A video acquisition system based on data cache reading

Abstract

The application relates to the field of video data collection, in particular to a video collection system based on data buffer reading, which comprises a target evaluation module, a dynamic analysis module, an analysis execution module, a steady-state analysis module and a buffer management module. The target evaluation module is used for periodically determining the buffer optimization strategy of each target collection area according to a region dynamic index and a collection interference index. The dynamic analysis module is used for determining the analysis execution mode of each first-type target collection area according to a dynamic difference coefficient. The analysis execution module is used for determining a transmission buffer parameter and determining whether the transmission buffer parameter needs to be adjusted. The steady-state analysis module is used for determining the transmission buffer parameter of each second-type target collection area based on an evaluation trend parameter and a resource execution parameter. The buffer management module is used for determining the buffer update strategy of each first-type target collection area according to the dynamic difference coefficient and determining a replacement video frame based on a key evaluation coefficient and a buffer retention index. The application guarantees the smoothness of the video data transmission and processing process in the video data collection stage.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of video data acquisition, and in particular to a video acquisition system based on data cache reading. BACKGROUND

[0002] In the process of video data acquisition, the smoothness of data transmission between other processing modules and the data processing efficiency can be guaranteed through cache technology. In industrial production processes, to ensure the effectiveness of the acquired video data, the acquired video data often has the characteristics of high resolution and high frame rate. Video data caching can significantly reduce data delay and loss during transmission. However, the existing video acquisition system often cannot dynamically adapt to the needs in the actual working process according to the fixed cache strategy, resulting in that the actual data transmission process cannot meet the real-time data transmission requirements. Therefore, how to dynamically optimize the data cache process based on the actual production operation situation to balance the system resource consumption and the smoothness of data transmission is a problem to be solved by those skilled in the art.

[0003] Chinese Patent Publication No. CN112506447A discloses a data stream lock-free cache method and server for a video monitoring device, comprising: initializing a cache area, the cache area including an index area and a data area; when a data stream is collected, the data stream is encapsulated to obtain encapsulated data; when the length of the frame data in the encapsulated data exceeds the preset maximum single frame length, the encapsulated data is discarded; when the length of the frame data in the encapsulated data does not exceed the preset maximum single frame length, the encapsulated data is stored in the data area, and the start and end position information of the encapsulated data in the data area is saved in the index area; when a data acquisition instruction from the cache area is received, the corresponding encapsulated data in the data area is obtained through the start and end position information saved in the index area, and the encapsulated data is sent. However, the above-mentioned scheme has the following problems: it fails to optimize the data cache process in the video acquisition process according to the changes in the actual production operation situation, resulting in that it is impossible to balance the system resource consumption while ensuring the real-time data transmission in the video acquisition process. SUMMARY

[0004] Therefore, the present application provides a video acquisition system based on data cache reading to overcome the problem in the prior art that the data cache process in the video acquisition process cannot be optimized according to the changes in the actual production operation situation, resulting in that it is impossible to balance the system resource consumption while ensuring the real-time data transmission in the video acquisition process.

[0005] To achieve the above-mentioned purpose, the present application provides a video acquisition system based on data cache reading, comprising:

[0006] The target evaluation module is configured to periodically determine a cache optimization strategy for each target collection region according to a region dynamic index and a collection interference index, and to optimize the video data cache process by using a dynamic analysis optimization strategy or a steady-state analysis optimization strategy.

[0007] The dynamic analysis module is connected to the target evaluation module and is configured to determine a dynamic difference coefficient according to an evaluation difference parameter and a key continuous proportion, and to determine an analysis execution mode for each first-type target collection region according to a transmission cache parameter determined according to an execution processing coefficient and a bandwidth application proportion, or a transmission cache parameter determined according to a key frame density and a reference key index.

[0008] The analysis execution module is connected to the dynamic analysis module and includes a first analysis execution unit and a second analysis execution unit, and is configured to determine a transmission cache parameter according to the analysis execution mode determined by the dynamic analysis module, and to determine whether to adjust the transmission cache parameter according to a stage trend change degree or an execution floating coefficient.

[0009] The steady-state analysis module is connected to the target evaluation module and is configured to determine a transmission cache parameter for each second-type target collection region based on an evaluation trend parameter and a resource execution parameter.

[0010] The cache management module is connected to the dynamic analysis module and the steady-state analysis module, and includes a first cache management unit and a second cache management unit, and is configured to determine a cache update strategy for each first-type target collection region according to a dynamic difference coefficient, and to determine a replacement video frame based on a key evaluation coefficient and a cache retention index.

[0011] The cache update strategy is to determine a replacement video frame based on an analysis correlation index and a cache retention index, or to determine a replacement video frame based on a key reference coefficient and a cache retention index.

[0012] Further, the optimization analysis condition responded by the target evaluation module is that the region dynamic index of the target collection region is greater than a preset region dynamic index or the collection interference index is greater than a preset collection interference index, and it is determined that the dynamic analysis optimization strategy is used to optimize the cache process of the video data of the target collection region.

[0013] The target collection region whose region dynamic index is greater than a preset region dynamic index or whose collection interference index is greater than a preset collection interference index is recorded as a first-type target collection region.

[0014] Further, the optimization analysis condition responded by the target evaluation module is that the region dynamic index of the target collection region is less than or equal to a preset region dynamic index and the collection interference index is less than or equal to a preset collection interference index, and it is determined that the steady-state analysis optimization strategy is used to optimize the cache process of the video data of the target collection region.

[0015] The target acquisition region with the area dynamic index less than or equal to the preset area dynamic index and the acquisition interference index less than or equal to the preset acquisition interference index is recorded as a second-class target acquisition region.

[0016] Further, the dynamic analysis module determines a dynamic difference coefficient according to the evaluation difference parameter and the key continuous proportion in response to a dynamic analysis condition, and determines an analysis execution mode of the first-class target acquisition region according to the dynamic difference coefficient.

[0017] The dynamic difference coefficient is in a positive correlation with the evaluation difference parameter and the key continuous proportion, respectively.

[0018] The dynamic analysis condition is that any first-class target acquisition region is determined by the target evaluation module to be optimized by a dynamic analysis optimization strategy for a video data caching process.

[0019] Further, the analysis execution condition to which the dynamic analysis module responds is that the dynamic difference coefficient of a first-class target acquisition region is greater than a preset dynamic difference coefficient, and then it is determined that a first analysis execution unit periodically determines a transmission caching parameter of the first-class target acquisition region according to an execution processing coefficient and a bandwidth application proportion, and determines whether to increase the transmission caching parameter according to an execution floating coefficient.

[0020] The first caching reservation condition to which the first analysis execution unit responds is that the execution floating coefficient is greater than a preset execution floating coefficient, and then it is determined that the transmission caching parameter is increased according to the execution floating coefficient.

[0021] The first caching reservation condition to which the first analysis execution unit responds is that the execution floating coefficient is less than or equal to a preset execution floating coefficient, and then it is determined that the transmission caching parameter is not increased.

[0022] The increase value of the transmission caching parameter is in a positive correlation with the execution floating coefficient.

[0023] Further, the analysis execution condition to which the dynamic analysis module responds is that the dynamic difference coefficient of a first-class target acquisition region is less than or equal to a preset dynamic difference coefficient, and then it is determined that a second analysis execution unit periodically determines a transmission caching parameter of the first-class target acquisition region according to a key frame density and a reference key index, and determines whether to increase the transmission caching parameter according to a stage trend change degree.

[0024] The second caching reservation condition to which the second analysis execution unit responds is that the stage trend change degree is greater than a preset stage trend change degree, and then it is determined that the transmission caching parameter is increased according to the stage trend change degree.

[0025] The second cache reserving condition responded by the second analysis executing unit is that the stage trend change degree is less than or equal to a preset stage trend change degree, and it is determined that the transmission cache parameter is not adjusted by increasing.

[0026] The increasing value of the transmission cache parameter and the stage trend change degree are in a positive correlation relationship.

[0027] Further, the first cache management unit responds to a first management executing condition, and determines a cache updating strategy of a target acquisition region of a first type according to a dynamic difference coefficient;

[0028] The updating analysis condition responded by the first cache management unit is that the dynamic difference coefficient is greater than a preset dynamic difference coefficient, and it is determined that a replacement video frame is determined based on an analysis correlation index and a cache retention index for cache updating;

[0029] The first management executing condition is that there is any target acquisition region of the first type that completes the determination of the analysis executing mode.

[0030] Further, the updating analysis condition responded by the first cache management unit is that the dynamic difference coefficient is less than or equal to a preset dynamic difference coefficient, and it is determined that a replacement video frame is determined based on a key reference coefficient and a cache retention index for cache updating;

[0031] The determined number of replacement video frames is a first replacement executing number, and the first replacement executing number and the dynamic difference coefficient of the target acquisition region of the first type are in a positive correlation relationship.

[0032] Further, the steady state analysis module responds to a steady state analysis condition, and determines a transmission cache parameter of a target acquisition region of a second type based on an evaluation trend parameter and a resource executing parameter;

[0033] The transmission cache parameter and the evaluation trend parameter are in a positive correlation relationship, and the transmission cache parameter and the resource executing parameter are in a negative correlation relationship;

[0034] The steady state analysis condition is that there is any target acquisition region of the second type that is determined by the target evaluation module to be optimized by a steady state analysis optimization strategy for the video data cache process.

[0035] Further, the second cache management unit responds to a second management executing condition, and periodically determines a replacement video frame based on a key evaluation coefficient and a cache retention index for cache updating;

[0036] The determined number of replacement video frames is a second replacement executing number, and the second replacement executing number and the evaluation trend parameter of the target acquisition region of the second type are in a positive correlation relationship.

[0037] The second management execution condition is that there is any second type target acquisition area to complete the determination of the transmission buffer parameter.

[0038] Compared with the prior art, the beneficial effects of the present application are that the technical scheme of the present application determines the buffer optimization strategy of the target acquisition area according to the region dynamic index and the acquisition interference index, so as to make targeted settings and optimization for the buffer process of the target acquisition area in a subsequent stage, ensure that the buffer settings actually used in the video data acquisition and transmission process are more in line with the video data situation collected in the actual working scene, and further ensure the smoothness of the video data transmission and processing process in the video data acquisition stage.

[0039] Further, in the present application, for a first type target acquisition area, a dynamic difference coefficient is determined according to an evaluation difference parameter and a key continuous proportion, and an analysis execution mode of the first type target acquisition area is determined according to the dynamic difference coefficient. Since the region dynamic index or the acquisition interference index of this type of target acquisition area is large, it indicates that the analysis and processing of the video frames obtained in the recent stage are more complex, so the delay of the video data acquisition and transmission process is more strict and needs to be optimized in real time according to the buffer capacity, so as to ensure the smoothness of the video data transmission and processing process in the actual video data acquisition stage.

[0040] Further, in the present application, the analysis execution mode of the first type target acquisition area is determined according to the dynamic difference coefficient. The dynamic difference coefficient represents the variation of the region dynamic index and the acquisition interference index of the first type target acquisition area, and based on the dynamic difference coefficient, the setting mode of the transmission buffer parameter and the updating mode of the buffer content are determined, so as to ensure that the determined buffer capacity can adapt to the needs of the actual working scene, and to prevent sudden increase of the number of video frames needing to be buffered due to sudden changes, the transmission buffer parameter is adjusted accordingly. The present application improves the adaptation degree of the buffer process in the video data acquisition stage to the actual working scene, ensures the data transmission quality, and avoids excessive consumption of data processing resources.

[0041] Further, the region dynamic index and the collection interference index of the second type target collection region are both small, which indicates that there is no obvious motion feature of the monitored region or object in the near future, and the monitoring process of the target collection region is less affected by the smoke and dust, the analysis and processing of the obtained video frames are relatively simple, the real-time requirement of the video data transmission process is relatively small and stable, the transmission buffer parameters in a long time range are determined based on the evaluation trend parameters and the resource execution parameters, the excessive consumption of data processing resources is avoided, and the replacement video frames are periodically determined based on the key evaluation coefficients and the buffer retention index to maintain the stability of the buffer process, and the smoothness of the video data transmission and processing process in the video data collection stage is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0042] Fig. 1 The module connection diagram of the video collection system based on data buffer reading of the present application;

[0043] Fig. 2 The module structure diagram of the analysis execution module of the present application;

[0044] Fig. 3 The flow chart of the target evaluation module of the present application for determining the buffer optimization strategy of each target collection region according to the region dynamic index and the collection interference index;

[0045] Fig. 4 The flow chart of the dynamic analysis module of the present application for determining the analysis execution mode of the first type target collection region according to the dynamic difference coefficient. DETAILED DESCRIPTION

[0046] In order to make the objects and advantages of the present application more clear, the present application will be further described below in combination with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the protection scope of the present application.

[0047] The preferred embodiments of the present application will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application, and are not used to limit the protection scope of the present application.

[0048] It should be noted that in the description of the present application, the terms indicating the direction or position relationship such as "up", "down", "left", "right", "in", "out" and the like are based on the direction or position relationship shown in the drawings, which is only for the convenience of description, and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application.

[0049] Moreover, it needs to be explained that in the description of the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be fixed connection, or detachable connection, or integrally connected; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through intermediate medium, or the internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0050] Please refer to Figs. 1 to 4 As shown in the figure, the present application provides a video acquisition system based on data cache reading, comprising:

[0051] The target evaluation module is used to periodically determine the cache optimization strategy of each target acquisition area according to the area dynamic index and the acquisition interference index, and to optimize the video data cache process by adopting dynamic analysis optimization strategy or steady-state analysis optimization strategy.

[0052] The dynamic analysis module is connected with the target evaluation module, and is used to determine the dynamic difference coefficient according to the evaluation difference parameter and the key continuous proportion, and to determine the analysis execution mode of each type of target acquisition area, that is, to determine the transmission cache parameter according to the execution processing coefficient and the bandwidth application proportion, or to determine the transmission cache parameter according to the key frame density and the reference key index.

[0053] The analysis execution module is connected with the dynamic analysis module, and comprises a first analysis execution unit and a second analysis execution unit, which are used to determine the transmission cache parameter according to the analysis execution mode determined by the dynamic analysis module, and to determine whether to adjust the transmission cache parameter according to the stage trend change degree or the execution floating coefficient.

[0054] The steady-state analysis module is connected with the target evaluation module, and is used to determine the transmission cache parameter of each type of target acquisition area based on the evaluation trend parameter and the resource execution parameter.

[0055] The cache management module is connected with the dynamic analysis module and the steady-state analysis module respectively, and comprises a first cache management unit and a second cache management unit, which are used to determine the cache update strategy of each type of target acquisition area according to the dynamic difference coefficient, and to determine the replacement video frame based on the key evaluation coefficient and the cache retention index.

[0056] The cache update strategy is to determine the replacement video frame based on the analysis correlation index and the cache retention index, or to determine the replacement video frame based on the key reference coefficient and the cache retention index.

[0057] Among them, the application is used for real-time optimization of the buffering process in the video data acquisition stage for the industrial production area, to ensure that the smoothness of the video data transmission and processing process in the actual video data acquisition stage meets the work requirements, the industrial production area that needs to be monitored is recorded as the target monitoring area, the part of the target monitoring area that needs to be video data acquisition is divided to obtain several target acquisition areas, users can divide the target acquisition area according to the work scene, which is easy for those skilled in the art to understand, and will not be repeated here;

[0058] In the application, there are several buffering optimization records, and each buffering optimization record records at least one of the following: region dynamic index, acquisition interference index, dynamic difference coefficient, execution floating coefficient, key evaluation coefficient, stage trend change degree and buffering update coefficient in the video data buffering optimization process for the target acquisition area, and each buffering optimization record corresponds to a qualified mark, which records whether the smoothness of the data transmission and processing process of the video data acquisition stage meets the user's demand.

[0059] Specifically, the optimization analysis condition responded by the target evaluation module is that the region dynamic index of the target acquisition area is greater than the preset region dynamic index or the acquisition interference index is greater than the preset acquisition interference index, and then it is determined that the dynamic analysis optimization strategy is used to optimize the buffering process of the video data of the target acquisition area.

[0060] The target acquisition area with the region dynamic index greater than the preset region dynamic index or the acquisition interference index greater than the preset acquisition interference index is recorded as a type of target acquisition area.

[0061] Specifically, the optimization analysis condition responded by the target evaluation module is that the region dynamic index of the target acquisition area is less than or equal to the preset region dynamic index and the acquisition interference index is less than or equal to the preset acquisition interference index, and then it is determined that the steady-state analysis optimization strategy is used to optimize the buffering process of the video data of the target acquisition area.

[0062] The target acquisition area with the region dynamic index less than or equal to the preset region dynamic index and the acquisition interference index less than or equal to the preset acquisition interference index is recorded as a type of target acquisition area.

[0063] Among them, the application applies a region evaluation period, the length of the region evaluation period can be determined by the user, the higher the user's requirement for the smoothness of the data transmission and processing process of the video data acquisition stage, the shorter the length of the region evaluation period, a value of the length of the region evaluation period is provided, the length of the region evaluation period is 10 min, at the end of each region evaluation period, the region dynamic index and the acquisition interference index of each target acquisition area are detected.

[0064] For a single target acquisition area, if the current time is the end time of a region evaluation period, the region evaluation period is the target analysis stage, and the video frames acquired for the target acquisition area in the target analysis stage are extracted, and the extracted video frames are recorded as dynamic analysis frames. If there are two dynamic analysis frames between which there are no other dynamic analysis frames, the above two dynamic analysis frames are recorded as a set of adjacent analysis frames, and the interval time between the acquisition time corresponding to the above two dynamic analysis frames is recorded as the extraction interval parameter of the set of adjacent analysis frames. The extraction interval parameters of each set of adjacent analysis frames are consistent. Users can set the extraction interval parameter according to the actual working scene. The higher the user's requirement for the smoothness of the data transmission and processing process of the video data acquisition stage, the smaller the extraction interval parameter. The region dynamic index is the average value of the video frame dynamic indexes of each dynamic analysis frame extracted for the target acquisition area in the target analysis stage. The acquisition interference index is the average value of the interference reference values of each video frame acquired for the target acquisition area in the target analysis stage. For a single video frame, the interference reference value = the dynamic range index of the dynamic analysis frame / the signal-to-noise ratio of the dynamic analysis frame. The dynamic range index = the maximum brightness value of the dynamic analysis frame / the minimum brightness value of the dynamic analysis frame.

[0065] For a single dynamic analysis frame, the video frame dynamic index is the frame difference reference value between the dynamic analysis frame and its analysis reference frame. The analysis reference frame is the dynamic analysis frame whose corresponding acquisition time is before the corresponding acquisition time of the dynamic analysis frame and whose interval time between the corresponding acquisition time and the corresponding acquisition time of the dynamic analysis frame is the shortest. By discrete wavelet transform technology, the dynamic analysis frame and its analysis reference frame are transformed to obtain four subbands respectively. The obtained subbands correspond to different frequencies and directions respectively. The two subbands with the same frequency and direction obtained by transforming the dynamic analysis frame and the analysis reference frame are recorded as a set of frame difference analysis combinations. For a single set of frame difference analysis combinations, the pixel difference values between the pixel points at the same pixel position in the difference value image corresponding to the two subbands in the frame difference analysis combination are detected. The average value of the pixel difference values corresponding to each pixel position is recorded as the combined frame difference value of the set of frame difference analysis combinations. The frame difference reference value is the average value of the combined frame difference values of each set of frame difference analysis combinations. How to determine the frame difference reference value based on the above process is easily understood by those skilled in the art, and will not be described here.

[0066] The user can determine the values of the preset region dynamic index and the preset collection interference index according to the actual working scene. For example, the user can set the values according to the cache optimization record. The higher the user's requirement for the smoothness of the data transmission and processing process in the video data collection stage, the smaller the value of the preset region dynamic index, and the smaller the value of the preset collection interference index. A method for determining the value of the preset region dynamic index is provided. The cache optimization record optimized by the dynamic analysis optimization strategy for the video data cache process of the target collection region is recorded as an evaluation reference record. The average value of the region dynamic index of the target collection region in the evaluation reference record that meets the user's requirement for the smoothness of the data transmission and processing process in the video data collection stage is recorded as the preset region dynamic index. A method for determining the value of the preset collection interference index is provided. The average value of the collection interference index of the target collection region in the evaluation reference record that meets the user's requirement for the smoothness of the data transmission and processing process in the video data collection stage is recorded as the preset collection interference index.

[0067] Specifically, the dynamic analysis module determines a dynamic difference coefficient according to the evaluation difference parameter and the key continuous proportion in response to a dynamic analysis condition, and determines an analysis execution mode of a type of target collection region according to the dynamic difference coefficient.

[0068] The dynamic difference coefficient is positively correlated with the evaluation difference parameter and the key continuous proportion, respectively.

[0069] The dynamic analysis condition is that any type of target collection region is determined by the target evaluation module to use the dynamic analysis optimization strategy to optimize the video data cache process.

[0070] Among them, for a type of target collection region, because the region dynamic index of this type of target collection region is larger or the collection interference index is larger, it indicates that there are significant motion features of the monitored region or object of the target collection region in the near stage or the monitoring process of the target collection region is greatly affected by smoke and dust, resulting in more complex analysis and processing of the acquired video frames after video information transmission. Therefore, the delay of the video data collection and transmission process is more strict and needs to be optimized in real time for the cache capacity to ensure the smoothness of the subsequent analysis and processing. The transmission cache parameter is dynamically optimized in a subsequent target evaluation period. The transmission cache parameter is optimized in real time based on the acquired video frames and the execution of the video collection task in the above target evaluation period. The analysis execution mode of the type of target collection region is determined according to the dynamic difference coefficient.

[0071] For a single one-class target collection region, the first region evaluation period after the region evaluation period of the one-class target collection region is recorded as a prediction analysis period, and the analysis execution mode of the one-class target collection region in the prediction analysis period is determined according to a dynamic difference coefficient, the dynamic difference coefficient = ln (evaluation difference parameter x key continuous proportion), the evaluation difference parameter n is the number of region evaluation periods in the trend evaluation range, xi is the region dynamic index determined by the i-th region evaluation period in the trend evaluation range for the one-class target collection region, is the average value of the region dynamic index determined by each region evaluation period in the trend evaluation range for the one-class target collection region, yi is the collection interference index determined by the i-th region evaluation period in the trend evaluation range for the one-class target collection region, is the average value of the collection interference index determined by each region evaluation period in the trend evaluation range for the one-class target collection region, the key continuous proportion = the number of times that the one-class target collection region is determined as a one-class target collection region in the trend evaluation range / the number of region evaluation periods in the trend evaluation range, the end time of the trend evaluation range is the end time of the optimization analysis period, the duration of the trend evaluation range, which can be set by the user according to the actual working scene. The longer the duration of the trend evaluation range, the higher the user's requirement for the smoothness of the data transmission and processing process of the video data collection stage. A duration of the trend evaluation range is provided, and the duration of the trend evaluation range is ten times the duration of the region evaluation period.

[0072] Specifically, the analysis execution condition to which the dynamic analysis module responds is that the dynamic difference coefficient of the one-class target collection region is greater than the preset dynamic difference coefficient, then the first analysis execution unit determines the transmission cache parameter of the one-class target collection region according to the execution processing coefficient and the bandwidth application proportion, and determines whether to increase the transmission cache parameter according to the execution floating coefficient.

[0073] The first cache reservation condition to which the first analysis execution unit responds is that the execution floating coefficient is greater than the preset execution floating coefficient, then it is determined that the transmission cache parameter is increased according to the execution floating coefficient.

[0074] The first cache reservation condition to which the first analysis execution unit responds is that the execution floating coefficient is less than or equal to the preset execution floating coefficient, then it is determined that the transmission cache parameter is not increased.

[0075] The increase value of the transmission cache parameter and the execution floating coefficient are in a positive correlation.

[0076] If the dynamic difference coefficient of the single one-class target collection area is greater than the preset dynamic difference coefficient, it indicates that the region dynamic index and the collection interference index of the one-class target collection area in a long time range are often large and have large fluctuations, indicating that the processing difficulty of the video frames obtained in the process of executing the video information collection task of the one-class target collection area is large, and the fluctuation of the processing interference factors is also large. In the process of real-time optimization of the transmission buffer parameters of the one-class target collection area, more emphasis is placed on the resources of the video frame processing process to ensure that the data buffer condition can timely adapt to the actual situation.

[0077] The value of the preset dynamic difference coefficient can be determined by the user according to the actual working scene. For example, the user can set it according to the buffer optimization record. A method for providing a value of a preset dynamic difference coefficient is provided. The buffer optimization record for determining the transmission buffer parameters of the one-class target collection area based on the key frame density and the reference key index is recorded as the difference reference record. The maximum value of the dynamic difference coefficient in the difference reference record that meets the user's requirement for the smoothness of the data transmission and processing process in the video data collection stage is recorded as the preset dynamic difference coefficient.

[0078] For a single dynamic difference coefficient greater than a preset dynamic difference coefficient, a type of target acquisition region, periodically for the execution processing coefficient and bandwidth application ratio of the type of target acquisition region, the application executes an evaluation period, the length of the execution evaluation period can be determined by the user, the higher the user's requirement for the smoothness of the data transmission and processing process in the video data acquisition stage, the shorter the length of the execution evaluation period, a value of the length of the execution evaluation period is provided, the length of the execution evaluation period is 15s, at the starting time of each execution evaluation period, the execution processing coefficient and the bandwidth application ratio of the type of target acquisition region are detected to determine the execution evaluation coefficient of each execution evaluation period, for a single execution evaluation period, the transmission cache parameter in the video data acquisition task process and the execution evaluation coefficient in the execution evaluation period are in a positive correlation, the transmission cache parameter is the maximum data amount that can be cached in the corresponding time range, the execution evaluation coefficient = ln( execution processing coefficient x bandwidth application ratio), the execution processing coefficient is the average value of the interval time length between the time when the analysis processing node receives each video frame and the time when the video frame analysis processing is completed in the last execution evaluation period of the execution evaluation period, the analysis processing node is used to receive and process the video frame obtained by each target acquisition region, the bandwidth application ratio = the maximum bandwidth used in the process of video data transmission / the maximum bandwidth that can be used in the process of video data transmission in the last execution evaluation period of the execution evaluation period, the execution floating coefficient is the average value of the execution processing difference of each execution evaluation period in the floating analysis range of the execution evaluation period, for a single execution evaluation period, the execution processing difference = | the execution processing coefficient determined by the last execution evaluation period of the execution evaluation period-the execution processing coefficient determined by the execution evaluation period | / the execution processing coefficient determined by the last execution evaluation period of the execution evaluation period, the end time of the floating analysis range is the end time of the execution evaluation period, the length of the floating analysis range, the user can set according to the actual working scene, the higher the user's requirement for the smoothness of the data transmission and processing process in the video data acquisition stage, the longer the length of the floating analysis range, a length of the floating analysis range is provided, the length of the floating analysis range is ten times the length of the execution evaluation period;

[0079] The preset execution floating coefficient value can be determined by the user according to the actual working scenario. For example, the user can set the preset execution floating coefficient value according to the cache optimization record. The higher the user's requirement for the smoothness of the data transmission and processing process in the video data acquisition stage, the smaller the preset execution floating coefficient value. A method for determining the preset execution floating coefficient value is provided. The cache optimization record in which the transmission cache parameter is adjusted according to the execution floating coefficient is recorded as a floating reference record. The average value of the execution floating coefficient in the floating reference record that meets the user's requirement for the smoothness of the data transmission and processing process in the video data acquisition stage is recorded as the preset execution floating coefficient.

[0080] Specifically, the dynamic analysis module responds to an analysis execution condition that a dynamic difference coefficient of a type of target acquisition region is less than or equal to a preset dynamic difference coefficient, and determines that the second analysis execution unit periodically determines the transmission cache parameter of the type of target acquisition region according to the key frame density and the reference key index, and determines whether to adjust the transmission cache parameter according to the stage trend change degree.

[0081] The second cache reservation condition to which the second analysis execution unit responds is that the stage trend change degree is greater than a preset stage trend change degree, and it is determined that the transmission cache parameter is adjusted according to the stage trend change degree.

[0082] The second cache reservation condition to which the second analysis execution unit responds is that the stage trend change degree is less than or equal to a preset stage trend change degree, and it is determined that the transmission cache parameter is not adjusted.

[0083] The increase value of the transmission cache parameter and the stage trend change degree have a positive correlation.

[0084] For a single type of target acquisition region, if the dynamic difference coefficient of the type of target acquisition region is less than or equal to a preset dynamic difference coefficient, it indicates that the difference between the region dynamic index and the acquisition interference index of the type of target acquisition region in a long time range is small and only a small part of the video frames obtained in a small part of the time have a large processing difficulty. However, it can be shown that the interference degree of the type of target acquisition region gradually increases in the above time range. Therefore, in the process of real-time optimization of the transmission cache parameter of the type of target acquisition region, the change of the factors affecting the processing difficulty of the video frame needs to be focused on to determine the transmission cache parameter. Whether the transmission cache parameter is adjusted according to the stage trend change degree is determined to reserve the cache capacity to avoid the interruption of the video data acquisition and transmission process due to the burst of video frame processing.

[0085] For a single dynamic difference coefficient is less than or equal to the preset dynamic difference coefficient of a kind of target acquisition area, periodically for the key frame density and reference key index of the target acquisition area, the application applies an evaluation update cycle, the length of the dynamic update cycle can be determined by the user, the higher the user's requirement for the smoothness of the data transmission and processing process of the video data acquisition stage, the shorter the length of the dynamic update cycle, provide a kind of dynamic update cycle length of value, the length of the dynamic update cycle is 30s, at the starting time of each dynamic update cycle, the key frame density and reference key index of the target acquisition area are detected to determine the transmission buffer parameter of each dynamic update cycle, for a single dynamic update cycle, the transmission buffer parameter in the video data acquisition task process in the dynamic update cycle and stage evaluation coefficient are positively correlated, the stage evaluation coefficient is the product of key frame density and reference key index, the key frame density = the number of dense evaluation video frames in the last dynamic update cycle of the dynamic update cycle / the number of key video frames in the last dynamic update cycle of the dynamic update cycle, the reference key index is the average value of the key evaluation coefficient of each dense evaluation video frame in the last dynamic update cycle of the dynamic update cycle, the key video frame is the video frame with key evaluation coefficient greater than the preset key evaluation coefficient, the dense evaluation video frame is the key video frame with the interval time between the acquisition time in the last dynamic update cycle of the dynamic update cycle and the starting time of the dynamic update cycle less than the preset dense interval time, the value of the preset dense interval time can be set by the user according to the actual working scene, provide a kind of preset dense interval time value, the preset dense interval time is 15s, for a single video frame, the key evaluation coefficient = video frame dynamic index / preset area dynamic index + interference reference value / preset acquisition interference index, the stage trend change degree = the absolute value of the difference between the maximum value and the minimum value of the key evaluation coefficient of the dense evaluation video frame in the last dynamic update cycle of the dynamic update cycle / reference key index;

[0086] The preset key evaluation coefficient and the preset stage trend change degree can be determined by the user according to the actual working scene. For example, the user can set the preset key evaluation coefficient and the preset stage trend change degree according to the cache optimization record. The higher the user's requirement for the smoothness of the data transmission and processing process of the video data acquisition stage is, the smaller the preset key evaluation coefficient is, and the smaller the preset stage trend change degree is. A method for determining the preset key evaluation coefficient is provided. The average value of the key evaluation coefficients of the key video frames in the cache optimization record that meets the user's requirement for the smoothness of the data transmission and processing process of the video data acquisition stage is recorded as the preset key evaluation coefficient. A method for determining the preset stage trend change degree is provided. The cache optimization record in which the transmission cache parameter of a type of target acquisition area is adjusted according to the stage trend change degree is recorded as the change reference record. The average value of the stage trend change degrees in the change reference record that meets the user's requirement for the smoothness of the data transmission and processing process of the video data acquisition stage is recorded as the preset stage trend change degree.

[0087] Specifically, the first cache management unit determines the cache update strategy of a type of target acquisition area according to the dynamic difference coefficient in response to a first management execution condition.

[0088] The update analysis condition to which the first cache management unit responds is that the dynamic difference coefficient is greater than a preset dynamic difference coefficient, and then it is determined to determine the replacement video frame based on the analysis correlation index and the cache retention index to perform cache update.

[0089] The first management execution condition is that there is any type of target acquisition area that completes the determination of the analysis execution mode.

[0090] Specifically, the update analysis condition to which the first cache management unit responds is that the dynamic difference coefficient is less than or equal to a preset dynamic difference coefficient, and then it is determined to determine the replacement video frame based on the key reference coefficient and the cache retention index to perform cache update.

[0091] The number of the determined replacement video frames is the first replacement execution number, and the first replacement execution number and the dynamic difference coefficient of a type of target acquisition area have a positive correlation.

[0092] In the process of video data collection for the one-class target collection area, the cache area needs to be reserved to avoid interruption of transmission or analysis processing. In actual process, there is a mutation in cache demand. Therefore, to ensure the smoothness of the cache process, part of the video data information needs to be cleared in time. When the cache update condition is met, if there is any video frame that needs to be cached, the cache update coefficient of each cached video frame is detected. The cached video frame is the video frame that has completed caching. The cached video frame with a cache update coefficient greater than a preset cache update coefficient is recorded as a replacement video frame. Data transmission is performed for the replacement video frame. For a single cached video frame, the cache update coefficient is determined according to an analysis correlation index and a cache retention index. The cache update coefficient = cache retention index / analysis correlation index. The cache retention index is the interval time length between the time when the cached video frame completes caching and the current time. The analysis correlation index is the minimum value of the interval time length between the time when the cached video frame is obtained and the time when each analysis execution frame is obtained. The analysis execution frame is a video frame that has completed transmission but has not been analyzed. The value of the preset cache update coefficient can be determined by the user according to the actual working scene. For example, the user can set it according to the cache optimization record. The higher the user's requirement for the smoothness of the data transmission and processing process in the video data collection stage, the greater the value of the preset cache update coefficient. A method for determining the value of the preset cache update coefficient is provided. The cache optimization record based on the analysis correlation index and the cache retention index for cache update is recorded as an update reference record. The average value of the cache update coefficients of each replacement video frame in the update reference record that meets the user's requirement for the smoothness of the data transmission and processing process in the video data collection stage is recorded as the preset cache update coefficient.

[0093] For a type of target collection area determined by a single complete analysis execution mode, if the dynamic difference coefficient of the type of target collection area is less than or equal to a preset dynamic difference coefficient, at this time, in the process of collecting video data for the type of target collection area, although there is a large demand for cache capacity, the demand changes relatively smoothly, therefore, the cache update coefficient of each cache video frame is detected periodically to determine the replacement video frame of the replacement execution number, the replacement video frame of the first replacement execution number is determined from large to small based on the cache update coefficient, and data transmission is performed for the replacement video frame, the application uses a replacement analysis period, the length of the replacement analysis period is in a negative correlation relationship with the evaluation difference parameter, at the starting time of each replacement analysis period, the cache update coefficient of each cache video frame is detected, for a single cache video frame, the cache update coefficient is determined according to a key reference coefficient and a cache retention index, the cache update coefficient = cache retention index / key reference coefficient, and the key reference coefficient is the average value of the absolute value of the difference between the key evaluation coefficient of the cache video frame and the key evaluation coefficient of each analysis execution frame.

[0094] Specifically, the steady-state analysis module responds to a steady-state analysis condition, determines a transmission cache parameter of a second type of target collection area based on an evaluation trend parameter and a resource execution parameter;

[0095] The transmission cache parameter and the evaluation trend parameter are in a positive correlation relationship, and the transmission cache parameter and the resource execution parameter are in a negative correlation relationship;

[0096] The steady-state analysis condition is that any second type of target collection area is determined by a target evaluation module to use a steady-state analysis optimization strategy to optimize the video data cache process.

[0097] Among them, for the second type of target collection area, because the region dynamic index and the collection interference index of this type of target collection area are relatively small, it is indicated that there is no obvious motion feature of the monitored region or object of the target collection area in the near stage, and the monitoring process of the target collection area is also less affected by smoke dust, and the analysis and processing process of the obtained video frame after video information transmission is relatively simple, therefore, the real-time demand of the video data transmission process is relatively small and stable, therefore, based on the actual situation of the previous video data, the cache situation in the subsequent stage is set;

[0098] For a single second-class target collection area, the first region evaluation period after the region evaluation period of the second-class target collection area is recorded as the optimization analysis period. The transmission buffer parameter of the second-class target collection area in the optimization analysis period is set. The transmission buffer parameter of the second-class target collection area in the optimization analysis period is positively correlated with the region estimation coefficient. The region estimation coefficient = evaluation trend parameter / resource execution parameter. The evaluation trend parameter is the sum of the region dynamic difference and the collection interference difference. The region dynamic difference = (the maximum value of the region dynamic index of each region evaluation period in the trend evaluation range - the minimum value of the region dynamic index of each region evaluation period in the trend evaluation range) / the minimum value of the region dynamic index of each region evaluation period in the trend evaluation range. The collection interference difference = (the maximum value of the collection interference index of each region evaluation period in the trend evaluation range - the minimum value of the collection interference index of each region evaluation period in the trend evaluation range) / the minimum value of the collection interference index of each region evaluation period in the trend evaluation range. The resource execution parameter is the sum of the CPU usage rate and the memory usage rate multiplied by the corresponding execution evaluation weight coefficient. The execution evaluation weight coefficient corresponding to the CPU usage rate and the memory usage rate can be set by the user according to the actual working scene. A CPU usage rate and a memory usage rate corresponding to the execution evaluation weight coefficient are provided. The execution evaluation weight coefficient corresponding to the CPU usage rate is 0.6, and the execution evaluation weight coefficient corresponding to the memory usage rate is 0.4. How to determine the CPU usage rate and the memory usage rate is easy for those skilled in the art to understand, and is not described here.

[0099] Specifically, the second cache management unit determines the replacement video frame based on the key evaluation coefficient and the cache retention index periodically in response to the second management execution condition to perform cache update.

[0100] The determined number of replacement video frames is the second replacement execution number, and the second replacement execution number is positively correlated with the evaluation trend parameter of the second-class target collection area.

[0101] The second management execution condition is that any second-class target collection area completes the determination of the transmission buffer parameter.

[0102] Wherein, for any complete transmission buffer parameter determination two kinds of target acquisition area, the application applies a replacement processing cycle, the length of the replacement processing cycle is negatively related to the evaluation trend parameter, at the starting time of each replacement processing cycle, the buffer update coefficient of each buffer video frame is detected, the replacement video frame of the second replacement execution quantity is determined from large to small based on the buffer update coefficient, data transmission is carried out for the replacement video frame, for a single buffer video frame, the buffer update coefficient = buffer retention index / key evaluation coefficient.

[0103] So far, the technical scheme of the present application has been described in combination with the preferred embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical schemes after the changes or replacements will fall within the protection scope of the present application.

[0104] The above description is only the preferred embodiments of the present application and is not used to limit the present application; for those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A video acquisition system based on data cache reading, characterized in that: include: A target evaluation module is used to periodically determine the cache optimization strategy of each target acquisition area based on the regional dynamic index and the acquisition interference index, and to optimize the video data cache process by adopting a dynamic analysis optimization strategy or a steady-state analysis optimization strategy; a dynamic analysis module connected to the target evaluation module, configured to determine a dynamic difference coefficient based on the evaluation difference parameter and the key persistence ratio, and to determine the analysis execution mode of each first-class target acquisition area by determining a transmission buffer parameter based on the execution processing coefficient and the bandwidth application ratio, or determining the transmission buffer parameter based on the key frame density and the reference key index; an analysis execution module connected to the dynamic analysis module, comprising a first analysis execution unit and a second analysis execution unit, configured to determine transmission buffer parameters according to the analysis execution mode determined by the dynamic analysis module, and to determine whether to adjust the transmission buffer parameters according to the stage trend change degree or the execution floating coefficient; a steady-state analysis module connected to the target evaluation module, for determining transmission buffer parameters of each second-category target acquisition area based on evaluation trend parameters and resource execution parameters; a cache management module, connected to the dynamic analysis module and the steady-state analysis module, respectively, comprising a first cache management unit and a second cache management unit, configured to determine a cache update strategy for each type of target acquisition area according to a dynamic difference coefficient, and to determine a replacement video frame based on a key evaluation coefficient and a cache retention index; The cache update strategy is to determine the replacement video frame based on analyzing the correlation index and the cache retention index, or to determine the replacement video frame based on the key reference coefficient and the cache retention index; The target assessment module responds to an optimization analysis condition where the regional dynamic index of the target acquisition area is greater than a preset regional dynamic index or the acquisition interference index is greater than a preset acquisition interference index, and then determines to use a dynamic analysis optimization strategy to optimize the caching process of the video data of the target acquisition area; A target acquisition area whose regional dynamic index is greater than a preset regional dynamic index or whose acquisition interference index is greater than a preset acquisition interference index is recorded as a first-class target acquisition area; The target assessment module responds to an optimization analysis condition where the regional dynamic index of the target acquisition area is less than or equal to a preset regional dynamic index and the acquisition interference index is less than or equal to a preset acquisition interference index, and then determines to use a steady-state analysis optimization strategy to optimize the caching process of the video data of the target acquisition area; The target acquisition area whose regional dynamic index is less than or equal to the preset regional dynamic index and whose acquisition interference index is less than or equal to the preset acquisition interference index is recorded as a second-class target acquisition area.

2. The video acquisition system based on data cache reading according to claim 1, characterized in that: The dynamic analysis module responds to the dynamic analysis conditions, determines a dynamic difference coefficient based on the evaluation difference parameter and the key continuous proportion, and determines an analysis execution mode for a type of target acquisition area based on the dynamic difference coefficient; The dynamic difference coefficient is positively correlated with the evaluation difference parameter and the key continuous proportion respectively; The dynamic analysis condition is that there is any type of target acquisition area determined by the target evaluation module to adopt a dynamic analysis optimization strategy to optimize the video data caching process.

3. The video acquisition system based on data cache reading according to claim 2, characterized in that: If the analysis execution condition responded by the dynamic analysis module is that there is a type of target collection area whose dynamic difference coefficient is greater than the preset dynamic difference coefficient, it is determined that the first analysis execution unit periodically determines the transmission cache parameter of the type of target collection area according to the execution processing coefficient and the bandwidth application ratio, and determines whether to increase the transmission cache parameter according to the execution floating coefficient; If the first cache reservation condition responded by the first analyzing and executing unit is that the execution float coefficient is greater than the preset execution float coefficient, it is determined to increase and adjust the transmission cache parameter according to the execution float coefficient; If the first cache reservation condition responded by the first analysis execution unit is that the execution float coefficient is less than or equal to the preset execution float coefficient, it is determined that the transmission cache parameter is not increased; The increase value of the transmission buffer parameter is positively correlated with the execution floating coefficient.

4. The video acquisition system based on data cache reading according to claim 3, characterized in that: If the analysis execution condition responded by the dynamic analysis module is that there is a type of target acquisition area whose dynamic difference coefficient is less than or equal to the preset dynamic difference coefficient, then the second analysis execution unit is determined to periodically determine the transmission buffer parameter of the type of target acquisition area based on the key frame density and the reference key index, and determine whether to increase the transmission buffer parameter based on the stage trend change degree; If the second cache reservation condition responded by the second analysis execution unit is that the stage trend change degree is greater than a preset stage trend change degree, it is determined to increase the transmission cache parameter according to the stage trend change degree; If the second cache reservation condition responded by the second analysis execution unit is that the stage trend change degree is less than or equal to the preset stage trend change degree, it is determined that the transmission cache parameter is not increased; The increase value of the transmission buffer parameter is positively correlated with the stage trend change degree.

5. The video acquisition system based on data cache reading according to claim 4, characterized in that: The first cache management unit determines a cache update strategy for a type of target acquisition area according to a dynamic difference coefficient in response to the first management execution condition; If the update analysis condition responded by the first cache management unit is that the dynamic difference coefficient is greater than the preset dynamic difference coefficient, it is determined based on the analysis correlation index and the cache retention index to replace the video frame for cache update; The first management execution condition is that there is any type of target collection area to complete the determination of the analysis execution mode.

6. The video acquisition system based on data cache reading according to claim 5, characterized in that: If the update analysis condition responded by the first cache management unit is that the dynamic difference coefficient is less than or equal to the preset dynamic difference coefficient, then a decision is made to determine a replacement video frame based on the key reference coefficient and the cache retention index to perform a cache update; The determined number of replacement video frames is a first replacement execution number, and the first replacement execution number is positively correlated with a dynamic difference coefficient of a type of target acquisition area.

7. The video acquisition system based on data cache reading according to claim 1, characterized in that: The steady-state analysis module determines the transmission buffer parameters of the second type of target acquisition area based on the evaluation trend parameter and the resource execution parameter in response to the steady-state analysis condition; The transmission cache parameter is positively correlated with the evaluation trend parameter, and the transmission cache parameter is negatively correlated with the resource execution parameter; The steady-state analysis condition is that there are any two types of target acquisition areas, which are determined by the target evaluation module and optimized using a steady-state analysis optimization strategy for the video data caching process.

8. The video acquisition system based on data cache reading according to claim 7, characterized in that: The second cache management unit, in response to the second management execution condition, periodically determines a replacement video frame based on a key evaluation coefficient and a cache retention index to perform cache update; The determined number of replacement video frames is a second replacement execution number, and the second replacement execution number is positively correlated with the evaluation trend parameter of the second type of target acquisition area; The second management execution condition is that there are any two types of target collection areas to complete the determination of the transmission buffer parameters.

Images