Image security transmission method and system for smart glasses

Abstract

The present application belongs to the technical field of image security transmission, and particularly relates to an image security transmission method and system for smart glasses, which comprises the following steps: synchronously collecting video stream data, motion posture data and system state data through a multi-source sensor, and performing space-time feature alignment; calculating a perception disturbance coupling factor corresponding to each frame of image according to the motion posture data, the current remaining percentage of power and the channel received signal strength indication; calculating an encryption mask threshold of each frame of image based on the perception disturbance coupling factor and the current remaining percentage of power; and selectively encrypting the frequency domain coefficients of each frame of image by using the encryption mask threshold and a chaotic key stream. By constructing a dynamic mapping mechanism of the perception disturbance coupling factor and the encryption mask threshold, the present application realizes the adaptive adjustment of the encryption strength to the wearer's motion state and the device resource state, thereby significantly prolonging the device endurance time while ensuring the core privacy security.

Description

Technical Field

[0001] This invention relates to the field of secure image transmission technology. More specifically, this invention relates to a method and system for secure image transmission in smart glasses. Background Technology

[0002] With the rapid development of wearable technology, smart glasses, as an important carrier for augmented reality (AR) and first-person perspective recording, have been widely used in fields such as security patrol, telemedicine, and personal life recording. In these application scenarios, the video streams collected in real time by smart glasses often contain sensitive privacy information such as faces and environmental locations, so the data must be encrypted to prevent leakage.

[0003] However, existing video encryption technologies have the following limitations:

[0004] (1) The contradiction between the high energy consumption of full-disk encryption and the device's battery life: Traditional high-strength encryption algorithms such as AES have high computational complexity, while smart glasses are usually lightweight, have limited battery capacity and small heat dissipation space; if full-band high-strength encryption is performed on each frame of image, it is very easy to cause device computational overload, serious heat generation and a significant reduction in battery life, which cannot meet the needs of long-term wear.

[0005] (2) Fixed strategies cannot adapt to dynamic scenarios: Existing lightweight encryption solutions usually use a fixed encryption ratio or a preset region of interest, which cannot be dynamically adjusted according to the wearer's real-time movement status (such as being stationary, walking, or running vigorously) and the current device status (such as battery level and network signal strength). For example, during vigorous exercise, screen shake causes a sharp increase in information entropy. If a high encryption ratio is maintained at this time, it will cause unnecessary waste of computing power. When the battery is extremely low, there is a lack of effective degradation protection mechanism, which may cause the device to shut down prematurely and lose critical data.

[0006] Therefore, there is an urgent need for an adaptive control method that dynamically adjusts encryption strength to achieve the optimal balance between ensuring core privacy security and extending device battery life. Summary of the Invention

[0007] To address the technical problem that existing video encryption technologies cannot achieve the optimal balance between ensuring core privacy security and extending device battery life, this invention provides solutions in the following aspects.

[0008] In a first aspect, the present invention provides a method for secure image transmission for smart glasses, comprising: synchronously acquiring video stream data, motion posture data, and system status data through multi-source sensors integrated in the smart glasses, and performing spatiotemporal feature alignment on the acquired data; calculating a perceptual perturbation coupling factor corresponding to each frame of image based on the motion posture data, the current remaining battery percentage in the system status data, and the channel received signal strength indication; calculating an encryption mask threshold for each frame of image based on the perceptual perturbation coupling factor and the current remaining battery percentage; and selectively encrypting the frequency domain coefficients of each frame of image using the encryption mask threshold and a preset chaotic key stream to obtain an encrypted video frame.

[0009] Preferably, the simultaneous acquisition of video stream data, motion posture data, and system status data via multi-source sensors integrated in the smart glasses includes: acquiring real-time video streams at a preset frame rate using the front-facing camera of the smart glasses, converting each frame of image data from the RGB color space to the YCrCb color space, and extracting the luminance component Y; acquiring three-axis acceleration and three-axis angular velocity using the inertial measurement unit built into the smart glasses; reading the current remaining battery percentage using the power management chip of the smart glasses, and reading the channel received signal strength indication using the wireless communication module.

[0010] This invention targets the human eye's high sensitivity to changes in brightness by extracting and processing the brightness component. This allows for the achievement of maximum visual obfuscation with minimal computational cost in subsequent encryption steps. Simultaneously, it comprehensively collects motion and system state data, providing data support for multi-dimensional contextual awareness.

[0011] Preferably, the formula for calculating the perceptual perturbation coupling factor corresponding to each frame of image is: In the formula: For the first The perceptual perturbation coupling factor corresponding to the frame image; This is the preset scaling factor; It is the natural logarithm function; For axial index, the value is... , , ; For the first The first frame image corresponding to The absolute value of the axial angular velocity; It is a natural constant; For the first The first frame image corresponding to The rate of change of axial acceleration; For the first The current remaining battery percentage at the time of frame image acquisition; For the first Channel received signal strength indication during frame image acquisition.

[0012] This invention constructs a perceptual disturbance coupling factor, introducing energy and channel state as adjustment levers. When the power decreases or the signal deteriorates, the denominator decreases, resulting in a significant increase in the value of the perceptual disturbance coupling factor. This amplifies the impact weight of external motion disturbances on the system, driving the system to make more sensitive policy adjustments in low-resource, high-risk scenarios, and avoiding rigid fixed encryption strategies.

[0013] Preferably, the first Rate of change of axial acceleration equal to the The first frame image corresponding to Axial acceleration and the first The first frame image corresponding to The ratio of the difference in axial acceleration to the time interval between acquiring two adjacent frames of images.

[0014] Preferably, the encryption mask threshold is calculated using the following formula: In the formula, For the first Encryption mask threshold for frame images; This indicates the floor function; As the benchmark coefficient; For the first The current remaining battery percentage at the time of frame image acquisition; It is the hyperbolic tangent function; For the first The perceptual perturbation coupling factor corresponding to the frame image; This indicates taking the maximum value.

[0015] This invention utilizes the threshold calculation formula to construct a dynamic balancing mechanism, in which the power consumption item acts as the main controller, automatically shrinking the encryption range to save energy when the power is low; the disturbance perception item acts as the adjustment, appropriately adjusting the threshold to expand the encryption range when violent movement or environmental disturbance is detected. This mechanism not only prevents the power from being exhausted due to over-protection, but also prevents the leakage of critical privacy information due to violent screen shaking.

[0016] Preferably, the selective encryption of the frequency domain coefficients of each frame of the image using the encryption mask threshold and a preset chaotic key stream includes: generating a chaotic key stream of the same length as the image data using a preset chaotic system; and then... The luminance component of the frame image is subjected to block discrete cosine transform to obtain multiple frequency domain coefficient matrices; for each frequency domain coefficient matrix, the frequency domain coefficients are traversed in Zigzag scanning order, and the frequency domain coefficients of each frame image are selectively encrypted.

[0017] Preferably, the selective encryption of the frequency domain coefficients of each frame of image includes: for frequency domain coefficients determined to be key privacy coefficients, performing XOR operation and sign bit flipping operation using the chaotic key stream; for frequency domain coefficients determined to be non-key high-frequency details, keeping the values ​​unchanged; and performing inverse discrete cosine transform on the processed frequency domain coefficients to generate encrypted video frames.

[0018] Preferably, the method for determining the frequency domain coefficient includes: extracting the scan number of the current frequency domain coefficient; if the scan number is less than the encryption mask threshold, determining that the frequency domain coefficient is a critical privacy coefficient; if the scan number is greater than or equal to the encryption mask threshold, determining that the frequency domain coefficient is a non-critical high-frequency detail.

[0019] This invention combines the energy concentration characteristics of discrete cosine transform with a dynamic threshold, processing only low-frequency coefficients containing key image information; by using XOR operations to obfuscate numerical values ​​and flipping the sign bit to destroy directionality, it can completely destroy the visual structure of the image, achieving efficient and low-latency encryption and avoiding the resource waste caused by full-disk encryption.

[0020] In a second aspect, the present invention provides an image secure transmission system for smart glasses, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the aforementioned image secure transmission method for smart glasses is implemented.

[0021] By adopting the above technical solution, a computer program for secure image transmission in smart glasses is generated and stored in a memory for loading and execution by a processor. This allows for the creation of a terminal device based on the memory and processor, facilitating its use.

[0022] The beneficial effects of this invention are as follows:

[0023] This invention constructs a perceptual disturbance coupling factor, introducing energy and channel state as adjustment levers. When battery power decreases or signal deteriorates, the impact weight of external motion disturbances on the system is amplified, driving the system to make more sensitive strategy adjustments in low-resource, high-risk scenarios. Furthermore, based on this perceptual disturbance coupling factor and the current remaining battery power, a dynamic balancing mechanism for the encryption mask threshold is constructed. The battery power plays a controlling role, automatically shrinking the encryption range to save energy when low power is detected. Simultaneously, the perceptual disturbance adjustment function appropriately adjusts the threshold to expand the encryption range when violent motion is detected, preventing the leakage of critical information due to severe image jitter. Finally, combining the energy concentration characteristics after discrete cosine transform, the dynamically calculated encryption mask threshold is used to perform chaotic encryption only on low-frequency coefficients containing key image information. XOR operations are used to confuse numerical values ​​and reverse sign bits to destroy directionality. While completely destroying the image's visual structure to ensure core privacy and security, this avoids the resource waste associated with full-screen encryption. This achieves adaptive adjustment of encryption strength to the wearer's motion state and device resource state, significantly extending the device's battery life. Attached Figure Description

[0024] The above and other objects, features, and advantages of the present invention will become readily apparent from the following detailed description of exemplary embodiments, accompanied by the accompanying drawings. In the drawings, several embodiments of the invention are illustrated by way of example and not limitation, and like or corresponding reference numerals denote like or corresponding parts, wherein:

[0025] Figure 1 This is a flowchart illustrating a method for secure image transmission for smart glasses according to the present invention;

[0026] Figure 2 This is a schematic diagram illustrating the adaptive encryption mask threshold under perceived perturbation and energy constraints;

[0027] Figure 3 This is a schematic diagram illustrating the performance differences between the present invention and the prior art. Detailed Implementation

[0028] 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, not all, of the embodiments of the present invention. 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.

[0029] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0030] This invention discloses a method for secure image transmission in smart glasses, referring to... Figure 1This includes steps S1 to S4:

[0031] S1: Through the multi-source sensors integrated into the smart glasses, video stream data, motion posture data, and system status data are collected simultaneously.

[0032] It should be noted that due to the differences in the operating frequency and sampling clock of different sensors, the multi-source data collected cannot be strictly aligned in the time dimension, resulting in misalignment between the perceived state and the video frame. Therefore, this invention establishes a unified spatiotemporal benchmark to synchronously collect and align the features of multi-source data, thereby ensuring the accuracy of the input to the decision model.

[0033] Specifically, the smart glasses use their front-facing camera to capture real-time video streams at a preset frame rate. For each captured frame of image data, the image is converted from the RGB color space to the YCrCb color space, and the luminance component is extracted. Further processing is required because the human eye is most sensitive to changes in brightness, and encrypting the brightness components can achieve the greatest visual confusion effect at the lowest cost.

[0034] Simultaneously, the inertial measurement unit built into the smart glasses synchronously collects six-axis data from the smart glasses, including three-axis acceleration. and triaxial angular velocity These data directly reflect the wearer's head movements and overall motion status; the power management chip of the smart glasses reads the current remaining battery percentage. Read the channel received signal strength indication via the wireless communication module. .

[0035] Furthermore, spatiotemporal feature alignment is performed on the above data to ensure that the first Frame images strictly correspond to the first Motion posture data at any time, current remaining battery percentage and channel received signal strength indication .

[0036] S2: Calculate the sensing disturbance coupling factor for each frame of image based on motion posture data, current remaining battery percentage, and channel received signal strength indication.

[0037] It should be noted that simple motion data cannot fully reflect the urgency of encryption for the device under current resource constraints. This may cause the system to crash due to excessive encryption in low-power, high-risk scenarios, or provide insufficient protection against privacy leaks during vigorous movements when the power is sufficient. Therefore, this invention constructs a perceptual disturbance coupling factor, introducing energy and channel state as adjustment levers. When the power decreases or the signal deteriorates, the impact weight of external disturbances on the system is amplified, thereby driving the system to make more sensitive policy adjustments.

[0038] Specifically, according to the first Triaxial acceleration at time t Triaxial angular velocity Current remaining battery percentage and channel received signal strength indication Calculate the first The perceptual perturbation coupling factor corresponding to the frame image is calculated using the following formula:

[0039]

[0040] In the formula: For the first The perceptual perturbation coupling factor corresponding to the frame image; This is the preset scaling factor; It is the natural logarithm function; For axial index, the value is... , , ; For the first The first frame image corresponding to The absolute value of the axial angular velocity, characterizing the instantaneous rotational intensity, is normalized to... ; It is a natural constant; For the first The first frame image corresponding to The rate of change of the axial acceleration is equal to the first Frame image and the first The first frame image corresponding to The ratio of the difference in axial acceleration to the time interval between acquiring two adjacent frames of images; For the first The current remaining battery percentage at the time of frame image acquisition, with a value range of [value range missing]. ; For the first Channel received signal strength indication during frame image acquisition, normalized to .

[0041] Among them, the numerator term Characterizing the intensity of the wearer's exercise using an exponential function The rate of change of acceleration is processed because sudden shocks have a greater impact on image quality than steady acceleration. As motion intensifies, the numerator increases, leading to a higher perturbation coupling factor. Increase; denominator term Characterizing the energy and communication redundancy of the system, when the power... Reduce or signal When the denominator decreases, under the same intensity of motion, the calculated value becomes... The value increases significantly, indicating that the system's sensitivity to external motion disturbances is amplified when the power is low or the signal is weak.

[0042] Among them, the preset scaling factor Used to adjust the overall amplitude of the disturbance factor, if If the value is too small, the change in the sensing disturbance coupling factor is not significant, and it cannot drive the system out of steady state. If the scaling factor is too large, it can easily lead to system saturation and loss of sensitivity; therefore, the scaling factor should be adjusted accordingly. The range of values ​​is In this embodiment, Set as In other embodiments, implementers may set the parameters according to the actual implementation situation. .

[0043] It should be noted that this invention breaks away from the rigid mode of traditional fixed encryption strength. By establishing a perception perturbation coupling model, the encryption strength, i.e. the encryption mask threshold, can automatically decrease as the battery power decreases. In low-power scenarios, the system automatically discards the encryption of high-frequency details and only protects the core low-frequency contours, thereby significantly reducing the amount of computation and extending the battery life of smart glasses in critical moments.

[0044] S3: Calculate the encryption mask threshold for each frame of image based on the perceived perturbation coupling factor and the current remaining battery percentage.

[0045] It should be noted that since full-frame high-intensity encryption can lead to heat buildup and rapid power consumption, thereby shortening the device's battery life, this invention utilizes the energy concentration characteristics after discrete cosine transform to control the encryption frequency band range by dynamically adjusting the encryption mask threshold. This adjustment follows the power-dominated principle and the motion compensation principle, that is, when the power is low, the encryption range is automatically reduced to survive, but when violent motion is detected, the threshold is appropriately adjusted back to expand the encryption range to prevent the leakage of key information due to violent screen shaking.

[0046] Specifically, based on the first Perceptual perturbation coupling factor corresponding to the frame image and current remaining battery percentage Calculate the first The encryption mask threshold for a frame image is calculated using the following formula:

[0047]

[0048] In the formula, For the first The encryption mask threshold of the frame image represents the cutoff index of the Zigzag scan order, and its value is an integer with a range of [0,63]. This indicates the floor function; As the benchmark coefficient; For the first The current remaining battery percentage at the time of frame image acquisition is used as the main control item and the threshold is directly linearly scaled. It is a hyperbolic tangent function that maps the input to the (0,1) interval, providing smooth saturation properties; For the first The perceptual perturbation coupling factor corresponding to the frame image; This indicates taking the maximum value, used to set a minimum lower limit of 3 to ensure that even when the power is extremely low, the most core privacy, such as the DC component, is still encrypted.

[0049] Among them, the calculation formula of the encryption mask threshold constructs a dynamic balance mechanism: (1) As the main control item, it reflects the main control role of the power: when When the value decreases, the baseline value Directly reduce, force Reduced, which means that the system automatically shrinks the encryption range when the power is low, and only encrypts the most critical low-frequency coefficients, thereby greatly saving computing power and power; (2) For adjustment: when strenuous exercise occurs or the sensitivity is amplified due to low battery, Increase, making Increase, and then through multiplication. right Forward compensation, hyperbolic tangent function Utilizing its nonlinear saturation characteristics, it prevents excessive threshold expansion under extreme disturbances; this is a crisis response mechanism: even with low battery, if extreme environmental disturbances or motion risks are detected, the system will appropriately adjust the threshold and expand the encryption range to prevent scene information under violent movement, such as sudden privacy intrusions that are not effectively masked. However, when the battery is extremely low, due to... The multiplicative effect of this callback is limited, thus ensuring that the system does not run out of power due to overprotection.

[0050] Among them, modulation factor The weights used to control the impact of perceived disturbances on the threshold are as follows: If it is too small, the effect of motor compensation will not be obvious. If the modulation factor is too large, it will cause the threshold correction amplitude to be too large under low power conditions, thus negating the energy-saving effect; therefore, the modulation factor... The range of values ​​is In this embodiment, Set as In other embodiments, implementers may set the parameters according to the actual implementation situation. .

[0051] It should be noted that by introducing IMU data to construct a perception disturbance coupling factor, the system can keenly sense the wearer's violent movements. When the movement changes abruptly, the encryption mask threshold is appropriately increased through nonlinear modulation logic to prevent the leakage of key information caused by violent screen shaking or rapid scene switching, thus solving the security blind spot of the single power control strategy in dynamic scenarios.

[0052] S4: Selectively encrypt the frequency domain coefficients of each frame of the image using an encryption mask threshold and a chaotic key stream.

[0053] It should be noted that traditional encryption methods often treat all data equally, leading to wasted resources. Therefore, this invention utilizes the pseudo-randomness and sensitivity to initial values ​​of chaotic systems, combined with dynamically calculated encryption mask thresholds, to process only low-frequency coefficients containing key image information, achieving efficient and low-latency encryption.

[0054] Specifically, a chaotic key stream of the same length as the image data is generated using a pre-set chaotic system, which includes, but is not limited to, Logistic mapping and Chen's system.

[0055] Furthermore, regarding the first The luminance component of the frame image is subjected to block discrete cosine transform, with a block size of 8×8, to obtain multiple 8×8 frequency domain coefficient matrices.

[0056] Furthermore, for each frequency domain coefficient matrix, the frequency domain coefficients are traversed according to the Zigzag scan order. Under the Zigzag scan order, low-frequency coefficients are concentrated at the beginning of the sequence, containing the main contour information of the image, while high-frequency coefficients are at the end, containing detailed textures. The scan sequence number of the current frequency domain coefficient is extracted. :like The frequency domain coefficient is determined to be a key privacy coefficient. A chaotic keystream is used to perform an XOR operation and sign bit flipping on this coefficient. The XOR operation obfuscates the numerical value of the coefficient, and the sign bit flipping destroys the directionality of the coefficient, thus completely destroying the visual structure of the image. The frequency domain coefficients are determined to be non-critical high-frequency details, so they are not processed and are directly retained.

[0057] Finally, the processed frequency domain coefficients are subjected to inverse discrete cosine transform to generate encrypted video frames.

[0058] It should be noted that instead of complex full-frame pixel-level encryption, selective encryption of frequency domain coefficients is adopted. Combined with the energy concentration characteristics of Zigzag scanning, only a few low-frequency coefficients need to be XORed and flipped to achieve the effect of destroying image visibility. This greatly reduces the time complexity and computing power consumption of the algorithm, making it very suitable for embedded devices with limited computing resources, such as smart glasses.

[0059] Figure 2 This diagram illustrates the adaptive encryption mask threshold under the conditions of perceived disturbance and energy constraints. It includes three main curves: the green dashed line represents the linear decrease in remaining device power over time; the blue background waveform represents the wearer's movement intensity, with perceived disturbance sources calculated from IMU data showing significant movement peaks in the 50-80 and 140-170 frame intervals; the red solid line represents the encryption mask threshold calculated by this invention, which generally decreases with decreasing power, reflecting a low-power energy-saving strategy; however, in the intense movement interval where the blue waveform spikes, the red solid line rises instead of decreasing, showing a significant pulse-like increase. This indicates that under the influence of perceived disturbance, even with low power, once intense wearer movement is detected, the system immediately and forcibly increases the encryption bandwidth to prioritize security.

[0060] Figure 3 The graph shows a performance difference between the present invention and the prior art, comparing the energy consumption growth trends of the two technologies. In the graph, the gray dashed line represents the prior art, whose energy consumption increases rapidly in a straight line, while the orange solid line represents the present invention, whose slope gradually slows down with time and power consumption decrease. Moreover, the final cumulative value of the orange solid line is significantly lower than that of the gray dashed line, proving that the hierarchical encryption strategy of the present invention can effectively reduce computational overhead and extend device battery life.

[0061] This invention also discloses an image secure transmission system for smart glasses, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement an image secure transmission method for smart glasses according to the present invention.

[0062] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

Claims

1. A method for secure image transmission in smart glasses, characterized in that, include: The smart glasses integrate multiple sensors to simultaneously collect video stream data, motion posture data, and system status data, and perform spatiotemporal feature alignment on the collected data. Based on the motion posture data, the current remaining battery percentage in the system status data, and the channel received signal strength indication, the perceptual disturbance coupling factor corresponding to each frame of the image is calculated. In the formula: For the first The perceptual perturbation coupling factor corresponding to the frame image; This is the preset scaling factor; It is the natural logarithm function; For axial index, the value is... , , ; For the first The first frame image corresponding to The absolute value of the axial angular velocity; It is a natural constant; For the first The first frame image corresponding to The rate of change of axial acceleration; For the first The current remaining battery percentage at the time of frame image acquisition; For the first Channel received signal strength indication during frame image acquisition; Based on the perceived perturbation coupling factor and the current remaining battery percentage, calculate the encryption mask threshold for each frame of the image. In the formula, For the first Encryption mask threshold for frame images; This indicates the floor function; As the benchmark coefficient; It is the hyperbolic tangent function; This indicates taking the maximum value; Using a pre-set chaotic key stream, for the first The luminance components of the frame image are subjected to block discrete cosine transform to obtain multiple frequency domain coefficient matrices; For each frequency domain coefficient matrix, the frequency domain coefficients are traversed according to the Zigzag scanning order. The frequency domain coefficients of each frame image are selectively encrypted to obtain the encrypted video frame. The selective encryption includes: if the scan number of the frequency domain coefficient is less than the encryption mask threshold, the frequency domain coefficient is XORed and the sign bit is flipped using a chaotic key stream.

2. The image secure transmission method for smart glasses according to claim 1, characterized in that, The multi-source sensors integrated through the smart glasses simultaneously collect video stream data, motion posture data, and system status data, including: The front-facing camera of the smart glasses captures real-time video streams at a preset frame rate, and converts each frame of image data from the RGB color space to the YCrCb color space to extract the luminance component Y. The triaxial acceleration and triaxial angular velocity are collected by the inertial measurement unit built into the smart glasses; The power management chip of the smart glasses reads the current remaining battery percentage, and the wireless communication module reads the channel reception signal strength indication.

3. The image secure transmission method for smart glasses according to claim 1, characterized in that, The first Rate of change of axial acceleration equal to the The first frame image corresponding to Axial acceleration and the first The first frame image corresponding to The ratio of the difference in axial acceleration to the time interval between acquiring two adjacent frames of images.

4. The image secure transmission method for smart glasses according to claim 1, characterized in that, The method for obtaining the chaotic key stream includes: A chaotic key stream of the same length as the image data is generated using a pre-set chaotic system.

5. The image secure transmission method for smart glasses according to claim 1, characterized in that, The selective encryption of the frequency domain coefficients of each frame of image further includes: If the scan number of the frequency domain coefficients is greater than or equal to the encryption mask threshold, the value of the frequency domain coefficients remains unchanged. The processed frequency domain coefficients are subjected to inverse discrete cosine transform to generate encrypted video frames.

6. An image secure transmission system for smart glasses, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement an image secure transmission method for smart glasses according to any one of claims 1-5.

Images