Video quality assessment methods, apparatus, storage media and computer equipment

Abstract

This application provides a video quality assessment method, apparatus, storage medium, and computer equipment. The method includes: extracting video frames from a monitoring video of a workbench, wherein the surface of the workbench is provided with an identifier of a target shape; extracting the contour of the video frames according to a target contour extraction rule to obtain a target extraction result, wherein the target contour extraction rule is used to extract the contour corresponding to the target shape; and determining the video quality of the monitoring video based on the target extraction result. This application determines the video quality of the monitoring video based on the contour extraction result of the target shape, which can weaken the influence of environmental factors such as position, distance, shooting angle, shooting light, and video color difference on the video quality assessment result, thereby improving the assessment accuracy and ensuring that the assessed video quality accurately reflects the actual video quality of the monitoring video.

Description

Technical Field

[0001] This application relates to the field of warehouse management technology, and in particular to a video quality assessment method, apparatus, storage medium, and computer equipment. Background Technology

[0002] In e-commerce sales, in order to clarify the status of goods during the shipping process and provide after-sales service to customers, some e-commerce sales companies install monitoring systems in warehouses and on workbenches to film the warehouse and workbenches respectively, so that the monitoring video can record the status of goods during the shipping process.

[0003] To ensure that surveillance video can provide evidence for pre-sales and after-sales transactions, e-commerce sales companies need to establish monitoring and alarm mechanisms for video quality, surveillance system hardware, and / or the hardware status of the surveillance system. This allows for the early detection of problems with the surveillance video and the dispatch of work orders to notify maintenance personnel for intervention. Video quality is used to assess whether the surveillance video clearly records the status of the goods. For example, video quality can reflect whether the camera angle of the surveillance system is directly facing the workbench when the hardware is functioning correctly, thus determining whether the surveillance video clearly records the status of the goods.

[0004] However, existing technologies are too susceptible to environmental influences when assessing the video quality of surveillance videos, resulting in low accuracy of assessment results. Summary of the Invention

[0005] The purpose of this application is to at least address one of the aforementioned technical deficiencies, particularly the technical deficiency of low accuracy in evaluation results in the prior art.

[0006] In a first aspect, embodiments of this application provide a video quality assessment method, the method comprising:

[0007] Video frames are extracted from the monitoring video of the workbench, the surface of which is provided with an identifier of the target shape;

[0008] The video frame is subjected to contour extraction according to the target contour extraction rule to obtain the target extraction result. The target contour extraction rule is used to extract the contour corresponding to the target shape.

[0009] The video quality of the surveillance video is determined based on the target extraction results.

[0010] In one embodiment, the step of extracting video frames from the monitoring video of the workbench includes:

[0011] The latest video frame is periodically extracted from the monitoring video, which is the real-time monitoring video of the workbench;

[0012] The step of extracting contours from the video frame according to the target contour extraction rules to obtain the target extraction result includes:

[0013] For each newly extracted video frame, the latest video frame is subjected to contour extraction according to the target contour extraction rules to obtain the target extraction result corresponding to the latest video frame;

[0014] The step of determining the video quality of the surveillance video based on the target extraction result includes:

[0015] The video quality of the surveillance video is determined based on the target extraction results corresponding to the latest extracted video frame.

[0016] In one embodiment, the step of determining the video quality of the surveillance video based on the target extraction result corresponding to the latest extracted video frame includes:

[0017] If the target extraction result corresponding to the latest extracted video frame is empty, and the target extraction result corresponding to the latest extracted video frame in the previous two extractions is empty, then the video quality of the monitoring video is determined to be unqualified.

[0018] If the target extraction result corresponding to the latest extracted video frame is not empty, then the video quality of the monitoring video is determined to be qualified.

[0019] In one embodiment, the target shape is a five-pointed star, the object color of the object being identified is red, and the target contour extraction rules include color extraction rules, area extraction rules, and angle number extraction rules.

[0020] The step of extracting contours from the video frame according to the target contour extraction rules to obtain the target extraction result includes:

[0021] Contours that simultaneously satisfy the color extraction rule and the area extraction rule are extracted from the video frames to obtain preliminary contour extraction results; wherein, the color extraction rule is used to extract the region contour corresponding to the enclosed area that matches the object color of the object being identified, and the area extraction rule is used to extract the region contour corresponding to the enclosed area that matches the object area of ​​the object being identified.

[0022] If the preliminary contour extraction result is not empty, then curve fitting is performed on each contour in the preliminary contour extraction result to obtain each fitted contour.

[0023] The fitted contours are filtered based on the angle number extraction rules to obtain the angle rule filtering results, and the target extraction results are obtained based on the angle rule filtering results.

[0024] In one embodiment, the step of obtaining the target extraction result based on the filtering result includes:

[0025] For each fitted contour in the angle rule filtering result, the image moment corresponding to the fitted contour is calculated. If the image moment corresponding to the fitted contour matches the preset image moment, convex hull detection is performed on the fitted contour. Based on the detection result of convex hull detection, it is determined whether to add the fitted contour to the target extraction result. If so, the fitted contour is added to the target extraction result.

[0026] In one embodiment, the method further includes:

[0027] Obtain the region ratio, and determine the region of interest in the video frame based on the region ratio;

[0028] The regions in the video frame other than the region of interest are converted to grayscale to obtain a grayscale-processed video frame.

[0029] The step of extracting contours from the video frame that simultaneously satisfy the color extraction rule and the area extraction rule to obtain preliminary contour extraction results includes:

[0030] Contours that simultaneously satisfy the color extraction rules and the area extraction rules are extracted from the grayscale video frames to obtain the preliminary contour extraction results.

[0031] In one embodiment, the method further includes: if the video quality of the monitored video is unqualified, sending a work order to the maintenance personnel.

[0032] Secondly, embodiments of this application provide a video quality assessment device, the device comprising:

[0033] A video frame extraction module is used to extract video frames from the monitoring video of the workbench, the surface of which is provided with a target shape identifier;

[0034] The target extraction result acquisition module is used to extract the contour of the video frame according to the target contour extraction rule to obtain the target extraction result. The target contour extraction rule is used to extract the contour corresponding to the target shape.

[0035] The video quality acquisition module is used to determine the video quality of the monitoring video based on the target extraction results.

[0036] Thirdly, embodiments of this application provide a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the video quality assessment method described in any of the above embodiments.

[0037] Fourthly, embodiments of this application provide a computer device, including: one or more processors, and a memory;

[0038] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the video quality assessment method described in any of the above embodiments.

[0039] In the video quality assessment method, apparatus, storage medium, and computer equipment provided in this application, the surface of the workbench is provided with an identifier of the target shape. The computer equipment can extract video frames from the monitoring video of the workbench as monitoring images, and perform contour extraction on the monitoring images using target contour extraction rules to obtain extraction results. These extraction results can be used to determine whether a contour corresponding to the target shape exists in the monitoring image. Based on these results, it can be determined whether the shooting angle of the monitoring system is directly facing the workbench, thereby determining whether the monitoring video can clearly record the state of the goods, and thus obtaining the video quality of the monitoring video. Since the accuracy of contour extraction is less affected by environmental factors, this application determines the video quality of the monitoring video based on the contour extraction results of the target shape. This can weaken the influence of environmental factors such as position, distance, shooting angle, shooting light, and video color difference on the video quality assessment results, thereby improving the assessment accuracy and ensuring that the assessed video quality accurately reflects the actual video quality of the monitoring video. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art 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 based on these drawings without creative effort.

[0041] Figure 1 This is one of the flowcharts illustrating a video quality assessment method in one embodiment;

[0042] Figure 2 This is a second flowchart illustrating a video quality assessment method in one embodiment;

[0043] Figure 3 This is a flowchart illustrating the contour extraction steps based on target contour extraction rules in one embodiment.

[0044] Figure 4 This is the third flowchart illustrating a video quality assessment method in one embodiment;

[0045] Figure 5 This is a structural block diagram of a video quality assessment device in one embodiment;

[0046] Figure 6 This is a schematic structural diagram of a computer device in one embodiment. Detailed Implementation

[0047] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0048] As mentioned in the background section, existing technologies for evaluating the video quality of surveillance videos are greatly affected by environmental factors, resulting in low accuracy. The inventors discovered that this problem stems from the fact that existing technologies use QR codes affixed to the workbench to assess video quality. If the QR code is recognized, the video quality is deemed acceptable, the surveillance system's camera angle is directly facing the workbench, and the video clearly records the product's condition. If the QR code cannot be recognized, the video quality is deemed unacceptable, the surveillance system's camera angle is inaccurate, and the video fails to clearly record the product's condition.

[0049] However, the recognition results of QR code recognition are greatly affected by the environment. Factors such as the placement of the QR code, the location of the monitoring system, the distance between the monitoring system and the QR code, the shooting angle of the monitoring system, the lighting conditions when the monitoring video is captured, and the color difference of the monitoring video can all affect the QR code recognition results, greatly reducing the accuracy of QR code recognition and resulting in inaccurate evaluation results.

[0050] To address the aforementioned problems, embodiments of this application provide a video quality assessment method, apparatus, storage medium, and computer equipment. A target shape identifier is provided on the surface of a workbench. The computer equipment can extract video frames from the monitoring video of the workbench as monitoring images, and perform contour extraction on the monitoring images using target contour extraction rules to obtain extraction results. These extraction results can be used to determine whether a contour corresponding to the target shape exists in the monitoring image. Based on these results, it can be determined whether the shooting angle of the monitoring system is directly facing the workbench, thereby determining whether the monitoring video can clearly record the state of the goods, and thus obtaining the video quality of the monitoring video. Since the accuracy of contour extraction is less affected by environmental factors, this application determines the video quality of the monitoring video based on the contour extraction results of the target shape. This weakens the influence of environmental factors such as position, distance, shooting angle, shooting light, and video color difference on the video quality assessment results, thereby improving the assessment accuracy and ensuring that the assessed video quality accurately reflects the actual video quality of the monitoring video.

[0051] In one embodiment, this application provides a video quality assessment method. The following embodiments illustrate the application of this method to a computer device. It can be understood that the computer device may be, but is not limited to, various types of terminals or servers. When the computer device is a cloud platform server, this application can horizontally expand containers by accessing the cloud platform server to improve recognition efficiency in a multi-threaded manner.

[0052] like Figure 1 As shown, the video quality assessment method of this application specifically includes the following steps:

[0053] S102, extract video frames from the monitoring video of the workbench, the surface of which is provided with an identifier of the target shape.

[0054] The surface of the workbench is provided with an identifier of the target shape. It is understood that the specific shape of the target shape can be determined according to actual needs, and this application does not impose specific limitations in this regard. Furthermore, the color of the identifier can also be determined according to actual needs, and this application does not impose specific limitations in this regard either. In one example, to reduce misidentification and improve the accuracy of the evaluation results, the target shape can be a five-pointed star, and the color of the identifier can be red; in other words, the identifier is a red five-pointed star.

[0055] Meanwhile, the identifier can be placed on the surface of the workbench used for storing or transporting goods, and the specific placement of the identifier can be determined according to the location of the goods on the workbench; this application does not impose specific limitations in this regard. In one example, the identifier can be placed on the surface furthest from the ground.

[0056] The monitoring video of the workbench refers to the video obtained after the workbench is captured by monitoring equipment. This monitoring video may include multiple video frames, and this application can extract one or more video frames from the monitoring video. In one embodiment, the computer device can extract any video frame from the monitoring video and process the extracted video frames using the following steps to obtain the video quality of the monitoring video. In one embodiment, the computer device can extract video frames from the monitoring video according to pre-set video frame extraction rules and process the extracted video frames using the following steps to obtain the video quality of the monitoring video.

[0057] S104, Perform contour extraction on the video frame according to the target contour extraction rule to obtain the target extraction result. The target contour extraction rule is used to extract the contour corresponding to the target shape.

[0058] Specifically, after extracting the video frames, the computer device can perform contour extraction on the extracted video frames to attempt to extract the contour corresponding to the target shape from the video frames and obtain the target extraction result. It can be understood that when the monitoring equipment's shooting angle is directly facing the workbench, the monitoring video can capture the entire object to be identified, ensuring that the extracted video frames contain a complete image of the object. In this case, the computer device can extract the contour corresponding to the target shape from the video frames, and the target extraction result will not be empty.

[0059] When the camera angle of the surveillance equipment deviates from its intended position, the surveillance video may not capture the entire object to be identified. Therefore, the object may not be present in the extracted video frames, or only a portion of the object may be captured. In such cases, the computer equipment cannot extract the contour corresponding to the target shape from the video frames.

[0060] It is understood that contour extraction can be achieved using any type of contour extraction algorithm based on any principle in the prior art, and this application does not impose any specific restrictions on it.

[0061] S106, Determine the video quality of the surveillance video based on the target extraction result.

[0062] Specifically, the computer equipment can determine whether the monitoring video is directly facing the workbench based on the target extraction results, and thus judge the video quality. When the video quality is acceptable, it indicates that the monitoring equipment's shooting angle is directly facing the workbench; when the video quality is unacceptable, it indicates that the monitoring equipment's shooting angle is off-center, making it difficult for the monitoring video to record the state of the goods.

[0063] In the video quality assessment method provided in this application embodiment, the surface of the workbench is provided with an identifier of the target shape. The computer device can extract video frames from the monitoring video of the workbench as monitoring images, and use target contour extraction rules to extract contours from the monitoring images to obtain extraction results. These extraction results can be used to determine whether a contour corresponding to the target shape exists in the monitoring image. Based on these results, it can be determined whether the shooting angle of the monitoring system is directly facing the workbench, thereby determining whether the monitoring video can clearly record the state of the goods, and thus obtaining the video quality of the monitoring video. Since the accuracy of contour extraction is less affected by environmental factors, this application determines the video quality of the monitoring video based on the contour extraction results of the target shape. This can weaken the influence of environmental factors such as position, distance, shooting angle, shooting light, and video color difference on the video quality assessment results, thereby improving the assessment accuracy and ensuring that the assessed video quality accurately reflects the actual video quality of the monitoring video.

[0064] In one embodiment, such as Figure 2 As shown, the step of extracting video frames from the monitoring video of the workbench includes:

[0065] S202, periodically extract the latest video frame from the monitoring video, which is the real-time monitoring video of the workbench.

[0066] The step of extracting contours from the video frame according to the target contour extraction rules to obtain the target extraction result includes:

[0067] S204. For each newly extracted video frame, perform contour extraction on the newly extracted video frame according to the target contour extraction rules to obtain the target extraction result corresponding to the newly extracted video frame.

[0068] The step of determining the video quality of the surveillance video based on the target extraction result includes:

[0069] S206, Determine the video quality of the surveillance video based on the target extraction result corresponding to the latest extracted video frame.

[0070] Specifically, the monitoring video of the workbench can be a real-time monitoring video, meaning that the monitoring video continuously captures the workbench's situation to continuously record the status of the goods on it. In this case, the computer equipment can periodically extract the latest video frame from the monitoring video and perform contour extraction on the latest extracted video frame to determine whether the latest extracted video frame contains the contour of the target shape. The latest video frame can be the video frame corresponding to the latest moment in the monitoring video at the time of video frame extraction. The duration of the video frame extraction period can be determined according to the actual situation, and this application does not impose specific limitations on it.

[0071] For example, if the current time is T0 and the video frame extraction period is ΔT, then at time T0, the computer device can extract the video frame corresponding to time T0 from the surveillance video, and perform contour extraction on the video frame corresponding to time T0 according to the target contour extraction rule to obtain the target extraction result. Based on the target extraction result, the video quality of the surveillance video at time T0 is judged. When the time (T0+ΔT) arrives, the computer device can extract the video frame corresponding to time (T0+ΔT) from the surveillance video, and perform contour extraction on the video frame corresponding to time (T0+ΔT) according to the target contour extraction rule to obtain the target extraction result. Based on the target extraction result, the video quality of the surveillance video at time (T0+ΔT) is judged. Whenever the time (T0+NΔT) arrives, the computer device can extract the video frame of the corresponding time and perform video quality evaluation according to the aforementioned steps. Here, N is a positive integer.

[0072] In this embodiment, when the monitoring video is real-time, the computer device can periodically extract the latest video frames from the monitoring video on the workbench and perform contour extraction on the extracted video frames to obtain target extraction results. Based on the target extraction results, the video quality of the monitoring video is evaluated. In this way, periodic quality assessments can be performed on the real-time monitoring video to accurately and promptly identify quality problems, ensuring that the monitoring video can provide support for pre-sales and after-sales video evidence.

[0073] In one embodiment, the step of determining the video quality of the surveillance video based on the target extraction result corresponding to the latest extracted video frame includes:

[0074] If the target extraction result corresponding to the latest extracted video frame is empty, and the target extraction result corresponding to the latest extracted video frame in the previous two extractions is empty, then the video quality of the monitoring video is determined to be unqualified.

[0075] If the target extraction result corresponding to the latest extracted video frame is not empty, then the video quality of the monitoring video is determined to be qualified.

[0076] Specifically, for the most recently extracted video frame, if the target extraction result corresponding to the latest video frame is not empty, it indicates that the most recently extracted video frame records a complete identifiable object, and the camera angle of the monitoring system is directly facing the workbench. Therefore, it can be determined that the video quality of the monitoring video is acceptable.

[0077] If the target extraction result corresponding to the latest video frame is empty, in order to accurately determine whether the video quality of the surveillance video is unqualified, the computer equipment can combine the target extraction structure corresponding to the latest video frames extracted in the previous two extractions to make a judgment. When the target extraction result corresponding to the latest video frame extracted in the Mth extraction is empty, the target extraction result corresponding to the latest video frame extracted in the (M-1)th extraction is empty, and the target extraction result corresponding to the latest video frame extracted in the (M-2)th extraction is empty, it indicates that none of the latest extracted video frames in the three consecutive extractions recorded a complete identifiable object. Therefore, it can be determined that the video quality of the surveillance video is unqualified. Here, M is the number of extractions corresponding to this extraction, and M is a positive integer greater than or equal to 3.

[0078] In one embodiment, the video quality assessment method of this application further includes: if the video quality of the surveillance video is unqualified, sending a work order to the maintenance personnel. The computer equipment can dispatch work order information via means such as email or SMS to notify the relevant maintenance personnel to intervene and promptly repair the surveillance equipment or adjust its shooting angle, thereby ensuring that the surveillance video can provide support for pre-sales and after-sales video evidence.

[0079] In one embodiment, the target shape is a five-pointed star, the object color of the object being identified is red, and the target contour extraction rules include color extraction rules, area extraction rules, and angle number extraction rules.

[0080] like Figure 3 As shown, the step of extracting contours from the video frame according to the target contour extraction rules to obtain the target extraction result includes:

[0081] S302, extract contours from the video frames that simultaneously satisfy the color extraction rule and the area extraction rule to obtain preliminary contour extraction results; wherein, the color extraction rule is used to extract the region contours corresponding to the enclosed areas that match the object color of the identified object, and the area extraction rule is used to extract the region contours corresponding to the enclosed areas that match the object area of ​​the identified object.

[0082] S304, if the preliminary contour extraction result is not empty, then curve fitting is performed on each contour in the preliminary contour extraction result to obtain each fitted contour.

[0083] S306, filter each of the fitted contours based on the angle quantity extraction rule to obtain the angle rule filtering result, and obtain the target extraction result based on the angle rule filtering result.

[0084] Specifically, the computer device can combine the object's color and area size to extract contours, obtaining preliminary contour extraction results. If the preliminary contour extraction results are not empty, the color corresponding to the enclosed area of ​​each contour in the preliminary contour extraction results matches the object's color, and the area corresponding to the enclosed area of ​​each contour satisfies the area extraction rules. In this way, preliminary screening can be performed by combining color difference and the area of ​​the contour's enclosed area to filter out contours with excessive color difference or areas that do not meet the requirements.

[0085] When the initial contour extraction result is not empty, curve fitting is performed on each contour in the initial contour extraction result to obtain each fitted contour. Each fitted contour corresponds one-to-one with each contour in the initial contour extraction result. Then, the computer device filters each fitted contour according to the angle number extraction rule to exclude fitted contours whose angle number does not meet the angle number extraction rule, and obtains the target extraction result based on the filtering result. In one embodiment, the angle number extraction rule can be used to extract contours whose actual angle number is a first preset number or a second preset number, where the first preset number and the second preset number can be determined according to the actual situation. When the first preset number is 6 and the second preset number is 7, the computer device can exclude fitted contours whose actual angle number is neither equal to 6 nor equal to 7, and retain fitted contours whose actual angle number is equal to 6 or whose actual angle number is equal to 7 as the angle rule filtering result, and obtain the target extraction result based on the angle rule filtering result.

[0086] In this embodiment, the computer device uses color extraction rules, area extraction rules, and angle number extraction rules to obtain target extraction results from video frames, thereby improving the extraction accuracy of the contour corresponding to the target shape and further improving the evaluation accuracy.

[0087] In one embodiment, the step of obtaining the target extraction result based on the angle rule filtering result includes: for each fitted contour in the angle rule filtering result, calculating the image moment corresponding to the fitted contour; if the image moment corresponding to the fitted contour matches a preset image moment, performing convex hull detection on the fitted contour; determining whether to add the fitted contour to the target extraction result based on the detection result of the convex hull detection; if so, adding the fitted contour to the target extraction result.

[0088] The preset image moment is the image moment corresponding to the pentagram shape. Specifically, for each fitted contour in the angle rule filtering results, the computer device can calculate the image moment (i.e., HU moment) of the fitted contour and perform similarity matching between this image moment and the image moment corresponding to the pentagram shape (i.e., the preset image moment). If the two match, convex hull detection can be performed on the fitted contour, and based on the detection result of the convex hull detection, it can be accurately determined whether the fitted contour is a pentagram shape, thereby obtaining the target extraction result.

[0089] In one embodiment, if the image moments of the fitted contour match preset image moments, the fitted contour can be processed by corner and side calculation, convex hull detection, and center distance drawing to accurately determine whether the fitted contour is a pentagram shape. In one embodiment, the computer device can upload the target extraction results to a cloud storage platform.

[0090] In this embodiment of the application, for each contour in the angle rule filtering result, the computer device can further determine whether the contour is the contour corresponding to the pentagram shape by image moment matching and convex hull detection, thereby improving the extraction accuracy of the contour corresponding to the target shape and further improving the evaluation accuracy.

[0091] In one embodiment, before performing step S302, this application may further include the following steps:

[0092] Obtain the region ratio, and determine the region of interest in the video frame based on the region ratio;

[0093] The regions in the video frame other than the region of interest are converted to grayscale to obtain a grayscale video frame.

[0094] The step of extracting contours from the video frame that simultaneously satisfy the color extraction rules and the area extraction rules to obtain preliminary contour extraction results includes: extracting contours from the grayscale processed video frame that simultaneously satisfy the color extraction rules and the area extraction rules to obtain the preliminary contour extraction results.

[0095] In one embodiment, the region ratio can be obtained by receiving user input parameters, and the region ratio is adjustable.

[0096] Specifically, computer devices can determine the Region of Interest (ROI) in a video frame based on the region ratio. For example, the computer device can use the center point coordinates of the video frame as the center point coordinates of the ROI, and select regions whose aspect ratio meets a preset aspect ratio and whose area proportion meets the region ratio as ROIs.

[0097] Computer equipment can perform grayscale processing on regions in a video frame other than the ROI (Region of Interest) to obtain a grayscale-processed video frame. In other words, for a certain region in a video frame, if the region does not fall within an ROI, it can be grayscale-processed to obtain a grayscale-processed video frame. During preliminary contour extraction, the computer equipment can extract contours that simultaneously satisfy both color extraction rules and area extraction rules from the grayscale-processed video frame to obtain preliminary contour extraction results. This reduces the computational workload of the computer equipment and improves the efficiency of video quality assessment.

[0098] In one example, the video quality assessment method of this application can be as follows: Figure 4 As shown, the method specifically includes the following steps:

[0099] S402, obtain video quality monitoring configuration information, and extract the workbench identifier from the video quality monitoring configuration information.

[0100] S404, for each workstation identifier, periodically extract the latest video frame from the monitoring video of the workstation corresponding to that identifier. For each extracted latest video frame, execute steps S406 to S424.

[0101] S406, determine the region of interest in the latest video frame being processed based on the configurable region ratio, and perform grayscale processing on the regions in the latest video frame being processed other than the region of interest to obtain a grayscale video frame.

[0102] S408 filters the grayscale video frames using an image filtering algorithm to obtain filtered video frames. Specifically, the computer device can sequentially perform filtering using a Gaussian filtering algorithm, a bilateral filtering algorithm, and opening / closing operations to obtain filtered video frames.

[0103] S410, extract contours from the filtered video frames that simultaneously satisfy both color extraction rules and area extraction rules to obtain preliminary contour extraction results. The color extraction rules and area extraction rules can be as described in the above embodiments, and will not be repeated here.

[0104] S412, If the preliminary contour extraction result is not empty, then perform curve fitting on each contour in the preliminary contour extraction result to obtain each fitted contour.

[0105] S414 filters each fitted contour based on the angle quantity extraction rule to obtain the angle rule filtering result.

[0106] S416: For each fitted contour in the angle rule filtering results, calculate the image moment corresponding to the fitted contour. If the image moment corresponding to the fitted contour matches a preset image moment, then perform corner and side calculation, convex hull detection, and center distance drawing on the fitted contour. Based on the processing results, determine whether to add the fitted contour to the target extraction results. In this way, the fitted contours in the angle rule filtering results can be filtered to obtain the target extraction results. Furthermore, the target extraction results can be uploaded to a cloud storage platform.

[0107] S418, determine whether the target extraction result is empty. If yes, proceed to step S420; otherwise, proceed to step S422.

[0108] S420, determine whether the target extraction result corresponding to the latest video frame extracted in each of the previous two extractions is empty. If yes, proceed to step S424; otherwise, proceed to step S422.

[0109] S422, the video quality of the surveillance video is determined to be acceptable.

[0110] S424 indicates that the video quality of the surveillance video is unqualified and sends a work order to the maintenance personnel.

[0111] The video quality assessment apparatus provided in the embodiments of this application is described below. The video quality assessment apparatus described below can be referred to in correspondence with the video quality assessment method described above.

[0112] In one embodiment, this application provides a video quality assessment device 500, such as... Figure 5 As shown, the device 500 specifically includes a video frame extraction module 510, a target extraction result acquisition module 520, and a video quality acquisition module 530. Wherein:

[0113] The video frame extraction module 510 is used to extract video frames from the monitoring video of the workbench, the surface of which is provided with a target shape identifier.

[0114] The target extraction result acquisition module 520 is used to extract the contour of the video frame according to the target contour extraction rule to obtain the target extraction result. The target contour extraction rule is used to extract the contour corresponding to the target shape.

[0115] The video quality acquisition module 530 is used to determine the video quality of the monitoring video based on the target extraction result.

[0116] In one embodiment, the video frame extraction module 510 includes a latest video frame extraction unit, which periodically extracts the latest video frame from the monitoring video, wherein the monitoring video is the real-time monitoring video of the workbench. The target extraction result acquisition module 520 includes a target shape contour extraction unit, which performs contour extraction on the latest video frame for each extraction according to target contour extraction rules to obtain the target extraction result corresponding to the latest video frame. The video quality acquisition module 530 includes a quality evaluation unit, which determines the video quality of the monitoring video based on the target extraction result corresponding to the latest extracted video frame.

[0117] In one embodiment, the quality assessment unit includes a first judgment unit and a second judgment unit. The first judgment unit is used to determine that the video quality of the surveillance video is unqualified when the target extraction result corresponding to the latest extracted video frame is empty, and the target extraction result corresponding to each of the two previous extracted latest video frames is also empty. The second judgment unit is used to determine that the video quality of the surveillance video is qualified when the target extraction result corresponding to the latest extracted video frame is not empty.

[0118] In one embodiment, the target shape is a pentagram, the object color is red, and the target contour extraction rules include color extraction rules, area extraction rules, and angle number extraction rules. The target extraction result acquisition module 520 includes a preliminary contour extraction unit, a fitting unit, and an angle number filtering unit. The preliminary contour extraction unit extracts contours from the video frame that simultaneously satisfy the color extraction rules and the area extraction rules to obtain a preliminary contour extraction result. The color extraction rules extract the region contours corresponding to enclosed areas matching the object color, and the area extraction rules extract the region contours corresponding to enclosed areas matching the object area. The fitting unit performs curve fitting on each contour in the preliminary contour extraction result when the preliminary contour extraction result is not empty to obtain each fitted contour. The angle number filtering unit filters each fitted contour based on the angle number extraction rules to obtain an angle rule filtering result, and obtains the target extraction result based on the angle rule filtering result.

[0119] In one embodiment, the angle quantity filtering unit includes a matching unit. This matching unit is used to calculate the image moment corresponding to each fitted contour in the angle rule filtering result. If the image moment corresponding to the fitted contour matches a preset image moment, convex hull detection is performed on the fitted contour. Based on the detection result of the convex hull detection, it is determined whether to add the fitted contour to the target extraction result. If so, the fitted contour is added to the target extraction result.

[0120] In one embodiment, the video quality assessment apparatus 500 of this application further includes a region of interest (ROI) determination module and a grayscale processing module. The ROI determination module is used to obtain a region ratio and determine the ROI in the video frame based on the region ratio. The grayscale processing module is used to perform grayscale processing on the regions in the video frame other than the ROI to obtain a grayscale-processed video frame. The preliminary contour extraction unit includes a grayscale video frame extraction unit, which is used to extract contours from the grayscale-processed video frame that simultaneously satisfy the color extraction rules and the area extraction rules to obtain the preliminary contour extraction result.

[0121] In one embodiment, the video quality assessment device 500 of this application further includes a work order information sending module. This work order information sending module is used to send work order information to maintenance personnel when the video quality of the monitored video is unsatisfactory.

[0122] In one embodiment, this application also provides a storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the video quality assessment method as described in any of the above embodiments.

[0123] In one embodiment, this application also provides a computer device. The computer device stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the video quality assessment method as described in any of the above embodiments.

[0124] Indicatively, Figure 6 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. In one example, the computer device can be a server. (Refer to...) Figure 6The computer device 900 includes a processing component 902, which further includes one or more processors, and memory resources represented by memory 901 for storing instructions, such as application programs, that can be executed by the processing component 902. The application programs stored in memory 901 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 902 is configured to execute instructions to perform the steps of the video quality assessment method described in any of the above embodiments.

[0125] The computer device 900 may also include a power supply component 903 configured to perform power management of the computer device 900, a wired or wireless network interface 904 configured to connect the computer device 900 to a network, and an input / output (I / O) interface 905. The computer device 900 may operate on an operating system stored in memory 901, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0126] Those skilled in the art will understand that the internal structure of the computer device shown in this application is merely a block diagram of a portion of the structure related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0127] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. In this document, "a," "an," "the," "the," and "its" may also include plural forms unless the context clearly indicates otherwise. "Multiple" refers to at least two, such as 2, 3, 5, or 8, etc. "And / or" includes any and all combinations of the related listed items.

[0128] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0129] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A video quality assessment method, characterized in that, The method includes: Video frames are extracted from the monitoring video of the workbench, the surface of which is provided with an identifier of a target shape; the target shape is a five-pointed star, and the object of the identifier is red. The video frame is subjected to contour extraction according to the target contour extraction rules to obtain the target extraction result; the target contour extraction rules are used to extract the contour corresponding to the target shape, including color extraction rules, area extraction rules and angle number extraction rules; The video quality of the surveillance video is determined based on the target extraction results; The step of extracting contours from the video frame according to the target contour extraction rules to obtain the target extraction result includes: Contours that simultaneously satisfy the color extraction rule and the area extraction rule are extracted from the video frames to obtain preliminary contour extraction results; wherein, the color extraction rule is used to extract the region contour corresponding to the enclosed area that matches the object color of the identified object, and the area extraction rule is used to extract the region contour corresponding to the enclosed area that matches the object area of ​​the identified object. If the preliminary contour extraction result is not empty, then curve fitting is performed on each contour in the preliminary contour extraction result to obtain each fitted contour. The fitted contours are filtered based on the angle number extraction rules to obtain the angle rule filtering results, and the target extraction results are obtained based on the angle rule filtering results.

2. The video quality assessment method according to claim 1, characterized in that, The step of extracting video frames from the monitoring video of the workbench includes: The latest video frame is periodically extracted from the monitoring video, which is the real-time monitoring video of the workbench; The step of extracting contours from the video frame according to the target contour extraction rules to obtain the target extraction result includes: For each newly extracted video frame, the latest video frame is subjected to contour extraction according to the target contour extraction rules to obtain the target extraction result corresponding to the latest video frame; The step of determining the video quality of the surveillance video based on the target extraction result includes: The video quality of the surveillance video is determined based on the target extraction results corresponding to the latest extracted video frame.

3. The video quality assessment method according to claim 2, characterized in that, The step of determining the video quality of the surveillance video based on the target extraction result corresponding to the latest extracted video frame includes: If the target extraction result corresponding to the latest extracted video frame is empty, and the target extraction result corresponding to the latest extracted video frame in the previous two extractions is empty, then the video quality of the monitoring video is determined to be unqualified. If the target extraction result corresponding to the latest extracted video frame is not empty, then the video quality of the monitoring video is determined to be qualified.

4. The video quality assessment method according to claim 1, characterized in that, The step of obtaining the target extraction result based on the angle rule filtering result includes: For each fitted contour in the angle rule filtering result, the image moment corresponding to the fitted contour is calculated. If the image moment corresponding to the fitted contour matches the preset image moment, convex hull detection is performed on the fitted contour. Based on the detection result of convex hull detection, it is determined whether to add the fitted contour to the target extraction result. If so, the fitted contour is added to the target extraction result.

5. The video quality assessment method according to claim 1 or 4, characterized in that, The method further includes: Obtain the region ratio, and determine the region of interest in the video frame based on the region ratio; The regions in the video frame other than the region of interest are converted to grayscale to obtain a grayscale-processed video frame. The step of extracting contours from the video frame that simultaneously satisfy the color extraction rule and the area extraction rule to obtain preliminary contour extraction results includes: Contours that simultaneously satisfy the color extraction rules and the area extraction rules are extracted from the grayscale video frames to obtain the preliminary contour extraction results.

6. The video quality assessment method according to any one of claims 1 to 4, characterized in that, The method further includes: If the video quality of the monitored video is unsatisfactory, a work order will be sent to the maintenance personnel.

7. A video quality assessment device, characterized in that, The device includes: The video frame extraction module is used to extract video frames from the monitoring video of the workbench, the surface of which is provided with a target-shaped identification object; the target shape is a five-pointed star, and the object is red in color; The target extraction result acquisition module is used to extract the contour of the video frame according to the target contour extraction rules to obtain the target extraction result; the target contour extraction rules are used to extract the contour corresponding to the target shape, including color extraction rules, area extraction rules and angle number extraction rules; The video quality acquisition module is used to determine the video quality of the surveillance video based on the target extraction result; The target extraction result acquisition module includes: A preliminary contour extraction unit is used to extract contours from the video frame that simultaneously satisfy the color extraction rule and the area extraction rule to obtain a preliminary contour extraction result; wherein, the color extraction rule is used to extract the region contour corresponding to the enclosed area that matches the object color of the object to be identified, and the area extraction rule is used to extract the region contour corresponding to the enclosed area that matches the object area of ​​the object contour of the object to be identified. The fitting unit is used to perform curve fitting on each contour in the preliminary contour extraction result if the preliminary contour extraction result is not empty, so as to obtain each fitted contour. An angle quantity filtering unit is used to filter each of the fitted contours based on the angle quantity extraction rules to obtain the angle rule filtering result, and to obtain the target extraction result based on the angle rule filtering result.

8. A storage medium, characterized in that, The storage medium stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the video quality assessment method as described in any one of claims 1 to 6.

9. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions that, when executed by the one or more processors, perform the steps of the video quality assessment method as described in any one of claims 1 to 6.

Images