Video decoding method and device based on sampling, medium and equipment

Through sampling and clustering technology, video decoding is directly performed in the decoding unit, solving the problem of low video decoding efficiency in the prior art and achieving efficient video data processing.

CN119031135AInactive Publication Date: 2024-11-26ZHEJIANG MEIRI HUDONG NETWORK TECH CO LTD
View PDF 11 Cites 0 Cited by

Patent Information

Application Number
CN202411458306.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-18
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing video decoding methods are inefficient when decoding a large amount of video data, making it difficult to achieve real-time decoding.

Method used

Using a sampling-based video decoding method, by obtaining the resolution, frame rate and event type list of the video acquisition device, sampling to obtain the internal encoded frame list, and clustering and splicing them to reduce the amount of decoded data, and directly perform it in the decoding unit. decoding.

Benefits of technology

The video decoding time is reduced, the video decoding efficiency is improved, the data download steps are avoided, and the efficiency of the decoding unit is enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119031135A_ABST
    Figure CN119031135A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of video decoding, in particular to a sampling-based video decoding method and device, a medium and equipment, and the method comprises the steps: sampling to obtain an internal coding frame list, clustering internal coding frames to obtain K internal coding frame sets, and splicing all internal coding frames in each internal coding frame set to obtain K internal coding frame sets; according to the method and the device, K pieces of spliced data are acquired, the K pieces of spliced data are sent to the decoding unit to be decoded, and the target decoded image is acquired, so that the initial video is sampled to obtain the internal coded frame directly based on the initial video stored in the data storage unit and the related parameters, the data volume needing to be decoded is reduced, video data do not need to be downloaded, and the decoding efficiency is improved. According to the method and the device, the video decoding time is greatly reduced, the internal coding frames are further clustered and spliced, and the spliced data composed of a plurality of internal coding frames with consistent resolution and relatively high relevance is obtained, so that the spliced data is decoded based on the decoding unit, and the video decoding efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of video decoding technology, and in particular to a sampling-based video decoding method, device, medium and equipment. Background Art

[0002] Digital video signals generally use the IPB (Inter-Frame Prediction Bitrate) compression method to eliminate the redundancy of the image in the temporal and spatial domains. Correspondingly, the video decoding method is that the decoder reads the encoded data in the video file and restores the compressed video data to the original video image and sound signal according to the decoding rules of the video encoding format.

[0003] Specifically, in the prior art, the encoder usually encodes multiple images and produces them into segments of GOP (Group of Pictures). GOP is a group of continuous pictures, consisting of an I frame (intra-coded frame) and several B frames (bidirectional interpolated frames) / P frames (forward predicted frames). It is the basic unit accessed by video image encoders and decoders.

[0004] During the video decoding process, the compressed data to be decoded is usually downloaded to the decoding end, and the decoder reads each GOP segment for decoding. However, when faced with the task of decoding a large amount of video data, the data downloading and decoding processes in the above video decoding method both take a lot of time, resulting in low efficiency of video decoding and difficulty in achieving real-time decoding of a large amount of video data.

[0005] Therefore, when faced with the task of decoding a large amount of video data, how to reduce the time of video decoding and improve the efficiency of video decoding becomes a problem that needs to be solved urgently. Summary of the invention

[0006] In view of the above technical problems, the technical solution adopted by the present invention is a sampling-based video decoding method, which includes the following steps: S1, obtain M video acquisition devices in the video platform, the resolution, frame rate, position and initial video corresponding to each video acquisition device, as well as the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, wherein each event type list includes several event types, and M refers to an integer greater than 0.

[0007] S2, according to the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, sample from each initial video to obtain an internal coding frame list corresponding to each event type, wherein each internal coding frame list includes a number of internal coding frames.

[0008] S3, clustering all the internal coding frames according to the resolution and position corresponding to each video acquisition device, and obtaining K sets of internal coding frames, where K is an integer greater than 1.

[0009] S4, splicing all the internal coding frames in each internal coding frame set respectively to obtain K splicing data.

[0010] S5, sending the K spliced ​​data to a decoding unit in the video platform for decoding, and obtaining a target decoded image corresponding to each internally encoded frame.

[0011] The present invention also provides a sampling-based video decoding device, the video decoding device comprising: The data acquisition module is used to obtain M video acquisition devices in the video platform, the resolution, frame rate, position and initial video corresponding to each video acquisition device, as well as the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, wherein each event type list includes several event types, and M refers to an integer greater than 0.

[0012] The sampling module is used to sample from each initial video to obtain an internal coding frame list corresponding to each event type according to the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, wherein each internal coding frame list includes a number of internal coding frames.

[0013] The clustering module is used to cluster all internal coding frames according to the resolution and position corresponding to each video acquisition device, and obtain K sets of internal coding frames, where K refers to an integer greater than 1.

[0014] The splicing module is used to splice all the internal coding frames in each internal coding frame set to obtain K splicing data.

[0015] The decoding module is used to send the K spliced ​​data to the decoding unit in the video platform for decoding, and obtain the target decoded image corresponding to each internal coding frame.

[0016] The present invention also provides a non-transitory computer-readable storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or at least one program is loaded and executed by a processor to implement the above-mentioned sampling-based video decoding method.

[0017] The present invention also provides an electronic device, comprising a processor and the above-mentioned non-transitory computer-readable storage medium.

[0018] The present invention has at least the following beneficial effects: according to the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, a list of internal coding frames corresponding to each event type is sampled from each initial video, and according to the resolution and position corresponding to each video acquisition device, all internal coding frames are clustered to obtain K sets of internal coding frames, and all internal coding frames in each set of internal coding frames are spliced ​​to obtain K spliced ​​data, and the K spliced ​​data are sent to a decoding unit in a video platform for decoding, and a target decoded image corresponding to each internal coding frame is obtained. It can be seen that the initial video is directly sampled based on the initial video and related parameters stored in the data storage unit to obtain the internal coding frame, which reduces the amount of data that needs to be decoded, and there is no need to download video data, which greatly reduces the time for video decoding, and the internal coding frames are further clustered and spliced ​​to obtain spliced ​​data consisting of several internal coding frames with consistent resolution and high correlation, so that the spliced ​​data is decoded based on the decoding unit, thereby improving the efficiency of video decoding. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 A flowchart of a sampling-based video decoding method provided by an embodiment of the present invention; Figure 2 A schematic structural diagram of a sampling-based video decoding device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It is understood that, where appropriate, the above-mentioned terms used to distinguish similar objects can be interchanged so that the present invention can also implement other embodiments other than the above-mentioned illustrated embodiments or described embodiments. In addition, the terms "including" and "having" and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] Embodiment 1 This embodiment 1 provides a sampling-based video decoding method, which includes the following steps: Figure 1 As shown: S1, obtain M video acquisition devices in the video platform, the resolution, frame rate, position and initial video corresponding to each video acquisition device, as well as the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, wherein each event type list includes several event types, and M refers to an integer greater than 0.

[0024] The video acquisition device may refer to a camera, which is used to acquire corresponding videos when various events occur. The location of the video acquisition device may refer to the geographical coordinates corresponding to the video acquisition device, including longitude and latitude.

[0025] The specific event type can be set by the implementer according to the actual situation. The video captured by the video acquisition device can facilitate relevant personnel to check the actual situation of the event and provide data support for solving related problems.

[0026] The event durations corresponding to different event types may be different. Therefore, this embodiment obtains the preset acquisition time corresponding to each event type, and extracts the video data corresponding to each event from the video acquisition device based on the preset acquisition time corresponding to each event type, thereby improving the accuracy of the video data corresponding to each event and providing more powerful assistance for relevant personnel to solve related problems.

[0027] In a specific implementation, S1 further includes the following steps: S110, obtaining historical videos corresponding to each video acquisition device within a preset historical time period.

[0028] S120, sampling from the corresponding historical videos according to the frame rate corresponding to each video acquisition device to obtain a first internal coding frame list corresponding to each historical video, wherein each first internal coding frame list includes a plurality of internal coding frames.

[0029] S130: Input the first internal coding frame list into the trained recognition model to obtain a plurality of second internal coding frame lists and an event type corresponding to each second internal coding frame list.

[0030] S140, according to each second internal coding frame list, obtaining the mixed event image frame in each second internal coding frame list.

[0031] S150, removing the corresponding mixed event image frame from each second internal coding frame list, and acquiring the corresponding third internal coding frame list.

[0032] S160: According to the first time corresponding to each third internal coding frame list and the event type corresponding to the third internal coding frame list, obtain a first time set corresponding to each event type.

[0033] S170: Acquire a preset collection time corresponding to each event type according to the first time set corresponding to each event type.

[0034] The specific time of the preset historical time period may be set by the implementer according to actual conditions. For example, the preset historical time period may refer to a time period of one, two, or three years before the current time point.

[0035] Historical video refers to a video obtained by compressing and encoding the video collected by a video acquisition device within a preset historical time period.

[0036] According to the frame rate corresponding to each video acquisition device, all first internal coding frames corresponding to each historical video are sampled from the corresponding historical video, and then a first internal coding frame list is formed according to the chronological order of the corresponding acquisition time. Among them, the internal coding frame is also called a key frame, which is an independent frame with all information and can be independently decoded without referring to other images. In this embodiment, the internal coding frame is collected from the historical video as the analysis object, so as to complete the evaluation of the preset acquisition time through the sampled data in the video, thereby improving the acquisition efficiency of the preset acquisition time.

[0037] The trained recognition model includes an encoder and a fully connected layer, wherein the encoder is used to extract features from each first internal coding frame in the input first internal coding frame list, and the fully connected layer is used to identify the event type corresponding to each first internal coding frame, and then according to the recognition result of the event type, a second internal coding frame list is composed of several first internal coding frames with consistent event types and adjacent acquisition times, and then several second internal coding frame lists are extracted from the first internal coding frame list, and the event type corresponding to each second internal coding frame list is obtained.

[0038] Among them, the occurrence time of different events may overlap, and limited by the accuracy of the recognition model, different second internal coding frame lists may include the same first internal coding frame, that is, the same internal coding frame is identified as corresponding to multiple event types. Therefore, in order to improve the accuracy of the preset acquisition time, this embodiment determines the internal coding frames identified as corresponding to multiple event types as mixed event image frames, and removes the corresponding mixed event image frames from each second internal coding frame list to obtain the corresponding third internal coding frame list, thereby improving the accuracy of the third internal coding frame list.

[0039] For the first time set corresponding to each event type, the duration of each event type can be analyzed, and then the preset acquisition time corresponding to each event type can be obtained to clarify the acquisition time corresponding to the internal coding frame list of each event type when sampling from the initial video, thereby improving the accuracy of the internal coding frame list.

[0040] In the above, based on the historical videos corresponding to the video acquisition device in the historical time period, the third internal coding frame list corresponding to each event type is identified, and then combined with the first time set corresponding to each event type, the duration of each event type is analyzed, and then the preset acquisition time corresponding to each event type is obtained, and the acquisition time corresponding to the internal coding frame list of each event type when sampled from the initial video is clarified, thereby improving the accuracy of the internal coding frame list.

[0041] In a specific implementation, S140 further includes the following steps: S141, for any second intra-coded frame, if the current second intra-coded frame belongs to two or more second intra-coded frame lists, determine the current second intra-coded frame as a mixed event image frame.

[0042] S142, traverse all second intra-coded frames to obtain all mixed event image frames in each intra-coded frame list.

[0043] Among them, if the current second internal coding frame belongs to two or more second internal coding frame lists, that is, the current second internal coding frame is identified as corresponding to multiple event types, then the current second internal coding frame is determined as a mixed event image frame to remove the corresponding mixed event image frame from each second internal coding frame list, thereby reducing the impact of the second internal coding frame with unclear event type on the preset acquisition time, thereby improving the accuracy of the preset acquisition time.

[0044] In a specific implementation, S170 further includes the following steps: S171, for any event type, according to the first time set corresponding to the current event type, obtain the first quantity of each first time corresponding to the current event type and the total quantity of the first time in the first time set.

[0045] S172: For any first time corresponding to the current event type, determine the ratio of the first quantity corresponding to the current first time to the total quantity as the quantity proportion corresponding to the current first time.

[0046] S173, obtaining a preset collection time corresponding to the current event type according to each first time and the quantity ratio corresponding to each first time.

[0047] Among them, the higher the proportion of the number corresponding to the current first time, the more important the current first time is in measuring the duration corresponding to the current event type. Therefore, the duration corresponding to the current event type is analyzed in combination with each first time and the proportion of the number corresponding to each first time, and then the preset acquisition time corresponding to the current event type is obtained, which serves as the basis for sampling the internal encoding frame list of each event type from the initial video.

[0048] In a specific implementation, the total number of event types is set to L, the total number of first time in the first time set corresponding to the ith event type is set to Ri, and the jth first time T corresponding to the ith event type is set to ij The first quantity is set to P ij , then according to the Ri first times corresponding to the i-th event type and the quantity proportion corresponding to each first time, the preset collection time Q corresponding to the i-th event type is obtained. i , Q i Meet the following conditions: Q i =Σ j=1 Ri ((P ij / L)×T ij ).

[0049] In a specific implementation, S1 further includes the following steps: S180, obtaining the corresponding captured video of each video capture device within a preset current time period.

[0050] S190, compress each collected video to obtain a corresponding initial video.

[0051] The specific time of the preset current time period may be set by the implementer according to actual conditions. For example, the preset current time period may refer to a time period of one month, two months, three months, etc. before the current time point.

[0052] Video uses the principle of human visual persistence to give the human eye a sense of motion by playing a series of pictures. Since the amount of video to be transmitted is very large, the existing network and storage are unable to bear the corresponding amount of data. Therefore, in order to facilitate video transmission and storage, this embodiment compresses the captured video according to a preset video compression method, and obtains the corresponding initial video for storage and decoding.

[0053] The preset video compression method can be set by the implementer according to the actual situation. For example, the preset video compression method can be MPEG-1, MPEG-2, MPEG-4, H.261, H.263, H.264 and other compression methods.

[0054] Among them, those skilled in the art know that any video compression method in the prior art falls within the protection scope of the present invention, and will not be described in detail here.

[0055] In a specific implementation, S1 further includes the following steps: S10, obtaining the trained prediction model; S20, inputting each initial video into the trained prediction model to obtain the key image frame corresponding to each initial video and the event type corresponding to each key image frame; S30, acquiring an event type list corresponding to each initial video according to the video time and event type corresponding to each key image frame.

[0056] Among them, the trained prediction model is used to extract and map features of the input initial video, analyze the image features in the initial video, obtain the key image frames in the initial video, and determine the event type corresponding to each key image frame. Then, combined with the video time corresponding to each key image frame in the initial video, the event type corresponding to each initial video is sorted, and the corresponding event type list is obtained to provide a data basis for sampling the initial video, thereby reducing the amount of decoding data of the decoding unit and improving the efficiency of video decoding.

[0057] S2, according to the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, sample from each initial video to obtain an internal coding frame list corresponding to each event type, wherein each internal coding frame list includes a number of internal coding frames.

[0058] Among them, the internal coding frame, also known as the key frame, is an independent frame that carries all the information and can be decoded independently without referring to other images.

[0059] According to the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, this embodiment can obtain a preset acquisition time list, and segment each initial video according to the preset acquisition time list to obtain segmented videos, and then obtain the corresponding sampling interval based on the frame rate corresponding to the corresponding video acquisition device, and obtain the corresponding internal coding frame list by sampling from each segmented video based on the sampling interval. By using the internal coding frames sampled from the initial video as the basis for video decoding, the time required for video decoding is reduced and the video decoding efficiency is improved.

[0060] In order to further improve the efficiency of video decoding, the sampling interval may be increased based on actual conditions, thereby extracting fewer internal coding frames as the basis for video decoding.

[0061] The specific size of the sampling interval can be set by the implementer according to actual conditions. Generally, the sampling interval for sampling two internal coding frames can vary with the frame rate. For example, the sampling interval for sampling two internal coding frames can be set to twice the frame rate, and each segmented video can be sampled according to the sampling interval to obtain the corresponding internal coding frame list.

[0062] As described above, based on the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, a list of internal coded frames corresponding to each event type is sampled from each initial video, and the sampled internal coded frames are used as the basis for video decoding, thereby reducing the time required for video decoding and improving video decoding efficiency.

[0063] S3, clustering all the internal coding frames according to the resolution and position corresponding to each video acquisition device, and obtaining K sets of internal coding frames, where K is an integer greater than 1.

[0064] In a specific implementation, S3 further includes the following steps: S31, clustering corresponding internal coding frames with consistent resolutions into the same first clustering set according to the resolution corresponding to each video acquisition device.

[0065] S32, for any first cluster set, cluster the internal coding frames in the current first cluster set according to the position corresponding to each video acquisition device, and obtain k1 internal coding frame sets, where 0<k1≤K.

[0066] S33, traverse all first cluster sets to obtain K internal coding frame sets.

[0067] Among them, the internal coding frames in each first clustering set are clustered according to a preset clustering algorithm to obtain a corresponding number of internal coding frame sets, wherein each internal coding frame set includes a number of internal coding frames with consistent resolution and high correlation, which serve as the data basis for each decoding unit to perform video decoding.

[0068] The preset clustering algorithm can be set by the implementer according to the actual situation. For example, the preset clustering algorithm can be a clustering algorithm such as DBSCAN, k-mesas, etc. Those skilled in the art know that any video compression method in the prior art falls within the protection scope of the present invention, which will not be described in detail here.

[0069] In the above, according to the resolution and position corresponding to each video acquisition device, several internal coding frames with consistent resolution and high correlation are clustered into the same internal coding frame set, which provides a data basis for the decoding unit to perform video decoding.

[0070] S4, splicing all the internal coding frames in each internal coding frame set respectively to obtain K splicing data.

[0071] S5, sending the K spliced ​​data to a decoding unit in the video platform for decoding, and obtaining a target decoded image corresponding to each internally encoded frame.

[0072] Among them, the data platform includes several decoding units.

[0073] All the internal coding frames in each internal coding frame set are spliced ​​together to obtain spliced ​​data corresponding to each internal coding frame set. Each spliced ​​data includes a number of internal coding frames with consistent resolution and high correlation.

[0074] Each spliced ​​data is sent to a decoding unit in the video platform for decoding, and a target decoded image corresponding to each internal coding frame in the spliced ​​data is obtained.

[0075] As described above, each spliced ​​data includes several internal coding frames with consistent resolution and high correlation. One spliced ​​data is decoded based on one coding unit, that is, a batch of sampled internal coding frames are decoded simultaneously by one decoding unit, which reduces the decoding time of the decoding unit and improves the decoding efficiency of the decoding unit.

[0076] In a specific implementation, the video platform includes a data acquisition unit, a data storage unit, a data processing unit and a decoding unit.

[0077] The data acquisition unit includes M video acquisition devices.

[0078] The data storage unit stores the resolution corresponding to each video acquisition device, the initial video corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type.

[0079] The data processing unit stores a computer program, and when the computer program is executed by the processor, steps S1-S5 are implemented.

[0080] The decoding unit is used to decode the spliced ​​data and output a decoded target decoded image.

[0081] Among them, the data storage unit and the data processing unit are both set in the data platform. The data processing unit can directly sample, cluster and splice the initial video based on the initial video and related parameters stored in the data storage unit without downloading the video data, and further decode the spliced ​​data based on the decoding unit in the data platform.

[0082] As described above, the data processing unit directly performs data processing based on the initial video and related parameters stored in the data storage unit, and the decoding unit directly decodes the spliced ​​data, eliminating the step of downloading the initial video, greatly reducing the time required for data decoding, and improving the efficiency of data decoding.

[0083] In the above, according to the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, a list of internal coding frames corresponding to each event type is sampled from each initial video, and according to the resolution and position corresponding to each video acquisition device, all internal coding frames are clustered to obtain K sets of internal coding frames, and all internal coding frames in each set of internal coding frames are spliced ​​to obtain K spliced ​​data, and the K spliced ​​data are sent to the decoding unit in the video platform for decoding, and a target decoded image corresponding to each internal coding frame is obtained. It can be seen that the initial video is directly sampled based on the initial video and related parameters stored in the data storage unit to obtain the internal coding frame, which reduces the amount of data that needs to be decoded, and there is no need to download video data, which greatly reduces the time for video decoding, and further clusters and splices the internal coding frames to obtain spliced ​​data consisting of several internal coding frames with consistent resolution and high correlation, so as to decode the spliced ​​data based on the decoding unit, thereby improving the efficiency of video decoding.

[0084] Embodiment 2 This second embodiment provides a video decoding device based on sampling, such as Figure 2 As shown, the video decoding device includes: The data acquisition module 21 is used to obtain M video acquisition devices in the video platform, the resolution, frame rate, position and initial video corresponding to each video acquisition device, as well as the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, wherein each event type list includes several event types, and M refers to an integer greater than 0.

[0085] The sampling module 22 is used to sample from each initial video to obtain an internal coding frame list corresponding to each event type according to the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, wherein each internal coding frame list includes a plurality of internal coding frames.

[0086] The clustering module 23 is used to cluster all the internal coding frames according to the resolution and position corresponding to each video acquisition device, and obtain K sets of internal coding frames, where K is an integer greater than 1.

[0087] The splicing module 24 is used to splice all the internal coding frames in each internal coding frame set to obtain K splicing data.

[0088] The decoding module 25 is used to send the K spliced ​​data to the decoding unit in the video platform for decoding, and obtain the target decoded image corresponding to each internal coding frame.

[0089] In a specific implementation, the data acquisition module 21 further includes: The data acquisition submodule is used to acquire the historical video corresponding to each video acquisition device within a preset historical time period.

[0090] The sampling submodule is used to sample from the corresponding historical video according to the frame rate corresponding to each video acquisition device to obtain the first internal coding frame list corresponding to each historical video, wherein each first internal coding frame list includes a plurality of internal coding frames.

[0091] The recognition submodule is used to input the first internal coding frame list into the trained recognition model to obtain a plurality of second internal coding frame lists and the event type corresponding to each second internal coding frame list.

[0092] The mixed event image frame acquisition submodule is used to acquire the mixed event image frame in each second internal coding frame list according to each second internal coding frame list.

[0093] The mixed event image frame removal submodule is used to remove the corresponding mixed event image frame from each second internal coding frame list to obtain the corresponding third internal coding frame list.

[0094] The first time set acquisition submodule is used to acquire the first time set corresponding to each event type according to the first time corresponding to each third internal coding frame list and the event type corresponding to the third internal coding frame list.

[0095] The preset collection time acquisition submodule is used to acquire the preset collection time corresponding to each event type according to the first time set corresponding to each event type.

[0096] In a specific implementation, the mixed event image frame acquisition submodule further includes: The first mixed event image frame acquisition unit is used to determine, for any second internal coding frame, if the current second internal coding frame belongs to two or more second internal coding frame lists, the current second internal coding frame as a mixed event image frame.

[0097] The second mixed event image frame acquisition unit is used to traverse all second internal coding frames and acquire all mixed event image frames in each internal coding frame list.

[0098] In a specific implementation, the preset acquisition time acquisition submodule further includes: The time quantity acquisition unit is used to acquire, for any event type, the first quantity of each first time corresponding to the current event type and the total quantity of the first time in the first time set according to the first time set corresponding to the current event type.

[0099] The quantity ratio acquisition unit is used to determine, for any first time corresponding to the current event type, a ratio of the first quantity corresponding to the current first time to the total quantity as the quantity ratio corresponding to the current first time.

[0100] The preset collection time acquisition unit is used to acquire the preset collection time corresponding to the current event type according to each first time and the quantity ratio corresponding to each first time.

[0101] In a specific implementation, the data acquisition module 21 further includes: The collected video acquisition submodule is used to obtain the collected video corresponding to each video collection device in a preset current time period.

[0102] The video compression submodule is used to compress each captured video to obtain the corresponding initial video.

[0103] In a specific implementation, the clustering module 23 further includes: The first clustering submodule is used to cluster corresponding internal coding frames with consistent resolutions into the same first clustering set according to the resolution corresponding to each video acquisition device.

[0104] The second clustering submodule is used to cluster the internal coding frames in the current first clustering set according to the position corresponding to each video acquisition device for any first clustering set, and obtain k1 internal coding frame sets, where 0<k1≤K.

[0105] The third clustering submodule is used to traverse all first clustering sets to obtain K internal coding frame sets.

[0106] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0107] Embodiment 3 Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the non-transitory computer-readable storage medium, and the at least one instruction or at least one program is loaded and executed by a processor to implement the steps: S1, obtain M video acquisition devices in the video platform, the resolution, frame rate, position and initial video corresponding to each video acquisition device, as well as the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, wherein each event type list includes several event types, and M refers to an integer greater than 0.

[0108] S2, according to the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, sample from each initial video to obtain an internal coding frame list corresponding to each event type, wherein each internal coding frame list includes a number of internal coding frames.

[0109] S3, clustering all the internal coding frames according to the resolution and position corresponding to each video acquisition device, and obtaining K sets of internal coding frames, where K is an integer greater than 1.

[0110] S4, splicing all the internal coding frames in each internal coding frame set respectively to obtain K splicing data.

[0111] S5, sending the K spliced ​​data to a decoding unit in the video platform for decoding, and obtaining a target decoded image corresponding to each internally encoded frame.

[0112] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronization link DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0113] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0114] Embodiment 4 A fourth embodiment of the present invention provides an electronic device, which includes a processor and the non-transitory computer-readable storage medium in the third embodiment of the present invention.

[0115] The above are only preferred embodiments of the present invention, and are not intended to limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any technician familiar with the profession can make some changes or modifications to equivalent embodiments of equivalent changes using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A sampling-based video decoding method, characterized in that: The video decoding method comprises the following steps: S1, obtaining M video acquisition devices in the video platform, the resolution, frame rate, position and initial video corresponding to each video acquisition device, and the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, wherein each event type list includes a number of event types, and M refers to an integer greater than 0; S2, sampling from each initial video to obtain an internal coding frame list corresponding to each event type according to the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video, and the preset acquisition time corresponding to each event type, wherein each internal coding frame list includes a plurality of internal coding frames; S3, clustering all the internal coding frames according to the resolution and position corresponding to each video acquisition device to obtain K sets of internal coding frames, where K is an integer greater than 1; S4, splicing all the internal coding frames in each internal coding frame set respectively to obtain K splicing data; S5, sending the K spliced ​​data to a decoding unit in the video platform for decoding, and obtaining a target decoded image corresponding to each internally encoded frame.

2. The sampling-based video decoding method according to claim 1, characterized in that: The video platform includes a data acquisition unit, a data storage unit, a data processing unit and a decoding unit; Wherein, the data acquisition unit includes M video acquisition devices; The data storage unit stores the resolution corresponding to each video acquisition device, the initial video corresponding to each video acquisition device, the event type list corresponding to each initial video, and the preset acquisition time corresponding to each event type; The data processing unit stores a computer program, and when the computer program is executed by the processor, steps S1-S5 are implemented; The decoding unit is used to decode the spliced ​​data and output a decoded target decoded image.

3. The sampling-based video decoding method according to claim 1, characterized in that: S1 also includes the following steps: S110, obtaining historical videos corresponding to each video acquisition device within a preset historical time period; S120, sampling from the corresponding historical videos according to the frame rate corresponding to each video acquisition device to obtain a first internal coding frame list corresponding to each historical video, wherein each first internal coding frame list includes a plurality of internal coding frames; S130, inputting the first internal coding frame list into a trained recognition model to obtain a plurality of second internal coding frame lists and an event type corresponding to each second internal coding frame list; S140, according to each second internal coding frame list, acquiring a mixed event image frame in each second internal coding frame list; S150, removing the corresponding mixed event image frame from each second internal coding frame list, and acquiring the corresponding third internal coding frame list; S160, acquiring a first time set corresponding to each event type according to the first time corresponding to each third internal coded frame list and the event type corresponding to the third internal coded frame list; S170: Acquire a preset collection time corresponding to each event type according to the first time set corresponding to each event type.

4. The sampling-based video decoding method according to claim 3, characterized in that: S140 also includes the following steps: S141, for any second intra-coded frame, if the current second intra-coded frame belongs to two or more second intra-coded frame lists, determine the current second intra-coded frame as a mixed event image frame; S142, traverse all second intra-coded frames to obtain all mixed event image frames in each intra-coded frame list.

5. The sampling-based video decoding method according to claim 3, characterized in that: S170 also includes the following steps: S171, for any event type, according to the first time set corresponding to the current event type, obtain the first quantity of each first time corresponding to the current event type and the total quantity of the first time in the first time set; S172, for any first time corresponding to the current event type, determine a ratio of the first quantity corresponding to the current first time to the total quantity as a quantity proportion corresponding to the current first time; S173, obtaining a preset collection time corresponding to the current event type according to each first time and the quantity ratio corresponding to each first time.

6. The sampling-based video decoding method according to claim 1, characterized in that: S1 also includes the following steps: S180, obtaining the corresponding captured video of each video capture device within a preset current time period; S190, compress each collected video to obtain a corresponding initial video.

7. The sampling-based video decoding method according to claim 1, characterized in that: S3 also includes the following steps: S31, clustering corresponding internal coding frames with consistent resolutions into the same first clustering set according to the resolution corresponding to each video acquisition device; S32, for any first cluster set, clustering the internal coding frames in the current first cluster set according to the position corresponding to each video acquisition device, to obtain k1 sets of internal coding frames, where 0<k1≤K; S33, traverse all first cluster sets to obtain K internal coding frame sets.

8. A sampling-based video decoding device, characterized in that: The video decoding device comprises: A data acquisition module is used to acquire M video acquisition devices in the video platform, the resolution, frame rate, position and initial video corresponding to each video acquisition device, and an event type list corresponding to each initial video and a preset acquisition time corresponding to each event type, wherein each event type list includes a number of event types, and M refers to an integer greater than 0; A sampling module, used for sampling from each initial video to obtain an internal coding frame list corresponding to each event type according to the frame rate corresponding to each video acquisition device, the event type list corresponding to each initial video and the preset acquisition time corresponding to each event type, wherein each internal coding frame list includes a plurality of internal coding frames; A clustering module, used to cluster all internal coding frames according to the resolution and position corresponding to each video acquisition device, and obtain K sets of internal coding frames, where K is an integer greater than 1; A splicing module, used for splicing all the internal coding frames in each internal coding frame set respectively to obtain K splicing data; A decoding module is used to send the K spliced ​​data to a decoding unit in the video platform for decoding, and obtain a target decoded image corresponding to each internal coding frame.

9. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the non-transitory computer-readable storage medium, characterized in that: The at least one instruction or the at least one program is loaded and executed by the processor to implement the sampling-based video decoding method as described in any one of claims 1-7.

10. An electronic device, characterized in that: The invention comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 9.

Citation Information

Patent Citations

  • System and method for encoding video content using virtual intra-frames

    CN104641638A

  • Video data intelligent compression method and device based on multi-frame content

    CN113132730A

  • Video data transmission method and system, video data injection method and electronic equipment

    CN115665422A

  • Video processing method and device, computer readable storage medium and computer equipment

    CN116939212A

  • Decoding performance detection method and device, electronic equipment and storage medium

    CN117412026A