Monitoring data acquisition and analysis methods, systems, equipment and storage media

Abstract

This invention relates to the field of surveillance data processing technology, and particularly to methods, systems, devices, and storage media for surveillance data acquisition and analysis. Addressing the characteristic of fixed monitoring areas in surveillance equipment, where background changes are minimal and largely unchanging over a period of time, and the background is not a key monitoring element, this invention pre-stores background encoded data for a specific period. During real-time monitoring, foreground and background images are separated, with only the foreground encoded frame-by-frame, while the background image is encoded at intervals. The encoded information is compared with pre-stored background encoded information to adjust the background image encoding interval. Since frame-by-frame encoding and transmission of background image encoded data are no longer necessary, significant storage resources and data transmission volume are saved, and video decoding efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of surveillance data processing technology, and in particular to surveillance data acquisition and analysis methods, systems, devices and storage media. Background Technology

[0002] To improve the clarity of surveillance videos, high-definition camera-based surveillance equipment has been widely used. However, since surveillance equipment needs to store or transmit surveillance video information, although existing H.265 video encoding and decoding technology can compress surveillance video information and save storage resources and network bandwidth, the amount of surveillance video information that needs to be stored or transmitted is still very large in application scenarios that require long-term monitoring.

[0003] Chinese patent application CN201010553038 discloses a video encoding and decoding method and apparatus. This patent segments a video image into a foreground image and a background image, encodes the foreground image, and builds and encodes a background model when the background is updated. This scheme uses a foreground segmentation algorithm to separate the foreground from the background, transmitting the unchanged background only once, while encoding and transmitting the foreground region containing important content. This saves encoder resources and reduces bandwidth usage during video transmission; it also obviously saves storage space when storing the transmitted video stream at the backend.

[0004] However, in some scenarios of surveillance video, it is rare for the background to remain completely unchanged. For example, changes in the lighting and shadows in the monitored area, as well as the movement of people, will cause some changes in the background. If the above-mentioned patented solution is adopted, the amount of video data saved will be limited.

[0005] Therefore, there is an urgent need to improve the methods for collecting and analyzing surveillance video data, and further reduce the amount of data collected and analyzed. Summary of the Invention

[0006] The technical problem to be solved by this invention is: there is an urgent need to improve the methods for collecting and analyzing surveillance video data, and to further reduce the amount of data collected and analyzed from surveillance video data.

[0007] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0008] The monitoring data collection and analysis methods include the following steps:

[0009] S11: Obtain a surveillance background video;

[0010] S12: Divide the above-mentioned monitoring background video into multiple video segments according to the time period of the clock;

[0011] S13: Encode each video segment frame by frame to obtain the first encoded information;

[0012] S14: Store the first encoded information;

[0013] S21: Real-time acquisition of surveillance video;

[0014] S22: Separate the foreground and background of each frame in the surveillance video;

[0015] S23: Encode each frame of the foreground image separated from the surveillance video to obtain the second encoded information;

[0016] S24: Encode the isolated single-frame background image in the surveillance video at every first interval to obtain the third encoded information;

[0017] S25: Compare the third encoded information with the first encoded information of all frames in the corresponding time period. If the similarity is less than the preset threshold, then encode the single-frame background image separated in the monitoring video at the subsequent second interval time to obtain the third encoded information; repeat S25.

[0018] S26: Send the second encoded stream and the third encoded stream;

[0019] S27: Decode the second and third encoded streams to obtain the decoded background and foreground images. Merge the obtained background and foreground images to obtain the decoded surveillance video, wherein the background images of each frame are the same during the first or second time interval.

[0020] Furthermore, in the above-mentioned monitoring data collection and analysis method, in step S25, if the similarity is greater than a preset threshold, step S24 is repeated.

[0021] Furthermore, in the above-mentioned monitoring data collection and analysis method, S11 specifically refers to: acquiring monitoring background videos with a time span of more than one day.

[0022] Furthermore, in the above-mentioned monitoring data collection and analysis method, in step S12, the span of each time period is 1-3 hours.

[0023] Furthermore, in the above-mentioned monitoring data collection and analysis method, the first encoded information, the second encoded information, and the third encoded information are based on the H.265 protocol.

[0024] Furthermore, in the above-mentioned monitoring data collection and analysis method, S11 specifically involves: acquiring monitoring background videos with a time span of one day.

[0025] The present invention also provides a monitoring data acquisition and analysis system, comprising:

[0026] The camera module is used to acquire a surveillance background video.

[0027] The separation module is used to separate the above-mentioned monitoring background video into multiple video segments;

[0028] The first encoding module is used to encode each video segment frame by frame to obtain the first encoding information;

[0029] The storage module is used to store the first encoded information;

[0030] The camera module is used to acquire surveillance video.

[0031] The separation module is used to separate the foreground and background of each frame in the surveillance video.

[0032] The second encoding module is used to separate the foreground and background of each frame in the surveillance video.

[0033] The third encoding module is used to encode the single-frame background image separated in the monitoring video at every first interval to obtain the third encoding information;

[0034] The comparison module is used to compare the third encoding information with the first encoding information of all frames in the corresponding time period. If the similarity is less than the preset threshold, the third encoding information is obtained by encoding the single-frame background image separated in the monitoring video at the subsequent second interval time.

[0035] The sending module is used to send the second encoded stream and the third encoded stream;

[0036] The decoding module is used to decode the second and third encoded streams to obtain the decoded background and foreground images. The obtained background and foreground images are then fused to obtain the decoded surveillance video.

[0037] The present invention also provides a monitoring data acquisition and analysis device, including a memory, a processor, and a monitoring data acquisition and analysis program stored in the memory and executable on the processor, wherein the monitoring data acquisition and analysis program is configured to implement the steps of the above-described monitoring data acquisition and analysis method.

[0038] The present invention also provides a storage medium storing a monitoring data acquisition and analysis program, wherein the monitoring data acquisition and analysis program, when executed by a processor, implements the steps of the above-described monitoring data acquisition and analysis method.

[0039] The beneficial effects of this invention are as follows: In the monitoring data acquisition and analysis method of this invention, considering the fixed monitoring area of ​​the monitoring equipment, the change in the monitoring background is very small and basically unchanged over a period of time, and the background is not the key point of monitoring. This invention pre-stores background encoded data for a period of time, and then separates the foreground image and background image during real-time monitoring. Only the foreground image is encoded frame by frame, while the background image is encoded once at intervals. The encoded information is compared with the pre-stored background encoded information, which serves as the basis for adjusting the background image encoding interval. Since it is no longer necessary to encode and send background image encoded data frame by frame, a large amount of storage resources and data transmission are saved, and the video decoding efficiency is also improved. Attached Figure Description

[0040] Figure 1 A flowchart illustrating a monitoring data acquisition and analysis method according to a specific embodiment of the present invention; Detailed Implementation

[0041] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0042] Please refer to Figure 1 The present invention relates to a method for monitoring data acquisition and analysis, comprising the following steps:

[0043] S11: Retrieve surveillance background video with a time span of 1 day;

[0044] S12: Divide the above-mentioned monitoring background video into multiple video segments according to the time period of the clock, with each time period spanning 2 hours;

[0045] S13: Encode each video segment frame by frame to obtain the first encoded information;

[0046] S14: Store the first encoded information;

[0047] S21: Real-time acquisition of surveillance video;

[0048] S22: Separate the foreground and background of each frame in the surveillance video; the algorithm for separating the foreground and background of a single frame is existing technology;

[0049] S23: Encode each frame of the foreground image separated from the surveillance video to obtain the second encoded information;

[0050] S24: Encode a single-frame background image separated from the surveillance video at every first interval to obtain third encoded information; the background image frame is an I-frame.

[0051] S25: Compare the third encoded information with the first encoded information of all frames in the corresponding time period. If the similarity is less than the preset threshold, then encode the single-frame background image separated in the monitoring video at the subsequent second interval time to obtain the third encoded information; repeat S25; if the similarity is greater than the preset threshold, repeat S24.

[0052] For example, the first interval is 40 seconds, corresponding to 3:20 PM. This is compared with the first encoded information of all frames within the pre-stored 2:00 PM - 4:00 PM timeframe. If the similarity is 80% or higher, the background frame is encoded every 40 seconds thereafter. If the similarity is less than 80%, the interval between encoding background frames is shortened to 1 second. The advantage of this method is that the pre-stored surveillance video spanning one day contains almost all background light and shadow changes. Generally, the pattern of background light and shadow changes is the same each day, with an error usually less than 2 hours. For example, if the background image encoding for a certain moment in the afternoon can find a highly similar background image in one of the pre-stored video encodings within two hours before or after, it indicates that the background has not changed significantly. Therefore, the background image can be encoded and sent at a lower frequency. If the real-time monitored video background image encoding cannot find a highly similar corresponding encoding in the pre-stored background encodings for the corresponding time period, it indicates an anomaly in the background image. An alarm can be issued to the terminal, and the interval for subsequent background video encoding can be significantly shortened to more fully monitor subsequent changes in the background image.

[0053] Preferably, in S11, the monitoring background video with a time span of one day can be multiple, including different weather conditions, such as cloudy days and sunny days, to make the comparison more comprehensive. The pre-stored background video encoding needs to be updated regularly, for example, after the background undergoes significant changes.

[0054] S26: Send the second encoded stream and the third encoded stream;

[0055] S27: Decode the second and third encoded streams to obtain the decoded background and foreground images. Merge the obtained background and foreground images to obtain the decoded surveillance video, wherein the background images of each frame are the same during the first or second time interval.

[0056] Since the background does not change significantly, it is not the focus of monitoring for surveillance video. Therefore, the same background image can be obtained by using the same encoding and decoding and then fused with the real-time foreground image within a specific time period. This greatly reduces the amount of data storage, transmission, and processing.

[0057] The first, second, and third encoded information are based on the H.265 protocol.

[0058] The present invention also relates to a monitoring data acquisition and analysis system, comprising:

[0059] The camera module is used to acquire a surveillance background video.

[0060] The separation module is used to separate the above-mentioned monitoring background video into multiple video segments;

[0061] The first encoding module is used to encode each video segment frame by frame to obtain the first encoding information;

[0062] The storage module is used to store the first encoded information;

[0063] The camera module is used to acquire surveillance video.

[0064] The separation module is used to separate the foreground and background of each frame in the surveillance video.

[0065] The second encoding module is used to separate the foreground and background of each frame in the surveillance video.

[0066] The third encoding module is used to encode the single-frame background image separated in the monitoring video at every first interval to obtain the third encoding information;

[0067] The comparison module is used to compare the third encoding information with the first encoding information of all frames in the corresponding time period. If the similarity is less than the preset threshold, the third encoding information is obtained by encoding the single-frame background image separated in the monitoring video at the subsequent second interval time.

[0068] The sending module is used to send the second encoded stream and the third encoded stream;

[0069] The decoding module is used to decode the second and third encoded streams to obtain the decoded background and foreground images. The obtained background and foreground images are then fused to obtain the decoded surveillance video.

[0070] The present invention also relates to a monitoring data acquisition and analysis device, including a memory, a processor, and a monitoring data acquisition and analysis program stored in the memory and executable on the processor, wherein the monitoring data acquisition and analysis program is configured to implement the steps of the above-described monitoring data acquisition and analysis method.

[0071] The present invention also relates to a storage medium storing a monitoring data acquisition and analysis program, wherein the monitoring data acquisition and analysis program, when executed by a processor, implements the steps of the above-described monitoring data acquisition and analysis method.

[0072] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for monitoring data acquisition and analysis, characterized in that, Includes the following steps: S11: Obtain a surveillance background video; S12: Divide the above-mentioned monitoring background video into multiple video segments according to the time period of the clock; S13: Encode each video segment frame by frame to obtain the first encoded information; S14: Store the first encoded information; S21: Real-time acquisition of surveillance video; S22: Separate the foreground and background of each frame in the surveillance video; S23: Encode each frame of the foreground image separated from the surveillance video to obtain the second encoded information; S24: Encode the single-frame background image separated from the surveillance video at every first interval to obtain the third encoded information; S25: Compare the third encoding information with the first encoding information of all frames in the corresponding time period. If the similarity is less than the preset threshold, then encode the single-frame background image separated in the monitoring video at the subsequent second interval time to obtain the third encoding information; repeat S25. S26: Send the second encoded stream and the third encoded stream; S27: Decode the second and third encoded streams to obtain the decoded background and foreground images. Merge the obtained background and foreground images to obtain the decoded surveillance video, wherein the background images of each frame are the same during the first or second time interval. In S12, each time period spans 1-3 hours.

2. The monitoring data acquisition and analysis method according to claim 1, characterized in that, In step S25, if the similarity is greater than a preset threshold, then step S24 is repeated.

3. The monitoring data acquisition and analysis method according to claim 1, characterized in that, Specifically, S11 involves acquiring surveillance background videos with a time span of more than one day.

4. The monitoring data acquisition and analysis method according to claim 1, characterized in that, The first, second, and third encoded information are based on the H.265 protocol.

5. The monitoring data acquisition and analysis method according to claim 1, characterized in that, Specifically, S11 involves acquiring surveillance background video with a time span of one day.

6. A monitoring data acquisition and analysis system, characterized in that, include: The camera module is used to acquire a surveillance background video. The separation module is used to separate the above-mentioned monitoring background video into multiple video segments; The first encoding module is used to encode each video segment frame by frame to obtain the first encoding information; The storage module is used to store the first encoded information; The camera module is used to acquire surveillance video. The separation module is used to separate the foreground and background of each frame in the surveillance video. The second encoding module is used to separate the foreground and background of each frame in the surveillance video. The third encoding module is used to encode the single-frame background image separated in the monitoring video at every first interval to obtain the third encoding information; The comparison module is used to compare the third encoding information with the first encoding information of all frames in the corresponding time period. If the similarity is less than the preset threshold, the third encoding information is obtained by encoding the single-frame background image separated in the monitoring video at the subsequent second interval time. The sending module is used to send the second encoded stream and the third encoded stream; The decoding module is used to decode the second and third encoded streams to obtain the decoded background and foreground images. The obtained background and foreground images are then fused to obtain the decoded surveillance video.

7. Monitoring data acquisition and analysis equipment, characterized in that, The method includes a memory, a processor, and a monitoring data acquisition and analysis program stored in the memory and executable on the processor, the monitoring data acquisition and analysis program being configured to implement the steps of the monitoring data acquisition and analysis method as described in any one of claims 1 to 5.

8. A storage medium, characterized in that, The storage medium stores a monitoring data acquisition and analysis program, which, when executed by a processor, implements the steps of the monitoring data acquisition and analysis method as described in any one of claims 1 to 5.

Images