Frame rate control method and apparatus, and computer device and storage medium

Abstract

Provided in the present application are a frame rate control method and apparatus, and a computer device and a storage medium. The method comprises: acquiring a target number of images, which are uploaded by a target camera according to an inspection interval; selecting a reference image from among the target number of images, determining the similarity between each of the remaining images and the reference image, and obtaining a comprehensive similarity on the basis of each similarity; if the comprehensive similarity is greater than a similarity threshold value and a collection frame rate of the target camera is greater than a first frame rate, adjusting the collection frame rate to the first frame rate, and adjusting the inspection interval to a first interval; and if the product of the comprehensive similarity and a sensitivity factor is less than the similarity threshold value and the collection frame rate of the target camera is less than a second frame rate, adjusting the collection frame rate to the second frame rate, and adjusting the inspection interval to a second interval. The solution optimizes the parameters of a camera in real time on the basis of a change condition of a scene, so as to reduce the bandwidth and storage pressure of a media server, and can also respond to changes in the scene in a timely manner to avoid frame loss and delay.

Description

Frame rate control method, device, computer equipment and storage medium Technical Field

[0001] The present application relates to the field of video surveillance technology, and in particular to a frame rate control method, apparatus, computer equipment, and storage medium. Background Art

[0002] With the advancement of communications and video surveillance technologies, many residential communities and homes are now equipped with surveillance cameras, allowing users to remotely monitor specific areas as needed. Currently, most cameras on the market use fixed parameters. In actual operation, when the captured image remains static for extended periods, the camera continues to transmit the video stream using fixed parameters, resulting in a certain amount of bandwidth and storage resource waste. Furthermore, when the captured image contains a lot of motion, the fixed parameters may be set too low, resulting in frame dropouts and delays. Summary of the Invention

[0003] The purpose of this application is to solve at least one of the above-mentioned technical defects, especially the problem of bandwidth and storage resource waste or frame loss and delay caused by the difficulty of the camera in the existing technology to adapt to the specific circumstances of the shooting picture.

[0004] In a first aspect, the present application provides a frame rate control method, applied to a media server, comprising:

[0005] Obtain the target number of images uploaded by the target camera according to the inspection interval;

[0006] A reference image is selected from a target number of images, and similarities between the remaining images and the reference image are determined respectively, and a comprehensive similarity is obtained based on the respective similarities;

[0007] If the comprehensive similarity is greater than the similarity threshold, and the acquisition frame rate of the target camera is greater than the first frame rate, the acquisition frame rate is adjusted to the first frame rate, and the inspection interval is adjusted to the first interval;

[0008] If the product of the comprehensive similarity and the sensitivity factor is less than the similarity threshold, and the acquisition frame rate of the target camera is less than the second frame rate, the acquisition frame rate is adjusted to the second frame rate and the inspection interval is adjusted to the second interval; the second frame rate is greater than the first frame rate, and the second interval is less than the first interval.

[0009] In one embodiment, the target camera creates a first topic on the message server, and the target camera takes screenshots of the video stream collected by itself according to the inspection interval and publishes the screenshots through the first topic;

[0010] Get the target number of images uploaded by the target camera according to the inspection interval, including:

[0011] Subscribe to the first topic on the message server to obtain and store images published by the target camera until the number of images obtained reaches the target number.

[0012] In one embodiment, the frame rate control method further includes:

[0013] Create a second topic on the message server;

[0014] The acquisition frame rate adjustment instruction and the inspection interval adjustment instruction are issued to the target camera through the second topic.

[0015] In one embodiment, determining the similarity between the remaining images and the reference image includes:

[0016] The image to be compared with the reference image is determined as the target image, and the grayscale values ​​of the target image and the reference image are respectively obtained;

[0017] According to the grayscale values ​​of the target image and the reference image, brightness comparison parameters, contrast comparison parameters and structure comparison parameters are obtained respectively;

[0018] The similarity between the target image and the reference image is obtained according to the brightness comparison parameter, the contrast comparison parameter and the structure comparison parameter.

[0019] In one embodiment, a comprehensive similarity is obtained based on the respective similarities, including:

[0020] The average of each similarity is calculated to obtain the comprehensive similarity.

[0021] In one embodiment, a comprehensive similarity is obtained based on the respective similarities, including:

[0022] A weight is set for the similarity based on the distance between the image corresponding to the similarity and the reference image; the greater the distance, the greater the weight;

[0023] Each similarity is weighted and summed according to the corresponding weight, and the average of the summed results is calculated to obtain the comprehensive similarity.

[0024] In one embodiment, before the target camera publishes the screenshot through the first topic, the screenshot is grayscaled, size-compressed, and encoded according to a preset code system.

[0025] Obtain the target number of images uploaded by the target camera according to the inspection interval, including:

[0026] The image acquired through the first theme is decoded according to a preset code system.

[0027] In a second aspect, the present application provides a frame rate control device, applied to a media server, comprising:

[0028] A data acquisition module is used to obtain images of a target quantity uploaded by the target camera according to the inspection interval;

[0029] An image analysis module is used to select a reference image from a target number of images, determine the similarities between the remaining images and the reference image, and obtain a comprehensive similarity based on the similarities;

[0030] a first instruction issuing module, configured to adjust the acquisition frame rate to the first frame rate and the inspection interval to the first interval if the comprehensive similarity is greater than the similarity threshold and the acquisition frame rate of the target camera is greater than the first frame rate;

[0031] The second instruction issuing module is used to adjust the acquisition frame rate to the second frame rate and the inspection interval to the second interval if the product of the comprehensive similarity and the sensitivity factor is less than the similarity threshold and the acquisition frame rate of the target camera is less than the second frame rate; the second frame rate is greater than the first frame rate, and the second interval is less than the first interval.

[0032] In a third aspect, the present application provides a computer device comprising one or more processors and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the steps of the frame rate control method in any of the above embodiments are performed.

[0033] In a fourth aspect, the present application provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the frame rate control method in any of the above embodiments.

[0034] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0035] Based on the frame rate control method in this application, the target camera regularly takes screenshots and uploads the captured video stream according to the inspection interval. After the media server receives a group of images that reaches the target number, it selects one of them as the baseline image, calculates the similarity between the other images and the baseline image, and generates a comprehensive similarity index for the group of images. Based on the comprehensive similarity level, the current acquisition frame rate and other conditions, it is judged whether the group of images belongs to a static scene or a dynamic scene. If it belongs to a static scene, the acquisition frame rate of the target camera is lowered and the inspection interval is reduced. If it belongs to a dynamic scene, the acquisition frame rate of the target camera is increased and the inspection interval is increased. This solution can optimize the working parameters of the camera in real time according to the changes in the scene, saving storage and bandwidth resources while ensuring the quality of monitoring. In static scenes, it can reduce the bandwidth and storage pressure of the media server, and perform scene judgment at a higher frequency to respond to scene changes in a timely manner. In dynamic scenes, important information is captured as much as possible to avoid frame loss and delay, and reduce the data processing pressure of the media server. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0037] FIG1 is a flow chart of a frame rate control method provided by one embodiment of the present application;

[0038] FIG2 is an application scenario diagram of a frame rate control method according to an embodiment of the present application;

[0039] FIG3 is a module diagram of a frame rate control device provided by an embodiment of the present application;

[0040] FIG4 is a diagram showing the internal structure of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0042] This application provides a frame rate control method. The devices in this application scenario include a camera and a media server. The camera continuously captures images to generate a video stream. The camera uploads the video stream to the media server. Users can view the video stream by accessing the media server. Traditional cameras have fixed acquisition frame rates, such as 25fps or 30fps. In video surveillance scenarios, many periods of video contain repetitive, unchanging images. Using this fixed acquisition frame rate will still result in the camera uploading many duplicate and useless images, which wastes the media server's storage resources. When the images become dynamic, the fixed frame rate is relatively low, resulting in a smaller number of key frames that can be captured. A group of pictures is a set of images that can be fully played, consisting of a key frame (I frame) and multiple predicted frames (P frames) predicted based on the key frame. A video stream is composed of continuous groups of pictures. When the number of key frames is small, the length of the group of pictures will be longer. Longer groups of pictures will increase latency during video encoding, transmission, and decoding, potentially resulting in frame loss and lag. To solve the above problems, the frame rate control method in this application will be applied to a media server. Please refer to FIG. 1 . The method includes steps S102 to S108 .

[0043] S102, obtaining a target number of images uploaded by the target camera according to the inspection interval.

[0044] It will be understood that the target camera here refers to the camera that requires frame rate adjustment. The verification interval is the interval at which the target camera takes screenshots of the video stream it captures. The media server in this embodiment analyzes the images uploaded by the target camera in batches. The total number of images in each batch is the target number. In this step, the media server obtains the most recent set of images uploaded by the target camera. The interval between adjacent images in this set is the verification interval. When the total number of images reaches the target number, there are sufficient samples for subsequent image similarity analysis.

[0045] In traditional technology, the media server is often required to extract key frames from the video stream, perform image similarity analysis on the extracted frames, remove duplicate images, and reconstruct the video. This will place a heavy burden on both the CPU and GPU of the media server. This embodiment mainly utilizes the data processing capabilities of the target camera end to reduce the difficulty of image similarity analysis and the data processing pressure. The camera takes screenshots of the video stream at intervals to avoid the difficulties in image similarity analysis caused by too many samples. In addition, when it is found that there may be a large number of duplicate images in the video stream, the output of the duplicate images is controlled from the source (i.e., the target camera) without processing the video stream at the media server, which reduces the data processing burden of the media server.

[0046] S104 , selecting a reference image from the target number of images, determining similarities between the remaining images and the reference image, and obtaining a comprehensive similarity based on the similarities.

[0047] As you can understand, the reference image serves as the benchmark for image similarity analysis. The similarity of each image reflects the degree of similarity between the current image and the reference image. Numerous algorithms exist for calculating similarity, and the appropriate algorithm can be selected as needed. The selection of the reference image can also be customized. Generally speaking, to speed up processing, the first image can be selected as the reference image. This is because the target camera uploads images at regular intervals. When the first image is used as the reference image, the similarity calculation between the second image and the reference image can be started immediately upon the arrival of the second image. For each subsequent image that arrives, if the similarity calculation for the previous image has been completed, the similarity calculation for this image can be started immediately. When the last image in the batch arrives, the similarity calculation for that last image is sufficient to obtain the overall similarity. This method maximizes the time required to upload the entire batch of images, improving processing efficiency.

[0048] After calculating the similarity of each image other than the baseline image, a comprehensive analysis of all similarities within the batch is performed to determine the overall similarity of the images. A higher overall similarity indicates a higher probability that the images in this group are static. Conversely, a lower overall similarity indicates a higher probability that the images in this group are dynamic. Based on this overall similarity, the image category of the image captured by the current target camera can be determined. Based on this determination, different parameter combinations can be selected for the target camera.

[0049] Furthermore, the target number setting here must consider the media server's image analysis efficiency. It must ensure that image messages do not accumulate, allowing for timely responses when switching between dynamic and static images. This means that when a new batch of images arrives, the previous batch must have been processed. The interval between batches is the verification interval. If image analysis begins only after all images in a batch have arrived, the total time required to calculate all similarities and combined similarities must be less than or equal to the verification interval. If the method described in the example above is used, the total time required to calculate a single similarity and combined similarity must be less than or equal to the verification interval. The calculation time for a single similarity and combined similarity can be calculated by running a trial run and taking the average time.

[0050] S106: If the comprehensive similarity is greater than the similarity threshold and the acquisition frame rate of the target camera is greater than the first frame rate, the acquisition frame rate is adjusted to the first frame rate and the inspection interval is adjusted to the first interval.

[0051] As you can understand, the similarity threshold is used to determine whether a batch of images is static. If the combined similarity exceeds the threshold, the batch of images is sufficiently similar to be considered static. The first frame rate is the lower limit of the target camera's adjustable acquisition frequency range. When the current image is determined to be static, the target camera can be controlled to use the lowest acquisition frame rate to reduce bandwidth and storage usage. Since the image is static, reducing the acquisition frame rate will not lose important information.

[0052] Therefore, if the acquisition frame rate of the target camera is greater than the first frame rate, it means that the acquisition frame rate still has room to decrease, and the acquisition frame rate can be further lowered to the first frame rate. If the acquisition frame rate is already the first frame rate when the comprehensive similarity is greater than the similarity threshold, it is only necessary to maintain the acquisition frame rate at the first frame rate. The first interval is the preset upper limit of the inspection interval. In order to cope with the sudden switch between static and dynamic, when controlling the target camera to use the first frame rate for acquisition, the inspection interval needs to be increased to the first interval. This allows the media server to perform dynamic and static judgments at a higher frequency, so that it can detect the arrival of dynamic images more quickly and increase the acquisition frame rate again as soon as possible.

[0053] S108: If the product of the comprehensive similarity and the sensitivity factor is less than the similarity threshold, and the acquisition frame rate of the target camera is less than a second frame rate, the acquisition frame rate is adjusted to the second frame rate and the inspection interval is adjusted to a second interval. The second frame rate is greater than the first frame rate, and the second interval is less than the first interval.

[0054] As you can understand, the sensitivity factor here is used to adjust the sensitivity of image change judgment. The sensitivity factor is generally a value greater than or equal to 1. When set to 1, it indicates high sensitivity to image changes. If the comprehensive similarity fluctuates, frequent adjustments to the target camera parameters may be necessary. When set to greater than 1, the sensitivity can be reduced as needed, creating a buffer between the dynamic and static image discrimination ranges. When the comprehensive similarity exceeds the similarity threshold, the image is determined to be static, and the acquisition frequency needs to be reduced and the test interval increased. However, when the comprehensive similarity is less than the similarity threshold but greater than the similarity threshold divided by the sensitivity factor, the image falls between the two judgment ranges. In this case, the original parameters can be maintained. When the comprehensive similarity is less than the similarity threshold divided by the sensitivity factor (that is, the product of the comprehensive similarity and the sensitivity factor is less than the similarity threshold), the image is determined to be dynamic, and the acquisition frequency needs to be increased and the test interval reduced.

[0055] Specifically, the second frame rate is the upper limit of the target camera's adjustable acquisition frequency range. When the current image is determined to be dynamic, the target camera can be controlled to use the highest acquisition frame rate to ensure that important information in the dynamic image is not lost. Therefore, if the target camera's acquisition frame rate is less than the second frame rate, it indicates that there is room for improvement and can be further increased to the second frame rate. If the product of the comprehensive similarity and the sensitivity factor is less than the similarity threshold, the acquisition frame rate is already at the second frame rate, and the acquisition frame rate only needs to be maintained at the second frame rate. The second interval is the preset lower limit of the verification interval. To reduce the processing pressure on the media server, when the target camera is controlled to use the second frame rate for acquisition, since it is now operating in a mode adapted for dynamic images and does not lose important information, the media server can reduce the frequency of dynamic and static judgments to reduce its own data processing burden. Therefore, it is necessary to reduce the verification interval to the second interval.

[0056] Based on the frame rate control method in this application, the target camera regularly takes screenshots and uploads the captured video stream according to the inspection interval. After the media server receives a group of images that reaches the target number, it selects one of them as the baseline image, calculates the similarity between the other images and the baseline image, and generates a comprehensive similarity index for the group of images. Based on the comprehensive similarity level, the current acquisition frame rate and other conditions, it is judged whether the group of images belongs to a static scene or a dynamic scene. If it belongs to a static scene, the acquisition frame rate of the target camera is lowered and the inspection interval is reduced. If it belongs to a dynamic scene, the acquisition frame rate of the target camera is increased and the inspection interval is increased. This solution can optimize the working parameters of the camera in real time according to the changes in the scene, saving storage and bandwidth resources while ensuring the quality of monitoring. In static scenes, it can reduce the bandwidth and storage pressure of the media server, and perform scene judgment at a higher frequency to respond to scene changes in a timely manner. In dynamic scenes, important information is captured as much as possible to avoid frame loss and delay, and reduce the data processing pressure of the media server.

[0057] In one embodiment, please refer to Figure 2, the interaction between the target camera and the media server is based on the MQTT protocol. In the MQTT protocol, the target camera and the media server are both MQTT clients, and the message server is an MQTT server. Data interaction between clients can be carried out through the message server in a subscription and publishing mode. In this embodiment, the media server needs to obtain data from the target camera, and the target camera needs to act as a publisher to create and upload a first topic related to screenshots on the message server. After the target camera takes a screenshot of the video stream collected by itself according to the inspection interval, the screenshot can be published through the first topic. On this basis, obtaining the target number of images uploaded by the target camera according to the inspection interval includes: subscribing to the first topic on the message server to obtain and store the images published by the target camera until the number of images obtained reaches the target number.

[0058] Specifically, the media server subscribes to the first topic as a subscriber. When the target camera publishes a new screenshot, the media server can obtain the new screenshot from the message server and store it locally. When the number of images stored locally reaches the target number, the same batch of images for image analysis has been obtained and persisted locally. When the next screenshot arrives, in order to save space in the media server, the stored images can be cleared and the next batch of images can be received. This embodiment utilizes an MQTT-based message subscription and publishing mechanism to receive images, which solves the disorder problem of traditional technology using UDP for transmission and ensures the accuracy of image similarity judgment.

[0059] In one embodiment, before the target camera publishes the screenshot through the first topic, the screenshot is gray-scaled, size-compressed, and encoded according to a preset code system.

[0060] It is understandable that subsequent similarity analysis often requires analysis of the grayscaled image. This embodiment further utilizes the data processing capabilities of the target camera, allowing the target camera to grayscale the screenshot. To reduce bandwidth requirements, the screenshot can also be resized. Generally, the image can be compressed to 100*100, and ultimately compressed to a grayscale thumbnail of less than 20KB. To facilitate data transmission, the target camera also encodes the compressed screenshot. The code system used can be preset in advance by the administrator. Base64 is generally selected. Based on this, after the media server receives data from the message server, it needs to decode it according to the preset code system before starting image analysis.

[0061] In one embodiment, please continue to refer to FIG2 , the frame rate control method further includes:

[0062] (1) Create a second topic on the message server.

[0063] (2) Issue acquisition frame rate adjustment instructions and inspection interval adjustment instructions to the target camera through the second topic.

[0064] It's understandable that in order to adjust the target camera's parameters, the media server needs to send data to the target camera. In this case, the media server acts as a publisher, creating a second topic on the message server related to issuing instructions. The target camera, in turn, subscribes to this second topic as a subscriber. When the media server needs to adjust the target camera's parameters, it can issue acquisition frame rate adjustment instructions and test interval adjustment instructions to the target camera via the second topic. The acquisition frame rate adjustment instruction is used to adjust the target camera's acquisition frequency. The test interval adjustment instruction is used to adjust the target camera's test interval.

[0065] In one embodiment, determining the similarity between the remaining images and the reference image includes:

[0066] (1) The image to be compared with the reference image is determined as the target image, and the grayscale values ​​of the target image and the reference image are obtained respectively.

[0067] It can be understood that the same batch of images can be classified into reference images and remaining images, and each of the remaining images needs to be compared with the reference image separately. Each image that needs to be compared is called a target image. The algorithm used in this embodiment to evaluate the similarity between two images is the SSIM (Structural Similarity Index) algorithm. The analysis basis of this algorithm is the grayscale value of the image, so it is necessary to obtain the grayscale value of the target image and the reference image. If the screenshot obtained is a grayscale image, it can be obtained directly, otherwise it needs to be grayscaled by the media server first.

[0068] (2) According to the grayscale values ​​of the target image and the reference image, the brightness comparison parameter, contrast comparison parameter and structure comparison parameter are obtained respectively.

[0069] As you can understand, the SSIM algorithm evaluates the similarity between two images along three dimensions: brightness, contrast, and structure. Each dimension corresponds to a comparison parameter: brightness, contrast, and structure. The brightness comparison parameter reflects the difference in brightness distribution between the target image and the reference image. The contrast comparison parameter reflects the difference in contrast distribution between the target image and the reference image. The structure comparison parameter reflects the difference in structure and detail between the two images. The brightness comparison parameter is based on the mean of the grayscale values ​​of the two images. The contrast comparison parameter is based on the standard deviation of the grayscale values ​​of the two images. The structure comparison parameter is based on the covariance and standard deviation of the grayscale values ​​of the two images.

[0070] (3) According to the brightness comparison parameter, contrast comparison parameter and structure comparison parameter, the similarity between the target image and the reference image is obtained.

[0071] After obtaining the three detailed comparison parameters above, they need to be combined to produce a summary similarity score. In the SSIM algorithm, each of the three comparison parameters is raised to a power, with the power representing the weight of that comparison parameter. The similarity score is obtained by multiplying these three power terms together. Generally, each power is set to 1, which is equivalent to simply multiplying the three comparison parameters together to obtain the similarity score.

[0072] In one embodiment, obtaining a comprehensive similarity based on the respective similarities includes: averaging the respective similarities to obtain the comprehensive similarity.

[0073] It can be understood that when multiple similarities are obtained, the average value can reflect the average level of the multiple similarities, thereby reflecting the overall similarity of the entire group of images.

[0074] In one embodiment, a comprehensive similarity is obtained based on the respective similarities, including:

[0075] (1) According to the distance between the image corresponding to the similarity and the reference image, a weight is set for the similarity. The greater the distance, the greater the weight.

[0076] It can be understood that the distance here refers to the time interval between each target image and the reference image on the time axis. After calculating the similarity of a group of target images with the reference image, since these images are distributed sequentially in time, the distance factor can be considered to set different weights for their similarities. Considering that the relevance of still images gradually decreases over time, the similarity of images with a longer temporal distance can be given a larger proportion, increasing the impact of the similarity of images with lower relevance on the overall similarity. This approach prevents several images close to the reference image from overly dominating the final similarity result, avoiding distortion.

[0077] (2) Each similarity is weighted and summed according to the corresponding weight, and the average of the summed results is calculated to obtain the comprehensive similarity.

[0078] After setting a weight for each target image, we can calculate the overall similarity of the entire batch of images. This is done by multiplying the similarity of each image by its weight and summing the results. The final sum is then divided by the total number of images to obtain the overall similarity. This averaging approach ensures that every image is considered, eliminating any information loss. The distance weights also help regulate the importance of images in different time sequences.

[0079] The present application provides a frame rate control device, which is applied to a media server. Please refer to Figure 3 and includes a data acquisition module 310, an image analysis module 320, a first instruction issuing module 330 and a second instruction issuing module 340.

[0080] The data acquisition module 310 is used to acquire a target number of images uploaded by the target camera according to the inspection interval.

[0081] The image analysis module 320 is used to select a reference image from a target number of images, determine similarities between the remaining images and the reference image, and obtain a comprehensive similarity based on the similarities.

[0082] The first instruction issuing module 330 is configured to adjust the acquisition frame rate to the first frame rate and the inspection interval to the first interval if the comprehensive similarity is greater than the similarity threshold and the acquisition frame rate of the target camera is greater than the first frame rate.

[0083] The second instruction issuing module 340 is configured to adjust the acquisition frame rate to the second frame rate and the inspection interval to the second interval if the product of the comprehensive similarity and the sensitivity factor is less than the similarity threshold and the acquisition frame rate of the target camera is less than a second frame rate. The second frame rate is greater than the first frame rate, and the second interval is less than the first interval.

[0084] In one embodiment, a target camera creates a first topic on a messaging server. The target camera takes screenshots of its captured video stream at a check interval and publishes the screenshots via the first topic. The data acquisition module 310 is configured to subscribe to the first topic on the messaging server to acquire and store images published by the target camera until the target number of images is reached.

[0085] In one embodiment, the frame rate control device includes a topic creation module. The topic creation module is configured to create a second topic on the message server. The first instruction issuing module 330 or the second instruction issuing module 340 is configured to issue an acquisition frame rate adjustment instruction and a check interval adjustment instruction to the target camera via the second topic.

[0086] In one embodiment, the image analysis module 320 is configured to identify an image to be compared with a reference image as a target image, obtain grayscale values ​​of the target image and the reference image, respectively, obtain brightness comparison parameters, contrast comparison parameters, and structure comparison parameters based on the grayscale values ​​of the target image and the reference image, and obtain similarity between the target image and the reference image based on the brightness comparison parameters, contrast comparison parameters, and structure comparison parameters.

[0087] In one embodiment, the image analysis module 320 is configured to average the similarities to obtain a comprehensive similarity.

[0088] In one embodiment, the image analysis module 320 is used to set a weight for the similarity based on the distance between the image corresponding to the similarity and the reference image; wherein, the greater the distance, the greater the weight; each similarity is weighted and summed according to the corresponding weight, and the average of the summed results is calculated to obtain a comprehensive similarity.

[0089] For the specific definition of the frame rate control device, please refer to the definition of the frame rate control method above, which will not be repeated here. The various modules in the above-mentioned frame rate control device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0090] The present application provides a computer device comprising one or more processors and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the steps of the frame rate control method in any of the above embodiments are performed.

[0091] Schematically, as shown in FIG4 , FIG4 is a schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. Referring to FIG4 , computer device 400 includes a processing component 402, which further includes one or more processors, and a memory resource represented by memory 401 for storing instructions executable by processing component 402, such as an application. The application stored in memory 401 may include one or more modules, each corresponding to a set of instructions. In addition, processing component 402 is configured to execute instructions to perform the steps of the frame rate control method of any of the above embodiments.

[0092] The computer device 400 may further include a power supply component 403 configured to perform power management of the computer device 400 , a wired or wireless network interface 404 configured to connect the computer device 400 to a network, and an input / output (I / O) interface 405 .

[0093] Those skilled in the art will understand that the structure shown in FIG4 is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0094] The present application provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the frame rate control method in any of the above embodiments.

[0095] The various embodiments in this specification are described in a progressive manner, and 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 referenced to each other.

[0096] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A frame rate control method, characterized in that: Applicable to media servers, including: Obtain the target quantity images uploaded by the target camera according to the inspection interval; Selecting a reference image from the target number of images, determining similarities between the remaining images and the reference image, and obtaining a comprehensive similarity based on the similarities; If the comprehensive similarity is greater than the similarity threshold, and the acquisition frame rate of the target camera is greater than the first frame rate, adjusting the acquisition frame rate to the first frame rate, and adjusting the inspection interval to the first interval; If the product of the comprehensive similarity and the sensitivity factor is less than the similarity threshold, and the acquisition frame rate of the target camera is less than the second frame rate, the acquisition frame rate is adjusted to the second frame rate, and the inspection interval is adjusted to the second interval; the second frame rate is greater than the first frame rate, and the second interval is less than the first interval.

2. The frame rate control method according to claim 1, characterized in that: The target camera creates a first topic on the message server, and the target camera takes a screenshot of the video stream collected by itself according to the inspection interval, and publishes the screenshot through the first topic; The step of obtaining the target number of images uploaded by the target camera according to the inspection interval includes: Subscribe to the first topic on the message server to obtain and store images published by the target camera until the number of images obtained reaches the target number.

3. The frame rate control method according to claim 1, characterized in that: Also includes: Create a second topic on the messaging server; The acquisition frame rate adjustment instruction and the inspection interval adjustment instruction are issued to the target camera through the second topic.

4. The frame rate control method according to claim 1, characterized in that: The respectively determining the similarities between the remaining images and the reference image comprises: Determine the image to be compared with the reference image as the target image, and obtain the grayscale values ​​of the target image and the reference image respectively; According to the grayscale values ​​of the target image and the reference image, a brightness comparison parameter, a contrast comparison parameter and a structure comparison parameter are obtained respectively; The similarity between the target image and the reference image is obtained according to the brightness comparison parameter, the contrast comparison parameter and the structure comparison parameter.

5. The frame rate control method according to claim 1, characterized in that: The comprehensive similarity is obtained according to each of the similarities, including: The average of the similarities is calculated to obtain the comprehensive similarity.

6. The frame rate control method according to claim 1, characterized in that: The comprehensive similarity is obtained according to each of the similarities, including: Setting a weight for the similarity according to the distance between the image corresponding to the similarity and the reference image; wherein the greater the distance, the greater the weight; The similarities are weighted and summed according to the corresponding weights, and the average of the summed results is calculated to obtain the comprehensive similarity.

7. The frame rate control method according to claim 2, characterized in that: Before the target camera publishes the screenshot through the first topic, the screenshot is grayed, compressed and encoded according to a preset code system in sequence; The step of obtaining the target number of images uploaded by the target camera according to the inspection interval also includes: The image acquired through the first subject is decoded according to the preset code system.

8. A frame rate control device, characterized in that: Applicable to media servers, including: A data acquisition module is used to acquire images of a target quantity uploaded by a target camera according to an inspection interval; An image analysis module, used for selecting a reference image from the target number of images, determining the similarities between the remaining images and the reference image, and obtaining a comprehensive similarity based on the similarities; A first instruction issuing module, configured to adjust the acquisition frame rate to the first frame rate and the inspection interval to the first interval if the comprehensive similarity is greater than a similarity threshold and the acquisition frame rate of the target camera is greater than a first frame rate; The second instruction issuing module is used for adjusting the acquisition frame rate to the second frame rate and the inspection interval to the second interval if the product of the comprehensive similarity and the sensitivity factor is less than the similarity threshold and the acquisition frame rate of the target camera is less than the second frame rate; the second frame rate is greater than the first frame rate, and the second interval is less than the first interval.

9. A computer device, characterized in that: The system comprises one or more processors and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the frame rate control method according to any one of claims 1 to 7 are executed.

10. A storage medium, characterized in that: The storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the frame rate control method according to any one of claims 1 to 7.

Images