Intelligent community access control response system

Abstract

This invention relates to a smart community access control system, which includes a layered construction device, a step-by-step enhancement device, a real-time imaging device, an intelligent judgment mechanism, and a monitoring and processing mechanism. The smart community access control system of this invention is logically reliable and intelligently designed.

Description

Technical Field

[0001] This invention relates to the field of smart access control, and more particularly to a smart community access control system. Background Technology

[0002] Smart access control systems effectively protect people's safety. Through high-definition cameras and facial recognition technology, they accurately identify the identity of each person entering and exiting, ensuring that only authorized personnel can enter the building. Furthermore, the system can monitor the access area in real time, promptly detecting and recording any suspicious behavior, providing reliable evidence for subsequent investigations. Whether in residential areas, commercial areas, or office areas, smart access control systems effectively protect the safety of every resident and employee. Smart access control systems bring great convenience to people's lives. Traditional access control systems require keys or cards to enter, which undoubtedly brings some inconvenience to people's lives. Smart access control systems connect to the system via a mobile app, allowing users to open the door with just a tap. Wherever you are, as long as you have your mobile phone, you can remotely open the door without worrying about forgetting your keys or card. In addition, the system also supports visitor access; visitors only need to submit an application on the app, and after being authorized, they can enter and exit the building independently, saving time for both parties by eliminating the need for visitors to search for those they are visiting.

[0003] The intelligent access control system saves real-time surveillance footage. Obviously, the amount of surveillance data saved in this way is huge, and it is necessary to select and store video footage of different importance in different ways. For example, if it is necessary to store the video footage when two adjacent human targets in the video footage are in contact, the video footage should be stored as the key footage; otherwise, the video footage should be stored as the non-key footage. However, it is difficult to accurately determine whether two adjacent human targets in the video footage are in physical contact. Summary of the Invention

[0004] To address technical issues in related fields, this invention provides a smart community access control response system. By using two adjacent human targets in the enhanced scene image (i.e., the step-by-step enhanced image) of the smart access control system as the first target and the second target respectively, the system determines the overall depth of field of the area occupied by the first target in the step-by-step enhanced image, the overall depth of field of the area occupied by the second target in the step-by-step enhanced image, and the imaging focal length of the real-time imaging device. The system also sets the horizontal and vertical coordinates of each pixel in the area occupied by the first target, the second target, and the first target in the step-by-step enhanced image. The overall depth of field of the area occupied by the first target in the image, the overall depth of field of the area occupied by the second target in the step-by-step enhanced image, and the imaging focal length of the real-time imaging device are input in parallel into the convolutional neural network model. The convolutional neural network model is executed to obtain the state identifier output by the convolutional neural network model, indicating whether the first target and the second target are in physical contact. When it is determined that two adjacent human targets in the step-by-step enhanced image are in contact, the step-by-step enhanced image is stored as a key image. Otherwise, the step-by-step enhanced image is stored as a non-key image. The storage time of the key image is longer than that of the non-key image, thereby completing the intelligent sorting and handling of key images.

[0005] According to the present invention, a smart community access control system is provided, the system comprising:

[0006] The device is constructed layer by layer to perform multiple learning operations on the convolutional neural network to obtain a convolutional neural network after multiple learning operations and output it as a convolutional neural network model. The number of learning operations of the convolutional neural network is positively correlated with the resolution of the real-time imaging device of the smart access control.

[0007] Step-by-step enhancement equipment is used to perform sharpening operations, piecewise linear grayscale transformation operations, and histogram correction operations on the image of a real-time imaging device step by step to obtain and output the corresponding step-by-step enhanced image.

[0008] A real-time imaging device is installed above the smart access control system and connected to the step-by-step enhancement device. It is used to perform real-time imaging operations on the scene in front of the smart access control system during operation, so as to obtain and output the corresponding imaging image.

[0009] An intelligent judgment mechanism, connected to both the step-by-step enhancement device and the layer-by-layer construction device, is used to identify two adjacent human targets in the step-by-step enhancement image as the first target and the second target, respectively. It determines the overall depth of field of the area occupied by the first target in the step-by-step enhancement image, the overall depth of field of the area occupied by the second target in the step-by-step enhancement image, and the imaging focal length of the real-time imaging device. The mechanism inputs the horizontal and vertical coordinates of each pixel in the area occupied by the first target in the step-by-step enhancement image, the horizontal and vertical coordinates of each pixel in the area occupied by the second target in the step-by-step enhancement image, the overall depth of field of the area occupied by the first target in the step-by-step enhancement image, and the imaging focal length of the real-time imaging device into the convolutional neural network model in parallel, and executes the convolutional neural network model to obtain a state identifier output by the convolutional neural network model indicating whether the first target and the second target are in physical contact.

[0010] The monitoring and processing mechanism, connected to the intelligent judgment mechanism, is used to store the step-by-step enhanced image as a key image when the received status identifier indicates that the first target and the second target entities are in contact; otherwise, it stores the step-by-step enhanced image as a non-key image.

[0011] Specifically, when the received status identifier indicates that the first target and the second target entities are in contact, the step-by-step enhanced screen is stored as a key storage screen; otherwise, the step-by-step enhanced screen is stored as a non-key storage screen, which includes the storage duration of the key storage screen being longer than the storage duration of the non-key storage screen.

[0012] Therefore, it can be seen that the present invention has the following four significant technical effects:

[0013] Technical Effect A: The two adjacent human targets in the front scene of the smart access control system after targeted enhancement processing, namely the step-by-step enhanced image, are respectively taken as the first target and the second target. The overall depth value of the area occupied by the first target in the step-by-step enhanced image, the overall depth value of the area occupied by the second target in the step-by-step enhanced image, and the imaging focal length of the real-time imaging device are determined, thereby obtaining targeted basic data for subsequent intelligent identification.

[0014] Technical Effect B: The horizontal and vertical coordinates of each pixel in the area occupied by the first target in the step-by-step enhanced image, the horizontal and vertical coordinates of each pixel in the area occupied by the second target in the step-by-step enhanced image, the overall depth of field of the area occupied by the first target in the step-by-step enhanced image, the overall depth of field of the area occupied by the second target in the step-by-step enhanced image, and the imaging focal length of the real-time imaging device are input in parallel into the convolutional neural network model, and the convolutional neural network model is executed to obtain the state identifier output by the convolutional neural network model, indicating whether the first target and the second target are in physical contact;

[0015] Technical effect C: When two adjacent human target entities are in contact in the step-by-step enhanced image, the step-by-step enhanced image is stored as a key image; otherwise, the step-by-step enhanced image is stored as a non-key image. The storage time of the key image is longer than that of the non-key image, thereby completing the intelligent sorting and handling of key images.

[0016] Technical Effect D: The targeted enhancement processing of the front scene image of the smart access control system involves performing sharpening operations, piecewise linear grayscale transformation operations, and histogram correction operations on the image image of the real-time imaging device step by step, thereby providing higher quality image content for subsequent intelligent judgment.

[0017] The smart community access control system of this invention is logically reliable and intelligently designed. Because it uses an artificial intelligence model to intelligently identify whether two adjacent human targets in a surveillance video are in physical contact based on multiple visual information, and then adopts different image storage strategies, it improves the intelligence level of surveillance video storage management. Attached Figure Description

[0018] The embodiments of the present invention will now be described with reference to the accompanying drawings, wherein:

[0019] Figure 1 This is a schematic diagram of the internal structure of a smart community access control system according to the first embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the internal structure of a smart community access control system according to a second embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram of the internal structure of a smart community access control system according to a third embodiment of the present invention. Detailed Implementation

[0022] The embodiments of the smart community access control system of the present invention will now be described in detail with reference to the accompanying drawings.

[0023] Figure 1 This is a schematic diagram of the internal structure of a smart community access control system according to a first embodiment of the present invention. The system includes:

[0024] The device is constructed layer by layer to perform multiple learning operations on the convolutional neural network to obtain a convolutional neural network after multiple learning operations and output it as a convolutional neural network model. The number of learning operations of the convolutional neural network is positively correlated with the resolution of the real-time imaging device of the smart access control.

[0025] Specifically, the layer-by-layer construction device is used to perform multiple learning operations on a convolutional neural network to obtain a convolutional neural network after multiple learning operations and output it as a convolutional neural network model. The number of learning operations of the convolutional neural network is positively correlated with the resolution of the real-time imaging device of the smart access control system. The layer-by-layer construction device can be implemented using an FPGA chip to perform multiple learning operations on a convolutional neural network to obtain a convolutional neural network after multiple learning operations and output it as a convolutional neural network model. The number of learning operations of the convolutional neural network is positively correlated with the resolution of the real-time imaging device of the smart access control system.

[0026] Step-by-step enhancement equipment is used to perform sharpening operations, piecewise linear grayscale transformation operations, and histogram correction operations on the image of a real-time imaging device step by step to obtain and output the corresponding step-by-step enhanced image.

[0027] A real-time imaging device is installed above the smart access control system and connected to the step-by-step enhancement device. It is used to perform real-time imaging operations on the scene in front of the smart access control system during operation, so as to obtain and output the corresponding imaging image.

[0028] An intelligent judgment mechanism, connected to both the step-by-step enhancement device and the layer-by-layer construction device, is used to identify two adjacent human targets in the step-by-step enhancement image as the first target and the second target, respectively. It determines the overall depth of field of the area occupied by the first target in the step-by-step enhancement image, the overall depth of field of the area occupied by the second target in the step-by-step enhancement image, and the imaging focal length of the real-time imaging device. The mechanism inputs the horizontal and vertical coordinates of each pixel in the area occupied by the first target in the step-by-step enhancement image, the horizontal and vertical coordinates of each pixel in the area occupied by the second target in the step-by-step enhancement image, the overall depth of field of the area occupied by the first target in the step-by-step enhancement image, and the imaging focal length of the real-time imaging device into the convolutional neural network model in parallel, and executes the convolutional neural network model to obtain a state identifier output by the convolutional neural network model indicating whether the first target and the second target are in physical contact.

[0029] The monitoring and processing mechanism, connected to the intelligent judgment mechanism, is used to store the step-by-step enhanced image as a key image when the received status identifier indicates that the first target and the second target entities are in contact; otherwise, it stores the step-by-step enhanced image as a non-key image.

[0030] Specifically, when the received status identifier indicates that the first target and the second target entities are in contact, the step-by-step enhanced screen is stored as a key storage screen; otherwise, the step-by-step enhanced screen is stored as a non-key storage screen, including: the storage duration of the key storage screen is longer than the storage duration of the non-key storage screen.

[0031] Specifically, the horizontal and vertical coordinates of each pixel in the area occupied by the first target in the step-by-step enhanced image, the horizontal and vertical coordinates of each pixel in the area occupied by the second target in the step-by-step enhanced image, the overall depth value of the area occupied by the first target in the step-by-step enhanced image, the overall depth value of the area occupied by the second target in the step-by-step enhanced image, and the imaging focal length of the real-time imaging device are input in parallel into the convolutional neural network model, and the convolutional neural network model is executed to obtain the status identifier output by the convolutional neural network model indicating whether the first target and the second target are in physical contact. This includes converting the horizontal and vertical coordinates of each pixel in the area occupied by the first target in the step-by-step enhanced image, the horizontal and vertical coordinates of each pixel in the area occupied by the second target in the step-by-step enhanced image, the overall depth value of the area occupied by the first target in the step-by-step enhanced image, the overall depth value of the area occupied by the second target in the step-by-step enhanced image, and the imaging focal length of the real-time imaging device into binary values ​​and then synchronously inputting them into the convolutional neural network model.

[0032] Figure 2 This is a schematic diagram of the internal structure of a smart community access control system according to a second embodiment of the present invention.

[0033] Compared to Figure 1 The smart community access control system according to the second embodiment of the present invention may further include:

[0034] The power detection mechanism includes multiple power detection units, which are used to detect the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, respectively.

[0035] The power detection mechanism includes multiple power detection units, which are used to detect the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device respectively. This includes calculating the current input power of each of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device by measuring the real-time power supply voltage and the real-time power supply current respectively.

[0036] The calculation of the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device by measuring the real-time power supply voltage and real-time power supply current respectively includes: for any device of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, its current input power is the product of its real-time power supply voltage and its real-time power supply current;

[0037] The power detection mechanism includes multiple power detection units for detecting the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, respectively. It also includes: the multiple power detection units used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device have the same upper limit threshold and lower limit threshold for power measurement.

[0038] The power detection mechanism includes multiple power detection units for detecting the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, respectively. It also includes multiple power detection units with identical internal structures used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device.

[0039] Figure 3 This is a schematic diagram of the internal structure of a smart community access control system according to a third embodiment of the present invention.

[0040] Compared to Figure 1 The smart community access control system shown in the third embodiment of the present invention may further include:

[0041] The on-site notification mechanism is connected to multiple power detection units respectively used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, and is used to perform corresponding power notification operations based on the power measurement results of the multiple power detection units respectively used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device;

[0042] The on-site notification mechanism is connected to multiple power detection units respectively used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device. It is used to perform corresponding power notification operations based on the power measurement results of the multiple power detection units used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device. The on-site notification mechanism includes a static storage unit for storing power notification thresholds.

[0043] The on-site notification mechanism, which is connected to multiple power detection units respectively used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, and is used to perform corresponding power notification operations based on the power measurement results of the multiple power detection units respectively used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, further includes: the on-site notification mechanism performing corresponding power notification operations for the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device respectively based on a voice playback mode.

[0044] In addition, in the smart community access control response system, the horizontal and vertical coordinate values ​​of each pixel in the area occupied by the first target in the step-by-step enhanced image, the horizontal and vertical coordinate values ​​of each pixel in the area occupied by the second target in the step-by-step enhanced image, the overall depth value of the area occupied by the first target in the step-by-step enhanced image, the overall depth value of the area occupied by the second target in the step-by-step enhanced image, and the imaging focal length of the real-time imaging device are input in parallel into the convolutional neural network model, and the convolutional neural network model is executed to obtain the state identifier output by the convolutional neural network model indicating whether the first target and the second target are in physical contact. Furthermore, the state identifier output by the convolutional neural network model indicating whether the first target and the second target are in physical contact is represented in binary numerical form.

[0045] Those skilled in the art should understand that the embodiments of the present invention shown in the above description are merely examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and described in the embodiments, and any variations or modifications may be made to the implementation of the present invention without departing from the stated principles.

Claims

1. A smart community access control system, characterized in that, The system includes: The device is constructed layer by layer to perform multiple learning operations on the convolutional neural network to obtain a convolutional neural network after multiple learning operations and output it as a convolutional neural network model. The number of learning operations of the convolutional neural network is positively correlated with the resolution of the real-time imaging device of the smart access control. Step-by-step enhancement equipment is used to perform sharpening operations, piecewise linear grayscale transformation operations, and histogram correction operations on the image of a real-time imaging device step by step to obtain and output the corresponding step-by-step enhanced image. A real-time imaging device is installed above the smart access control system and connected to the step-by-step enhancement device. It is used to perform real-time imaging operations on the scene in front of the smart access control system during operation, so as to obtain and output the corresponding imaging image. An intelligent judgment mechanism, connected to both the step-by-step enhancement device and the layer-by-layer construction device, is used to identify two adjacent human targets in the step-by-step enhancement image as the first target and the second target, respectively. It determines the overall depth of field of the area occupied by the first target in the step-by-step enhancement image, the overall depth of field of the area occupied by the second target in the step-by-step enhancement image, and the imaging focal length of the real-time imaging device. The mechanism inputs the horizontal and vertical coordinates of each pixel in the area occupied by the first target in the step-by-step enhancement image, the horizontal and vertical coordinates of each pixel in the area occupied by the second target in the step-by-step enhancement image, the overall depth of field of the area occupied by the first target in the step-by-step enhancement image, and the imaging focal length of the real-time imaging device into the convolutional neural network model in parallel, and executes the convolutional neural network model to obtain a state identifier output by the convolutional neural network model indicating whether the first target and the second target are in physical contact. The monitoring and processing mechanism, connected to the intelligent judgment mechanism, is used to store the step-by-step enhanced image as a key image when the received status identifier indicates that the first target and the second target entities are in contact; otherwise, it stores the step-by-step enhanced image as a non-key image. Specifically, when the received status identifier indicates that the first target and the second target entities are in contact, the step-by-step enhanced screen is stored as a key storage screen; otherwise, the step-by-step enhanced screen is stored as a non-key storage screen, which includes the storage duration of the key storage screen being longer than the storage duration of the non-key storage screen.

2. The smart community access control system as described in claim 1, characterized in that: The horizontal and vertical coordinates of each pixel in the area occupied by the first target in the step-by-step enhanced image, the horizontal and vertical coordinates of each pixel in the area occupied by the second target in the step-by-step enhanced image, the overall depth value of the area occupied by the first target in the step-by-step enhanced image, the overall depth value of the area occupied by the second target in the step-by-step enhanced image, and the imaging focal length of the real-time imaging device are input in parallel into the convolutional neural network model, and the convolutional neural network model is executed to obtain the status identifier output by the convolutional neural network model indicating whether the first target and the second target are in physical contact. This includes converting the horizontal and vertical coordinates of each pixel in the area occupied by the first target in the step-by-step enhanced image, the horizontal and vertical coordinates of each pixel in the area occupied by the second target in the step-by-step enhanced image, the overall depth value of the area occupied by the first target in the step-by-step enhanced image, the overall depth value of the area occupied by the second target in the step-by-step enhanced image, and the imaging focal length of the real-time imaging device into binary values ​​and then synchronously inputting them into the convolutional neural network model.

3. The smart community access control system as described in claim 2, characterized in that, The system also includes: The power detection mechanism includes multiple power detection units, which are used to detect the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, respectively. The power detection mechanism includes multiple power detection units, which are used to detect the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device respectively. This includes calculating the current input power of each of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device by measuring the real-time supply voltage and real-time supply current respectively.

4. The smart community access control system as described in claim 3, characterized in that: The calculation of the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device by measuring the real-time supply voltage and real-time supply current respectively includes: for any device of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, its current input power is the product of its real-time supply voltage and its real-time supply current.

5. The smart community access control system as described in claim 3, characterized in that: The power detection mechanism includes multiple power detection units for detecting the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, respectively. It further includes ensuring that the multiple power detection units used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device each have the same upper and lower power measurement thresholds.

6. The smart community access control system as described in claim 5, characterized in that: The power detection mechanism includes multiple power detection units for detecting the current input power of the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device, respectively. It also includes multiple power detection units with identical internal structures used by the monitoring and processing mechanism, the intelligent judgment mechanism, the step-by-step enhancement device, and the layer-by-layer construction device.

Images