Epilating device and epilating method

Abstract

An epilating apparatus that performs epilation treatment using light emitted from a light source, the epilating apparatus including: a light source section that includes a light source; a camera section that is capable of capturing an image of a treatment target region of skin; a pore determination section that determines pores present in the treatment target region based on image data of the treatment target region captured by the camera section; a deviation amount detection section that detects a deviation amount of a position of the pores that accompanies a positional deviation of the epilating apparatus from a time of the capturing; and an irradiation position correction section that corrects an irradiation position of the light on the pores based on the deviation amount detected by the deviation amount detection section.

Description

Technical Field

[0001] This invention relates to hair removal devices and methods. Background Technology

[0002] Previously, laser hair removal devices were known to remove body hair by irradiating it with lasers. However, existing commercially available hair removal devices that use lasers or flashlights are designed to target only about 1% of the skin's surface area with light. These devices, which irradiate the entire skin with strong light, are inefficient, bulky, and pose a greater risk of skin damage. Furthermore, they irradiate light regardless of hair root thickness, hair color, or skin color or density, and therefore cannot be considered the most suitable option.

[0003] In recent years, to address this issue, laser hair removal devices that irradiate the hair root specifically for body hair have been proposed (Patent Document 1, etc.). The laser hair removal device in Patent Document 1 is configured to determine the thickness (coarseness) of the hair root and the color of the hair based on an image taken of the skin to be treated, and then determine the dosage of the irradiated laser based on this determined thickness and color. According to this laser hair removal device in Patent Document 1, a laser dosage suitable for the hair to be treated can be irradiated, thus enabling effective hair removal.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Publication No. 2005-500879 Summary of the Invention

[0007] The problem that the invention aims to solve

[0008] However, the laser hair removal device described in Patent Document 1 has the following problem: there is a time lag between taking a picture of the skin to be treated and actually irradiating it with laser. If the position of the laser hair removal device deviates during this period, it will be impossible to irradiate the hair to be treated with laser.

[0009] Therefore, the object of the present invention is to provide a hair removal apparatus and a hair removal method capable of reliably irradiating the hair to be treated with a light beam.

[0010] Methods for solving problems

[0011] To achieve this objective, the hair removal device of the present invention is a hair removal device that uses light irradiated from a light source for hair removal processing. The device is characterized by comprising: a light source unit having the light source; a camera unit capable of capturing images of a target area of ​​skin; a pore determination unit that determines pores present in the target area based on image data of the target area captured by the camera unit; a deviation detection unit that detects the deviation in pore position caused by a positional deviation of the hair removal device from the time of capturing the image; and an irradiation position correction unit that corrects the irradiation position of the light on the pores based on the deviation detected by the deviation detection unit.

[0012] Alternatively, the hair removal device of the present invention may also include a motion detection unit, which is capable of detecting the relative movement of the hair removal device relative to the treatment target area. The deviation detection unit is configured such that the motion detection unit detects the relative movement of the hair removal device relative to the treatment target area from the time of shooting, thereby enabling the deviation detection unit to detect the deviation based on the relative movement.

[0013] In the hair removal apparatus of the present invention, the movement detection unit may be configured to detect the relative movement of the hair removal apparatus in the planar direction of the treatment target area and the relative rotation of the hair removal apparatus in a direction parallel to the planar direction.

[0014] In the hair removal apparatus of the present invention, the deviation detection unit may be configured to detect the deviation based on first image data and second image data, wherein the first image data is captured for determining pores using the pore determination unit, and the second image data is captured again before irradiating the pore with light.

[0015] In the hair removal apparatus of the present invention, the deviation detection unit may be configured to detect the deviation based on a cut pore image extracted from the first image data and the second image data, wherein the cut pore image is an image cut from the first image data in a manner that includes the pore determined by the pore determination unit and the skin surrounding the pore.

[0016] In the hair removal device of the present invention, the second image data may have fewer pixels than the first image data, or have a larger pixel size than the image of the cut pores.

[0017] In the hair removal device of the present invention, the irradiation position correction unit may be configured to determine whether to perform light irradiation position correction based on the deviation amount detected by the deviation amount detection unit.

[0018] In the hair removal apparatus of the present invention, the irradiation position correction unit may be configured to determine whether to notify an error based on the deviation amount detected by the deviation amount detection unit.

[0019] In the hair removal device of the present invention, the pore determination unit may also use AI image recognition to determine the pores.

[0020] Furthermore, the hair removal method of the present invention utilizes light irradiated from a light source for hair removal processing. The method is characterized by comprising the following steps: an imaging step, which captures an image of a target area of ​​skin; a pore determination step, which determines pores present within the target area based on image data of the target area captured by the imaging step; a deviation detection step, which detects a deviation related to the pore position from the time of imaging; and an irradiation position correction step, which corrects the irradiation position of the light on the pores based on the deviation detected by the deviation detection step.

[0021] Invention Effects

[0022] According to the present invention, a hair removal apparatus and a hair removal method are provided that can reliably irradiate the hair to be treated with a light beam. Attached Figure Description

[0023] Figure 1 This is a diagram that schematically illustrates the structure of a hair removal device according to one embodiment of the present invention.

[0024] Figure 2 This is a diagram that roughly shows the structure of the illumination position control mechanism.

[0025] Figure 3 This is a diagram that schematically illustrates an example of the structure of the first sensor of the motion detection sensor.

[0026] Figure 4 This is a diagram that schematically illustrates an example of the structure of the second sensor of the motion detection sensor.

[0027] Figure 5 This is a diagram that schematically illustrates an example of the structure of the third sensor of the motion detection sensor.

[0028] Figure 6 It is a diagram that roughly shows the structure of the control unit.

[0029] Figure 7 (a) is a diagram showing the original image captured by the camera unit. Figure 7 (b) is a diagram showing the state of the pores as determined by the pore determination section.

[0030] Figure 8(a) is a graph showing image data after pores have been determined. Figure 8 (b) is a diagram showing the cut images of the pores cut according to the defined pores. Figure 8 (c) is a diagram showing part or all of a re-image taken before the beam of light was applied. Figure 8 (d) is a diagram that shows the position of the original pore image and the position of the pore image in the re-enhanced image together.

[0031] Figure 9 This diagram illustrates the processing of a device for detecting the overall deviation of a camera range using the positional deviation of multiple template images for positional deviation detection.

[0032] Figure 10 This diagram illustrates the process of detecting the deviation of each pore location using the overall device deviation of the camera's field of view.

[0033] Figure 11 This is a flowchart that schematically illustrates the process of a hair removal method according to one embodiment of the present invention.

[0034] Figure 12 It is a flowchart that roughly shows the process from the pore identification step to the irradiation step.

[0035] Figure 13 This is a diagram that schematically illustrates the processing sequence of the hair removal method of this embodiment.

[0036] Figure 14 It is Figure 13 The diagram shows an enlarged view of part A. Detailed Implementation

[0037] Hereinafter, preferred embodiments for carrying out the present invention will be described using the accompanying drawings. Furthermore, the following embodiments do not limit the invention as claimed in each claim, and the combinations of features described in the embodiments are not necessarily all necessary for the solution of the invention. Additionally, the accompanying drawings are schematic diagrams with appropriate emphasis or omission and adjusted proportions for illustrating the present invention, and may differ from actual shapes, positional relationships, or proportions.

[0038] The hair removal device 1 of this embodiment is a hair removal device that permanently or permanently removes body hair (performs hair removal treatment) by irradiating the body hair present on human skin with light from a light source.

[0039] Specifically, such as Figure 1As shown, the hair removal device 1 includes a housing 10 that can be held by a user, a light source 20 housed within the housing 10, an irradiation position control mechanism (e.g., a control mechanism that controls the irradiation position of the light beam in the XY direction by arranging an electric scanner with a rotating mirror along two axes XY), a camera unit 40, and a control unit 100 (see reference 100) that controls the light source 20 and the irradiation position control mechanism 30 based on image data captured by the camera unit 40. Figure 6 Alternatively, the control unit 100 may be located inside the housing 10 or in another terminal that can be connected to the housing 10 via wired or wireless data communication.

[0040] [Structure of the outer shell]

[0041] like Figure 1 As shown, the outer casing 10 has a gripping part 11 for a user to hold and a head 12 continuously disposed on the front end side of the gripping part 11. In the hair removal device 1 of this embodiment, a light source 20 and an irradiation position control mechanism 30 are disposed in the gripping part 11, and a camera part 40 is disposed in the head 12, but it is not limited to these. In addition, the structure and shape of the outer casing 10 are not limited to the example shown and can be arbitrarily changed.

[0042] The holding part 11 is formed into any shape, such as a cylinder, with a diameter and length that can be held by a user, and is designed to be such that the skin-facing surface of the head 12 faces the skin to be treated. Therefore, within the housing 10, when the holding part 11 is held, the hair removal device 1 can be easily positioned on the skin to be treated, and the hair removal device 1 can be easily moved from the treated area to the untreated area. Furthermore, the holding part 11 is provided with an irradiation button 18 for switching the irradiation from the light source 20 on / off (see reference). Figure 6 ).

[0043] The head 12 has an opening 13 on its skin-facing surface (the lower surface in this embodiment), and a cover member 14 is provided to cover this opening 13. This skin-facing surface is opposite to the skin to be treated during hair removal. The opening 13 is larger than or equal to the area of ​​the skin to be treated by a single shot. The cover member 14 is dustproof enough to prevent dust and the like from entering the housing 10, and light-transmitting enough not to obstruct the illumination process of the light source 20 or the imaging process of the camera 40. For example, a transparent glass plate can be used as the cover member 14, but it is not limited to this.

[0044] Furthermore, a dichroic mirror 17 is provided inside the head 12. This dichroic mirror 17 further reflects the light beam from the light source 20, which has been deflected by the irradiation position control mechanism 30, toward the outside of the opening 13. The dichroic mirror 17 is inclined at an angle of approximately 45 degrees relative to the opening 13 and is configured to efficiently reflect the irradiation light, which is infrared light with a longer wavelength. This reflective surface effectively reflects the light beam from the light source 20, which has been deflected by the irradiation position control mechanism 30, toward the outside of the opening 13 (the area of ​​skin to be treated). On the other hand, the dichroic mirror 17 allows visible light with a shorter wavelength to be transmitted with a higher transmittance. An imaging unit 40 is arranged on the side of this transmission surface. Thus, the imaging unit 40 can capture images of the outside of the opening 13 (the area of ​​skin to be treated) with less loss via the dichroic mirror 17.

[0045] Furthermore, an illumination component (not shown) is provided inside the head 12 to illuminate the area of ​​the skin being treated via the opening 13. This illumination component is configured to light up when the camera unit 40 takes a picture, thereby illuminating the treated area of ​​the skin through the opening 13. Various arbitrary light sources, such as general-purpose LEDs, can be used as this illumination component.

[0046] In addition, the head 12 is provided with a motion detection sensor 15 (motion detection unit) on the skin-facing surface for detecting the relative movement amount (hereinafter referred to as "horizontal movement amount") of the hair removal device 1 (head 12) relative to the skin (treatment target area) in the XY plane direction and the relative rotation amount (hereinafter referred to as "horizontal rotation amount") of the hair removal device 1 (head 12) in a direction parallel to the XY plane direction. The motion detection sensor 15 is provided in a position not obstructed by the user's hand when the user is holding the grip 11, for example, near the opening 13. The motion detection sensor 15 can be any optical mouse sensor, accelerometer, gyroscope sensor, etc. Examples of the structure of the motion detection sensor 15 are the first sensor structure example 15A to the third sensor structure example 15C.

[0047] 【Sensor Structure Example 15A】

[0048] like Figure 3As shown, the first sensor structure example 15A combines an accelerometer 15a capable of detecting the acceleration of the head 12 with a gyroscope 15b capable of detecting the angular velocity of the head 12 to form a motion detection sensor 15. These accelerometer 15a and gyroscope 15b can be disposed at any position near the opening 13. According to this first sensor structure example 15A, the horizontal movement can be measured by integrating the acceleration detected by the accelerometer 15a twice, and the horizontal rotation can be measured by integrating the angular velocity detected by the gyroscope 15b.

[0049] [Example 15B of Sensor 2 Structure]

[0050] like Figure 4 As shown, the second sensor structure example 15B is an example of combining two or more optical mouse sensors (first optical mouse sensor 15c and second optical mouse sensor 15d) capable of detecting the horizontal movement of the head 12 into a movement detection sensor 15. The first optical mouse sensor 15c and the second optical mouse sensor 15d are preferably arranged separately along the X and / or Y directions. In the second sensor structure example 15B, the first optical mouse sensor 15c and the second optical mouse sensor 15d are arranged separately from each other with an opening 13 between them. According to this second sensor structure example 15B, the horizontal movement can be measured by each optical mouse sensor 15c, 15d. Furthermore, the horizontal rotation can be measured by the difference (small movement) between the measured value of the first optical mouse sensor 15c and the measured value of the second optical mouse sensor 15d.

[0051] 【Example 15C of the third sensor structure】

[0052] like Figure 5 As shown, the third sensor structure example 15C combines an optical mouse sensor 15e capable of detecting the horizontal movement of the head 12 with a gyroscope sensor 15f capable of detecting the angular velocity of the head 12 to form a motion detection sensor 15. These optical mouse sensors 15e and gyroscope sensors 15f can be positioned anywhere near the opening 13. According to this third sensor structure example 15C, the horizontal movement can be measured using the optical mouse sensor 15e, and the horizontal rotation can be measured by integrating the angular velocity detected by the gyroscope sensor 15b.

[0053] Furthermore, the motion detection sensor 15 is not limited to the first sensor structure example 15A to the third sensor structure example 15C described above, and various known structures can be adopted. Additionally, the motion detection sensor 15 may also be a structure capable only of detecting horizontal movement.

[0054] Furthermore, a display panel 16 is provided on the side of the head 12 facing the user during hair removal (the upper surface in this embodiment). The display panel 16 is configured to display real-time images (live feeds) captured by the camera unit 40 when the hair removal device 1 is moved from the treated area to the untreated area. Thus, by displaying live feeds on the display panel 16, the movement of the hair removal device 1 relative to the untreated area can be assisted. For example, a liquid crystal panel can be used as the display panel 16, but it is not limited to this.

[0055] [Structure of the light source section]

[0056] The light source unit 20 has a beam-shaped high-brightness light source (not shown) with an irradiation intensity (energy density) capable of sufficiently damaging the hair root to permanently or long-term remove hair (perform hair removal treatment). Such a light source can be, for example, any known light source such as a laser, semiconductor laser, semiconductor-excited solid-state laser, solid-state laser, or ultra-high brightness LED.

[0057] The light beam emitted from the light source preferably has a diameter on the irradiated surface that is necessary and sufficiently large relative to a hair root. That is, considering the image recognition accuracy or the positioning accuracy (positional deviation) of the scanner, the diameter of the light beam emitted from the light source is preferably set to be larger than the diameter of a hair root or pore.

[0058] The light source unit 20 is preferably configured to be able to operate within a specified range (e.g., 1 to 100 J / cm). 2 The irradiation intensity (power, dose) of the light source is adjusted. In particular, the light source unit 20 is preferably configured to select the optimal irradiation intensity corresponding to the size of the pores, the color of the hair, and the color of the skin surrounding the pores of each hair being treated, and to irradiate each hair with a light beam. Furthermore, various known methods, such as controlling the power output itself or controlling the pulse width, can be used as methods for controlling the irradiation intensity of the light source. Additionally, in this specification, the term "pore size" includes any of the following: the size (thickness) of the pore itself, the thickness of the hair, or the size obtained by combining the size of the pores and the hair.

[0059] Furthermore, the light source unit 20 preferably has multiple (e.g., three) light sources with different wavelengths, and has a beam combiner (not shown) for properly combining the light irradiated from these multiple light sources. In this case, the multiple light sources may also be configured to include: a first light source (not shown) capable of irradiating a relatively short wavelength (e.g., about 755 nm) light beam that is easily absorbed by melanin, which is abundant in hair; a third light source (not shown) capable of irradiating a relatively long wavelength (e.g., about 1064 nm) light beam that is less absorbed by melanin and is skin-friendly; and a second light source capable of irradiating a light beam with a wavelength (e.g., about 810 nm) between the first and third light sources. In addition, as the beam combiner, various known components such as wavelength selective mirrors (diffuse mirrors), wavelength selective prisms (diffuse prisms), polarizing beam splitters (PBS), and polarizing plates can be used.

[0060] Based on this structure, multiple wavelengths of light can be irradiated in a combined state with arbitrary intensity. Therefore, based on information such as the size of the pores of each hair being treated, the color of the hair, and the color of the skin around the pores, the optimal combination of irradiation intensity and wavelength can be selected, and the light beam can be irradiated onto each hair.

[0061] In addition, Figure 1 In the example shown, the light source 20 is disposed inside the holding part 11 of the housing 10, but it is not limited to this. As long as light can be irradiated from the opening 13 of the housing 10 via the irradiation position control mechanism 30, it can be disposed at any position inside the housing 10.

[0062] [Structure of the Irradiation Position Control Mechanism]

[0063] The irradiation position control mechanism 30 is a beam deflection member (scanning member) used to position the beam of light irradiated from the light source unit 20 at any position (X, Y) on the treatment area of ​​the skin (the XY plane that forms the treatment range). Specifically, as... Figure 1 and Figure 2 As shown, the irradiation position control mechanism 30 includes: an X-direction deflection section 34 for moving the light beam irradiated from the light source section 20 along the X direction (first direction) on the treatment target area of ​​the skin; and a Y-direction deflection section 32 for moving the light beam along the Y direction (second direction perpendicular to the first direction) on the treatment target area of ​​the skin.

[0064] like Figure 1 and Figure 2As shown, the Y-direction deflection section 32 and the X-direction deflection section 34 have reflectors 32a and 34a capable of reflecting the light beam, and drive sections 32b and 34b for changing the tilt angle of the reflectors 32a and 34a. The Y-direction deflection section 32 is configured to reflect the light beam irradiated from the light source section 20 toward the X-direction deflection section 34, and the X-direction deflection section 34 is configured to further reflect the light beam reflected by the Y-direction deflection section 32 toward the dichroic mirror 17. In addition, these Y-direction deflection sections 32 and X-direction deflection sections 34 are configured such that the rotation axis of the reflector 32a of the Y-direction deflection section 32 is perpendicular to the rotation axis of the reflector 34a of the X-direction deflection section 34. With this structure, the irradiation position control mechanism 30 is configured to be able to position the beam of light irradiated from the light source 20 at any position (X, Y) on the treatment target area of ​​the skin (the XY plane that becomes the treatment range) by controlling the tilt angle of the reflectors 32a, 34a of the X-direction deflection section 34 and the Y-direction deflection section 32 respectively.

[0065] As the X-direction deflection unit 34 and the Y-direction deflection unit 32, for example, an electric scanner (electromagnetic method), a servo motor (electromagnetic method), a MEMS mirror (electromagnetic force or electrostatic force), or other deflectors that use electromagnetic force or electrostatic force to tilt the mirror can be used. In addition, various known structures such as AO (Acousto-Optics) deflectors (acousto-optic components) can also be used.

[0066] [Structure of the camera unit]

[0067] like Figure 1 As shown, the camera unit 40 is disposed on the transmissive surface side of the dichroic mirror 17, configured to capture images of the skin treatment target area via the dichroic mirror 17 and the opening 13. The camera unit 40 is preferably a 4K camera with 4K resolution, but is not limited to this; it is acceptable as long as it has a sufficient number of pixels to capture pores within the field of view at adequate resolution. As the camera unit 40, various known imaging components such as CMOS sensors, CCD sensors, array sensors, and camera tubes can be arbitrarily used.

[0068] [Structure of the Control Department]

[0069] like Figure 6As shown, the control unit 100 includes: external interfaces 102, 104, and 106 for connecting to devices such as the camera unit 40, the motion detection sensor 15, and the irradiation button 18; a main control unit 110 for performing calculations to operate the hair removal device 1; a control mechanism drive control unit 122 for controlling the irradiation position control mechanism 30; a light source control unit 124 for controlling the light source unit 20; a display control unit 126 for controlling the display panel 16; and a storage unit 130 for storing various data and information required for the hair removal process. Additionally, the control unit 100 also includes a communication processing unit (not shown) capable of communicating with an external network.

[0070] External interface 102 is used to connect to the camera unit 40, external interface 104 is used to connect to the motion detection sensor 15, and external interface 106 is used to connect to the illumination button 18. However, the external interfaces provided on the hair removal device 1 are not limited to these interfaces and can be arbitrarily set according to the connected device. Furthermore, these external interfaces 102, 104, and 106 can use known interfaces corresponding to the connected device, therefore detailed descriptions are omitted.

[0071] The storage unit 130 is, for example, a memory composed of RAM, ROM, etc., storing programs containing instructions for the main control unit 110 to operate, learning result data for setting the learner (pore determination unit 112 and irradiation condition determination unit 114 described later) after learning is completed. Alternatively, the storage unit 130 may also be composed of RAM and ROM included in the main control unit 110.

[0072] The main control unit 110 is configured to include a CPU, RAM, ROM, etc., which serve as hardware processors. The program stored in the storage unit 130 is expanded in the RAM, and the CPU interprets and executes the program, thereby realizing the functions of the pore determination unit 112, the irradiation condition determination unit 114, the deviation detection unit 116, and the irradiation position correction unit 118, which will be described later. Furthermore, the CPU is preferably a high-performance processor (high-speed CPU) capable of performing deep learning (DL). Alternatively, the main control unit 110 may include multiple hardware processors, which may be configured as GPUs (including CPU-integrated GPUs), FPGAs, etc.

[0073] The pore determination unit 112 is configured to determine pores existing within the processing target area based on image data of the processing target area captured by the imaging unit 40. Specifically, the pore determination unit 112 is configured as follows: Figure 7As shown in (a), image data I of the processing target area TA captured by the camera unit 40 is acquired via the external interface 102. Based on the preprocessing performed on this image data I as needed, such as... Figure 7 As shown in (b), candidate pores (pore candidates P) existing in the processing target region TA are extracted from the image data I through image analysis. Preprocessing may include, for example, applying minimum filtering to a 4K image to emphasize pores, or removing unnecessary information to reduce the burden of subsequent processing, thus creating a 2K image, but is not limited to these methods. Furthermore, in the case referred to as "image data I" (where "image data" is labeled "I"), it includes not only the original image data captured by the camera unit 40, but also the processed image data obtained after preprocessing in the pore determination unit 112.

[0074] Here, the extraction of pore candidates P by the pore determination unit 112 is preferably performed by image processing (AI image recognition) driven by AI such as DL (Deep Learning). Specifically, since the original image data is a huge image such as 4K×2K pixels, keeping it as is is not suitable for DL ​​processing. Therefore, the image is divided into small regions (units) such as 256×256 pixels and inference is performed. The pore determination unit 112 can be configured as a learner (neural network) that has been trained to minimize an objective function consisting of the inferred values ​​of the XY coordinates of the pores in the small region and the inferred values ​​of the confidence that they are pores. Images of small regions (units) obtained by segmenting the image data I of the processing target region TA captured by the camera unit 40 are sequentially input into the learner. The learner obtains the confidence and coordinates of pore candidates with high confidence in the pores contained in the images of the small regions, thereby extracting pore candidates P. This method completely avoids the binarization-based image processing common in existing technologies, thus its detection accuracy is less affected by factors such as the brightness of the captured image or the orientation of the pores. It can also detect pores of various shapes and sizes with high precision. In this way, by extracting pore candidate P through AI image recognition, it is also possible to identify small pores (fine hairs, etc.) with low contrast that are difficult to measure or even detect in conventional image processing.

[0075] In this embodiment, an example of a learned learner is a convolutional neural network (e.g., ResNet-50) trained using ImageNet or similar methods, fine-tuned; however, this is not a limitation. Furthermore, the objective function described above (comprising the inferred XY coordinates of pores within a small region and the inferred confidence level of the presence of pores) is exemplified by the following objective function. In this objective function, the first term on the right relates to the pore location (XY coordinates of the pore) and is a function obtained using MSE (mean square error). The second term on the right relates to the determination of whether a pore exists within a certain region and is a function obtained using binary cross-entropy. The third term on the right is a regularization term used to prevent overlearning.

[0076] objective function

[0077]

[0078] In the objective function (the first term on the right), λ coord This is the weight of the first term on the right relative to the second term on the right; in this embodiment, it can be fixed to 1.0, for example. Furthermore, in the above objective function (the first and second terms on the right), s... 2 "A" refers to all units, and "B" refers to all the smaller regions within a unit. Furthermore,

[0079] It is an indicator function that is 1 when an object is present and 0 otherwise. ij It is the inferred X-coordinate value of the position of the object in region j of unit i.

[0080] That is the aforementioned objective. Additionally, y ij It is the inferred Y-coordinate value of the position of the object in region j of unit i.

[0081] That is the aforementioned goal.

[0082] Furthermore, in the objective function mentioned above (the second term on the right), c ij It is the inferred confidence value regarding the existence of objects in region j of unit i.

[0083] This is the aforementioned objective. Furthermore, λ noconf It is the weight of the case without objects relative to the case with objects. In this embodiment, there are more areas that are not pores, so it can be set to 0.05 for example.

[0084] Furthermore, in the above objective function (the third term on the right), λ L2 is the weight of the L2 norm, which in this implementation can be set to, for example, 0.001. Additionally, w is a parameter for all kernels.

[0085] Furthermore, for coarse black pores with sufficient contrast, the pore determination unit 112 can arbitrarily adopt methods such as binarization processing or threshold determination of image data I of the processing target area TA captured by the camera unit 40 to extract pore candidate P, instead of the structure of extracting pore candidate P by AI image recognition.

[0086] The irradiation condition determination unit 114 is configured to determine the irradiation conditions (irradiation intensity and wavelength, etc.) of the light beam from the light source unit 20 for each pore (pore candidate P) determined by the pore determination unit 112. Specifically, the irradiation condition determination unit 114 is configured as follows: Figure 8 (a) and Figure 8 As shown in (b), firstly, for each pore (pore candidate P) determined by the pore determination unit 112, an image containing the pore and the skin around the pore is cut from the image data I (pore image CI is cut), and the pore image CI is classified into any one of the multiple standard model images with the highest confidence among those with different pore size, hair color and skin color around the pore.

[0087] Here, each pore-extracting image CI is formed with a pore candidate P located approximately at the center, surrounded by skin. Similarly, the standard model image, like the pore-extracting image CI, is an image containing one or more pores and the skin surrounding those pores. Regarding the standard model images, multiple standard model images with different pore sizes, hair colors, and skin colors surrounding those pores are prepared in advance and stored in storage unit 130, etc. Each standard model image is associated with the most suitable beam irradiation conditions (irradiation intensity and wavelength, etc.) for the treatment subject with the pore size, hair color, and skin color surrounding those pores, based on considerations of hair removal efficiency and safety (burn risk). Furthermore, the irradiation intensity tends to be set to a higher value for coarser pores, lighter hair color, and lighter skin color, while the wavelength tends to be set to a shorter wavelength for coarser pores, lighter hair color, and lighter skin color.

[0088] Furthermore, the irradiation condition determination unit 114 is configured to determine the irradiation conditions (irradiation intensity and wavelength, etc.) of the beam light that is preset for the classified standard model image as the irradiation conditions (irradiation intensity and wavelength, etc.) of the beam light for the pores (pore candidates P) of the cut pore image CI.

[0089] Furthermore, the classification performed by the illumination condition determination unit 114 is preferably performed through image processing (AI image recognition) driven by AI such as DL (Deep Learning). Specifically, the illumination condition determination unit 114 can be configured to have a learner (neural network) that has been trained to minimize an objective function consisting of inference values ​​representing hair thickness (pore size), inference values ​​representing hair color, and inference values ​​representing skin color. A pore image CI is input to the learner, and information about pore candidates P contained in the pore image CI and the standard model image with the highest confidence (highest score) is obtained from the learner. The pore image CI is then classified as any one of the standard model images that is most similar among multiple standard model images. In this case, the learner may also perform the following processing: if all standard model images are significantly lower than the predetermined confidence, it is determined that there is no standard model image similar to the pore image CI among the pre-prepared standard model images (cannot be classified), and thus the pore candidate P of the pore image CI is determined not to be a pore (false judgment).

[0090] The deviation detection unit 116 is configured to detect the deviation (δx, δy) of the pore position caused by the positional deviation of the hair removal device 1 from the moment it takes a picture of the area to be treated. Specifically, the deviation detection unit 116 is configured to detect the device deviation (ΔX, ΔY, Δθ) of the head 12 during the time delay from when the object area is photographed by the camera unit 40 to when the pore position and irradiation conditions for the pore are determined by the pore determination unit 112 and the irradiation condition determination unit 114, and calculate the deviation (δx, δy) of each pore position based on this value, or directly detect the deviation (δx, δy) of each pore position. Examples of such deviation detection unit 116 structures include those using a motion detection sensor 15 to detect the deviation and those using a camera unit 40 to detect the deviation.

[0091] [Structure that uses a motion detection sensor to detect deviation]

[0092] First, the structure of the deviation detection unit 116 when the deviation is detected using the motion detection sensor 15 will be described. In this case, the deviation detection unit 116 first uses the motion detection sensor 15 (first sensor structure example 15A to third sensor structure example 15C, etc.) to detect the horizontal movement and horizontal rotation of the head 12 during the period from when the camera unit 40 captures the entire treatment target area to when the pore determination unit 112 and the irradiation condition determination unit 114 determine the position of the pore and the irradiation conditions for that pore. Next, based on the horizontal movement and horizontal rotation, the device deviation (ΔX, ΔY, Δθ) within the entire field of view of the camera unit 40 of the hair removal device 1 from the time of capture is calculated, and the deviation (δx, δy) for each pore position is calculated using this value. Furthermore, the position is corrected for each pore position as described later using the deviation determined in this way.

[0093] Based on this structure, the following advantages are available: the movement of the head 12 can be directly detected in real time by the motion detection sensor 15 to determine the amount of deviation, thus enabling real-time correction processing with simple equipment.

[0094] [Structure that uses a camera to detect deviation]

[0095] Next, the structure of the deviation detection unit 116 when the camera unit 40 is used to detect the deviation will be described. As examples of the structure of the deviation detection unit 116 when the camera unit 40 is used to detect the deviation, the following first camera detection structure example to the third camera detection structure example are shown.

[0096] [Example of a camera-detected structure]

[0097] In general, the deviation detection unit 116 of the first camera detection structure example is an example of directly detecting the deviation amount (δx, δy) of each pore position accompanying the position deviation of the hair removal device 1 based on the image data (first image data) captured by the pore determination unit 112 to determine the pores and the image data captured again before the beam light irradiation (second image data).

[0098] Specifically, firstly, the deviation detection unit 116 of the first camera detection structure example registers the cut pore image CI generated by the irradiation condition determination unit 114 (or pore determination unit 112) as a template image in the storage unit 130. Furthermore, as described above, this cut pore image CI is an image cut from image data I (first image data) captured for determining pores using the pore determination unit 112, in a manner that includes the pores determined by the pore determination unit 112 and the skin surrounding those pores.

[0099] Furthermore, in the first example of the camera detection structure, the deviation detection unit 116, when the irradiation position control mechanism 30 moves to the position of the pore determined by the pore determination unit 112 (the predetermined irradiation position) or before or after, uses the camera unit 40 to re-capture a predetermined range including the predetermined irradiation position. In this case, the re-capture range (pixel size of the second image data) is not particularly limited as long as it is larger than the cropped pore image CI. However, from the viewpoint of high-speed processing, compared to the image data I (first image data) of the entire field of view captured by the pore determination unit 112 to determine the pores, image recognition is not required. Therefore, a sufficiently coarse resolution (fewer pixels in the same capture range) necessary for detecting positional deviations can be used. More preferably, it is a narrower range including the predetermined irradiation position (the pore being irradiated) rather than the entire field of view. For example, when the size of the cropped pore image CI (template image) is 64×64 pixels, the re-capture range of the camera unit 40 can be set to 128×128 pixels.

[0100] Moreover, such as Figure 8 As shown in (c), the deviation detection unit 116 of the first camera detection structure example detects the position of the cut pore image CI' in the re-enhanced image using various matching methods such as grayscale search, based on the cut pore image CI and the re-enhanced image I'. Additionally, as... Figure 8 As shown in (d), the deviation detection unit 116 of the first camera detection structure example is configured to calculate the difference between the position of the original cut pore image CI and the position of the cut pore image CI′ in the re-captured image, and determine the difference as the deviation amount (δx, δy) of the pore position accompanying the position deviation of the hair removal device 1 from the time of shooting.

[0101] The deviation detection unit 116 of this first camera detection structure example has the following advantages: it can directly photograph the pores that are the object being irradiated to determine the deviation amount, and thus can also correct local stretching or deformation of the skin, thereby achieving a high-precision correction process.

[0102] [Example of a second camera-detected structure]

[0103] In general, the deviation detection unit 116 of the second camera detection structure example calculates the device deviation (ΔX, ΔY, Δθ) within the entire field of view of the camera unit 40 of the hair removal device 1 from the time of shooting based on the image data (first image data) captured by the pore determination unit 112 to determine the pores and two or more image data captured again before the beam light irradiation (second image data), and uses this value to detect the deviation (δx, δy) of each pore position.

[0104] Specifically, firstly, the deviation detection unit 116 of the second camera detection structure example registers image ranges containing two or more pore images generated by the irradiation condition determination unit 114 (or pore determination unit 112) that are separated from each other as position deviation detection template images T1 and T2 in the storage unit 130. The position deviation detection template images T1 and T2 with two or more locations can be images related to different positions within the processing target area, but from the viewpoint of high-precision detection of the horizontal movement and horizontal rotation of the head 12, images related to positions separated in both the X and Y directions are preferred. Furthermore, in cases of large position deviations, to prevent misidentification as other pore locations, the position deviation detection templates T1 and T2 are preferably not a narrow range containing only one pore as in the first camera detection structure example, but rather a larger range containing multiple pores or a larger blank area around the pores.

[0105] Next, before irradiating each pore with a beam of light, the deviation detection unit 116 of the second camera detection structure example uses the camera unit 40 to re-image the specified range of the templates T1 and T2 used for position deviation detection registered in the storage unit 130. Furthermore, the re-image range (pixel size of the second image data) in this case is similar to that of the first camera detection structure example; it is not particularly limited as long as it is larger than the template images T1 and T2 used for position deviation detection.

[0106] Next, as Figure 9 As shown, the deviation detection unit 116 of the second camera detection structure example uses various matching methods such as grayscale search to detect the position deviation (δx1, δy1 and δx2, δy2) of the template images T1 and T2 used for position deviation detection.

[0107] Furthermore, the deviation detection unit 116 of the second camera detection structure example is configured to: use the position deviations (δx1, δy1 and δx2, δy2) of the template images T1 and T2 used for position deviation detection, calculate the device deviation (ΔX, ΔY, Δθ) within the entire field of view of the camera unit 40 of the hair removal device 1 from the time of shooting, and as follows... Figure 10 As shown, this value is used to determine the deviation (δx, δy) for each pore location.

[0108] [Example of a third camera-detected structure]

[0109] The deviation detection unit 116 of the third camera detection structure example uses the same principle as the second sensor structure example 15B described above. It uses the camera unit 40 to extract two or more separate measurement ranges within the field of view. Similar to the optical mouse sensors (first optical mouse sensor 15c and second optical mouse sensor 15d) described above, it measures not only pores but also the movement of tiny skin textures. Thus, it achieves an example of the same operation as the second sensor structure example 15B described above.

[0110] According to this third camera detection structure example, similarly to the second sensor structure example 15B described above, it is possible to detect the horizontal movement amount (positional deviation) and the horizontal rotation amount (rotational deviation). Furthermore, based on this horizontal movement amount and horizontal rotation amount, it is possible to calculate the device deviation amount (ΔX, ΔY, Δθ) within the entire field of view of the camera unit 40 of the hair removal device 1 from the time of shooting, and use this value to determine the deviation amount (δx, δy) for each pore position. Additionally, in this third camera detection structure example, to improve the contrast of skin texture, it is also possible to use illumination that is different from normal lighting conditions, such as side illumination that is more enhanced, or illumination with a different wavelength.

[0111] In the case where the camera unit 40 is used to detect the deviation amount as in the deviation amount detection unit 116 of the first to third camera detection structure examples described above, any deviation in each pore within the entire field of view can be detected by a single sensing member (camera) as a deviation amount summarized as δx and δy. As long as the deviation amount (δx, δy) of each pore position is reflected in the control signal to the illumination position control mechanism 30, it is not necessary to install a separate camera unit. Figures 3-5 The types of sensors described can reduce the number of components.

[0112] In particular, when a CMOS sensor is used in the camera unit 40, high-speed shooting at a high frame rate (e.g., more than 1,000 frames per second) can be performed by limiting the shooting range. By utilizing this, the same level of real-time performance (responsiveness) as the motion detection sensor 15 can be achieved while reducing the number of components.

[0113] Furthermore, the deviation detection unit 116 can be a structure that detects the deviation amount for each pore, or it can be a structure that detects the deviation amount for each predetermined period (each of multiple hairs), and performs pore position correction after interpolation by using the deviation amount of the most recent pore that is nearby and has passed less time as a reference. That is, the deviation detection unit 116 can arbitrarily set the detection frequency of the deviation amount.

[0114] Furthermore, simply put, the positional deviation can be detected not every time during the irradiation of the beam, but after the irradiation of all pores in the field of view is completed or after the pre-set number of irradiations is completed, the image is taken again to calculate the deviation of one or several pores in the field of view. Only when there is a deviation exceeding the specified limit, the pores are identified again and the laser irradiation is repeated, or as described later, after an error is displayed, the operator is urged to perform the operation again.

[0115] The irradiation position correction unit 118 is configured to determine the position of the actual irradiated beam based on the position of the pores (predetermined irradiation position) during the shooting and the subsequent deviation amount. Specifically, the irradiation position correction unit 118 is configured to use the deviation amount (δx, δy) of each pore detected by the deviation amount detection unit 116 to correct the position of the pores (i.e., the predetermined irradiation position) (X, Y) determined by the pore determination unit 112, thereby determining the corrected coordinate position (X′, Y′) of each pore.

[0116] Alternatively, the irradiation position correction unit 118 may be configured to determine whether to perform correction of the predetermined irradiation position and / or whether to notify an error based on the deviation amount (δx, δy) detected by the deviation amount detection unit 116.

[0117] Specifically, the irradiation position correction unit 118 may also be configured to determine whether the deviation amount (δx, δy) detected by the deviation amount detection unit 116 is smaller than the beam diameter, and if it is determined to be smaller than the beam diameter (i.e., even if the beam light is irradiated at the predetermined irradiation position, the pores will still be irradiated by the beam light), then no correction is performed on the predetermined irradiation position. Furthermore, regarding the condition of "smaller than the beam diameter," for example, the case where the composite vector of the deviation amounts (δx, δy) is smaller than the radius of the beam diameter is given, but the determination criterion is not limited to this.

[0118] Alternatively, the irradiation position correction unit 118 may be configured to determine whether the deviation amount (δx, δy) detected by the deviation amount detection unit 116 exceeds a predetermined correctable range. If it is determined that the deviation amount exceeds the range, the unit will not correct the predetermined irradiation position, but will instead perform a warning process such as notifying the error through an alarm sound or display, thereby urging the unit to irradiate again.

[0119] Furthermore, the irradiation position correction unit 118 may be configured to correct the irradiation position (X, Y) based on the deviation amount (δx, δy) when it is determined that the deviation amount is larger than the beam diameter (i.e., when the beam light is irradiated at the predetermined irradiation position, the pores are not irradiated by the beam light or the irradiation is insufficient) and the deviation amount (δx, δy) is determined to be within a predetermined correctable range.

[0120] The control mechanism drive control unit 122 is configured to control the irradiation position control mechanism 30 to irradiate the light beam from the light source unit 20 one by one into the pores that have been determined by the pore determination unit 112 of the main control unit 110 and whose positions have been corrected by the irradiation position correction unit 118 as needed.

[0121] Specifically, the control mechanism drive control unit 122 is configured such that, when the irradiation predetermined position is not corrected using the irradiation position correction unit 118, it sequentially controls the tilt angles of the reflector 32a of the Y-direction deflection unit 32 and the reflector 34a of the X-direction deflection unit 34 to sequentially irradiate the irradiation predetermined position (X, Y) of each pore determined by the pore determination unit 112. On the other hand, the control mechanism drive control unit 122 is configured such that, when the irradiation predetermined position is corrected using the irradiation position correction unit 118, it sequentially controls the tilt angles of the reflector 32a of the Y-direction deflection unit 32 and the reflector 34a of the X-direction deflection unit 34 to sequentially irradiate the irradiation beam at the corrected coordinate position (X′, Y′) of each pore determined by the irradiation position correction unit 118.

[0122] Such a control mechanism drives the control unit 122, which can be controlled, for example, by digital PID control performed by a dedicated, inexpensive embedded microcomputer, but is not limited to this.

[0123] Thus, in the hair removal device 1 of this embodiment, the position (X, Y) of each pore is determined with high precision by AI image recognition in the pore determination unit 112, and the irradiation position is corrected based on the deviation (δx, δy) of the pore position caused by the position deviation of the head 12 generated from the capture of the camera unit 40 to the actual irradiation of the beam light. The irradiation position control mechanism 30 is controlled to irradiate the beam light precisely toward each pore. Therefore, the beam light can be irradiated only at the position very close to the pore, thereby improving efficiency and safety.

[0124] Specifically, the aforementioned positional correction ensures high irradiation accuracy, eliminating the need to consider positional deviations that could lead to an excessively large beam diameter. Therefore, while maintaining the same irradiation power density, a lower-power laser can be flexibly used as the light source, enabling miniaturization or cost reduction of the device. Furthermore, using the same light source, compared to devices with a coarser beam, the irradiation time per pore can be significantly shortened, thus enabling high-speed hair removal. For example, by reducing the beam diameter from 1mm to 0.5mm, the required light source power can be reduced to approximately one-quarter. Additionally, with the same light source power, the irradiation time can be significantly reduced to approximately one-quarter, further contributing to higher speeds.

[0125] The light source control unit 124 is configured to control the light source unit 20 according to each pore, so as to irradiate the light source unit 20 with irradiation conditions determined by the irradiation condition determination unit 114 of the main control unit 110. Specifically, the light source control unit 124 is configured to perform selection control of the light source (first light source to third light source) and output control of the light source according to each pore, so as to produce a light beam with the determined irradiation conditions (irradiation intensity and wavelength, etc.). In addition, the light source control unit 124 may also be able to perform control of an illumination member (not shown) that can irradiate illumination light toward the opening 13.

[0126] The display control unit 126 is configured to perform processing that transmits and displays real-time images (live images) captured by the camera unit 40 onto the display panel 16. Various known control methods can be employed in such a display control unit 126, therefore detailed descriptions are omitted.

[0127] 【Hair removal method】

[0128] Next, use Figures 11-14 The hair removal method using the hair removal device 1 of this embodiment will be described. Figure 11 This is a flowchart that schematically illustrates the overall process of the hair removal method of this embodiment. Figure 12 This is a flowchart that roughly illustrates the process of treating one pore (from the pore determination step to the irradiation step) for a single pore determined by the pore determination unit 112. Additionally, Figure 13 This is a diagram that schematically illustrates the overall processing sequence of the hair removal method of this embodiment. Figure 14 It is Figure 13 The diagram shows an enlarged view of part A. Furthermore, the hair removal method described below is executed using programs and learning result data stored in the storage unit 130 of the hair removal device 1.

[0129] In general, the hair removal method of this embodiment is a hair removal method that uses light irradiated from a light source for hair removal treatment, and includes the following steps: an imaging step (S4) to photograph the skin target area; a pore determination step (S5-1 to S5-n) to determine the pores existing in the target area based on the image data of the target area photographed by the imaging step; a deviation detection step (S7) to detect the deviation (δx, δy) related to the pore position since the time of photographing; and an irradiation position correction step (S10) to correct the irradiation position (X, Y) of the light on the pore based on the deviation (δx, δy) detected by the deviation detection step.

[0130] The following is a detailed explanation of the hair removal method that includes these steps.

[0131] When starting the hair removal method of this embodiment, firstly, the main power supply of the hair removal device 1 is turned on to start the device. When the hair removal device 1 is started, the real-time image (live image) captured by the camera unit 40 is displayed on the display panel 16. Thus, even when the opening 13 is pressed against the skin (in the process of triggering movement by a person), the area to be treated can be visually confirmed through the live image on the display panel 16. In addition, the hair removal device 1 can be operated by the person being treated or by someone other than the person being treated (such as a medical professional). Hereinafter, the person operating the hair removal device 1 will be referred to as the "user".

[0132] With the hair removal device 1 activated, such as Figure 11 and Figure 13 As shown, the user positions the hair removal device 1 so that the opening 13 of the outer casing 10 is located in the treatment target area (S1). After positioning, the user turns on the irradiation button 18 (S2). When the irradiation button 18 is turned on, the display panel 16 is turned off (S3), and the camera unit 40 takes a picture of the treatment target area of ​​the skin (S4: imaging process). Then, the image data captured by the camera unit 40 is sent to the main control unit 110 of the control unit 100, and the image data is preprocessed as needed using the function of the pore determination unit 112 in the main control unit 110. Then, the pores (pore candidates P) existing in the treatment target area are determined sequentially (S5-1 to S5-n: pore determination process).

[0133] In addition, in parallel with the determination of pores, the determined pores are sequentially subjected to determination of irradiation conditions (irradiation intensity and wavelength, etc.), correction processing of the irradiation position of the light beam, and irradiation processing. That is, when the main control unit 110 determines the first pore candidate P through the function of the pore determination unit 112 described above, as follows: Figure 12 and Figure 14 As shown, the irradiation condition determination process for the first pore candidate P is performed independently (in parallel) from the determination process for the second pore candidate P. Furthermore, when determining the second pore candidate P, the main control unit 110 performs the irradiation condition determination process for the second pore candidate P independently (in parallel) from the determination process for the first pore candidate P and the determination process for the third pore candidate P. The main control unit 110 performs this parallel processing until the last (nth) pore candidate P. In this way, by performing the sequence of pore identification and beam irradiation in parallel, the identification processing time can be ensured without making one cycle too long.

[0134] The determination of irradiation conditions (irradiation intensity and wavelength, etc.) for each pore candidate P (S6: irradiation condition determination process) is performed by the function of the irradiation condition determination unit 114 in the main control unit 110. Furthermore, in this irradiation condition determination process, if it is determined that no matching standard model image exists, it can also be determined that the pore candidate P is not a pore, and the processing for the pore candidate P ends without proceeding to the next process (irradiation of the pore candidate P by the light beam is not performed).

[0135] Furthermore, after or in parallel with the step of determining the irradiation conditions, the deviation amount detection unit 116 in the main control unit 110 is used to detect the deviation amount (δx, δy) related to the pore position since the time of shooting (S7: deviation amount detection step). In addition, the deviation amount detection step can be performed separately for each pore candidate P, or it can be performed according to each predetermined period (each of the multiple pores).

[0136] When a deviation (δx, δy) is detected by the deviation detection process, a determination is made as to whether the deviation exceeds a predetermined correctable range (S8). Then, if it is determined that the deviation exceeds the predetermined correctable range, a warning process is performed, such as notifying the error through an alarm sound or display, and urging re-irradiation (S9′).

[0137] On the other hand, if it is determined that the deviation amount (δx, δy) detected by the deviation amount detection process is within a predetermined correctable range, a determination is made as to whether the deviation amount is to the extent that correction is required (S9). In this determination, for example, it may be determined whether the deviation amount is smaller than the beam diameter (i.e., whether the pores will be irradiated with the beam light even if the beam light is irradiated at the predetermined irradiation position).

[0138] Furthermore, if the determination process determines that correction is required, the irradiation position correction unit 118 in the main control unit 110 corrects the irradiation position (X, Y) of the pores based on the deviation (δx, δy) detected by the deviation detection process (S10: Irradiation position correction process). After the irradiation position correction process, the process proceeds to the irradiation process (S11). Conversely, if the determination process determines that correction is not required, the process proceeds to the irradiation process (S11) without going through the irradiation position correction process (S10).

[0139] When the process moves to the irradiation step (S11), the control unit 122 drives the irradiation position control mechanism 30 to control the irradiation position, so that a beam of light from the light source unit 20 is irradiated onto the coordinate position (X, Y) of the pore candidate P determined by the pore determination step or the corrected coordinate position (X′, Y′) of the pore candidate P after correction by the irradiation position correction step. Then, after this irradiation position control, a beam of light with irradiation conditions (irradiation intensity and wavelength, etc.) determined by the irradiation condition determination step is irradiated from the light source unit 20 onto the pore candidate P (S11: irradiation step). As a result, the hair root of the pore candidate P is heated, thereby being permanently or permanently removed.

[0140] Furthermore, the time required to control the irradiation position varies depending on factors such as the movement distance, but is generally around a few milliseconds. Additionally, the irradiation time of the light beam varies depending on factors such as the irradiation intensity, but is generally around a few milliseconds to tens of milliseconds. At this point, the irradiation conditions (irradiation intensity and wavelength, etc.) of the light beam are the most appropriate irradiation conditions (irradiation intensity and wavelength, etc.) assigned to the closest standard model image. Therefore, it is effective for this pore candidate P, and causes less damage to the surrounding skin, making it safe.

[0141] Furthermore, the above series of processes from pore identification to irradiation are performed on the last (nth) pore candidate P. When the processing for all pore candidates P is completed, as follows: Figure 11 and Figure 13 As shown, the illumination button 18 is turned off (S12), and the real-time image (live image) captured by the camera unit 40 is displayed on the display panel 16 again (S13).

[0142] The process described above, from the positioning (movement) of the hair removal device 1 to the completion of irradiation on all pores within the target area, is considered as one cycle. This cycle is performed sequentially and alternately throughout the entire desired target area, thereby performing hair removal. Furthermore, regarding the reference time (processing time of the hair removal device 1) from the activation of the irradiation button 18 to the completion of irradiation on the pores within the target area within one cycle, assuming there are 30 or fewer pores and an irradiation / movement time of 20 ms, this reference time is less than 1 second; assuming there are 100 or fewer pores and an irradiation / movement time of 20 ms, this reference time is less than 3 seconds. Thus, the hair removal device 1 of this embodiment can perform hair removal in a very short time.

[0143] [Advantages of the hair removal device in this embodiment]

[0144] As described above, the hair removal device 1 of this embodiment includes: a light source unit 20, which has a light source; an imaging unit 40, which is capable of taking pictures of the treatment target area of ​​the skin; a pore determination unit 112, which determines the pores existing in the treatment target area based on the image data of the treatment target area taken by the imaging unit 40; a deviation detection unit 116, which detects the deviation amount (δx, δy) of the pore position accompanying the positional deviation of the hair removal device 1 from the time of taking the picture; and an irradiation position correction unit 118, which corrects the irradiation position (X, Y) of the pores by the light based on the deviation amount (δx, δy) of each pore detected by the deviation detection unit 116.

[0145] According to the hair removal device 1 of this embodiment, configured in this way, the irradiation position (X, Y) can be corrected based on the deviation (δx, δy) of the pore position caused by the positional deviation of the hair removal device 1 from the time the camera unit 40 captures the image to the time the actual irradiation of the beam light. Therefore, even if the relative position of the hair removal device 1 (head 12) and the treatment area (skin) deviates after the camera unit 40 captures the image, the beam light can reliably and accurately irradiate each pore. In addition, this allows the beam light to be irradiated only into the pores, thus improving efficiency and safety.

[0146] [Variation Example]

[0147] The preferred embodiments of the present invention have been described above, but the technical scope of the present invention is not limited to the scope described in the above embodiments. Various modifications or improvements can be made to the above embodiments.

[0148] For example, in the above embodiment, the structure of the irradiation condition determination unit 114 is described in which the irradiation conditions corresponding to the size of the pores, the color of the hair, the color of the surrounding skin, etc. are determined by classifying the cut pore image CI into a standard model image. However, it is not limited to this and may also be a structure that does not change the irradiation conditions according to each pore.

[0149] Furthermore, in the above embodiment, the structure of temporarily identifying only pores using the pore determination unit 112 and then extracting the pore image CI from the image data I using the illumination condition determination unit 114 has been described, but it is not limited to this. The pore determination unit 112 can also apply inference values ​​such as the size of the pores, the color of the hair, and the color of the surrounding skin to the target function, thereby inferring the classification or position of images similar to the standard model image based on the images of small regions (units) obtained by segmenting the image data I, thereby directly obtaining the position of the pores, the size of the pores, the color of the hair, and the color of the surrounding skin.

[0150] Furthermore, in the above embodiments, a method for classifying cut pore images CI using AI image recognition has been described, but it is not limited to this. The following methods may also be used: quantify the feature quantities of the pore size, hair color, and skin color around the pore of the pore candidate P contained in the cut pore image CI, and compare them with the feature quantities of the standard model pre-registered in the database, thereby classifying it as the most approximate standard model.

[0151] Furthermore, in the above embodiment, the structure of the hair removal device 1 having a learned learner (neural network) for AI image recognition related to pore determination and pore image classification has been described, but it is not limited to this. A structure could also be adopted in which a learned learner is installed in another device connected to the hair removal device 1 via a high-speed communication network, and real-time communication (cloud computing) occurs between each hair removal device 1 and the other device. Additionally, in the hair removal device 1 or the other device, a structure could be adopted in which image data I of the processing target area TA is used as input data according to a predetermined learning procedure, and the certainty or coordinates of the pores are used as reference data to perform machine learning (AI learning process). Alternatively, a structure could be adopted in which images captured by multiple hair removal devices 1 are uploaded to the cloud, increasing the acquisition of images at an accelerated rate, thereby sharing learning data in real time to further improve recognition accuracy.

[0152] Furthermore, in the above embodiment, a structure was described in which a dichroic mirror 17 is provided inside the head 12, a light source 20 is arranged on the reflective side of the dichroic mirror 17, and an imaging unit 40 is arranged on the transmissive side, but this is not a limitation. For example, a structure in which the imaging unit 40 is arranged on the reflective side of the dichroic mirror 17 and the light source 20 is arranged on the transmissive side may also be used. Alternatively, a structure in which the dichroic mirror 17 is not provided may also be used. In addition, as a structure in which the dichroic mirror 17 is not provided, for example, a structure in which the imaging unit 40 is arranged perpendicularly to the opening 13 (the area of ​​skin to be treated), and the beam light from the light source 20, which is deflected by the irradiation position control mechanism 30, irradiates the opening 13 (the area of ​​skin to be treated) from an oblique direction; a structure in which the imaging unit 40 takes pictures of the opening 13 (the area of ​​skin to be treated) from an oblique direction, and the beam light from the light source 20, which is deflected by the irradiation position control mechanism 30, irradiates the opening 13 (the area of ​​skin to be treated) from an oblique direction, and so on, but this is not a limitation.

[0153] As can be seen from the claims, the aforementioned variations are included within the scope of this invention.

[0154] Label Explanation

[0155] 1: Hair removal device; 10: Outer shell; 11: Holding part; 12: Head; 13: Opening; 14: Cover part; 15: Motion detection sensor; 15A: Example of first sensor structure; 15B: Example of second sensor structure; 15C: Example of third sensor structure; 15a: Accelerometer sensor; 15b: Gyroscope sensor; 15c: First optical mouse sensor; 15d: Second optical mouse sensor; 15e: Optical mouse sensor; 15f: Gyroscope sensor; 16: Display panel; 17: Dichroic mirror; 18: Illumination button; 20: Light source part; 30: Irradiation position control mechanism; 32: Y-direction deflection unit; 32a: reflector; 32b: drive unit; 34: X-direction deflection unit; 34a: reflector; 34b: drive unit; 40: camera unit; 100: control unit; 102: external interface; 104: external interface; 106: external interface; 110: main control unit; 112: pore determination unit; 114: irradiation condition determination unit; 116: deviation detection unit; 118: irradiation position correction unit; 122: control mechanism drive control unit; 124: light source control unit; 126: display control unit; 130: storage unit.

Claims

1. A hair removal device that uses light irradiated from a light source to perform hair removal, characterized in that, This hair removal device has: A light source unit having the light source; The camera unit is capable of photographing the area of ​​skin to be treated; The pore determination unit determines the pores existing in the processing object area based on image data of the processing object area captured by the camera unit, thereby determining the predetermined position for light irradiation; The deviation detection unit detects the amount of deviation in the position of the hair removal device that occurs during the period from the time of the shooting by the camera unit to the time before the light is irradiated onto the pores determined by the pore determination unit. as well as The irradiation position correction unit corrects the predetermined irradiation position of the light on the pore based on the deviation detected by the deviation detection unit.

2. The hair removal device according to claim 1, characterized in that, The hair removal device also includes a motion detection unit capable of detecting the relative movement of the hair removal device relative to the area to be treated. The deviation detection unit is configured such that the movement detection unit detects the relative movement of the hair removal device relative to the area to be treated from the time of shooting, thereby enabling the deviation detection unit to detect the deviation based on the relative movement.

3. The hair removal device according to claim 2, characterized in that, The movement detection unit is configured to detect the relative movement of the hair removal device in the planar direction of the treatment target area and the relative rotation of the hair removal device in a direction parallel to the planar direction.

4. The hair removal device according to any one of claims 1 to 3, characterized in that, The deviation detection unit is configured to detect the deviation based on first image data and second image data, wherein the first image data is captured for determining pores using the pore determination unit, and the second image data is captured again before irradiating the pore with light.

5. The hair removal device according to claim 4, characterized in that, The deviation detection unit is configured to detect the deviation based on the cropped pore image extracted from the first image data and the second image data. The pore image is an image extracted from the first image data in a manner that includes the pore determined by the pore determination unit and the skin surrounding the pore.

6. The hair removal device according to claim 5, characterized in that, The second image data has fewer pixels than the first image data, or has a larger pixel size than the image of the cut pores.

7. The hair removal device according to any one of claims 1 to 3, characterized in that, The illumination position correction unit is configured to determine whether to correct the illumination position of the light based on the deviation detected by the deviation detection unit.

8. The hair removal device according to any one of claims 1 to 3, characterized in that, The irradiation position correction unit is configured to determine whether to notify an error based on the deviation amount detected by the deviation amount detection unit.

9. The hair removal device according to any one of claims 1 to 3, characterized in that, The pore determination unit uses AI image recognition to determine the pores.

10. A hair removal method, comprising using light irradiated from a light source for hair removal treatment, characterized in that, This hair removal method includes the following steps: The photography process involves photographing the area of ​​skin to be treated. The pore determination process involves determining the pores present in the processing object area based on image data captured by the imaging process, thereby determining the predetermined position for light irradiation. The deviation detection process detects the deviation related to the pore position that occurs during the period from the shooting in the imaging process to the illumination of light on the pores determined by the pore determination process. as well as The irradiation position correction process corrects the predetermined irradiation position of the light on the pores based on the deviation detected by the deviation detection process.

Images