Big data-based image optimization design management system and method

By using a big data-based image optimization design and management system, the system analyzes the status of personnel in the monitored area in real time, automatically adjusts the lossy compression ratio of image data, solves the problem of increased data storage space in the factory monitoring system, and achieves efficient image data storage and anomaly detection.

CN116996646BActive Publication Date: 2026-11-10XIAMEN HUAXIA UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311073499.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-08-24
Publication Date
2026-11-10
Estimated Expiration
2043-08-24

AI Technical Summary

Technical Problem

In factory monitoring systems, video and image data are taking up more and more space, leading to information loss. How can we extend the data retention time and preserve valuable information?

Method used

The image optimization design and management system, based on big data, analyzes the status of personnel in the monitored area in real time through image acquisition, analysis, monitoring center and alarm module, automatically adjusts the lossy compression ratio of image data, and optimizes image storage based on risk assessment results.

Benefits of technology

It effectively reduces the space occupied by image data, extends data retention time, improves storage efficiency, and promptly detects anomalies and triggers alarms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116996646B_ABST
    Figure CN116996646B_ABST
Patent Text Reader

Abstract

This invention discloses an image optimization design management system and method based on big data, relating to the field of image monitoring data optimization management technology. The system includes: an image acquisition module, an analysis module, a monitoring center, and an alarm module. The output of the image acquisition module is interconnected with the input of the analysis module to acquire image data within the factory. The output of the analysis module is interconnected with the inputs of the monitoring center and the alarm module to adjust the image compression ratio and reduce the space occupied by the images. The output of the monitoring center is interconnected with the input of the alarm module to monitor the factory's status. When the alarm module detects abnormal activity, it triggers an alarm and alerts the operators. This invention can increase the image compression ratio, reduce the space occupied by image data, store more image data using the same storage device, and extend the image data deletion cycle while ensuring safety.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of image surveillance data optimization technology, specifically to an image optimization design management system and method based on big data. Background Technology

[0002] There is a close relationship between video and images. Video can be viewed as a continuous sequence of images, displaying multiple image frames per second. Each image frame is a still image. Video is obtained by playing these image frames at a fast speed within consecutive time intervals. In factory monitoring systems, as time goes on, the space occupied by video and images increases, necessitating periodic deletion of video and images. Deleted image and video data may contain valuable information for the factory, which may need this information again due to unforeseen circumstances after deletion. How to extend the retention time of video and image data has become a problem that needs to be solved. Summary of the Invention

[0003] The purpose of this invention is to provide an image optimization design management system and method based on big data to solve the problems mentioned in the background art.

[0004] Firstly, a big data-based image optimization design management system is provided, comprising: an image acquisition module, an analysis module, a monitoring center, and an alarm module; the output of the image acquisition module is interconnected with the input of the analysis module to acquire image data within the factory; the output of the analysis module is interconnected with the inputs of the monitoring center and the alarm module to adjust the image compression ratio and reduce the space occupied by the image; the output of the monitoring center is interconnected with the input of the alarm module to monitor the factory's status; when the alarm module detects abnormal activity, it triggers an alarm and alerts the operators, while also recording relevant event information for subsequent processing.

[0005] The image acquisition module, analysis module, monitoring center, and alarm module also include a communication unit, which is used for data transmission between the image acquisition module, analysis module, monitoring center, and alarm module.

[0006] The analysis module automatically adjusts the lossy compression ratio of the image data based on the status of people in the monitored area, reducing the space occupied by the image data while ensuring safety, thereby extending the image data storage time.

[0007] Secondly, a big data-based image optimization design management method is provided, including the following steps:

[0008] S4-1, the monitoring area is divided into multiple sub-areas, and the image acquisition module monitors the operating status of the equipment in each sub-area through a camera; the monitoring area is divided into K octagons, each octagon being a sub-area; optionally, K is set to 6, and the monitoring area is divided into 6 octagons, each octagon having an angle of 60 degrees on the XOY plane. The human binocular field of vision is 124 degrees, so no matter what posture the human is in, the status of the equipment within at least one octagon within a 60-degree range can be seen. The number and range of octagons can also be adjusted according to the actual situation.

[0009] S4-2, The analysis module acquires image data within the monitored area, including the operating status of equipment and the status of people within the monitored area;

[0010] S4-3, the analysis module analyzes the image data within the monitored area and optimizes the image data based on the analysis results;

[0011] S4-4, the analysis module sends the optimized image data to the monitoring center;

[0012] S4-5, the monitoring center receives image data sent by the analysis module, reflecting the real-time status of the equipment.

[0013] The monitoring area is divided into K quadrants, with each quadrant representing a sub-area. The specific steps include:

[0014] S5-1, with the ground as the XOY plane, the vertical upward direction as the positive Z-axis, the center of the monitoring area as the coordinate origin O, the line connecting the coordinate origin O and the edge of the equipment as the positive X-axis, and the positive Y-axis rotated 90 degrees counterclockwise to establish a spatial rectangular coordinate system;

[0015] S5-2, Starting from the positive direction of the X-axis, use the XOZ plane to rotate counterclockwise around the Z-axis by deg degrees, repeat the rotation K times until the total number of degrees rotated by the XOZ plane reaches 360 degrees;

[0016] S5-3, following the order of rotation, set the regions obtained after each rotation as the first octagon, the second octagon, ..., the Kth octagon in sequence;

[0017] Where deg is the angle of each rotation, which is set by the user, and K×deg is greater than or equal to 360.

[0018] The analysis module acquires image data for each triad, assesses the risk of each triad, and determines the lossy compression ratio of the image data for each triad based on the risk assessment results. The specific analysis steps include the following:

[0019] S6-1, The analysis module calculates the risk value F for each criterion based on the image data. iF i =max,0,w1T1-w2T2-w3T 3i}; T1 represents the time when no one is in the monitored area, T2 represents the time when someone is in the monitored area, T 3i Let w1, w2, and w3 be the time intervals within the monitored area when someone's line of sight falls within the i-th octet, and let T1, T2, and T3 be T1, T2, and T3, respectively. 3i The weights are denoted by , where i is the quaternion number and its value is a positive integer between [1, K]. When there are people in the monitored area, they can handle or report any abnormal situations that occur in the equipment. Therefore, the longer a person stays in the monitored area, the lower the risk value. The risk value F for each quaternion is... i The risk value F of each hexagram is negatively correlated with the time T2 spent by a person in the monitored area. When someone is in the monitored area, they cannot simultaneously pay attention to the equipment status of all hexagrams. For hexagrams that are not noticed, they will only be discovered after a problem occurs. However, for hexagrams that are noticed, people can detect equipment abnormalities before problems occur and take preventative measures. Therefore, the risk value F of each hexagram is... i The duration T of attention for the i-th quaternion 3i There is a negative correlation; the longer the monitoring area is unoccupied, the lower the expected probability of detecting equipment malfunctions, and the risk value F for each criterion. i It is directly proportional to the time T1 during which no one is in the monitored area;

[0020] S6-2, After determining the risk value for each hexagram, the analysis module adjusts the lossy compression ratio P of each hexagram image based on the risk value. i P i =g(F i ), where g is F i To P i The mapping, P i With F i The reciprocal of F is positively correlated. i The larger P is, the better. i The smaller F i The smaller P is i The larger the value, the more likely the mapping g is to be; alternatively, the mapping g can take the form of a hyperbola, a logarithmic function, a polynomial function, or a composite function.

[0021] Specifically, it can be taken The function has the advantages of being continuous and smooth everywhere, and exhibits exponential decay; when F i When F is greater than or equal to zero, F can be used. iThe range is compressed to the interval [0,1], where e is the natural constant. When the risk value is 0, the lossy compression ratio is 100%, and the image quality is reduced to the maximum extent, which may result in distortion, blurring, and color loss. When the risk value approaches infinity, the compression ratio is 0%, and the image quality and file size are exactly the same as the original image. Meanwhile, Function in F i When the value is small, the decay rate is fast; as F... i Increase The function decay rate gradually decreases, which is consistent with the changing trend of the difficulty in reducing the risk value. When the risk value is high, it is easier to reduce the risk value. There is no need to deliberately arrange personnel to enter the monitoring area to monitor the equipment status. Increasing the flow of personnel in the monitoring area can effectively reduce the risk value. When the risk value is low, it is more difficult to reduce the risk value. In order to reduce the risk value, it is necessary to specially arrange personnel to enter the monitoring area to monitor the equipment status of each limit.

[0022] In step S6-1, the analysis module analyzes the time a person's gaze falls within each trigram as a basis for risk assessment, specifically including the following analysis steps:

[0023] S7-1: Initially, the direction the person is facing and the direction of their gaze belong to the same trigram. When the analysis module detects a change in the trigram of the direction the person is facing, if the trigram of the direction of their gaze is the same as the trigram of the direction the person is facing, proceed to S7-2; if the trigram of the direction of their gaze is different from the trigram of the direction the person is facing, proceed to S7-3. The trigrams of the direction the person is facing and the direction of their gaze may change simultaneously. In this case, the trigram of the i-th trigram the person was in before the change of the direction the person was facing, T... 3i The corresponding time period is when the direction the person is facing and the direction the person is looking are in the same trigram. The trigrams to which the direction the person is facing and the direction the person is looking may change asynchronously, so it is necessary to analyze two cases.

[0024] S7-2, If the direction a person is facing is within the i-th octagon, then the time T for the person's line of sight to fall within the i-th octagon is... 3i Increase Ts, where Ts is the time during which the trigram corresponding to the direction the body is facing does not change, and enter S7-11;

[0025] S7-3: If the person's line of sight belongs to the α octagon and the direction the person is facing belongs to the β octagon, the time *m* it takes for the person's line of sight to move from the α octagon to the β octagon is calculated. If the octagon to which the person's facing direction belongs does not change within time *m*, proceed to S7-4. If the octagon to which the person's facing direction belongs changes within time *m*, proceed to S7-5. Here, α ≠ β, and the values ​​of α and β are both positive integers between [1, K]. The time *m* required for the person's line of sight to move from the α octagon to the β octagon is calculated as follows: If the octagon to which the person's facing direction belongs does not change within this time *m*, then the person's line of sight passes through a potentially incomplete octagon α and a complete octagon between α and β. The time is calculated according to the method in S7-4. If the octagon to which the person's facing direction belongs changes within this time *m*, then two cases need to be analyzed.

[0026] S7-4, the time T during which a person's line of sight falls within the α-th octave. 3α Increase m%t, the octet between the α octet and the β octet, and the corresponding time T. 3i Increase t, where the octagon between α and β is the range of less than or equal to 180 degrees formed between the α and β octagons, t is the average time required for the direction of a person's line of sight to turn through a complete octagon, % is the remainder division operation, and proceed to S7-11;

[0027] S7-5: If the trigram corresponding to the direction the person is facing changes to trigram γ (between α and β), proceed to S7-7. If the trigram corresponding to the direction the person is facing changes to trigram δ (outside of α and β), proceed to S7-6. Trigrams outside of α and β are ranges greater than 180 degrees formed between trigrams α and β. Before the direction of the person's line of sight and the direction the person is facing belong to the same trigram, if the direction the person is facing changes outward to trigram δ, there is no path intersection. The person passes through a potentially incomplete trigram α and a complete trigram between α and δ, and the time is calculated according to S7-6. If the direction the person is facing changes inward to trigram γ, there is a path intersection, which is analyzed in three separate cases.

[0028] S7-6, obtain the time n taken for the person's gaze to move from the α octagon to the δ octagon, and the time T taken for the person's gaze to fall within the α octagon. 3α Increasing n%t, the time T is the time between the α octet and the δ octet. 3i Increase t to enter S7-11;

[0029] S7-7: The trigram corresponding to the direction the person's body is facing changes to the trigram γ between α and β over time u, where u is less than m. If the person's line of sight has not yet reached the γ trigram within time u, then proceed to S7-8; if the person's line of sight reaches the γ trigram within time u but has not entered the next trigram of the γ trigram, then proceed to S7-9; if the person's line of sight crosses the γ trigram within time u, then proceed to S7-10.

[0030] S7-8, if we obtain the time v that it takes for a person's gaze to move from the α octagon to the γ octagon, then the time T that the person's gaze falls within the α octagon. 3α Increase v%t, the octet between the α and γ octets, and the corresponding time T. 3i Increasing t leads to S7-11; in this case, there is no path intersection, passing through a potentially incomplete octet α and a complete octet between α and γ.

[0031] S7-9, the octet between the α and γ octets, corresponding to the time T. 3i Let y be the number of octets between the α and γ octets, and let T be an increment t. 3α and T 3γ All increase time Enter S7-11; in this case there is no path intersection, passing through a potentially incomplete octet α, a potentially incomplete octet γ, and a complete octet between α and γ;

[0032] S7-10, obtain the time r taken for the person's line of sight to move from the α octagon to the γ octagon for the second time. Then, find the time T corresponding to the octagons between the α and γ octagons, and the γ octagon itself. 3i Let t be the number of octaves between α and γ octaves, y-1, and the time T for the person's line of sight to fall within the α-th octave. 3α Increase The time T during which a person's gaze falls within the β octave 3β Increase Enter S7-11; in this case, the β octagon has a path intersection, passing through a possibly incomplete octagon α and the octagon between the complete α and γ, and passing through a possibly incomplete octagon β twice;

[0033] S7-11, the analysis module analyzes the acquired time T 3i Determine the risk and lossy compression ratio of each trigram, continue monitoring the trigram to which the body is facing, and return to step S7-1;

[0034] The number of quadrants between the α quadrant and the β quadrant does not include the α quadrant and the β quadrant themselves. When the α quadrant and the β quadrant are adjacent, the number of quadrants between the α quadrant and the β quadrant is 0.

[0035] The direction a person is facing is determined by acquiring the person's body features and establishing a K-classification model, outputting K different labels, each label corresponding to a trigram. The trigram to which the person's gaze direction belongs is determined by acquiring the person's eye features and establishing K binary classification models, each trigram corresponding to a binary classification model. The output of each binary classification model is either 0 or 1. If the output of the i-th binary classification model is 0, the person's gaze direction does not belong to the i-th trigram; if the output of the i-th binary classification model is 1, the person's gaze direction belongs to the i-th trigram.

[0036] The analysis module analyzes the trigram corresponding to the direction the person's body is facing in real time, but does not analyze the trigram corresponding to the direction the person's line of sight is in real time. When the trigram corresponding to the direction the person's body is facing does not change, the time T corresponding to the trigram corresponding to the direction the person's line of sight is calculated. 3i As time increases; when the octagon to which the direction the person's body is facing changes from α to β, firstly, the octagon to which the direction the person's body is facing is the same as the octagon to which the direction the person's body was facing before, that is, the α octagon. Then, the binary classification model of the β octagon is used to analyze the octagon to which the direction the person's gaze direction belongs. The time it takes for the binary classification result output to change from 0 to 1 is the time required for the octagon to which the direction the person's gaze direction belongs to be the same as the octagon to which the direction the body is facing. This is also the time m, n, v and r in steps S7-1 to S7-11.

[0037] Compared with the prior art, the beneficial effects achieved by the present invention are: calculating the risk value of each zodiac sign according to the zodiac sign to which the human line of sight belongs, automatically adjusting the lossy compression ratio of the image data of each zodiac sign according to the risk value, using a larger lossy compression ratio when the risk value is low, reducing the space occupied by the image data, storing more image data using the same storage device, and extending the deletion cycle of the image data. Attached Figure Description

[0038] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0039] Figure 1 This is a schematic diagram of the structure of an image optimization design management system based on big data according to an embodiment of the present invention;

[0040] Figure 2 This is a flowchart illustrating the analysis of the time it takes for a person's gaze to fall within each trigram in an embodiment of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] In an embodiment of the present invention, the center of the monitoring area is taken as the origin O, the ground is taken as the XOY plane, the vertical upward direction is taken as the positive Z-axis, the line connecting the origin O and the edge of the device is taken as the positive X-axis, and the positive Y-axis is taken as the positive Y-axis by rotating the positive X-axis counterclockwise by 90 degrees. A spatial rectangular coordinate system is established. Starting from the positive X-axis, the XOZ plane is rotated 60 degrees counterclockwise around the Z-axis, and this rotation is repeated 6 times. According to the order of rotation, the area obtained after each rotation is set as the first octet, the second octet, ..., the sixth octet.

[0043] The direction a person is facing is determined by acquiring their body features and building a 6-class classification model, outputting 6 different labels, each corresponding to a trigram. The trigram to which a person's gaze direction belongs is determined by acquiring their eye features and building 6 binary classification models, with each trigram corresponding to a binary classification model. The output of each binary classification model is either 0 or 1. If the output of the i-th binary classification model is 0, then the person's gaze direction does not belong to the i-th trigram; if the output of the i-th binary classification model is 1, then the person's gaze direction belongs to the i-th trigram.

[0044] When an employee enters the equipment monitoring area, the direction their body is facing and their line of sight belong to the first trigram. The analysis module acquires the employee's body characteristics in real time and analyzes the trigram to which the direction the body is facing belongs. The trigram to which the line of sight belongs is not analyzed in real time. When the trigram to which the direction the body is facing does not change, the time T corresponding to the first trigram to which the direction the body is facing is recorded. 31 Increase over time;

[0045] After a time interval (sec), when the analysis module detects that the direction the person is facing has changed from the 1st zodiac sign to the 3rd zodiac sign, the analysis module simultaneously acquires the person's eye features and analyzes the zodiac sign of the person's gaze direction. If the analysis module determines that the person's gaze direction is also in the 3rd zodiac sign, then the time T corresponding to the 1st zodiac sign... 31The value increases by sec; if the analysis module does not determine that the direction the person is facing has entered the second trigram, it means that the direction the person is facing has moved from the first trigram through the second trigram to the third trigram in a short period of time, and the time for the second trigram can be ignored; if the analysis module determines that the person's gaze direction is still in the first trigram, then the time m required for the analysis module to determine that the person's gaze direction has changed from the first trigram to the third trigram is obtained. Within the time m, the analysis module still obtains the person's body characteristics in real time and analyzes the trigram to which the direction the person is facing belongs. The trigram to which the direction the person is facing may change, and two cases are analyzed.

[0046] In the first scenario, if the direction the person is facing is still within the 3rd trigram, then the time T corresponding to the 1st trigram... 31 Increase m%t, the time T corresponding to the second octet 32 Add t; where t is the average time required for a person's line of sight to turn through a complete octagon; the second case is when the direction of the person's body changes from the 3rd octagon to the other octagons, and the two cases are analyzed according to the position of the octagon.

[0047] In the first scenario, if the direction the person is facing changes to the 4th octagon within time m, then the time n required for the analysis module to determine whether the person's gaze direction changes from the 1st to the 4th octagon is the same as the time T corresponding to the 1st octagon. 31 Increasing n%t, the time T corresponding to the second and third octets. 32 and T 33 Increase t; the second type of situation is that within time m, the trigram to which the direction the person's body is facing changes to the second trigram after time u, where u is less than m;

[0048] If, within time u, the person's gaze has not yet reached the second octagon and remains in the first octagon, then the analysis module determines that the time required for the person's gaze to change from the first octagon to the second octagon is v, and the time T corresponding to the first octagon is... 31 Increase v%t, at which point v is less than t, T 31 The increment is v; if, within time u, the person's line of sight enters the second octet but not the third octet, then T... 31 and T 32 All increased If a person's gaze enters the 3rd trigram, the analysis module determines that the time required for the gaze to change from the 1st trigram to the 2nd trigram for the second time is v. During this phase, the order of trigram changes for the person's gaze includes: 1st trigram, 2nd trigram, 3rd trigram, and 2nd trigram, T. 31 Increase T 33 Increase T 32 Increase t;

[0049] The time T of the zodiac sign to which the person's line of sight belongs 31 T 32 T 33 T 34 T 35 and T 36 As time progresses, the analysis module continuously updates the risk value F for each quaternion based on the time frame. i F i =max{2T1-0.5T2-T 3i}, T1 is the time when no one is in the monitored area, T2 is the time when someone is in the monitored area, T 3i For the time when someone is in the monitored area and their line of sight falls within the i-th octet, the risk value F for each octet is calculated. i Adjust and update the lossy compression ratio P of each octet image data. i , When the risk value F i When the risk value F is 0, the lossy compression ratio is 100%, and the image quality is reduced to the maximum extent, which may result in distortion, blurring, and color loss; when the risk value F is 0, the lossy compression ratio is 100%, and the image quality is reduced to the maximum extent, which may result in distortion, blurring, and color loss. i As the value approaches infinity, the compression ratio is 0%, and the image quality and file size are exactly the same as the original image; after the device has been running for one hour, for the following T... 31 T 32 T 33 T 34 T 35 and T 36 The sequence is: [0.1 0.1 0.3 0.05 0.0.5 0.1], T 31 T 32 T 33 T 34 T 35 and T 36The units are all hours. Therefore, the time with people in the monitored area is T2 = 0.1 + 0.1 + 0.3 + 0.05 + 0.05 + 0.1 = 0.7 hours, and the time with no one in the monitored area is T1 = 1 - 0.7 = 0.3 hours. According to the formula, the risk value sequence for each threshold is [0.15 0.15 0 0.2 0.2 0.15], and the compression ratio sequence for each threshold is [86 86 100 82 82 86]. During one hour of equipment operation, employees are in the monitored area for 70% of the time, resulting in a lower overall risk value and a relatively higher compression ratio. For another time sequence: [0.05 0.05 0.1 0.05 0.0.5]... [0.05], the time when there are people in the monitored area T2 = 0.0.5 + 0.0.5 + 0.1 + 0.05 + 0.05 + 0.0.5 = 0.35 hours, and the time when there are no people in the monitored area T1 = 1 - 0.35 = 0.65 hours. According to the formula, the risk value sequence of each limit is [1.075 1.075 1.025 1.075 1.075], and the compression ratio sequence of each limit is [34 34 36 34 34 34]. In one hour of equipment operation, there are employees in the monitored area for only 35% of the time, the overall risk value is high, and the compression ratio is relatively low.

[0050] The analysis module sends the compressed image data to the monitoring center. When staff in the monitoring area discover abnormal equipment operation, they can trigger an alarm through the alarm module. At the same time, the monitoring center reflects the equipment operation status within the monitoring area. When staff in the monitoring center discover abnormal equipment operation, they can also trigger an alarm through the alarm module.

[0051] When the device is off, T1, T2, T 31 T 32 T 33 T 34 T 35 and T 36 All values ​​are set to zero. When the device is restarted, T1, T2, and T... 31 T 32 T 33 T 34 T 35 and T 36 As the equipment uptime increases from zero.

[0052] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0053] Finally, it should be noted that the above descriptions are merely preferred embodiments 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 foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A big data-based image optimization design management method, characterized in that, Includes the following steps: S4-1 divides the monitoring area into multiple sub-areas, and the image acquisition module monitors the operating status of the equipment in each sub-area through cameras; S4-2, The analysis module acquires image data within the monitored area, including the operating status of equipment and the status of people within the monitored area; S4-3, the analysis module analyzes the image data within the monitored area, and compresses the image data based on the analysis results to reduce the space occupied by the image data; S4-4, The analysis module sends the compressed image data to the monitoring center; S4-5: The monitoring center receives and stores the compressed image data sent by the analysis module to reflect the real-time status of the equipment. It uses the same storage device to store more image data and extends the update cycle of image deletion. In step S4-1, the monitoring area is divided into K quadrants, with each quadrant serving as a sub-region. This includes the following steps: S5-1, with the ground as the XOY plane, the vertical upward direction as the positive Z-axis, the center of the monitoring area as the coordinate origin O, the line connecting the coordinate origin O and the edge of the equipment as the positive X-axis, and the positive Y-axis rotated 90 degrees counterclockwise to establish a spatial rectangular coordinate system; S5-2, Starting from the positive direction of the X-axis, use the XOZ plane to rotate counterclockwise around the Z-axis by deg degrees, repeat the rotation K times until the total number of degrees rotated by the XOZ plane reaches 360 degrees; S5-3, following the rotation order, set the regions obtained after each rotation as the first octagon, the second octagon, ..., the Kth octagon in sequence; Where deg is the angle of each rotation, which is set by the user, and K×deg is greater than or equal to 360; The analysis module acquires image data for each triad, assesses the risk of each triad, and determines the lossy compression ratio of the image data for each triad based on the risk assessment results. This includes the following analysis steps: S6-1, The analysis module calculates the risk value for each trigram based on the image data. , ; For the time when no one is in the monitored area, For the time when people are in the monitored area, The time during which someone is within the monitored area and their line of sight falls within the i-th octet. , and They are respectively , and The weights are given by i, where i is the quaternion number and the value range is the interval [0, 1]. Positive integers between; S6-2, After determining the risk value for each hexagram, the analysis module adjusts the lossy compression ratio of the image for each hexagram based on the risk value. , ,in for arrive The mapping, and The reciprocal of is positively correlated. The larger, the better The smaller, The smaller, The larger; In step S6-1, the analysis module analyzes the time a person's gaze falls within each trigram as a basis for risk assessment, including the following analysis steps: S7-1: Initially, the direction the person's body is facing and the direction of the person's gaze belong to the same trigram. When the analysis module detects a change in the trigram of the direction the person's body is facing, if the trigram of the person's gaze direction is the same as the trigram of the direction the person's body is facing, then proceed to S7-2. If the trigram of the person's gaze direction is different from the trigram of the direction the person's body is facing, then proceed to S7-3. S7-2, If the direction a person is facing is within the i-th octagon, then the time it takes for the person's line of sight to fall within the i-th octagon. Increase Ts, where Ts is the time during which the trigram corresponding to the direction the body is facing does not change, and enter S7-11; S7-3, if the person's line of sight belongs to The trigram represents the direction the body is facing. The trigram, the direction of the person's gaze. The hexagram enters The time taken for the trigram to be entered is m. If the trigram corresponding to the direction the person is facing does not change within time m, then proceed to S7-4. If the trigram corresponding to the direction the person is facing changes within time m, then proceed to S7-5. , and The range of values ​​is all within an interval. Positive integers between; S7-4, the person's line of sight falls on the... Time within the hexagram limit Increase , hexagram limits and The time corresponding to the intervals between the hexagrams Add t, where and The interval between the two quatrains is and The range of less than or equal to 180 degrees formed between the octagons, t is the average time required for a person's line of sight to turn around a complete octagon. For the remainder division operation, proceed to S7-11; S7-5, if the trigram corresponding to the direction the person's body is facing changes and The interval between the gua limits Then proceed to S7-7, if the trigram corresponding to the direction the person's body is facing changes. and Other than the hexagram limit Then proceed to S7-6. and The other trigrams are and The range greater than 180 degrees formed between the trigrams; S7-6, Obtaining the direction of the person's gaze from... The hexagram enters The time n spent on the hexagram, the direction of the person's gaze falls on the first... Time within the hexagram limit Increase , hexagram limits and The time corresponding to the intervals between the hexagrams Increase t to enter S7-11; S7-7, the trigram corresponding to the direction the person's body is facing changes over time u. and The interval between the gua limits If u is less than m, and the person's line of sight has not yet reached the destination within time u. If the hexagram is within the limit, then proceed to S7-8; if within time u, the direction of the person's line of sight reaches... The hexagram limit has not yet been reached. The next hexagram after the first hexagram enters S7-9; if within time u, the person's line of sight crosses... If the hexagram is in a certain range, then proceed to S7-10; S7-8, Obtaining the direction of the person's gaze from... The hexagram enters If the time spent on the hexagram is v, then the person's gaze will fall on the first hexagram. Time within the hexagram limit Increase , hexagram limits and The time corresponding to the intervals between the hexagrams Increase t to enter S7-11; S7-9, hexagram limits and The time corresponding to the intervals between the hexagrams Add t, record hexagram limits and The number of hexagrams between hexagrams is y. and All increase time Enter S7-11; S7-10, obtaining the direction of the person's gaze from... The second time the hexagram has entered If the time r spent on the hexagram is then... hexagram limits and The intervals between hexagrams, and The horoscope, the corresponding time. Add t, record hexagram limits and The number of trigrams between the trigrams is y-1, and the direction of the person's gaze falls on the y-1. Time within the hexagram limit Increase The person's gaze falls on the first Time within the hexagram limit Increase Enter S7-11; S7-11, the analysis module analyzes the time acquired. Determine the risk and lossy compression ratio of each trigram, continue monitoring the trigram to which the body is facing, and return to step S7-1; Among them, the hexagram limits and The intervals between the hexagrams do not include the intervals between the hexagrams. hexagram limits and The hexagram limits oneself, when hexagram limits and When the hexagrams are adjacent, hexagram limits and The number of hexagrams between hexagrams is 0.

2. The image optimization design management method based on big data according to claim 1, characterized in that, The analysis module analyzes the trigram corresponding to the direction the person's body is facing in real time, but does not analyze the trigram corresponding to the direction the person's line of sight is in real time. When the trigram corresponding to the direction the person's body is facing does not change, the time corresponding to the trigram corresponding to the direction the person's line of sight is... As time increases, when the trigram corresponding to the direction the person's body is facing changes, the analysis is performed on the trigram corresponding to the direction the person's line of sight is changed, and only the time required for the trigram corresponding to the direction the person's line of sight to be the same as the trigram corresponding to the direction the body is facing is obtained.

3. The image optimization design management method based on big data according to claim 2, characterized in that, The direction a person is facing is determined by acquiring the person's body features and establishing a K-classification model; the zodiac sign to which the person's gaze direction belongs is determined by acquiring the person's eye features and establishing K binary classification models, with each zodiac sign corresponding to a binary classification model.

4. A big data-based image optimization design management system, used to implement the big data-based image optimization design management method as described in claim 1, characterized in that, include: Image acquisition module, analysis module, monitoring center, and alarm module; The output of the image acquisition module is connected to the input of the analysis module to acquire image data within the factory. The output of the analysis module is connected to the input of the monitoring center and the alarm module to adjust the image compression ratio and reduce the space occupied by the image. The output of the monitoring center is connected to the input of the alarm module to monitor the factory's situation. When the alarm module detects abnormal activity, it will trigger an alarm and remind the operators. At the same time, the relevant event information will be recorded for subsequent processing.

5. The image optimization design management system based on big data according to claim 4, characterized in that, The image acquisition module, analysis module, monitoring center, and alarm module also include a communication unit, which is used for data transmission between the image acquisition module, analysis module, monitoring center, and alarm module.

6. The image optimization design management system based on big data according to claim 4, characterized in that, The analysis module automatically adjusts the lossy compression ratio of the image data based on the status of people in the monitored area, reducing the space occupied by the image data while ensuring safety.

Citation Information

Patent Citations

  • Method and device for adaptively compressing image data

    CN106256126A

  • Intelligent monitoring security data processing management system based on big data analysis

    CN112437280A