A non-contact AI laser shaving device and method

Abstract

This invention discloses a non-contact AI laser shaving device and method, relating to the fields of personal care appliances and laser application technology. The device includes an image acquisition unit, a processing unit, and a laser emission unit. The processing unit runs an artificial intelligence algorithm to process the acquired image data, identify the beard hairs, and determine their three-dimensional spatial position. Then, it controls the laser emission unit to target and remove the beard hairs at a non-contact distance. The laser power can be configured as needed to either a low-power multi-pass accumulation mode or a high-power single-cut mode, both of which meet safety standards. This invention does not limit the specific hardware type, algorithm model, laser wavelength, or ranging method. This invention solves the problems of skin damage and residual black spots caused by existing contact shaving, achieving non-contact, clean, and comfortable home shaving, with shaving effects reaching or exceeding those of electric shavers.

Description

Technical Field

[0001] This invention relates to the fields of personal care appliances, laser applications, artificial intelligence and machine vision technology, and specifically to a non-contact AI laser shaving device and method. Background Technology

[0002] Current shaving methods are divided into three categories: contact manual shaving, contact electric shaving, and skin-tight laser hair removal. Contact shaving can easily cause skin abrasions, redness, swelling, and allergies, and it cannot remove the hair shaft located in the epidermis (i.e., the black dots left after shaving). Skin-tight laser hair removal devices aim to destroy hair follicles to achieve permanent hair removal, and are not suitable for daily shaving; at the same time, they rely on skin-tight pressure sensors for unlocking, and cannot work safely in a non-contact state, and lack the ability to accurately identify, dynamically track, and non-contact three-dimensionally locate hair.

[0003] In existing technologies, lasers can be used for hair removal, and machine vision can be used for target recognition. However, no technology has yet formed a complete technical route from "non-contact image acquisition → AI recognition of beard → laser targeted removal of beard" to achieve non-contact, targeted, clean, and skin-friendly home shaving. Existing laser devices are either for skin-to-skin hair removal or industrial marking, neither of which meets the functionality and safety requirements for daily home shaving. Summary of the Invention

[0004] This invention provides a non-contact AI laser shaving device and method, aiming to protect the complete technical route of "non-contact acquisition → AI recognition and positioning → laser excision" and achieve a cleaner non-contact shaving experience than electric shavers.

[0005] A non-contact AI laser shaving device includes: An image acquisition unit is used to acquire image data containing beards at a non-contact distance; The processing unit runs artificial intelligence algorithms to process image data to identify beard hairs and determine the three-dimensional spatial position of the beard hairs relative to the device. Since the human face is a three-dimensional curved surface, the root depth of beard hairs varies in different locations, and the laser beam needs to be precisely focused on the root of each beard hair at a non-contact distance to achieve effective removal. Therefore, two-dimensional image coordinates alone cannot meet the requirements for safe and clean shaving, and three-dimensional spatial coordinates must be obtained. The laser emitting unit is used to emit a laser beam at a non-contact distance to remove the beard based on its three-dimensional spatial position.

[0006] The image acquisition unit can use any type of imaging device, including but not limited to monocular cameras, binocular cameras, multi-view cameras, global shutter cameras, rolling shutter cameras, visible light cameras, infrared cameras, and event cameras.

[0007] The processing unit can employ any type of artificial intelligence algorithm, including but not limited to deep learning algorithms, instance segmentation algorithms, object detection algorithms, multi-object tracking algorithms, and traditional computer vision algorithms. The processing unit can determine its 3D spatial position using methods including but not limited to binocular stereo vision, Time-of-Flight (ToF) depth sensing, monocular depth estimation, infrared structured light, visual SLAM, millimeter-wave radar, ultrasound, and purely visual triangulation.

[0008] The laser emitting unit can use a laser source of any wavelength, including but not limited to 755nm, 810nm, and 1064nm in the 750–1100nm band, and the water absorption peak wavelength in the 1900–2000nm band; the laser type includes but is not limited to pulsed lasers, continuous lasers, and Q-switched lasers.

[0009] The output power and operating mode of the laser emitting unit should be configured according to the specific product application. For consumer home products, the laser itself can meet the safety requirements of IEC60825-1 Class 1, Class 2 or Class 3R (e.g., continuous output power not exceeding 5mW), but such low-power configurations are usually insufficient for a single pulse to effectively remove beard hairs.

[0010] To achieve a clean shave, the actual product can employ any combination of one or more of the following methods: (a) using a high peak power, short pulse (nanosecond or picosecond level) laser, whose average power can still be controlled within safe limits, but whose peak power can reach hundreds of watts or more, sufficient to carbonize the beard; (b) employing multi-scan cumulative heating technology, gradually increasing the beard temperature to the carbonization point through multiple low-energy irradiations; (c) configuring a higher average power (e.g., 50mW to several watts) with additional engineering protection (such as human eye detection, distance locking, hardware security lock), and packaging according to Class 3B or Class 4 engineering control requirements. Regardless of the power configuration used, as long as the technical route of "non-contact acquisition → AI recognition and positioning → laser removal" is achieved, it falls within the protection scope of this invention. Any specific power values ​​in this specification are merely examples and do not constitute a limitation on the scope of protection.

[0011] Beneficial effects This invention offers the broadest scope of protection, covering all devices and methods that implement the "non-contact acquisition → AI recognition and positioning → laser excision" technical route, without limiting any specific hardware, algorithm, wavelength, ranging method, or power level. The device operates non-contactly, avoiding skin scratches, redness, and allergies; it can remove the hair shaft located within the epidermis (i.e., the blackheads commonly seen after shaving), leaving no visible residue on the skin surface after shaving, achieving a cleanliness level equal to or exceeding that of traditional electric shavers; the non-contact method avoids skin irritation caused by blade friction; and its cost is flexible, adaptable to different market positioning. Detailed Implementation

[0013] Definition of non-contact distance In this invention, "non-contact distance" refers to a state where there is no physical contact between the device and the user's skin surface. Typical working distances range from 1mm to 15cm, but are not limited to this. Any distance from which beard removal can be completed without the device contacting the skin is considered a non-contact distance.

[0014] Overall structure example In one specific embodiment, the device features a handheld ergonomic design. The upper part of the device houses the blade, integrating an image acquisition unit and a laser emission window; the middle section contains control buttons and a status display screen; the internal components house the processing unit, laser emission unit, and battery. The entire device is IPX4 waterproof.

[0015] Integration method of processing units In a preferred embodiment, the processing unit employs a single neural network model (such as YOLOv8-seg) to simultaneously perform pixel-level recognition and monocular depth estimation of the beard hairs, directly outputting the three-dimensional coordinates of each hair. This neural network model is trained on a large number of facial images labeled with beard position and depth information. The processing unit can be located locally on the device, on a remote server, or through a combination of local and cloud computing. In this invention, "processing unit" refers to a functional processing module, the physical implementation of which can be any combination of one or more chips.

[0016] Implementation of safety protection and power configuration This invention does not require the inclusion of a separate safety protection module. If the laser emitting unit uses a low-power laser conforming to IEC60825-1 Class 1, Class 2, or Class 3R (e.g., continuous output power ≤ 5mW), the single pulse energy is extremely low and cannot directly carbonize the beard. In such a configuration, the device must rely on other mechanisms to achieve the shaving function, for example: Using a short-pulse (nanosecond level) laser with high peak power (hundreds of watts or more), the average power can still be controlled within safe limits, but the peak power is enough to burn off a beard; Multi-pass scanning cumulative heating is employed: the laser beam is controlled to repeatedly irradiate the same beard hair multiple times, and the temperature of the beard is gradually increased by utilizing the avalanche effect until it is carbonized; By combining exogenous light absorbers (such as carbon microparticles) with a coating on the beard surface, the absorption efficiency of low-power lasers can be enhanced.

[0017] Preferably, the device can be further equipped with one or more safety mechanisms, including eye detection, distance locking, skin temperature monitoring, hardware failure lock, and current monitoring, thereby allowing for the configuration of lasers with higher average power (e.g., 50mW to 2W) while meeting home safety requirements. With such a configuration, a single pulse can effectively remove beard hairs.

[0018] For any power configuration, the overall optical radiation safety of the device shall comply with the requirements of GB44703-2024 and GB / T7247.1-2024. High-power versions (exceeding Class 3R limits) shall limit human exposure through engineering protection measures (such as closed optical paths and forced distance locking) and shall be marked and user-warned according to the corresponding safety level.

[0019] Workflow The working process of the device is as follows: Step 1: The device is powered on, and the image acquisition unit begins to acquire facial images at a non-contact distance; Step 2: The processing unit runs an artificial intelligence algorithm to identify the beard in the image and calculate the three-dimensional spatial coordinates of each beard hair; Step 3: The processing unit sends the three-dimensional spatial coordinates to the laser emitting unit; Step 4: The laser emitting unit emits a laser beam at a non-contact distance based on the coordinates to remove the beard. Step 5: Repeat the above steps until the shaving is complete.

[0020] Regulatory compliance statement This device is a non-contact laser shaving device, and its laser safety design complies with the requirements of GB / T7247.1-2024 (equivalent to IEC60825-1:2014). Depending on the actual power configuration, the device can achieve Class 1, Class 2, Class 3R or higher (requiring engineering protection). Because this device does not have physical contact with the skin during operation, it is not entirely subject to the special clause regarding "reliance on skin contact" in GB / T4706.119-2024 "Special Requirements for Cosmetic and Beauty Care Appliances Containing Lasers or Intense Light Sources (ILS)". The overall light radiation safety of the device also meets the mandatory requirements of GB44703-2024 "General Requirements for Light Radiation Safety". For products planned for market release, relevant safety testing and certification should be completed by a third-party testing institution based on the final power rating and intended use.

[0021] Optional variants All variations thereof—including but not limited to different cameras, different artificial intelligence algorithms, different ranging methods, different laser wavelengths, different power configurations (from extremely low power multi-pass accumulation to high peak power single-pass removal), local or cloud processing, whether or not a scanning mechanism is included, and whether or not a safety protection module is included—are all within the scope of protection of this invention. The core protection point of this invention lies in the technical route of "non-contact acquisition → artificial intelligence recognition and positioning → laser removal," rather than any specific implementation details or power parameters.

[0022] Argument for the Necessity of Three-Dimensional Coordinates The reason why this invention requires obtaining the three-dimensional spatial coordinates of the beard, rather than just the two-dimensional image coordinates, is based on the following technical considerations: The human face is not a flat surface; the distance between the beard root and the device varies with the curvature of the face (for example, the distance is closer at the jawline and farther at the cheek). If only two-dimensional coordinates are used, the laser beam cannot adjust the focus individually for each beard hair, which may result in incomplete removal of some beard hairs or accidental skin damage from laser energy.

[0023] This invention operates at a non-contact distance (1mm to 15cm), within which the depth of field is relatively large. However, in order to achieve precise removal with a spot diameter ≤0.5mm, it is still necessary to perform independent distance measurement and focus calibration for each whisker. Two-dimensional coordinates cannot provide distance information.

[0024] Any beard recognition scheme that only outputs two-dimensional pixel coordinates without calculating depth values ​​cannot achieve the shaving effect of removing the hair shaft located in the epidermis as described in this invention, and therefore does not belong to the technical route of this invention.

Claims

1. A non-contact AI laser shaving device, characterized in that, include: An image acquisition unit is used to acquire image data containing beards at a non-contact distance; The processing unit runs an artificial intelligence algorithm to process the image data to identify the beard and determine the three-dimensional spatial position of the beard relative to the device. A laser emitting unit is used to emit a laser beam at a non-contact distance to remove the beard based on the three-dimensional spatial position.

2. The non-contact AI laser shaving device according to claim 1, characterized in that, The processing unit integrates the functions of beard recognition and three-dimensional spatial location determination into the same algorithm or the same chip.

3. The non-contact AI laser shaving device according to claim 2, characterized in that, The image acquisition unit can use any type of imaging device, including but not limited to monocular cameras, binocular cameras, multi-view cameras, global shutter cameras, rolling shutter cameras, visible light cameras, infrared cameras, and event cameras.

4. The non-contact AI laser shaving device according to claim 3, characterized in that, The processing unit employs any type of artificial intelligence algorithm, including but not limited to deep learning algorithms, instance segmentation algorithms, object detection algorithms, multi-object tracking algorithms, and traditional computer vision algorithms.

5. The non-contact AI laser shaving device according to claim 4, characterized in that, The processing unit determines the three-dimensional spatial position in ways including but not limited to binocular stereo vision, ToF depth sensing, monocular depth estimation, infrared structured light, visual SLAM, millimeter-wave radar, ultrasound, and pure visual triangulation.

6. The non-contact AI laser shaving device according to claim 5, characterized in that, The laser emitting unit uses a laser source of any wavelength, including but not limited to 755nm, 810nm, and 1064nm in the 750–1100nm band, and the water absorption peak wavelength in the 1900–2000nm band; the laser type includes but is not limited to pulsed lasers, continuous lasers, and Q-switched lasers.

7. The non-contact AI laser shaving device according to claim 6, characterized in that, The output power and operating mode of the laser emitting unit can be configured as any one or more of the following: (a) a pulse mode with low average power but high peak power that meets the safety level of consumer laser products (Class 3R or lower); (b) a low power mode that achieves effective removal by cumulative heating through multi-pass scanning; (c) a higher average power configuration for single-pass beard removal, provided that additional engineering protection measures are in place.

8. The non-contact AI laser shaving device according to claim 7, characterized in that, It also includes a scanning mechanism for targeting and guiding the laser beam to the three-dimensional spatial position; the scanning mechanism includes, but is not limited to, a MEMS galvanometer, a galvanometer motor, a polygonal prism, and a liquid crystal beam deflector.

9. The non-contact AI laser shaving device according to claim 8, characterized in that, It also includes safety protection mechanisms to prevent accidental laser flare from harming the human body; these safety protection mechanisms include, but are not limited to, human eye detection, distance locking, skin temperature monitoring, hardware failure safety lock, current monitoring, and safety achieved through the laser's own power level.

10. The non-contact AI laser shaving device according to claim 9, characterized in that, The device is used for daily shaving, and after shaving, there are no visible black spots left on the skin surface, achieving a level of cleanliness comparable to an electric shaver.

11. The non-contact AI laser shaving device according to claim 10, characterized in that, The processing unit may be located locally on the device, on a remote server, or employ a combination of local and cloud computing.

12. A non-contact AI laser shaving method, characterized in that, Perform the following steps at a non-contact distance: Collect image data containing beard; The image data is processed using artificial intelligence algorithms to identify the beard and calculate its three-dimensional spatial coordinates. A laser beam is emitted toward the beard according to the three-dimensional spatial coordinates to remove the beard.

13. The non-contact AI laser shaving method according to claim 12, characterized in that, The artificial intelligence algorithm detects skin tone type at the same time or after recognizing beard, and automatically adjusts laser energy according to skin tone depth.

14. The non-contact AI laser shaving method according to claim 13, characterized in that, The artificial intelligence algorithm also determines the color of the beard when identifying it, and automatically switches the laser wavelength to the water absorption peak band when the beard is light-colored.

15. The non-contact AI laser shaving method according to claim 14, characterized in that, The calculation of the three-dimensional spatial coordinates and the recognition of the beard are accomplished by the same neural network model.