Safety protection method for performing AI visual identification and preventing candid photography by using camera

Abstract

The invention relates to the field of computer screen candid photography prevention, in particular to a safety protection method for preventing candid photography through AI visual recognition by means of a camera. According to the scheme, the method comprises the following steps: virtualizing an HDMI / VGA signal sent by a confidential computer into equipment of an optical camera on an operating system level, accessing the accessed equipment through a camera number, detecting an input signal in a virtual camera mode, and judging whether an HDMI / VGA line is correctly inserted and whether the signal is transmitted in; bearing information input by the HDMI / VGA and different prompt information through different views; screen locking and screen unlocking are achieved by switching different views, and switching of the different views is achieved in the mode that the transparency of the views is set to be full-transparent or non-transparent. Through a mode of capturing a picture by a camera, a picture of an HDMI / VGA input signal is captured on a view, and evidence obtaining processing is carried out in a non-invasive mode. The method is suitable for preventing candid shooting of the computer screen.

Description

Technical Field

[0001] The present invention relates to the field of anti-stealth shooting of computer screens, and specifically relates to a security protection method for anti-stealth shooting using AI visual recognition by a camera. Background Art

[0002] The control of HDMI (High Definition Multimedia Interface) / VGA (Video Graphics Array) information transmitted from a confidential computer is controlled by a hardware switch when passing through the Figure 1 host device shown in the figure. When an optical camera takes a picture and AI recognition on the host device detects mobile phone stealth shooting, the hardware switch of the host closes the HDMI channel information transmitted from the confidential computer to the display and switches to the reminder interface provided by the host device. When the stealth shooting scenario is lifted, the HDMI channel transmitted from the confidential computer to the display is reopened through the host hardware switch.

[0003] The disadvantages of the traditional method are:

[0004] The switching time is long, and the signal switching cannot be completed within an extremely short time, resulting in an increased risk of information leakage and not meeting the security and confidentiality standards.

[0005] It belongs to an intrusive switching method, which requires the hardware to perform intrusive control on the HDMI / VGA signal and the interaction protocol, including signal parsing and interaction protocol interception, which does not meet the requirements of confidentiality specifications.

[0006] It is difficult to effectively switch multiple channels in the case of the screen being locked. For example, it is difficult to quickly and effectively select multiple channels such as setting multiple channels for mobile phone stealth shooting prompts, object blocking camera prompts, camera not inserted prompts, HDMI no signal prompts, etc. using a switch. Summary of the Invention

[0007] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a security protection method for anti-stealth shooting using AI visual recognition by a camera, which realizes non-intrusive switching and reduces the switching time at the same time.

[0008] The present invention adopts the following technical solutions to achieve the above purpose. The present invention provides a security protection method for anti-stealth shooting using AI visual recognition by a camera, including:

[0009] S1. Virtualize the HDMI / VGA signal sent by the confidential computer into the signal of an optical camera at the operating system level, and configure the resolution of the HDMI / VGA signal in the view by configuring the input parameters of the camera.

[0010] S2. Access the connected device by the camera number, and detect the input signal in the way of virtual camera to judge whether the HDMI cable is correctly inserted and whether there is signal input.

[0011] S3. Carry the information input by HDMI / VGA and different prompt messages through different views.

[0012] S4. Realize locking and unlocking the screen by switching different views. The switching of different views is realized by setting the view transparency to be fully transparent or opaque.

[0013] S5. Capture the picture of the HDMI / VGA input signal on the view in the way of camera capture, so as to achieve the evidence collection process in a non-invasive way.

[0014] Furthermore, virtualizing the HDMI / VGA signal sent by the classified computer into the signal of an optical camera at the operating system level specifically includes:

[0015] Decoding the HDMI / VGA signal into the original video data;

[0016] Converting the format and compressing the original video data;

[0017] Encapsulating the compressed original video data into the data corresponding to the optical camera protocol;

[0018] Finally, outputting the standard optical camera interface data.

[0019] Furthermore, detecting the input signal in the way of virtual camera specifically includes:

[0020] Judging the input signal by setting parameter parsing, and the parameter corresponds to the format converted from the original video data.

[0021] The beneficial effects of the present invention are:

[0022] The HDMI / VGA signal of the present invention can configure its resolution on the view by configuring the input parameters of the camera, so as to avoid taking over and parsing the HDMI interaction protocol and meet the non-invasive security requirements.

[0023] The information input by HDMI and different prompt messages of the present invention can all be carried through different views. When it is necessary to lock and unlock the screen, it can be quickly switched in a short time by switching different views.

[0024] The present invention captures the picture of the signal input by HDMI on the view in the way of camera capture, so as to achieve the evidence collection process in a non-invasive way. Description of the Drawings

[0025] Figure 1 It is a schematic diagram of the screen anti-photography system provided by the present invention;

[0026] Figure 2 It is a flowchart of a security protection method for preventing secret photography by using AI visual recognition with a camera provided by the present invention. Specific embodiments

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0028] The present invention provides a security protection method for preventing secret photography by using AI visual recognition with a camera. As Figure 2 shown, it specifically includes:

[0029] S1. Virtually convert the HDMI / VGA signal sent by the confidential computer into the signal of an optical camera at the operating system level. The HDMI / VGA signal configures its resolution on the view by configuring the input parameters of the camera.

[0030] Specifically, virtually converting the HDMI / VGA signal into the signal of an optical camera includes:

[0031] Decode the HDMI / VGA signal into raw video data through an HDMI decoding chip.

[0032] Perform format conversion and compression on the video data through a video processing unit (such as converting it to the MJPEG or YUV format).

[0033] Package the processed video data into a Camera protocol (such as the UVC protocol) through a Camera signal generation module.

[0034] Through interface conversion: output the signal as a standard Camera interface (such as USB or MIPI CSI).

[0035] When configuring the resolution, the resolution can be viewed on the HDMI decoding chip and then the resolution on the view can be set.

[0036] S2. Access the connected device by the camera number and detect the input signal in the way of a virtual camera to judge whether the HDMI cable is correctly inserted and whether there is a signal input.

[0037] Specifically, inside the operating system, the ID and number of the camera hardware are corresponded one by one to form a list, and then the camera is accessed by traversing.

[0038] S3. Carry the information input via HDMI / VGA and different prompt messages through different views;

[0039] S4. Realize locking and unlocking the screen by switching between different views. The switching between different views is achieved by setting the transparency of the view to be fully transparent or opaque;

[0040] S5. Capture pictures of the HDMI / VGA input signal on the view by means of camera screenshot, so as to perform evidence collection processing in a non-invasive manner. Certain fixed areas can be blocked or changed non-invasively on the view of the HDMI / VGA input, such as blurring with mosaic, etc.

[0041] The above are only the preferred embodiments of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments. Instead, it can be used in various other combinations, modifications and environments, and can be changed within the scope of the concept described herein through the above teachings or the technology or knowledge in related fields. Any changes and variations made by those skilled in the art without departing from the spirit and scope of the present invention shall fall within the protection scope of the appended claims of the present invention.

Claims

1. A security protection method for preventing secret photographing by using AI visual recognition with a camera, characterized in that, Including: S1. Virtually convert the HDMI / VGA signal sent by the classified computer into the signal of an optical camera at the operating system level, and configure its resolution on the view by configuring the input parameters of the camera for the HDMI / VGA signal; S2. Access the connected device by the camera number, and detect the input signal in the way of a virtual camera to judge whether the HDMI cable is correctly inserted and whether there is an incoming signal; S3. Carry the information input by HDMI / VGA and different prompt messages through different views; S4. Realize locking and unlocking the screen by switching between different views, and realize the switching between different views by setting the view transparency to be fully transparent or opaque; S5. Capture the picture of the HDMI / VGA input signal on the view in the way of camera screenshot, so as to conduct forensic processing in a non-invasive way.

2. The security protection method for preventing secret photographing by using AI visual recognition with a camera according to claim 1, characterized in that, Specifically, virtually converting the HDMI / VGA signal sent by the classified computer into the signal of an optical camera at the operating system level includes: Decoding the HDMI / VGA signal into the original video data; Converting the format and compressing the original video data; Encapsulating the compressed original video data into the data corresponding to the optical camera protocol; Finally, outputting the standard optical camera interface data.

3. The security protection method for preventing secret photographing by using AI visual recognition with a camera according to claim 1, characterized in that, Specifically, detecting the input signal in the way of a virtual camera includes: Judging the input signal by setting parameter parsing, and the parameter corresponds to the format converted from the original video data.

Images