1. A safety guard system for security is characterized by comprising,

the acquisition module is used for acquiring video information acquired by the camera equipment in real time and is connected with the processing module;

the processing module is used for extracting video frames from the video information acquired by the acquisition module, carrying out image processing on the video frames and generating danger levels according to the image contents in the video frames, and is connected with the storage module;

the storage module is used for storing video frames containing danger levels and is connected with the alarm module;

the alarm module is used for prompting the current danger state according to the danger level;

after the processing module acquires a video frame, the processing module judges the danger level of the video frame according to the image content in the video frame, when the camera equipment is normal, the processing module acquires a face area image in the video frame and judges a head danger coefficient according to the average texture complexity A of the face area, and after the judgment of the head danger coefficient is finished, the processing module adjusts the head danger coefficient according to the number M of the face area in the video frame;

after the head danger coefficient is determined, the processing module acquires a human body graph area in a video frame, judges limb danger coefficients according to the composition graph of the human body graph area, and corrects the limb danger coefficients according to the number N of dangerous article graphs in the video frame after the limb danger coefficients are judged;

the processing module calculates a danger grade parameter D according to the adjusted head danger coefficient and the corrected limb danger coefficient, and judges the danger grade according to the danger grade parameter D;

and after the danger level judgment is finished, the alarm module carries out corresponding danger prompt according to the danger level.

2. The security guard system for security according to claim 1, wherein the processing module performs a risk level determination according to image contents when performing image processing on video frames, wherein,

when the video information cannot be acquired, the processing module judges that the camera equipment is damaged and the camera area is in a low-risk state;

when the complexity of the image texture of each part in the video frame is the same, the processing module judges that the camera shooting equipment is blocked and the camera shooting area is in a low-risk state;

and when different graphic texture complexity exists in the video frame, the processing module judges that the camera equipment is normal and judges the danger level according to the content of the video frame.

3. The security guard system for security according to claim 2, wherein when the image capturing apparatus is normal, the processing module performs area division on a video frame according to gray values, uses a plurality of image areas formed after division as reference images, and uses a reference image area having the same shape as a preset human face image as a face area, the processing module obtains an average texture complexity a of the face area, compares the average texture complexity a with each preset texture complexity, and performs head risk coefficient determination according to the comparison result, wherein,

when A is not more than A1, the processing module judges that the human face has a head cover and judges that the head danger coefficient is B1;

when A is greater than A1 and less than or equal to A2, the processing module judges that the human face is normal and the inside of the camera shooting area is in a safe state;

when A is larger than A2, the processing module judges that the face of the human body is provided with a mask and judges that the danger coefficient of the head is B2;

wherein A1 is a first predetermined texture complexity, A2 is a second predetermined texture complexity, A1 is greater than A2, B1 is a first predetermined head risk factor, B2 is a second predetermined head risk factor, and B1 is greater than B2.

4. The security guard system for security according to claim 3, wherein when the processing module determines the head risk coefficient, the processing module obtains the number M of face regions in the video frame, compares the obtained number M of face regions with the preset number M0 of face regions, and adjusts the head risk coefficient Bi according to the comparison result, and sets i =1,2,

when the processing module selects the jth adjusting coefficient mj to adjust the head risk coefficient Bi, the adjusted head risk coefficient is Bi ', and Bi' = Bi × mj is set, wherein,

when M is not more than M0, the processing module selects a first adjusting coefficient M1 to adjust Bi, M1 is a preset value, and M1 is more than 1 and less than 1.5;

when M is larger than M0, the processing module selects a second adjusting coefficient M2 to adjust Bi, and sets M2= M1 x [1+ (M-M0)/M0 ].

5. The security guard system for security according to claim 4, wherein after the determination of the head risk factor is completed, the processing module is provided with a human body shape curve therein, the processing module determines a human body figure region according to the reference figures connected to the face region, compares a plurality of reference figures constituting the human body figure region with a preset dangerous goods figure, and determines the limb risk factor according to the comparison result, wherein,

when the human body graphic area contains a control instrument graphic, the processing module judges that the limb risk coefficient is C1;

when the human body graphic area contains a firearm graphic, the processing module judges that the limb risk coefficient is C2;

wherein C1 is the first preset limb risk coefficient, C2 is the second preset limb risk coefficient, and C1 is less than C2.

6. The security guard system for security according to claim 5, wherein when the processing module determines the risk coefficient of limbs, the processing module obtains the number N of dangerous goods graphics in the video frame, compares the obtained number N of dangerous goods graphics with the preset number N0 of dangerous goods graphics, and corrects the risk coefficient of limbs Ci according to the comparison result, wherein i =1,2 is set,

when the processing module selects the j-th correction coefficient nj to correct the limb risk coefficient Ci, the corrected limb risk coefficient is Ci ', and Ci' = Ci × nj is set, wherein,

when N is less than or equal to N0, the processing module selects a first correction coefficient N1 to correct Ci, N1 is a preset value, and N1 is more than 1 and less than 1.5;

when N is larger than N0, the processing module selects a second correction coefficient N2 to correct Ci, and N2= N1 × [1+ (N-N0)/N0 ].

7. The security guard system for security according to claim 6, wherein after the determination of the limb risk coefficient is completed, a risk level parameter D is set in the processing module, and D =0.3 xBi '+ 0.7 xCi' is set, after the calculation of the risk level parameter D is completed, the processing module compares the calculated risk level parameter D with a preset risk level parameter D0, and performs the risk level determination according to the comparison result, wherein,

when D is not more than D0, the processing module judges that the camera shooting area is in a moderate dangerous state;

when D > D0, the processing module determines that the camera area is in a high risk state.

8. The safety guard system for security according to claim 7, wherein when the processing module determines that the camera area is in a low-risk state, the processing module obtains a video frame after t1 time, and repeatedly performs the risk level determination on the video frame, and if the video frame is determined to be in the low-risk state again, the alarm module performs a first-level risk prompt to prompt that the camera device is damaged or blocked.

9. The safety guard system for security according to claim 7, wherein when the processing module determines that the camera area is in the moderate risk state, the processing module obtains a video frame after t2 time, and repeatedly performs the risk level determination on the video frame, and if the video frame is determined to be in the moderate risk state or the high risk state, the alarm module performs secondary risk prompt to prompt that dangerous persons exist in the camera area.

10. The safety guard system for security according to claim 7, wherein when the processing module determines that the camera area is in a high risk state, the alarm module directly performs a third-level risk prompt to prompt the presence of dangerous personnel in the camera area.

Background

A shopping mall refers to a large store with a large area and relatively complete commodities. Along with the rapid development of economy in China, the living standard of people is greatly improved, all markets across the country are full of goods, abundant goods bring great convenience to the material demand of people, large-scale markets have many business residents, the flow of people is large in peak time, the safety work of the markets is very important, and at present, the insides of the markets in various places are provided with mature security systems.

At present, the equipartition is equipped with a large amount of cameras in the market to supply security personnel to monitor the personal safety in the market, but only monitor by manpower naked eye, the monitoring dynamics is low, the error is big, can't accurately acquire dangerous picture, and then leads to dangerous condition frequent, can't in time stop.

Disclosure of Invention

Therefore, the invention provides a safety guard system for security, which is used for solving the problem of low safety monitoring efficiency caused by the fact that a monitoring picture cannot be accurately analyzed in the prior art.

In order to achieve the above objects, the present invention provides a security guard system for security, comprising,

the acquisition module is used for acquiring video information acquired by the camera equipment in real time and is connected with the processing module;

the processing module is used for extracting video frames from the video information acquired by the acquisition module, carrying out image processing on the video frames and generating danger levels according to the image contents in the video frames, and is connected with the storage module;

the storage module is used for storing video frames containing danger levels and is connected with the alarm module;

the alarm module is used for prompting the current danger state according to the danger level;

after the processing module acquires a video frame, the processing module judges the danger level of the video frame according to the image content in the video frame, when the camera equipment is normal, the processing module acquires a face area image in the video frame and judges a head danger coefficient according to the average texture complexity A of the face area, and after the judgment of the head danger coefficient is finished, the processing module adjusts the head danger coefficient according to the number M of the face area in the video frame;

after the head danger coefficient is determined, the processing module acquires a human body graph area in a video frame, judges limb danger coefficients according to the composition graph of the human body graph area, and corrects the limb danger coefficients according to the number N of dangerous article graphs in the video frame after the limb danger coefficients are judged;

the processing module calculates a danger grade parameter D according to the adjusted head danger coefficient and the corrected limb danger coefficient, and judges the danger grade according to the danger grade parameter D;

and after the danger level judgment is finished, the alarm module carries out corresponding danger prompt according to the danger level.

Further, when the processing module processes the video frame, the processing module performs a risk level determination according to the image content, wherein,

when the video information cannot be acquired, the processing module judges that the camera equipment is damaged and the camera area is in a low-risk state;

when the complexity of the image texture of each part in the video frame is the same, the processing module judges that the camera shooting equipment is blocked and the camera shooting area is in a low-risk state;

and when different graphic texture complexity exists in the video frame, the processing module judges that the camera equipment is normal and judges the danger level according to the content of the video frame.

Further, when the camera device is normal, the processing module performs region division on the video frame according to a gray value, and uses a plurality of graphic regions formed after division as reference graphics, and uses a reference graphic region with the same shape as a preset human face graphic as a face region, the processing module obtains an average texture complexity A of the face region, compares the average texture complexity A with each preset texture complexity, and performs head danger coefficient judgment according to a comparison result, wherein,

when A is not more than A1, the processing module judges that the human face has a head cover and judges that the head danger coefficient is B1;

when A is greater than A1 and less than or equal to A2, the processing module judges that the human face is normal and the inside of the camera shooting area is in a safe state;

when A is larger than A2, the processing module judges that the face of the human body is provided with a mask and judges that the danger coefficient of the head is B2;

wherein A1 is a first predetermined texture complexity, A2 is a second predetermined texture complexity, A1 is greater than A2, B1 is a first predetermined head risk factor, B2 is a second predetermined head risk factor, and B1 is greater than B2.

Further, when the processing module determines the head risk coefficient, the processing module obtains the number M of face regions in the video frame, compares the obtained number M of face regions with a preset number M0 of face regions, and adjusts the head risk coefficient Bi according to the comparison result, and sets i =1,2, wherein,

when the processing module selects the jth adjusting coefficient mj to adjust the head risk coefficient Bi, the adjusted head risk coefficient is Bi ', and Bi' = Bi × mj is set, wherein,

when M is not more than M0, the processing module selects a first adjusting coefficient M1 to adjust Bi, M1 is a preset value, and M1 is more than 1 and less than 1.5;

when M is larger than M0, the processing module selects a second adjusting coefficient M2 to adjust Bi, and sets M2= M1 x [1+ (M-M0)/M0 ].

Further, after the judgment of the head danger coefficient is finished, a human body shape curve is arranged in the processing module, the processing module determines a human body figure region according to a reference figure connected with the face region, compares a plurality of reference figures forming the human body figure region with a preset dangerous article figure, and judges the limb danger coefficient according to a comparison result, wherein,

when the human body graphic area contains a control instrument graphic, the processing module judges that the limb risk coefficient is C1;

when the human body graphic area contains a firearm graphic, the processing module judges that the limb risk coefficient is C2;

wherein C1 is the first preset limb risk coefficient, C2 is the second preset limb risk coefficient, and C1 is less than C2.

Further, when the processing module determines the limb risk coefficient, the processing module obtains the number N of dangerous goods graphics in the video frame, compares the obtained number N of dangerous goods graphics with the preset number N0 of dangerous goods graphics, and corrects the limb risk coefficient Ci according to the comparison result, and sets i =1,2, wherein,

when the processing module selects the j-th correction coefficient nj to correct the limb risk coefficient Ci, the corrected limb risk coefficient is Ci ', and Ci' = Ci × nj is set, wherein,

when N is less than or equal to N0, the processing module selects a first correction coefficient N1 to correct Ci, N1 is a preset value, and N1 is more than 1 and less than 1.5;

when N is larger than N0, the processing module selects a second correction coefficient N2 to correct Ci, and N2= N1 × [1+ (N-N0)/N0 ].

Further, after the limb risk coefficient is judged, a risk level parameter D is set in the processing module, D =0.3 × Bi '+ 0.7 × Ci' is set, after the risk level parameter D is calculated, the processing module compares the calculated risk level parameter D with a preset risk level parameter D0, and judges a risk level according to a comparison result, wherein,

when D is not more than D0, the processing module judges that the camera shooting area is in a moderate dangerous state;

when D > D0, the processing module determines that the camera area is in a high risk state.

Further, when the processing module judges that the camera shooting area is in a low-risk state, the processing module obtains a video frame after t1 time, repeatedly judges the risk level of the video frame, and if the video frame is judged to be in the low-risk state again, the alarm module carries out first-level risk prompt to prompt that the camera shooting device is damaged or shielded.

Further, when the processing module determines that the camera area is in a moderate risk state, the processing module obtains a video frame after t2 time, repeatedly performs risk level determination on the video frame, and if the video frame is determined to be in the moderate risk state or in the high risk state, the alarm module performs secondary risk prompt to prompt that dangerous persons exist in the camera area.

Further, when the processing module judges that the camera shooting area is in a high-risk state, the alarm module directly carries out three-level risk prompt to prompt that dangerous personnel exist in the camera shooting area.

Compared with the prior art, the processing module acquires the video frames according to the video information, judges the danger level of the camera area according to the content of the acquired video frames, and carries out danger prompt according to the danger level by the alarm module so as to improve the safety monitoring efficiency; when the camera equipment is normal, the processing module acquires a face area image in a video frame, judges the head danger coefficient according to the average texture complexity A of the face area, divides the video frame according to the gray value to select the face area, effectively ensures the accuracy of face area selection, judges the head danger coefficient according to the average texture complexity A of the face area, can effectively distinguish whether the face is provided with a head cover or a mask according to the texture complexity, improves the accuracy of head danger coefficient judgment, adjusts the head danger coefficient according to the number M of the face areas in the video frame, increases the head danger coefficient if the number M of the face areas is larger, further improves the accuracy of the head danger coefficient by adjustment, and acquires a human body image area in the video frame after the head danger coefficient is determined, and the limb danger coefficient is judged according to the composition graph of the human body graph area, whether dangerous goods are carried or not is determined according to the composition graph of the human body graph area, the limb danger coefficient judgment is carried out according to different dangerous goods, the accuracy of the limb danger coefficient judgment is effectively ensured, the limb danger coefficient is corrected according to the number N of dangerous goods graphs in a video frame, the accuracy of the limb danger coefficient is further ensured through correction, the processing module calculates a danger grade parameter D according to the adjusted head danger coefficient and the corrected limb danger coefficient, and the danger grade is judged according to the danger grade parameter D, the accuracy of the danger grade judgment is effectively ensured by calculating the danger grade parameter D, therefore, the alarm module can prompt danger in time, the video monitoring efficiency is improved, and the safety in a camera shooting area is further ensured.

Particularly, when the video frame is subjected to image processing, the processing module judges the danger level according to the image content, and the accuracy of the danger level judgment is effectively guaranteed by analyzing and judging the acquired image content, so that the video monitoring efficiency is further improved, and the safety in a camera area is further guaranteed.

Especially, the processing module judges the head danger coefficient by comparing the average texture complexity A with each preset texture complexity, judges the head danger coefficient by the texture complexity, and can effectively judge whether the head is provided with a head cover or a mask, thereby further ensuring the accuracy of judging the head danger coefficient, further improving the video monitoring efficiency and further ensuring the safety in a camera area.

Especially, the processing module adjusts the head danger coefficient Bi by comparing the obtained face region number M with the preset face region number M0, and further ensures the accuracy of the head danger coefficient through adjustment, thereby further improving the video monitoring efficiency and further ensuring the safety in the camera area.

Particularly, the processing module compares a plurality of reference patterns forming the human body pattern area with preset dangerous article patterns to judge the danger coefficient of limbs, and the dangerous article patterns are compared to further ensure the accuracy of the judgment of the danger coefficient of limbs, so that the video monitoring efficiency is further improved, and the safety in a camera shooting area is further ensured.

Particularly, the processing module corrects the limb danger coefficient Ci by comparing the acquired dangerous article figure number N with the preset dangerous article figure number N0, and further ensures the accuracy of the limb danger coefficient by correcting the limb danger coefficient Ci, thereby further improving the video monitoring efficiency and further ensuring the safety in the camera area.

Particularly, the processing module compares the calculated danger level parameter D with the preset danger level parameter D0 to judge the danger level, and the accuracy of the danger level judgment is effectively guaranteed by calculating the danger level parameter D, so that the video monitoring efficiency is further improved, and the safety in a camera shooting area is further guaranteed.

Particularly, the processing module sets different waiting times according to different danger levels, acquires the video frames after the waiting time and judges the danger levels again, so that misjudgment can be effectively avoided, the error rate is reduced, the video monitoring efficiency is further improved, and the safety in a camera area is further ensured.

Drawings

Fig. 1 is a structural framework diagram of a security guard system for security protection according to the embodiment.

Detailed Description

In order that the objects and advantages of the invention will be more clearly understood, the invention is further described below with reference to examples; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and do not limit the scope of the present invention.

It should be noted that, in the description of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as being fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.

Referring to fig. 1, it is a structural frame diagram of a security guard system for security of the present embodiment, the system includes,

the acquisition module is used for acquiring video information acquired by the camera equipment in real time and is connected with the processing module;

the processing module is used for extracting video frames from the video information acquired by the acquisition module, carrying out image processing on the video frames and generating danger levels according to the image contents in the video frames, and is connected with the storage module;

the storage module is used for storing video frames containing danger levels and is connected with the alarm module;

and the alarm module is used for prompting the current danger state according to the danger level.

Specifically, when the processing module performs image processing on a video frame, the processing module performs danger level determination according to image content, wherein,

when the video information cannot be acquired, the processing module judges that the camera equipment is damaged and the camera area is in a low-risk state;

when the complexity of the image texture of each part in the video frame is the same, the processing module judges that the camera shooting equipment is blocked and the camera shooting area is in a low-risk state;

and when different graphic texture complexity exists in the video frame, the processing module judges that the camera equipment is normal and judges the danger level according to the content of the video frame.

Specifically, in the embodiment, when the video frame is subjected to image processing, the processing module performs risk level determination according to the image content, and the accuracy of risk level determination is effectively ensured by analyzing and determining the acquired image content, so that the video monitoring efficiency is further improved, and the safety in the camera area is further ensured.

Specifically, when the image pickup apparatus is normal, the processing module performs area division on a video frame according to a gray value, and uses a plurality of graphic areas formed after division as reference graphics, and uses a reference graphic area having the same shape as a preset face graphic as a face area, the processing module acquires an average texture complexity a of the face area, compares the average texture complexity a with each preset texture complexity, and performs head risk coefficient determination according to the comparison result, wherein,

when A is not more than A1, the processing module judges that the human face has a head cover and judges that the head danger coefficient is B1;

when A is greater than A1 and less than or equal to A2, the processing module judges that the human face is normal and the inside of the camera shooting area is in a safe state;

when A is larger than A2, the processing module judges that the face of the human body is provided with a mask and judges that the danger coefficient of the head is B2;

wherein A1 is a first predetermined texture complexity, A2 is a second predetermined texture complexity, A1 is greater than A2, B1 is a first predetermined head risk factor, B2 is a second predetermined head risk factor, and B1 is greater than B2.

Specifically, when the processing module determines the head risk coefficient, the processing module obtains the number M of face regions in the video frame, compares the obtained number M of face regions with a preset number M0 of face regions, and adjusts the head risk coefficient Bi according to the comparison result, where i =1,2 is set,

when the processing module selects the jth adjusting coefficient mj to adjust the head risk coefficient Bi, the adjusted head risk coefficient is Bi ', and Bi' = Bi × mj is set, wherein,

when M is not more than M0, the processing module selects a first adjusting coefficient M1 to adjust Bi, M1 is a preset value, and M1 is more than 1 and less than 1.5;

when M is larger than M0, the processing module selects a second adjusting coefficient M2 to adjust Bi, and sets M2= M1 x [1+ (M-M0)/M0 ].

Specifically, this embodiment the processing module adjusts head danger coefficient Bi by comparing the facial region number M that will acquire with preset facial region number M0, has further guaranteed the degree of accuracy of head danger coefficient through adjusting to further improve video monitoring efficiency, further guarantee the security in the camera shooting district.

Specifically, after the head danger coefficient is judged, a human body shape curve is arranged in the processing module, the processing module determines a human body figure region according to a reference figure connected with the face region, compares a plurality of reference figures forming the human body figure region with a preset dangerous goods figure, the dangerous goods figure comprises a control instrument figure and a firearm figure, and judges the limb danger coefficient according to the comparison result, wherein,

when the human body graphic area contains a control instrument graphic, the processing module judges that the limb risk coefficient is C1;

when the human body graphic area contains a firearm graphic, the processing module judges that the limb risk coefficient is C2;

wherein C1 is the first preset limb risk coefficient, C2 is the second preset limb risk coefficient, and C1 is less than C2.

Specifically, when the processing module determines the limb risk coefficient, the processing module obtains the number N of dangerous goods graphics in the video frame, compares the obtained number N of dangerous goods graphics with the preset number N0 of dangerous goods graphics, and corrects the limb risk coefficient Ci according to the comparison result, and sets i =1,2, wherein,

when the processing module selects the j-th correction coefficient nj to correct the limb risk coefficient Ci, the corrected limb risk coefficient is Ci ', and Ci' = Ci × nj is set, wherein,

when N is less than or equal to N0, the processing module selects a first correction coefficient N1 to correct Ci, N1 is a preset value, and N1 is more than 1 and less than 1.5;

when N is larger than N0, the processing module selects a second correction coefficient N2 to correct Ci, and N2= N1 × [1+ (N-N0)/N0 ].

Specifically, the processing module of this embodiment corrects the risk coefficient Ci of the limb by comparing the obtained number N of dangerous goods graphs with the preset number N0 of dangerous goods, and further ensures the accuracy of the risk coefficient of the limb by correcting the risk coefficient Ci of the limb, thereby further improving the video monitoring efficiency and further ensuring the security in the camera area.

Specifically, after the limb risk coefficient is determined, a risk level parameter D is set in the processing module, D =0.3 × Bi '+ 0.7 × Ci' is set, after the risk level parameter D is calculated, the processing module compares the calculated risk level parameter D with a preset risk level parameter D0, and determines a risk level according to a comparison result, wherein,

when D is not more than D0, the processing module judges that the camera shooting area is in a moderate dangerous state;

when D > D0, the processing module determines that the camera area is in a high risk state.

Specifically, when the processing module determines that the image pickup area is in a low-risk state, the processing module obtains a video frame after t1 time, repeatedly performs risk level determination on the video frame, and if the video frame is determined to be in the low-risk state again, the alarm module performs primary risk prompt to prompt that the image pickup device is damaged or blocked;

when the processing module judges that the camera shooting area is in a moderate dangerous state, the processing module acquires a video frame after t2 time, repeatedly judges the danger level of the video frame, and if the video frame is judged to be in the moderate dangerous state or the high dangerous state, the alarm module carries out secondary danger prompt to prompt dangerous persons in the camera shooting area;

when the processing module judges that the interior of the camera shooting area is in a high-risk state, the alarm module directly carries out three-level risk prompt to prompt that dangerous personnel exist in the camera shooting area;

wherein t1 is a first preset waiting time, t2 is a second preset waiting time, the unit is second, and t2 is more than 1 and t1 is more than 10.

Specifically, the processing module of this embodiment sets different waiting times according to different risk levels, and obtains the video frame after the waiting time to perform the risk level determination again, so as to effectively avoid erroneous determination, thereby reducing the error rate, further improving the video monitoring efficiency, and further ensuring the security in the camera area.

So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.