Computer digital video monitoring system and method based on cloud computing

Abstract

The invention discloses a computer digital video monitoring system and method based on cloud computing, relates to the technical field of video processing, and solves the technical problem that efficient data compression and storage optimization are difficult to realize in a complex scene. Similar frames, same frames and special frames are accurately distinguished, redundancy of traditional full-frame storage is avoided, only reference frames are stored, repeated counting is carried out, the data size is reduced, difference areas are extracted and converted into grey-scale maps, continuous pixel sequences are compressed through stroke length coding, compression efficiency is improved, original data are reserved, and completeness of key pictures is ensured. The resolution and the frame rate are dynamically adjusted according to scene requirements, the code rate is automatically adjusted based on the state of the virtual buffer area, the code rate is reduced to avoid overflow when the buffer area is close to the upper limit, the code rate is improved to fully utilize the bandwidth when the buffer area is close to the lower limit, quantization parameters are dynamically adjusted through a formula, and it is ensured that the code rate is matched with the network transmission capacity.

Description

Technical Field

[0001] The present invention relates to the field of video processing technology, and in particular to a computer digital video monitoring system and method based on cloud computing. Background Art

[0002] In recent years, the performance of computers and peripheral equipment has become increasingly higher while the price has become lower and lower. Traditional monitoring systems have been continuously improved and updated, and digital video monitoring systems have been used more and more widely, gradually moving from important security departments such as finance and high-end hotels to factories, mines, enterprises and even homes.

[0003] According to the patent application with publication number CN106507034A, a computer digital video monitoring system is disclosed, which includes a computer, a network communication device electrically connected to the computer, a digital camera transmitted to the network communication device via a network cable, and a pan-tilt decoder electrically connected to the digital camera; the computer is equipped with a storage device and a network card.

[0004] In the above patent, the digital camera directly transmits the digitized and compressed digital video image stream, and transmits the video stream to the computer through the network. This digital camera can also output a control signal to control the pan-tilt decoder.

[0005] However, while existing technologies offer compression methods based on inter-frame differences, they lack dynamic analysis and classification of video frame similarities, making efficient data compression and storage optimization difficult in complex scenarios. Furthermore, traditional systems lack flexibility in storage strategies for real-time and backup data, making it impossible to dynamically adjust resolution, frame rate, and bitrate based on scenario requirements, leading to irrational resource allocation. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the present invention provides a computer digital video surveillance system and method based on cloud computing, which solves the problem of difficulty in achieving efficient data compression and storage optimization in complex scenarios.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A computer digital video monitoring system based on cloud computing, comprising:

[0008] The video coding optimization processing module is used to segment the digital video image transmitted by the video information acquisition module to obtain unit video images, calculate the similarity of consecutive video frames in the unit video image, and compare it with the preset similarity to obtain similar video frames and special video frames. At the same time, the hash values ​​of all similar video frames are calculated, and similar video frames with the same hash values ​​are selected and recorded as the same video frames. The unit video coding information is generated and transmitted to the video compression optimization module;

[0009] The video compression optimization module is used to analyze the unit video coding information, compress the same video frames, retain the reference frame, and process consecutive identical video frames using a repeat counting method to generate the same compression information. It also compresses similar video frames, determines the difference area based on the difference elements, converts it into a grayscale image, identifies the original pixel sequence in the grayscale image, and compresses it using "value + quantity" to generate similar compression information, and transmits both to the video storage analysis module.

[0010] The video storage analysis module is used to store and process the compressed computer video, classify it to obtain backup data and real-time data, store and process the backup data, adjust the video resolution and frame rate according to the backup requirements, and store and generate backup storage information;

[0011] Perform storage processing on real-time data, generate a virtual buffer area, obtain its corresponding real-time remaining storage space and storage upper and lower limits, compare the sizes, and then calculate the quantization parameter based on the bit rate and buffer occupancy formula, and use it as a standard to adjust storage and generate real-time storage information;

[0012] At the same time, the generated storage information is transmitted to the video processing information display module.

[0013] As a further solution of the present invention, it also includes a video information acquisition module and a video processing information display module;

[0014] The video information acquisition module is used to collect digital video images of the monitoring area through multiple cameras and transmit them to the video encoding optimization processing module;

[0015] The video processing information display module is used to perform corresponding processing on the backup data and the real-time data according to the obtained backup storage information and real-time storage information.

[0016] As a further solution of the present invention, the specific method of the video encoding optimization processing module to screen and obtain similar video frames and special video frames is:

[0017] The digital video is divided into small segments according to time. All frames in each video are extracted and numbered in sequence. The first frame of the first video is used as a reference. The similarity between the subsequent connected frames and it is calculated and compared with the similarity threshold pre-set by the operator.

[0018] If the similarity between the current frame and the reference frame is higher than the threshold, it is marked as a similar video frame; if it is lower than the threshold, it is marked as a special video frame, and so on, completing the classification and marking of all video frames.

[0019] As a further solution of the present invention, the specific method of the video coding optimization processing module to generate unit video coding information is as follows:

[0020] Obtain all similar video frames and special video frames, and label the similar video frames as a, and a=1, 2, ..., b, where b represents the number of similar video frames, and calculate the hash value of the similar video frame a. Compare the hash values ​​in the order of labeling, filter out similar video frames with the same hash value and record them as the same video frames, and generate unit video coding information and transmit it to the video compression optimization module.

[0021] As a further solution of the present invention, the specific manner in which the video compression optimization module generates the same compression information is as follows:

[0022] All consecutive identical video frames are obtained, and their corresponding numbers are obtained. The first video frame is retained as a reference frame, and subsequent consecutive identical video frames are repeatedly counted to obtain compressed identical video frames and generate identical compression information.

[0023] As a further solution of the present invention, the specific manner in which the video compression optimization module generates similar compression information is as follows:

[0024] The original video frame corresponding to the similar video frame is obtained, and then the difference elements between the similar video frame and the original video frame are obtained. The corresponding difference area is obtained based on the difference elements, and the area is converted into a grayscale image. Then, the original pixel sequence of the grayscale image is obtained, and run-length encoding is performed according to the original pixel sequence to generate similar compressed information.

[0025] As a further solution of the present invention, the specific manner in which the video compression optimization module performs run-length encoding processing according to the original pixel sequence is:

[0026] Obtain continuous and non-continuous repeating sequences in the same group of original pixel sequences. For continuous repeating sequences, use the "value + quantity" method to perform compression processing, and for non-continuous repeating sequences, do not process them. Similarly, process all original pixel sequences in the same group to generate similar compression information.

[0027] As a further solution of the present invention, the specific manner in which the video storage analysis module generates the backup storage information is as follows:

[0028] Obtain all backup data and classify them with time T as the period to obtain time period data, and label them as r, where r = 1, 2, ..., e, where e represents the number of time period data, and obtain the video resolution and frame rate corresponding to the time period data r. Then obtain the backup requirements corresponding to the backup data, and the backup requirements here are expressed as the corresponding scene requirements. Based on the backup requirements, the video resolution and frame rate are adjusted to obtain adjustment information, and the adjustment information is stored to generate backup storage information.

[0029] As a further solution of the present invention, the specific manner in which the video storage analysis module generates real-time storage information is as follows:

[0030] Generate a virtual buffer area based on real-time data, obtain the storage upper limit and storage lower limit corresponding to the virtual buffer area, and obtain the corresponding real-time remaining storage space, and then compare the real-time remaining storage space with the storage upper limit and storage lower limit;

[0031] If the real-time remaining storage space is close to the storage upper limit, a bit rate reduction signal is generated; conversely, if the real-time remaining storage space is close to the storage lower limit, a bit rate increase signal is generated;

[0032] At the same time, according to the formula QP=QP base +ΔQP×f calculates the quantization parameter QP, where QP base is the basic quantization parameter, ΔQP is the quantization parameter adjustment amount, f is the buffer occupancy rate, and the bit rate of the real-time data is adjusted according to the calculated quantization parameter QP. At the same time, the real-time data is stored to generate real-time storage information.

[0033] A computer digital video monitoring method based on cloud computing, the method specifically comprising the following steps:

[0034] Step S1: collecting computer video and converting it into digital signals, and segmenting it to obtain unit video images;

[0035] Step S2: Calculate the similarity of consecutive video frames in the unit video image, and compare with the preset similarity to obtain similar video frames and special video frames;

[0036] Step S3: Calculate the hash values ​​of all similar video frames, select similar video frames with the same hash value and record them as the same video frames, and generate unit video coding information;

[0037] Step S4: compress the same video frames, retain the reference frame, and process consecutive same video frames in a repeat counting manner to generate the same compression information;

[0038] Step S5: compress similar video frames, determine the difference area based on the difference elements, convert it into a grayscale image, identify the original pixel sequence in the grayscale image, and compress it using "value + quantity" to generate similar compression information;

[0039] Step S6: Acquire computer video based on the compressed information, classify and obtain backup data and real-time data, and perform storage analysis on the two respectively to generate storage information.

[0040] The present invention provides a computer digital video surveillance system and method based on cloud computing. Compared with the existing technology, it has the following advantages:

[0041] The present invention calculates the similarity between video frames and compares it with a preset threshold to accurately distinguish similar frames, identical frames and special frames, avoiding the redundancy of traditional full-frame storage. It only stores reference frames and repeat counts to reduce the amount of data, extracts the difference area and converts it into a grayscale image, and uses run-length encoding (RLE) to compress continuous pixel sequences, improving compression efficiency, preserving the original data, and ensuring the integrity of key images.

[0042] The present invention dynamically adjusts the resolution and frame rate according to the scene requirements, and automatically adjusts the bit rate based on the virtual buffer status. When the buffer is close to the upper limit, the bit rate is reduced to avoid overflow. When it is close to the lower limit, the bit rate is increased to fully utilize the bandwidth. The quantization parameters are dynamically adjusted through the corresponding formula to ensure that the bit rate matches the network transmission capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a block diagram of the system principle of the present invention;

[0044] Figure 2 It is a diagram of the steps of the present invention. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0046] Example 1

[0047] See also Figure 1 The present application provides a computer digital video surveillance system based on cloud computing, including: a video information acquisition module, a video encoding optimization processing module, a video compression optimization module, a video storage analysis module and a video processing information display module, and combined with Figure 1 It can be known that the functional modules are electrically connected in a unidirectional manner.

[0048] The video information acquisition module is used to acquire computer video. Specifically, it collects digital video images of the monitored area through multiple cameras, converts them into digital signals, and transmits them to the video encoding optimization processing module.

[0049] A video encoding optimization processing module is used to optimize the obtained digital video image, segment the obtained digital video image to obtain unit video images, and here the segmentation is performed according to the unit time second, and at the same time, the video frames of the unit video image are obtained and labeled as i, where i=1, 2, ..., j, where j represents the number of video frames. Then, the similarity of the video frame with label i=1 is calculated as the standard, and the obtained similarity is compared with a preset similarity, and the specific value of the preset similarity is set by the operator;

[0050] The input digital video image is divided into multiple independent unit video images at equal intervals according to the unit time (such as 1 second). For each unit video image, the video frames it contains are obtained in chronological order and assigned a unique label i, where i = 1, 2, ..., j, where j represents the number of video frames. The video frame labeled i = 1 is used as the initial reference frame, and an efficient image similarity calculation algorithm (such as a perceptual hash algorithm based on a hash algorithm, a mean square error algorithm based on pixel differences, or a feature extraction algorithm based on deep learning) is used to calculate its similarity with subsequent adjacent video frames. The calculated similarity is compared with a preset similarity threshold set by the operator.

[0051] If the similarity is greater than the preset similarity, the two are marked as similar video frames. Otherwise, if the similarity is less than the preset similarity, they are not processed and marked as special video frames. Similarly, all video frames i are processed in the same way to obtain similar video frames and special video frames.

[0052] If the similarity between the current video frame and the reference frame is greater than a preset similarity, they are marked as similar video frames;

[0053] If the similarity is less than the preset value, the current video frame is marked as a special video frame and is not included in the similar video frame set;

[0054] Specifically, when analyzing the video frame labeled 1, the similarity between the video frames labeled 1 and 2 is first calculated. If the two are classified as similar video frames, the similarity between the video frames labeled 2 and 3 is calculated. If the two are also classified as similar video frames, the similarity between the video frames labeled 1 and 3 is further calculated. If the similarity between the two is less than the preset similarity, the video frames labeled 1 and 2 are marked as similar video frames, and the video frame labeled 3 is not classified into the two. If the similarity between the two is greater than the preset similarity, the video frames labeled 1, 2 and 3 are all classified as similar video frames. And so on, all video frames are processed in the same way.

[0055] All similar video frames and special video frames are obtained, and here they are represented as all video frames in a unit video image. At the same time, similar video frames are labeled as a, and a=1, 2, ..., b, where b represents the number of similar video frames. The hash value of similar video frame a is calculated, and the hash values ​​are compared in the order of the labels. Similar video frames with the same hash value are filtered and recorded as the same video frames. The remaining ones are not processed and are still recorded as similar video frames. At the same time, unit video coding information is generated and transmitted to the video compression optimization module. In order to deal with abnormal situations such as noise interference and sudden picture changes that may occur in video data, the module adds an abnormality detection mechanism. When the similarity calculation results of multiple consecutive video frames fluctuate drastically, it automatically triggers recalculation or adjustment of the parameters of the similarity calculation algorithm. At the same time, it supports operators to dynamically adjust the preset similarity threshold according to different video scenes (such as surveillance videos, film and television programs) to achieve the best video frame classification effect.

[0056] The video compression optimization module is used to analyze the obtained unit video encoding information, obtain the corresponding identical video frames and similar video frames, and compress the two respectively. The specific processing is as follows:

[0057] Compress the obtained identical video frames to obtain all consecutive identical video frames and their corresponding numbers, while retaining the first video frame as a reference frame, and performing repeat counting processing on subsequent consecutive identical video frames to obtain compressed identical video frames, generating identical compression information, and including reference frame index, repeat count list and timestamp information;

[0058] For example, let's say that after analyzing a traffic surveillance video, multiple consecutive identical video frames are detected within a 10-second video clip. From the 1st second to the 3rd second, there are three identical video frames. These are grouped together, and the video frame at the 1st second is retained as the reference frame, with a repeat count of 2. From the 5th to the 7th second, there are also three identical video frames, with the video frame at the 5th second being used as the reference frame, and a repeat count of 2. From the 8th to the 10th second, there are three identical video frames, with the video frame at the 8th second being used as the reference frame, and a repeat count of 2. The corresponding reference frame index, repeat count list, and timestamp information are as follows:

[0059] Reference frame index: [1, 5, 8] (indicates the position of the reference frame in the original 10 video frame sequence).

[0060] Repeat count list: [2, 2, 2].

[0061] Timestamp information: [[0, 3], [4, 7], [7, 10]] (representing the start and end time of each packet in the video, in seconds).

[0062] The obtained similar video frames are compressed to obtain the original video frames corresponding to the similar video frames, and the original video frames are represented as video frames for comparison with the similar video frames as a reference. Then, the difference elements between the similar video frames and the original video frames are obtained, and the corresponding difference areas are obtained based on the difference elements. The difference areas here include their corresponding positions, sizes, and encoding parameters. At the same time, they are converted into grayscale images. Then, the original pixel sequence of the grayscale image is obtained, and run-length encoding is performed according to the original pixel sequence. The specific processing method is as follows:

[0063] Obtain continuous and non-continuous repeating sequences in the same group of original pixel sequences. For continuous repeating sequences, use the "value + quantity" method to compress them, and do not process non-continuous repeating sequences. For example, the original pixel sequence is AAAAABBBCCD, and the compressed pixel sequence is A5B3C2D. Similarly, all original pixel sequences in the same group are processed to generate similar compression information. Specifically, special video frames are not processed and the original frame number is retained.

[0064] At the same time, the generated compressed information is transmitted to the video storage analysis module.

[0065] The video storage and analysis module is used to store and process computer videos based on the obtained compression information, classify computer videos according to time t to obtain backup data and real-time data, and the specific value of time t is set by the operator. Specifically, the video before time t is backed up, and the data after time t is recorded as real-time data. The real-time data generated by monitoring is stored and processed separately.

[0066] Perform storage processing on the backup data, obtain all backup data, and classify them based on time period T to obtain time period data. At the same time, label them as r, where r = 1, 2, ..., e, where e represents the number of time period data. The video resolution and frame rate corresponding to the time period data r are obtained. Then, the backup requirements corresponding to the backup data are obtained, and the backup requirements here are expressed as corresponding scene requirements. The video resolution and frame rate are adjusted based on the backup requirements to obtain adjustment information. The adjustment information is stored and generated into backup storage information.

[0067] Perform storage processing on real-time data, obtain real-time data, obtain the corresponding storage node, and obtain the network transmission rate and reading speed corresponding to the storage node. Combined with the two, select the storage node with the highest storage performance as the storage target. Then, dynamically adjust the bit rate according to the image complexity corresponding to the real-time data. The specific adjustment method is as follows:

[0068] Generate a virtual buffer area based on real-time data, obtain the storage upper limit and storage lower limit corresponding to the virtual buffer area, and obtain the corresponding real-time remaining storage space, and then compare the real-time remaining storage space with the storage upper limit and storage lower limit;

[0069] If the real-time remaining storage space approaches the storage limit, it means that the buffer is about to be filled. If it is not controlled, there will be no place to store the newly generated video data, which may cause data loss or playback interruption, and generate a signal to reduce the bit rate;

[0070] On the contrary, if the real-time remaining storage space is close to the storage lower limit, it means that when the buffer has more remaining space, it means that the data generation rate is lower than the transmission or processing rate at the current bit rate. At this time, the bandwidth resources are not fully utilized, and a bit rate increase signal is generated;

[0071] At the same time, according to the formula QP=QP base +ΔQP×f calculates the quantization parameter QP, where QP base is the basic quantization parameter, ΔQP is the quantization parameter adjustment amount, f is the buffer occupancy rate, and the bit rate of the real-time data is adjusted according to the calculated quantization parameter QP. At the same time, the real-time data is stored to generate real-time storage information.

[0072] The generated backup storage information and real-time storage information are transmitted to the video processing information display module.

[0073] The video processing information display module is used to perform corresponding processing on the backup data and the real-time data according to the obtained backup storage information and real-time storage information.

[0074] Example 2

[0075] See also Figure 2 The present application provides a computer digital video monitoring method based on cloud computing, which specifically includes the following steps:

[0076] Step S1: collect computer video and convert it into digital signals, and segment it to obtain unit video images. The specific processing method is the same as that of the video encoding optimization processing module in Example 1;

[0077] Step S2: Calculate the similarity of consecutive video frames in the unit video image, and compare with the preset similarity to obtain similar video frames and special video frames. The specific processing method is the same as that of the video encoding optimization processing module in Example 1.

[0078] Step S3: Calculate the hash values ​​of all similar video frames, select similar video frames with the same hash value and record them as the same video frames, generate unit video coding information, and the specific processing method is the same as the processing method of the video coding optimization processing module in Example 1;

[0079] Step S4: compress the same video frames, retain the reference frame, and process consecutive identical video frames in a repeat counting manner to generate the same compression information. The specific processing method is the same as that of the video compression optimization module in Example 1.

[0080] Step S5: compress similar video frames, determine the difference area based on the difference elements, convert the image into a grayscale image, identify the original pixel sequence in the grayscale image, and compress it using "value + quantity" to generate similar compressed information. The specific processing method is similar to that of the video encoding optimization processing module in Example 1.

[0081] Step S6: Obtain computer video based on the compressed information, classify it into backup data and real-time data, and perform storage analysis on the two respectively to generate storage information. The specific processing method is the same as that of the video storage analysis module in Example 1.

[0082] Some of the data in the above formulas are calculated based on their numerical values ​​and are not substituted into parameter units for calculation. At the same time, the contents not described in detail in this specification belong to the existing technology known to those skilled in the art.

[0083] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A computer digital video surveillance system based on cloud computing, characterized in that: include: The video coding optimization processing module is used to segment the digital video image transmitted by the video information acquisition module to obtain unit video images, calculate the similarity of consecutive video frames in the unit video image, and compare it with the preset similarity to obtain similar video frames and special video frames. At the same time, the hash values ​​of all similar video frames are calculated, and similar video frames with the same hash values ​​are selected and recorded as the same video frames. The unit video coding information is generated and transmitted to the video compression optimization module; The video compression optimization module is used to analyze the unit video encoding information, compress identical video frames, retain reference frames, and process consecutive identical video frames using a repeat counting method to generate identical compression information. It also compresses similar video frames, determines the difference area based on the difference elements, converts it into a grayscale image, identifies the original pixel sequence in the grayscale image, and compresses it using "value + quantity" to generate similar compression information. Both are transmitted to the video storage analysis module. The video storage analysis module is used to store and process the compressed computer video, classify it to obtain backup data and real-time data, store and process the backup data, adjust the video resolution and frame rate according to the backup requirements, and store and generate backup storage information; Perform storage processing on real-time data, generate a virtual buffer area, obtain its corresponding real-time remaining storage space and storage upper and lower limits, compare the sizes, and then calculate the quantization parameter based on the bit rate and buffer occupancy formula, and use it as a standard to adjust storage and generate real-time storage information; At the same time, the generated storage information is transmitted to the video processing information display module.

2. A computer digital video surveillance system based on cloud computing according to claim 1, characterized in that: It also includes a video information acquisition module and a video processing information display module; The video information acquisition module is used to collect digital video images of the monitoring area through multiple cameras and transmit them to the video encoding optimization processing module; The video processing information display module is used to perform corresponding processing on the backup data and the real-time data according to the obtained backup storage information and real-time storage information.

3. The computer digital video surveillance system based on cloud computing according to claim 1, characterized in that: The specific method of the video coding optimization processing module to screen and obtain similar video frames and special video frames is as follows: The digital video is divided into small segments according to time. All frames in each video are extracted and numbered in sequence. The first frame of the first video is used as a reference. The similarity between the subsequent connected frames and it is calculated and compared with the similarity threshold pre-set by the operator. If the similarity between the current frame and the reference frame is higher than the threshold, it is marked as a similar video frame; if it is lower than the threshold, it is marked as a special video frame, and so on, completing the classification and marking of all video frames.

4. The computer digital video surveillance system based on cloud computing according to claim 1, characterized in that: The specific method for the video coding optimization processing module to generate unit video coding information is as follows: Obtain all similar video frames and special video frames, and label the similar video frames as a, and a=1, 2, ..., b, where b represents the number of similar video frames, and calculate the hash value of the similar video frame a. Compare the hash values ​​in the order of labeling, filter out similar video frames with the same hash value and record them as the same video frames, and generate unit video coding information and transmit it to the video compression optimization module.

5. The computer digital video surveillance system based on cloud computing according to claim 1, characterized in that: The specific method for the video compression optimization module to generate the same compression information is as follows: All consecutive identical video frames are obtained, and their corresponding numbers are obtained. The first video frame is retained as a reference frame, and subsequent consecutive identical video frames are repeatedly counted to obtain compressed identical video frames and generate identical compression information.

6. The computer digital video surveillance system based on cloud computing according to claim 1, characterized in that: The specific method for the video compression optimization module to generate similar compression information is as follows: The original video frame corresponding to the similar video frame is obtained, and then the difference elements between the similar video frame and the original video frame are obtained. The corresponding difference area is obtained based on the difference elements, and the area is converted into a grayscale image. Then, the original pixel sequence of the grayscale image is obtained, and run-length encoding is performed according to the original pixel sequence to generate similar compressed information.

7. The computer digital video surveillance system based on cloud computing according to claim 6, characterized in that: The specific method of the video compression optimization module performing run-length encoding processing according to the original pixel sequence is as follows: Obtain continuous and non-continuous repeating sequences from the same group of original pixel sequences. For continuous repeating sequences, use the "value + quantity" method to perform compression processing, and for non-continuous repeating sequences, do not process them. Similarly, process all original pixel sequences in the same group to generate similar compression information.

8. The computer digital video surveillance system based on cloud computing according to claim 1, characterized in that: The specific method for the video storage analysis module to generate backup storage information is as follows: Obtain all backup data and classify them with time T as the period to obtain time period data, and label them as r, where r = 1, 2, ..., e, where e represents the number of time period data, and obtain the video resolution and frame rate corresponding to the time period data r. Then obtain the backup requirements corresponding to the backup data, and the backup requirements here are expressed as the corresponding scene requirements. Based on the backup requirements, the video resolution and frame rate are adjusted to obtain adjustment information, and the adjustment information is stored to generate backup storage information.

9. The computer digital video surveillance system based on cloud computing according to claim 1, characterized in that: The specific method for the video storage analysis module to generate real-time storage information is as follows: Generate a virtual buffer area based on real-time data, obtain the storage upper limit and storage lower limit corresponding to the virtual buffer area, and obtain the corresponding real-time remaining storage space, and then compare the real-time remaining storage space with the storage upper limit and storage lower limit; If the real-time remaining storage space is close to the storage upper limit, a bit rate reduction signal is generated; conversely, if the real-time remaining storage space is close to the storage lower limit, a bit rate increase signal is generated; At the same time, according to the formula QP=QP base +ΔQP×f calculates the quantization parameter QP, where QP base is the basic quantization parameter, ΔQP is the quantization parameter adjustment amount, f is the buffer occupancy rate, and the bit rate of the real-time data is adjusted according to the calculated quantization parameter QP. At the same time, the real-time data is stored to generate real-time storage information.

10. A computer digital video monitoring method based on cloud computing, executed by the computer digital video monitoring system according to any one of claims 1 to 9, characterized in that: The method specifically comprises the following steps: Step S1: collecting computer video and converting it into digital signals, and segmenting it to obtain unit video images; Step S2: Calculate the similarity of consecutive video frames in the unit video image, and compare with the preset similarity to obtain similar video frames and special video frames; Step S3: Calculate the hash values ​​of all similar video frames, select similar video frames with the same hash value and record them as the same video frames, and generate unit video coding information; Step S4: compress the same video frames, retain the reference frame, and process consecutive same video frames in a repeat counting manner to generate the same compression information; Step S5: compress similar video frames, determine the difference area based on the difference elements, convert it into a grayscale image, identify the original pixel sequence in the grayscale image, and compress it using "value + quantity" to generate similar compressed information; Step S6: Acquire computer video based on the compressed information, classify and obtain backup data and real-time data, and perform storage analysis on the two respectively to generate storage information.

Images