis without optimizing the existing video quality diagnosis algorithm or making additional modifications to ordinary cameras, and reducing or eliminating false alarms caused by the low accuracy of the diagnosis results of the existing video quality diagnosis algorithm itself. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a specific flow chart of camera video quality diagnosis according to an embodiment of the present invention.

[0010] Figure 2 It is the traditional system model diagram of the video management platform.

[0011] Figure 3 It is a structural diagram of a video management platform system according to an embodiment of the present invention.

[0012] Figure 4 yes Figure 3 The system shown implements a specific flow chart of diagnostic processing.

[0013] Figure 5 It is a structural block diagram of a camera video quality diagnosis device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0014] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0015] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present invention, and have no special meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.

[0016] like Figure 1 As shown, the embodiment of the present invention provides a camera video quality diagnosis method, such as Figure 1 As shown, the method may include the following steps:

[0017] Step S200: Obtain a first video file recorded by the camera to be tested and a second video file recorded by the drone camera, wherein the first video file and the second video file are respectively recorded by the camera to be tested and the drone camera under the same shooting conditions. Specifically, step 200 may include: the camera video quality diagnosis device first sends a recording instruction to the camera to be tested and the drone camera, so that the camera to be tested and the drone camera record the first video file and the second video file respectively according to the recording instruction, and then receives the first video file sent by the camera to be tested and the second video file sent by the drone camera.

[0018] The same shooting conditions may include: the drone camera is hovering near the camera to be tested, for example, side by side up and down or side by side left and right with the camera to be tested, and the orientation of the drone camera is consistent with the orientation of the camera to be tested; the same shooting conditions may also include: the focal length of the drone camera is consistent with the focal length of the camera to be tested. In this way, the drone camera can shoot the same shooting object as the camera to be tested, and the captured image is used as the diagnostic standard of the video quality of the camera to be tested.

[0019] Step S400: acquiring a first image and a second image within a preset time range from the first video file and the second video file respectively.

[0020] The preset time range may be a short time range, such as within 10 seconds or within 5 seconds. To improve diagnostic accuracy, the first and second images at the same time point are preferably obtained.

[0021] Step S600: Determine the video quality diagnosis result of the camera to be tested by performing feature analysis on the first image and the second image.

[0022] The camera video quality diagnosis device uses a first video quality diagnosis algorithm to perform feature analysis on the first image and the second image respectively to obtain a first image feature value and a second image feature value, and determines the video quality diagnosis result of the camera to be tested based on the difference between the first image feature value and the second image feature value. For example, if the difference between the first image feature value and the second image feature value is less than a preset range, it is determined that the video quality of the camera to be tested meets the requirements.

[0023] In order to improve the accuracy of camera video quality diagnosis, this embodiment may also adjust the drone camera before executing step S200.

[0024] Adjusting the drone camera may include adjusting the drone camera position and orientation, specifically: obtaining the position information and camera orientation information of the camera to be tested; sending a movement instruction carrying the position information and camera orientation information to the drone camera, so that the drone camera moves to a position near the camera to be tested (for example, up, down, left, and right side by side) and hovers according to the position information in the movement instruction, and adjusts the camera gimbal orientation to be consistent with the camera orientation of the camera to be tested according to the camera orientation information in the movement instruction.

[0025] Adjusting the drone camera may also include adjusting the focal length of the drone camera. There are various ways to adjust the focal length. For example, the focal length of the drone camera may be adjusted in advance; it may also be adjusted during the process of moving the drone camera to the camera to be tested. Specifically, the focal length information of the camera to be tested is obtained, and a movement instruction carrying the focal length information is sent to the drone camera, so that the drone camera adjusts the focal length according to the focal length information in the movement instruction; the focal length may also be adjusted after the drone camera is in place according to an image obtained by taking the drone camera and the camera to be tested of the same object.

[0026] Adjusting the drone camera may also include adjusting the camera pan / tilt of the drone camera, specifically: after sending a movement instruction carrying the position information and camera orientation information to the drone camera, receiving an arrival message sent by the drone camera after it moves to the vicinity of the position of the camera to be tested and the orientation is consistent with that of the camera to be tested; after receiving the arrival message, adjusting the camera pan / tilt of the drone camera to a specified position, specifically, sending a snapshot instruction to the camera to be tested and the drone camera; receiving a first snapshot image taken by the camera to be tested according to the snapshot instruction and multiple second snapshot images taken by the drone camera according to the snapshot instruction and corresponding multiple pan / tilt absolute positions, wherein each second snapshot image is a snapshot of the drone camera. The camera pan / tilt of the drone camera is obtained by shooting at different pan / tilt absolute positions; respectively determine the similarity between the first captured image and each of the second captured images, and obtain the pan / tilt absolute position corresponding to any second captured image whose similarity is within the similarity threshold range as the designated position. In order to improve the accuracy of video quality diagnosis, the pan / tilt absolute position corresponding to the second captured image with the highest similarity is preferably used as the designated position, wherein the pan / tilt absolute position is the position relative to the initial position of the pan / tilt, and the initial position of the pan / tilt may be the pan / tilt position when the device leaves the factory; send a moving instruction carrying the designated position information of the camera pan / tilt to the drone camera, so that the camera pan / tilt of the drone camera moves to the designated position, so that the camera pan / tilt records the second video file when at the designated position.

[0027] The respectively determining the similarity between the first snapshot image and each of the second snapshot images may include: determining the similarity between the first snapshot image and each of the second snapshot images by respectively comparing features between the first snapshot image and each of the second snapshot images.

[0028] Through the embodiments of the present invention, the accuracy of video quality diagnosis can be improved without optimizing the current video quality diagnosis algorithm and without additional modification of ordinary cameras. The implementation cost is low and the management and maintenance capabilities of the video management platform for cameras can be enhanced.

[0029] Take the video management platform in the security system as an example. Figure 2 As shown in the figure, the traditional security management platform and the corresponding video quality diagnosis system mainly include: management module, storage module, business processing module, access module, streaming media capability module, and video quality diagnosis module. The accuracy of the diagnosis in this implementation is determined by the introduction of the video quality diagnosis module, or more precisely, by the diagnosis algorithm of the video quality diagnosis module, and it is difficult to improve the accuracy of the algorithm itself. In addition, the accuracy of the diagnosis results of the algorithm itself is not high enough, and false alarms are still common.

[0030] The camera video quality diagnosis method of the present application is applied to a security system to form a security system that supports camera video quality diagnosis, which can effectively solve the above-mentioned limitations and shortcomings of the existing security system.

[0031] like Figure 3 As shown, the embodiment of the present invention provides a video management platform system (or management platform, video management system) including: a business processing module 10, a storage module 20, a camera management module 30, an access module 40, a streaming media capability module 50, a quality diagnosis module 60, a camera 70, and a drone camera 80. Among them, the camera 70 and the drone camera 80 are cameras that meet the access requirements of the video management platform.

[0032] The business processing module 10 is used to process business logic, such as recording, capturing and other logics, and has business capabilities such as recording and snapshot.

[0033] The camera management module 30 is used to manage the camera information, including the camera's latitude and longitude, installation height, camera pointing and other information.

[0034] Access module 40 is used for receiving and sending messages between the video management system and the camera, drone (or drone camera), wherein the messages are control messages of the camera, drone (or drone camera), specifically, for connecting to the control signaling of the camera, forwarding the control signaling sent to the camera by the business processing module, and receiving the control signaling reported by the camera to the business processing module.

[0035] The streaming media capability module 50 is used to connect to the camera media stream and forward the media stream to the user who needs to use it. It has the business capabilities of recording, snapshot, etc.

[0036] The quality diagnosis module 60 includes a quality diagnosis scheduling module for scheduling drone cameras and third-party video quality diagnosis systems, and also includes a video quality diagnosis module for analyzing image features. Specifically, a system that uses computer vision algorithms to extract image features usually requires GPU hardware and multiple vision algorithms.

[0037] The camera 70 is a device with video acquisition and network transmission capabilities, a spherical video encoder, and a pan / tilt head that can rotate 360 ​​degrees.

[0038] The drone camera 80 has wireless communication capabilities, GPS, etc. Its onboard camera has video acquisition capabilities and a pan / tilt that can rotate 360 ​​degrees for obtaining comparison pictures of the camera position to be tested.

[0039] Figure 3 The video management platform system (or management platform, video management system) shown in the figure performs a camera video quality diagnosis method, which includes the following steps: the management platform initiates camera video quality diagnosis; the management platform obtains the information of the camera to be tested, including latitude and longitude, camera direction, installation height and other information, and assigns the drone camera to move to the camera to be tested and put it side by side with the camera to be tested; the management platform uses the current picture of the camera to be tested as the standard, moves the drone camera pan to the same direction of the camera to be tested, so that the picture taken by it is consistent with the camera to be tested; the management platform simultaneously triggers the camera to be tested and the drone camera to record, and obtains the picture of the camera to be tested and the drone camera at the same time from the recording, the drone camera picture is the standard picture, and the camera to be tested is the picture to be diagnosed; the management platform uses the standard picture to perform feature analysis according to a certain algorithm to obtain the standard feature value. The management platform uses the picture to be tested to perform feature analysis according to the same algorithm to obtain the feature value to be tested; the management platform compares the feature value to be tested with the standard feature value to obtain the feature difference value, and if the feature difference value is similar, it means that the camera to be tested is normal.

[0040] This embodiment can improve the accuracy of diagnosis without optimizing the accuracy of the current algorithm. At the same time, no additional modification of ordinary cameras is required, the implementation cost is low, and the management and maintenance capabilities of the video management platform for cameras can be improved.

[0041] Figure 4 yes Figure 3 The specific flow chart of the system to implement the diagnostic processing is as follows: Figure 4 As shown, the following steps are included:

[0042] Step S100: Start video diagnosis for a certain camera, which is now the camera to be tested.

[0043] Step S101: the quality diagnosis module 60 requests the camera management module 30 for information about the camera to be tested.

[0044] Step S102: The quality diagnosis module 60 obtains the information of the camera to be tested and extracts the required key information, such as longitude and latitude, installation height, camera direction and other information.

[0045] Step S103: The quality diagnosis module 60 sends a drone movement instruction to the access module 40, which includes information about the camera to be tested, such as latitude and longitude, installation height, camera direction, etc.

[0046] Step S104: The access module 40 forwards the drone movement instruction to the drone camera 80.

[0047] Step S105: The drone camera 80 receives the drone movement instruction, moves to the destination, hovers after reaching the destination, and moves the camera pan to be consistent with the direction of the camera to be tested.

[0048] Step S106: The drone camera 80 returns an arrival message after arriving at the destination.

[0049] Step S107: the access module 40 forwards the arrival message to the quality diagnosis module 60.

[0050] Step S108: After the drone arrives at the designated location, the quality diagnosis module 60 sends a capture instruction to the camera to be tested and the drone camera at the same time.

[0051] Step S109: The access module 40 forwards the capture instruction to the drone camera 80.

[0052] Step S110: The access module 40 forwards the capture instruction to the camera 70.

[0053] Step S111: The drone camera 80 moves the pan / tilt within a certain range of the current pan / tilt position, while taking pictures and recording the current pan / tilt absolute position.

[0054] Step S112: The drone camera 80 uploads multiple pictures to the access module 40, and at the same time, the gimbal absolute position information corresponding to the pictures is attached.

[0055] Step S113: The camera 70 uploads the picture to the access module 40.

[0056] Step S114: the access module 40 returns the received pictures taken by the camera to be tested and the pictures taken by the multiple drone cameras 80 to the quality diagnosis module 60 at the same time.

[0057] Step S115: The quality diagnosis module 60 performs feature analysis and comparison on the received pictures, and takes the absolute position of the gimbal corresponding to the drone camera picture with the highest similarity as the best position.

[0058] Step S116: The quality diagnosis module 60 sends a drone camera pan / tilt movement instruction to the access module 40, and the instruction moves the pan / tilt of the mobile drone camera 80 to an optimal position.

[0059] Step S117: The access module 40 forwards the pan / tilt movement instruction.

[0060] Step S118: The drone camera 80 moves the pan / tilt to the optimal position.

[0061] Step S119: The drone camera 80 returns a response indicating that the pan / tilt head has reached the specified position.

[0062] Step S120: The access module 40 forwards the PTZ arrival instruction (ie, the response of the PTZ reaching the specified position) to the quality diagnosis module 60.

[0063] Step S121: The quality diagnosis module 60 sends a video recording instruction to the camera 70 and the drone camera 80 simultaneously.

[0064] Step S122: The access module 40 forwards the video recording instruction to the camera 70.

[0065] Step S123: The access module 40 forwards the video recording instruction to the drone camera 80.

[0066] Step S124: The drone camera 80 completes recording.

[0067] Step S125: The camera 70 completes recording.

[0068] Step S126: The access module 40 forwards the video recording completion message to the quality diagnosis module 60.

[0069] Step S127: The quality diagnosis module 60 obtains the pictures at the same time point in the video file of the camera to be tested and the video file of the drone camera, and obtains the picture of the camera to be tested and the picture of the drone camera (ie, the standard picture).

[0070] Step S128: The quality diagnosis module 60 performs feature analysis on the image to be tested to obtain a feature value to be tested.

[0071] Step S129: The quality diagnosis module 60 performs feature analysis on the standard image to obtain a standard feature value.

[0072] Step S130: The quality diagnosis module 60 compares the obtained characteristic value to be tested and the standard characteristic value, and determines whether the image quality of the camera to be tested is normal according to the difference between the two characteristic values.

[0073] Step S131: Complete video diagnosis.

[0074] This embodiment does not require optimization of the accuracy of the current algorithm, does not require additional modifications to ordinary cameras, has high diagnostic accuracy, and has low implementation costs, and can enhance the management and maintenance capabilities of the video management platform for cameras.

[0075] like Figure 5 As shown, an embodiment of the present invention provides a camera video quality diagnosis device 20, which includes a memory 21, a processor 22, a program stored in the memory and executable on the processor, and a data bus 23 for realizing connection and communication between the processor 21 and the memory 22. When the program is executed by the processor, the steps of the aforementioned camera video quality diagnosis method are realized.

[0076] An embodiment of the present invention further provides a storage medium storing one or more programs, which can be executed by one or more processors to implement the steps of the aforementioned camera video quality diagnosis method.

[0077] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0078] In hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0079] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the present invention is not limited thereby. Any modification, equivalent substitution and improvement made by those skilled in the art without departing from the scope and essence of the present invention shall be within the scope of the present invention.

Claims

1. A camera video quality diagnosis method, characterized in that: The method comprises the following steps: Obtain a first video file recorded by the camera to be tested and a second video file recorded by the drone camera, wherein the first video file and the second video file are respectively recorded by the camera to be tested and the drone camera under the same shooting conditions; Acquire a first image and a second image within a preset time range from the first video file and the second video file respectively; Determine the video quality diagnosis result of the camera to be tested by performing feature analysis on the first image and the second image; The same shooting conditions include: the drone camera and the camera to be tested are arranged side by side vertically or side by side horizontally, the orientation of the drone camera is consistent with the orientation of the camera to be tested, and the focal length of the drone camera is consistent with the focal length of the camera to be tested.

2. The method according to claim 1, characterized in that: Obtaining the first video file recorded by the camera to be tested and the second video file recorded by the drone camera includes: Sending a video recording instruction to the camera to be tested and the drone camera, so that the camera to be tested and the drone camera respectively record the first video file and the second video file according to the video recording instruction; Receive a first video file sent by the camera to be tested and a second video file sent by the drone camera.

3. The method according to claim 1, characterized in that The step of determining the video quality diagnosis result of the camera to be tested by performing feature analysis on the first image and the second image comprises: Using a first video quality diagnosis algorithm to perform feature analysis on the first image and the second image respectively, to obtain a first image feature value and a second image feature value; A video quality diagnosis result of the camera to be tested is determined according to a difference between the first image characteristic value and the second image characteristic value.

4. The method according to claim 3, characterized in that Determining the video quality diagnosis result of the camera to be tested according to the difference between the first image characteristic value and the second image characteristic value comprises: If the difference between the first image characteristic value and the second image characteristic value is less than a preset range, it is determined that the video quality of the camera to be tested meets the requirements.

5. The method according to any one of claims 1 to 4, characterized in that: Before obtaining the first video file recorded by the camera to be tested and the second video file recorded by the drone camera, the method further includes: Obtaining the position information and camera orientation information of the camera to be tested; A movement instruction carrying the position information and the camera orientation information is sent to the drone camera, so that the drone camera moves to the position of the camera to be tested and hovers according to the position information in the movement instruction, and adjusts the camera pan orientation to be consistent with the camera orientation of the camera to be tested according to the camera orientation information in the movement instruction.

6. The method according to claim 5, characterized in that After sending a movement instruction carrying the position information and the camera orientation information to the drone camera, the method further includes: Receiving an arrival message sent by the drone camera after the drone camera moves to the vicinity of the location of the camera to be tested and has the same orientation as the camera to be tested; After receiving the arrival message, the camera pan / tilt of the drone camera is adjusted to a specified position.

7. The method according to claim 6, characterized in that The step of adjusting the camera pan / tilt of the drone camera to a specified position comprises: Sending a capture instruction to the camera to be tested and the drone camera; Receive a first captured image taken by the camera to be tested according to the captured instruction and a plurality of second captured images taken by the drone camera according to the captured instruction and a plurality of corresponding pan-tilt absolute positions, wherein each second captured image is captured by the camera pan-tilt of the drone camera at a different pan-tilt absolute position; Determine the similarity between the first captured image and each of the second captured images respectively, and obtain the absolute position of the pan / tilt corresponding to any second captured image whose similarity is within a similarity threshold range as the designated position; A moving instruction of the designated position of the camera pan head is sent to the drone camera, so that the camera pan head of the drone camera moves to the designated position, so that the camera pan head records the second video file when it is at the designated position.

8. The method according to claim 7, characterized in that The respectively determining the similarity between the first captured image and each of the second captured images comprises: The similarity between the first snapshot image and each of the second snapshot images is determined by performing feature comparison between the first snapshot image and each of the second snapshot images.

9. A camera video quality diagnostic device, characterized in that: The device includes a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory. When the program is executed by the processor, the steps of the camera video quality diagnosis method as described in any one of claims 1 to 8 are realized.

10. A storage medium for computer-readable storage, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the camera video quality diagnosis method according to any one of claims 1 to 8.

Images